&EPA
         United States
         Environmental Protection
         Agency
Office of Air Quality
Planning and Standards
Research Triangle Park, NC 27711
EPA-453/R-94-082b
November 1994
         Air
          Chromium Emissions from Chromium
          Electroplating and Chromic acid
          Anodizing Operations -- Background
          Information for Promulgated Standards
              NESHAP

-------

-------
                           November 1994
Chromium Emissions from Chromium
Electroplating and Chromic Acid
Anodizing Operations-Background
Information for Promulgated
Standards

-------
This report has been reviewed by the Emission Standards Division
of the Office of Air Quality Planning and Standards, EPA, and
approved for publication.  Mention of trade names or commercial
products is not intended to constitute endorsement or
recommendation for use.  Copies of this report are available
through the Library Services Office  (MD-35), U. S. Environmental
Protection Agency, Research Triangle Park, North Carolina  27711;
from the Office of Air Quality Planning and Standards Technology
Transfer Network, U. S. Environmental Protection Agency, Research
Triangle Park, North Carolina  27711; or, for a fee, from the
National Technical Information Services, 5285 Port Royal Road,
Springfield, Virginia  22161.
                Publication No. EPA-453/R-94-082b
                               ii

-------
                    ENVIRONMENTAL PROTECTION AGENCY

           Chromium Emissions from Chromium Electroplating
           and Chromic Acid Anodizing Operations--Background
                 Information for Promulgated Standards
                             Prepared by:
/"-^Bruce JordaKX
   Director,^Mission Standards Division
   U. S. Environmental Protection Agency
   Research Triangle Park, North Carolina
                        (Date)
      27711
   1.-   The subject national emission standards limit emissions of
        chromium compounds from existing and new chromium
        electroplating and chromium anodizing tanks.  The standards
        implement section 112 of the Clean Air Act as amended in
        1990 and are based on the Administrator's determination of
        July 16, 1992 (57 FR 31576) that chromium electroplating and
        chromium anodizing tanks generate a large amount of chromium
        compounds, a hazardous air pollutant listed in
        section 112(b) of 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; the National Science Foundation; the Council on
        Environmental Quality; members of the State and Territorial
        Air Pollution Program Administrators; the Association of
        Local Air Pollution Control Officials; EPA Regional
        Administrators; Office of Management and Budget; and other
        interested parties.

   3.   For additional information contact:

        Mr. Laiit Banker  (MD-13)
        Organic Chemicals Group
        U. S. Environmental Protection Agency
        Research Triangle Park, NC  27711
        Telephone:   (919) 541-5420

   4.   Copies of this document may be obtained from:
        U. S. EPA Library  (MD-35)
        Research Triangle  Park, NC
27711
        National Technical  Information  Service
        5285  Port Royal Road
        Springfield, Virginia   22161
                                   iii

-------
IV

-------
                        TABLE OF CONTENTS
1.0 SUMMARY   .  .  .	
    1.1 SUMMARY OF  CHANGES SINCE PROPOSAL   .  .  .
    1.2 SUMMARY OF  IMPACTS OF PROMULGATED ACTIONS
2.0 SUMMARY OF PUBLIC COMMENTS
    2.1 NESHAP DECISION PROCESS   	  ......
    2.2 PROCESS DESCRIPTION AND DESCRIPTION OF
        CONTROL TECHNOLOGIES  ..... 	
    2.3 SELECTION OF POLLUTANTS AND SOURCE CATEGORIES
        TO BE REGULATED		
        2.3.1   Selection of  Pollutants to be  Regulated
        2.3.2   Selection of  Source Categories to be
                Regulated	 .  .
    2 .4 SELECTION OF MACT/GACT APPROACH   	
    2.5 SELECTION OF MACT FOR HARD CHROMIUM
        ELECTROPLATING TANKS  	  	
        2.5.1   Selection of  the MACT Floor  	
        2.5.2   Regulatory Alternatives Considered  .  .
        2.5.3   Selection of  MACT	
    2.6 SELECTION OF MACT FOR DECORATIVE  CHROMIUM
        ELECTROPLATING AND CHROMIUM ANODIZING TANKS
        2.6.1   Decorative Chromium Electroplating--
                Trivalent Chromium Electroplating
                Process	
        2.6.2   Decorative Chromium Electroplaters--
                Chromic Acid Electroplating Process .  .
        2.6.3   Chromium Anodizing Operations  	
    2.7 SELECTION OF THE  FORMAT  OF THE STANDARD   . .  .
    2.8 SELECTION OF THE  EMISSION LIMITS  	
    2.9 SELECTION OF DEFINITION  OF SOURCE  ......
    2.10   SELECTION OF COMPLIANCE DATES	  .
    2.11   SELECTION OF MONITORING REQUIREMENTS   . .  .
        2.11.1  Enhanced Compliance Monitoring  for
                Packed-Bed Scrubber and Composite
                Mesh-pad Systems   	
        2.11.2  Selection of Proposed Operation and
                Maintenance Requirements for
                Packed-Bed Scrubber and Composite
                Mesh-Pad Systems   	• •
        2.11.3  Frequency of Monitoring ........
        2.11.4  Alternate Compliance Monitoring for
                Sources Using Fume Suppressants  .  . . .
        2.11.5  Monitoring Requirements--General   . . .
    2.12  SELECTION OF TEST METHODS    ....  	
         2.12.1 Test Methods 306 and 306A  	
         2.12.2 Demonstrating Compliance When Using
                Fume Suppressants  	
         2.12.3 Demonstrating Compliance When Using
                Add-On Controls  	
 1-1
 1-1
 .1-3

 2-1
 2-4

 2-6

 2-12
 2-12

 2-14
 2-22
   •*

 2-26
 2-26
 2-30
 2-36

 2-47
 2-47

 2-53
 2-58
 2-63
 2-70
 2-73
 2-76
 2-83
 2-83
 2-92
2-101

2-106
2-116
2-119
2-119

2-127

2-129

-------
              TABLE OF CONTENTS (continued)
                   REPORTING AND RECORDKEEPING
       .13
        13
2.13  SELECTION OF
      REQUIREMENTS
      2.13.1  Reporting/Recordkeeping Associated
             With Fume Suppressants   	
             Reporting Frequency  	  .  .
             Reporting and Recordkeeping
             Requirements--General  	
             Requirements of the General
             Provisions   	
  14  OPERATING PERMIT PROGRAM  	
  15  WORDING OF THE REGULATION   	
2.16  MISCELLANEOUS	
     2.13.4
2
2
                                                         Page
2-137

2-137
2-138

2-140

2-146
2-148
2-150
2-155
                           VI

-------
                          LIST OF TABLES
TABLE 2-1.

TABLE 2-2.
PUBLIC COMMENTS RECEIVED
SUMMARY OF COMPLIANCE MONITORING
REQUIREMENTS AND WORK PRACTICE STANDARDS
FOR THE FINAL RULE .....  	
Page

 2-2



 2-88
                                VI1

-------

-------
                          1.0  SUMMARY

     On December 16,  1993,  the Environmental Protection Agency
(EPA) proposed national emission standards for hazardous air
pollutants (NESHAP)  for major and area sources performing hard
chromium electroplating, decorative chromium electroplating, and
chromium anodizing.   A total of 62 comment letters were received
from industry, trade associations, regulatory agencies, and
environmental groups during the public comment period.  As a
result of these comments, additional information was gathered,
evaluated, and incorporated into the EPA's responses to the
comments.  The comments and responses are summarized in this
document.  The comments and additional information gathered are
contained in the project docket and serve as the basis for the
revisions made to the standards between proposal and
promulgation.
1.1  SUMMARY OF CHANGES SINCE PROPOSAL
     In response to public comments and as a result of the EPA
reevaluation, changes have been made to the proposed standards.
Significant changes are summarized below.  All changes that have
been made to the regulation are explained fully in the responses
to the comments.
     The following are  expected to be the significant changes
discussed, listed in approximate  order from most to least
significant:
      1.  Emission limits based on the control techniques that
form the basis  for MACT have been revised.  The emission limit
based on the use of a composite mesh-pad system is
0.015 milligrams of total chromium per dry standard cubic meter
                                1-1

-------
 (mg/dscm)  of exhaust air.   The emission limit based on the use  of
 a fume suppressant is o.Ol mg/dscm.
       2.   Monitoring,  reporting,  and recordkeeping burden for
 decorative chromium "electroplating tanks using a  trivalent
 chromium process  that incorporates a wetting agent has been
 substantially reduced.
       3.   The compliance date for all hard chromium
 electroplaters and chromium anodizers is 2 years  after
 promulgation of the standards.  The  compliance date for
 decorative chromium electroplaters is 1 year after promulgation
 of the standards.
       4.   Monitoring requirements have been changed.   Monitoring
 provisions have been added for fiber-bed mist eliminators.  Also',
 the initial frequency of monitoring  surface tension and foam
 blanket thickness  has  not  been reduced,  but the final  rule allows
 a reduction in frequency if no exceedances occurred during a
 given  timeframe.
       5.   The operation  and maintenance requirements in the
 proposed rule have been  revised,  and are now work practice
 standards.
       6.   Reporting and  recordkeeping requirements have been
 revised.   The requirements  of  this rule and the interrelation of
 these  requirements with  the General  Provisions  have been
 clarified.  The relationship between subpart  N and subpart A is
 presented  in  Table 1 of  the rule.
       7.   Measurement  of hexavalent  chromium in lieu of total
 chromium is allowed to demonstrate compliance with the  emission
 limits.  A measurement method  for hexavalent  chromium  is
 specified  (ion chromatography using  a post  column reactor
 [ICPCR]).
       8.   Comments  were  solicited on regulating wastewater from
 control devices used to  comply with  the  standard.    Based on
 comments received,  no regulation  of  wastewater  is  included in the
 final rule.
      9.  The applicability of the standard has been clarified.
Hard chromium electroplating sources  can use  actual rectifier
                               1-2

-------
capacity to determine whether they are small or large if the
capacity is recorded by nonresettable ampere-hour (amp-hr)  meters
or if their source has accepted a Federally-enforceable limit in
the title V permit. • Also, the final rule exempts tanks used and
qualified for research and development purposes.
     10.  Definitions have been revised where necessary to
clarify requirements of the rule.
1.2  SUMMARY OF IMPACTS OF PROMULGATED ACTIONS
     The cost, economic, and environmental impacts to sources are
unchanged from proposal because the technologies that form the
basis of the MACT standards have not changed.  The monitoring,
reporting, and recordkeeping costs associated with the final rule
have been reduced from proposal and are presented in
Section 2.13.3.
                                1-3

-------

-------
                 2.0  SUMMARY OF PUBLIC COMMENTS

     A total of 62 letters commenting on the proposed regulation
and the background information document for the proposed
standards were received during the public comment period.  Three
late comments were also received and were considered in
finalizing this regulation.  Comments from the public hearing on
the proposed standards were recorded, and a transcript of the
hearing was placed in the project docket.  A list of commenters,
their affiliations, and the EPA document number of Docket A-88-02
assigned to their correspondence is given in Table 2-1.
     For the purpose of orderly presentation, the comments have
been categorized under the following topics:
      1.  NESHAP Decision Process;                         . .
      2.  Process Description and Description of Control
Technologies;
      3.  Selection of Pollutants and Source Categories to be
Regulated;
      4.  Selection of MACT/GACT Approach;
      5.  Selection of MACT for Hard Chromium Electroplating
Tanks;
      6.  Selection of MACT for Decorative Chromium
Electroplating and Chromium Anodizing Tanks;
      7.  Selection of the Format of the Standard;
      8.  Selection of the Emission Limits;
      9.  Selection of Definition of Source;
     10.  Selection of Compliance Dates;
•     11.  Selection of Monitoring Requirements;
     12.  Selection of Test Methods;
     13.  Selection of Reporting and Recordkeeping Requirements;
                               2-1

-------
TABLE 2-1.  PUBLIC COMMENTS RECEIVED
Commenter
M. Corbin
D. Martin
C. O'Neill
W. Schott
W. Wenske
A. Maclnnes
D. Driesen
C. Garland
S. letter
B. Bateman
J. Tyler
K. Tyler
W. Schott
K. Kurucz
J. O'Brien
L. Hill
L. Liszewski
G. Hannabarger
J. Walton
E. Blankenship
D. Theiler
B. Harthoorn
D. Foster
R. Schenker
L. Strohm
J. Lamancusa
J. Cullen
E. Smith
G. Bender
K. Yiin
A. Turnbull
D. Preston
J. Keittdey
M. Tyler
J. Keithley
A. Roy
R. Curtis
Affiliation
Puget Sound Air Pollution Control Agency
Plumbing Manufacturers Institute
Washington State Department of Ecology
Kimre, Inc.
Kohler Company
United Airlines
Natural Resources Defense Council
Arkansas Department of Pollution Control and Ecology
Fort Recovery Industries, Inc.
Bay Area Air Quality Management District
Crown Plating, Inc.
Crown Plating, Inc.
Kimre, Inc.
Lockheed Missiles and Space Company
Lockheed Corporation
Hill Air Force Base
Kodak
Rohr, Inc.
Tennessee Department of Environment and Conservation
Wisconsin Department of Natural Resources
Deere and Company
Delta Faucet Company
General Electric Company
General Motors
Lacks Enterprises, Inc.
Masco Corporation
Reynolds Metals Company
RR Donnelley and Sons Company
Stanley Mechanics Tools
Tumbull and Associates, Inc.
Varnum, Riddering, Schmidt, and Hewlett
Crown Plating, Inc.
Hard Chrome Specialties, Inc.
Hard Chrome Specialties, Inc.
Allied Signal
American Airlines
Document No.a
IV-D-01
IV-D-02
IV-D-03
IV-D-04
IV-D-05
IV-D-06
IV-D-07
IV-D-08
IV-D-09
IV-D-10
IV-D-11
IV-D-12
IV-D-13
IV-D-14
IV-D-15
IV-D-16
IV-D-17
IV-D-18
IV-D-19
IV-D-20
IV-D-21
IV-D-22
IV-D-23
IV-D-24
IV-D-25
IV-D-26
IV-D-27
IV-D-28
IV-D-29
IV-D-30
IV-D-31
IV-D-32
IV-D-33
IV-D-34
IV-D-35
IV-D-36
                 2-2

-------
                         TABLE  2-1.   (continued)
Commenter
E. Munsell
A.. Brooks
D. Drake
W. O'Sullivan
T. Allen
M. Bradley
D. Theiler
R. Colby
L. Carothers
L. Thorvig
C. Huston
G. Rountree
J. Graham
W. Ellsworth
H. Buffington
L. Zitko
H. Kimball
P. Leyden
B. Breisinger
G. Danielson
F. Chiappone
B. McGarvey
W. Sonntag
D. Martin
J. Keithley
R. Kennealy
R. Kennealy
R. Pelletier
D. Driesen
M. Lake
Affiliation
Department of the Navy
KCH Services, Inc.
Michigan Department of Natural Resources
Stew Jersey Department of Environmental Protection and
Energy
Hew York State Department of Environmental Conservation
Mortheast States for Coordinated Air Use Management
(NESCAUM)
State and Territorial Air Pollution Program Administrators,
and Association of Local Air Pollution Control Officials
(STAPPA/ALAPCO)
United Technologies
Minnesota Pollution Control Agency
Iowa Waste Reduction Center
Aerospace Industries Association
Atotech, Inc.
National Chromium
Snap-on Tools
ChromeTech, Inc.
Boeing
South Coast Air Quality Management District
Alcoa Company of America
Merlin Enterprises
Bausch and Lomb
National Aeronautics and Space Administration
National Association of Metal Finishers and The American
Electroplaters and Surface Finishers Society
Plumbing Manufacturers Institute
Crown Plating, Inc.
United Defense
United Defense
U. S. Department of Energy
Natural Resources Defense Council
San Diego County Air Pollution Control District
)ocument No.a
V-D-37
V-D-38
V-D-39
V-D-40
V-D-41
IV-D-42
IV-D-43
IV-D-44
IV-D-45
IV-D-46
IV-D-47
IV-D-48
IV-D-49
IV-D-50
IV-D-51
IV-D-52
IV-D-53
IV-D-54
IV-D-55
IV-D-56
IV-D-57
IV-D-58
IV-D-59
IV-D-60
IV-D-61
IV-D-62
W-G-01b
IV-G-02b
IV-G-03b
aDocket No. A-88-02.
bLate comment.
                                     2-3

-------
      14.   Operating Permit Program;
      15.   Wording of the Regulation;  and
      16.   Miscellaneous.
      In some instances,  several comments are presented
 individually,  but only one response  has been written to reply to
 all of the comments because the issues the comments raise are
 interrelated.
 2.1  NESHAP DECISION PROCESS
      Comment:   Four commenters (IV-D-07,  IV-D-43,,  IV-D-45,
 IV-G-02)  agreed with the EPA's interpretation of the statutory
 language  in section 112(d)(3)  of the  Clean Air Act (the Act)
 concerning the MACT floor for existing sources.   That is,  they
 agreed that the MACT floor for existing sources is properly
 determined by  first identifying the best performing 12 percent of
 the existing sources,  then determining the average emission
 limitation achieved by these sources.   One commeiiter (IV-G-02)
 noted that had Congress  intended for  the floor to be set at the
 88th percentile,  they would have stated this.   Five commenters
 (IV-D-14,  IV-D-15,  IV-D-17,  IV-D-28,  IV-D-36)  disagreed,  stating
 that the  MACT  floor should be  set by  looking at the average
 emission  limits achieved by each of the best performing
 12  percent of  existing sources,  and be based on the lowest
 (i.e.,  the 88th percentile).   Three commenters (IV-D-14,  IV-D-15,
 IV-D-17)  supported  this  interpretation by pointing out that the
 EPA could always  select  a MACT standard that is  more stringent
 than the  floor.   Another commenter (IV-D-36)  supported this
 interpretation by pointing  out that when  such an approach is
 taken,  one always arrives at an emission  limit that corresponds
 to  a technology (see  discussion of next comment).
      Comment:   Three  commenters  (IV-D-07,  IV-D-17,  IV-D-28)
 disagreed with the  EPA's  interpretation of  "average."   That is,
when determining  the MACT floor,  the average is  calculated  as  the
arithmetic mean.  When, however,  an average  calculated by this
method does not  correspond  to  a  technology,  the median is
considered the  average.   Two of  the commenters  (IV-D-17,  IV-D-28)
noted that this  interpretation seems contrary  to Congress'
                               2-4

-------
 intent.   Earlier versions of the Act identified the top 10 and
 15 percent.   In the latter case, it is unlikely that they would
:have meant the 93.5 percentile  (IV-D-17).   Likewise, this
', interpretation would infer that slight distributional variations
 among data points should dramatically impact the MACT floor
 (IV-D-28).  Finally, one commenter (IV-D-07) argued that if an
 arithmetic average does not correspond to a technology, the EPA
 may select a MACT standard more stringent than the floor.
      Response:  On June 6, 1994  (57 FR 29196), the EPA
 promulgated a final rule that presents the Agency's
 interpretation of section 112(d)(A) of the Act regarding the
 basis for the MACT floor.  Under this interpretation, which is
 referred to as the "Higher Floor Interpretation," the Agency
 considers the emission limitations achieved by the best
 performing 12 percent of existing sources and arrives at the MACT
 floor by averaging those emission limitations.  Furthermore, the
 Agency interprets  "average" to  mean a measure of central tendency
 such as  the arithmetic mean or  median.  For example, if a  source
 category included  fewer than  30 existing sources, the Agency
 would consider the five best  performing sources.  If those
 sources  achieve emission reductions of 99,  98, 95,  94, and
 93 percent, the arithmetic mean of the five values  would be
 95.8 percent.  Because this value does not  correspond  to an
 actual control efficiency being achieved, the Agency could set
 the MACT floor at  95 percent, which is the  median of the  five
 values.
      The Agency has concluded that  if  Congress had  intended  the
 EPA to set the MACT floor as  the lowest reduction achieved by  the
 .best performing  12 percent  of existing sources,  section  112(d)(A)
 would have been worded differently.   For example, Congress could
 have expressed  such a meaning by requiring  the  floor to  be set at
  "the emission limitation achieved by  all sources within  the  best
 .performing 12 percent" or at  "the emission  limitation achieved by
 ....the  least efficient member  of the best performing  12  percent."
 Therefore, the  Agency believes  that such an interpretation,  which
  is referred  to  as  the  "Lower  Floor Interpretation," requires that
                                 2-5

-------
 words  or concepts,  which are not stated explicitly in  the
 statute,  be added or inferred.   Furthermore,  the  Higher  Floor
 Interpretation is supported by section 112(d)(3)(B), which
 requires the MACT floor for existing sources  in categories  or
 subcategories with fewer than 30 sources to be no less stringent
 than the "average emission limitation achieved by the best
 performing 5 sources."   To infer a Lower Floor Interpretation
 from this provision,  Congress would have used language such as
 "the emission limitation achieved by the 5th  best performing
 source."
     Comment;   One  commenter (IV-D-02)  asked  for  an explanation
 of  the EPA's authority  for enforcing the final rule, and of the
 civil  and/or criminal penalties  for any violations of the
 standard.
     Response:  Section 113  of the Act  allows the EPA to assess a
 civil  penalty of  not  more than $25,000  per day per violation of
 the Act.   Section 113 also allows the EPA to  assess criminal
 penalties  for each  day  of a  violation for knowing violations of
 the Act.
 2.2  PROCESS  DESCRIPTION AND DESCRIPTION OF CONTROL TECHNOLOGIES
     Comment:   Eight  commenters  (IV-D-12, IV-D-22,  IV-D-32,
 IV-D-33,  IV-D-34, IV-D-47,  IV-D-52,  IV-D-56) provided comments on
 the process  description.   One commenter (IV-D-22)  questioned the
 statement  that  chromic  acid  is emitted  in significant quantities
 from all hard  chromium  electroplating and anodizing tanks,  and
most decorative chromium electroplating tanks, given the fact
 that the industry as  a  whole barely qualifies as  a major source.
     Two commenters  (IV-D-47, IV-D-52)  wished to  clarify the
EPA's discussion  of the boric-sulfuric  acid anodizing process
that the EPA identified as a potential  new process  to replace
chromium anodizing.   Both commenters  stated that  this process
cannot be assumed to  have widespread  applicability.  Also,  there
are other nonchromate processes besides boric-sulfuric acid
anodizing that are being  evaluated.   Based on this, the
commenters felt that  the  use of such processes should not be
                               2-6

-------
 mandated; jbhat  is,  air pollution  control  device  use should
 continue  to  be  allowed.
      One  commenter  (IV-D-56)  pointed out  a  distinction among
•decorative chromium electroplating processes:  black chromium and
.white chromium.   White chromium electroplating is  the decorative
 chromium  electroplating process described in the proposed rule.
 Black chromium  electroplating is  more like  hard  chromium
 electroplating  in terms of process parameters, and the commenter
 recommended  that black chromium electroplating be  regulated with
 hard chromium electroplating processes.
      The  other  commenters (IV-D-12, IV-D-32, IV-D-33, IV-D-34)
 pointed out  that the proposed rule does  not account for all hard
 chromium electroplating processes.  For  example, one hard
 chromium electroplating method uses low  amperage and a longer
 electroplating  time (less amperage per square foot than
.decorative electroplating processes) such that emissions are
 lower.  These facilities suggested that  the EPA has been led to
 believe that this is not a viable process by larger facilities,
 because mostly  smaller facilities use such a process.  This,group
 of commenters also noted that the terminology used in the
 proposed rule is incorrect.  Proper terminology would
 differentiate between industrial  chromium electroplating and
 decorative chromium electroplating.
      Response:   The process descriptions that were provided in
 the preamble to the proposed rule were intended to provide an
 overview.  More detailed process descriptions are provided  in
 Chapter 3 of the proposal background information document  (BID)
 developed for this standard.
      The EPA stated on page 65771  of the proposal preamble  that
 "... Chromic acid, a hexavalent chromium compound, is emitted in
 significant quantities from all hard chromium electroplating and
 anodizing tanks and from most  decorative.chromium electroplating
....tanks."  The EPA believes that this is a valid  statement,
^especially when the toxicity of hexavalent  chromium  is
 considered.  Even  though  the total quantity of  emissions may be
 small compared to  emissions of other pollutants, the cancer
                                2-7

-------
 potency of  hexavalent  chromium warrants  that the quantity  of
 emissions be  considered significant.
      In describing the chromium anodizing process,  the  EPA also
 described some  research that  focuses  on  the  use  of  a  sulfuric-
 boric acid  anodizing process  to replace  chromium anodizing in
 some  instances.   The EPA recognizes that research is  being
 conducted on  replacement processes other than this  one, and that
 such  processes  are not sufficiently developed for use in
 production.   So,  while the  EPA encourages continued research in
 this  area,  the  final rule does not mandate this  or  any  other
 replacement process for chromium anodizing.
      Comments were received regarding other  processes (e.g., the
 black decorative  chromium electroplating process and  the low-
 amperage hard chromium electroplating process) conducted by this
 source category that were not included in the process
 description.  In  developing the standards, the EPA  focused on the
 operations  that were widely regarded  as  being representative of
 the source  category.   The final rule  also discusses standards in
 terms  of hard chromium electroplating, decorative chromium
 electroplating, and chromium  anodizing.   However, the definitions
 of these processes  have been  expanded, and are now  expressed in
 terms  of process  parameters as well as by function.   For example,
 electroplating and  anodizing  operations  are  characterized  in
 terms  of electroplating thickness achieved,  current density, and
 electroplating time, as appropriate.   Therefore,  regardless of
 what name a facility has assigned to  its  process,, for the
 purposes of the regulation  the process will  be regulated
 according to  its  function,  bath operating parameters, and  desired
 plating  characteristics.  For example, based on  the description
 provided by the commenter,  black decorative  chromium
 electroplaters would be subject  to the standards  for  hard
 chromium electroplaters based on plating  characteristics.  The
 expanded definitions for each subcategory will most likely
alleviate confusion concerning the applicability  of the standard.
The EPA will provide States with additional  guidaince  on these
                               2-8

-------
 types of applicability issues in the enabling document for this
 rule.
'.:,     The commenters that use a low-amperage electroplating
•_process were concerned that such a process would not be allowed
 by the rule, even though emissions from this process are low.
 The process does differ from other hard chromium electroplating
 processes in that a lower amperage process is used.  However, the
 rule does not preclude the use of this process or any other
 technique to meet the applicable emission limit.  The rule does
 require that the technique be demonstrated through performance
 testing conducted in accordance with the test methods and
 procedures identified in the final rule, and that compliance
 monitoring be conducted to determine continuous compliance.
      Comment:  Seven commenters  (IV-D-04, IV-D-06, IV-D-10,
 IV-D-13, IV-D-43, IV-D-47, IV-D-55) provided comments on the
 description of the add-on air pollution control devices such as
 packed-bed scrubbers and composite mesh-pad systems.  One company
  (IV-D-13) noted that the proper terminology is one that refers to
 a.composite mesh-pad system, not a composite mesh pad, because
 high levels of control require a system of several composite mesh
 .pads.  The commenter goes on to provide a detailed description of
 a multistage system and the function of each stage.  This same
 commenter also stated that their company should be named in  the
 rule because the system described uses the company's patented
 design.
      Another commenter  (IV-D-06) noted an inaccuracy in the  EPA's
 control device description; the  statement that a fiber-bed mist
 eliminator is only installed at  a government facility is
 inaccurate.  This commenter identified a commercial facility that
 employs such a control device.
      Several commenters  (IV-D-10, IV-D-13, IV-D-43, and IV-D-47)
 stated that the control efficiencies identified in the proposed
..";rule are overstated.  According  to the commenters, removal
 ^efficiencies for packed-bed scrubbers may be as low as 50 percent
 -(IV-D-10, IV-D-43) and are only  as high as 90 percent  (IV-D-10,
 IV-D-43), 95 percent  (IV-D-47),  or 97 percent  (IV-D-13).  Two
                                2-9

-------
commenters  (IV-D-io, IV-D-43) also noted that the efficiency of
composite mesh-pad systems were overstated, and that they were
more likely in the range of  60 to 95 percent.
     One commenter  ('IV-D-55) stated that the proposed rule failed
to identify all of the possible control systems that could be
used to control chromium emissions from hard chromium
electroplating tanks.  This  commenter described the Merlin mist
control system and provided  test reports to demonstrate that this
system can achieve the control levels required by the proposed
rule.
     Response;  In the final preamble and rule, the EPA has used
the terms composite mesh pad and composite mesh-pad system
consistently.  Although the EPA gathered information based on
mesh pads developed by Kimre, Inc., it is not the EPA's policy to
endorse certain products or recommend them for use.  In response
to comments received on the efficiency of composite mesh pad and
packed-bed scrubber systems, the EPA recognizes that efficiencies
will vary depending on the actual system.  For exampl-e, a system
with a lower inlet loading will have a lower overall control
efficiency.  Also, the manner in which a system is operated will
affect its efficiency.  Given the potential problems in
identifying an overall control efficiency for these systems, the
final rule requires that an outlet concentration,, not an
efficiency, be achieved.  The control efficiencies specified in
the proposed rule are typical for well-operated and well-
maintained systems operating at typical conditions.  The EPA has
included some work practice standards in the fineil rule
[§ 63.342 (f)] to facilitate optimal control device operation to
meet the emission limits.
     A benefit of expressing the standard in terms of a
concentration format is that it provides sources with flexibility
in terms of choosing the control device to meet the standard.
Therefore,  if a system, such as the Merlin mist control system,
is tested using the specified method and procedures and meets the
applicable concentration limit, it can be used to comply with the
rule.  The EPA does not have the resources to test every possible
                               2-10

-------
 control  system that could be used to achieve  compliance with the
 standard;  testing focused on the major types  of  control equipment
:that were  thought to achieve the best emission reduction.
      Since the time of proposal, the EPA has  confirmed that other
 fiber-bed  mist eliminators,  in addition to the one that was
 discussed  at proposal, exist.  The existence  of  multiple fiber-
 bed mist eliminators is discussed in more detail in
 Section 2.5.1, Selection of  the MACT Floor for Hard Chromium
 Electroplating Tanks.
      Comment:  Four commenters  (IV-D-22, IV-D-32,  IV-D-34,
 IV-D-58) provided comments on chemical fume suppressants as a
 control technology.  One commenter (IV-D-22)  noted that, contrary
 to the EPA's description, his experience was  that electroplating
 baths that use fume suppressants do usually have ventilation
 systems.  The other three commenters disagreed with the EPA's
 opinion on the potential negative aspects of foam blanket-type
 fume suppressants.  Two commenters (IV-D-32,  IV-D-34) stated that
 the explosion potential of certain fume suppressants could be
 minimized by an effective safety training program.  Another
 commenter  (IV-D-58) disagreed that fume suppressants cause
 .process and product problems for hard chromium electroplaters and
 pointed to data that show that  inlet chromium emissions to a
 scrubber were effectively reduced by using fume suppressants in a
 hard chromium electroplating bath.
      Response:  In describing fume suppressants and tanks  in
 which fume suppressants are used in the preamble to the proposed
 rule and the BID, the EPA described what was found to be true
 based on their information gathering.  Thus, it may be  true that
 it  is one  facility's  experience that all sources that use  fume
 suppressants are ventilated.  However,  this facility may not be
 familiar with sources located in other  areas of the country, or
 of  smaller size, etc., that  do  not have ventilation.   Likewise,
 :when discussing the  use  of  fume suppressants in hard chromium
 .electroplating tanks, it has been the EPA's experience  that they
 •are not generally used because  of performance problems.  However,
 this does  not mean  that  fume suppressants  are never used,  or that
                                2-11

-------
 fume  suppressants  that  can be successfully used in hard  chromium
 electroplating tanks  do not exist.   In fact,  on page  65781 of the
 proposal  preamble,  the  EPA describes the  test results referred to
 by commenter IV-D-58".   Two systems  are described---one in which
 fume  suppressants  were  used prior to a mist eliminator and one in
 which a fume suppressant was used prior to a packed-bed  scrubber.
 When  the  fume suppressant was used  in the tank controlled by an
 older packed-bed scrubber,  the use  of the fume suppressant
 enhanced  the overall  control of the system.   In the case of the
 fume  suppressant being  used prior to a new mist eliminator, only
 a  slight  decrease  in  outlet emissions was observed.   These data
 suggest that fume  suppressants can  be used to enhance the control
 level achieved by  older control systems.   Because  the rule
 requires  sources to meet an emission standard rather  than
 specifying a particular type of control equipment,  nothing
 precludes the owner or  operator from using any control scenarios,
 such  as those described in the proposal preamble,  as  long as the
 standard  is  achieved.
      In the  preamble  to the proposed rule,  the EPA discusses the
 potential safety hazards of foam blankets,  a  type  of  fume
 suppressant.   Foam blankets trap the hydrogen gas  and chromic
 acid  mist in the foam layer.   When  heavy  foam layers  develop,
 hydrogen gases  can build up in the  foam;  if a spark is generated,
 a hydrogen explosion can occur.  The EPA  agrees  that  foam
 blankets can be used safely,  but also believes  that the  potential
 hazards of foam blankets should be  known.  The  rule does not
 preclude the use of foam blankets as long  as  the applicable
 emission limit  is met.
 2.3   SELECTION OF  POLLUTANTS  AND SOURCE CATEGORIES TO BE
      REGULATED
 2.3.1  Selection of Pollutants  to be Regulated
      Comment:  In presenting  the health effects data  that led to
 the listing  of the chromium electroplating and chromium anodizing
area  source  categories,   the EPA requested  comments on whether
 there was a basis for removing  any  of  the  categories and
subcategories covered by  the proposed  regulation from the source
                               2-12

-------
 category  List.   Comments  were  solicited primarily because the EPA
 proposed  regulation of  decorative  chromium electroplaters using
•the  trivalent  chromium  electroplating process.   The EPA was
 -specifically interested in whether the delisting criteria of
 section 112(c)(9)  could be met,  or alternatively,  if it would be
 appropriate  to conduct  a  health assessment under
 section 112(c)(3)  prior to creating this subcategory and adding
 it to the source category list.  Two commenters  (IV-D-26,
 IV-D-59)  responded specifically to this request  for comment and
 stated that  the EPA should qonduct a health assessment under
 section 112(c)(3)  to determine the health effects of trivalent
 chromium  prior to creating a subcategory regulating this process.
                                                                 •*
 Both commenters argued  that until  such an assessment is made, the
 trivalent chromium process should  not be addressed in the rule.
 Another commenter (IV-D-24)  asserted that the rule must
 acknowledge  the fundamental toxicological differences between
 trivalent and  hexavalent  chromium; the same commenter contends
 that there are no data  indicating  that trivalent chromium is
 carcinogenic.   In addition to  these three commenters, many other
 commenters disagreed with the  EPA's decision to  regulate
 decorative chromium electroplating sources that  use the trivalent
 chromium  electroplating processes.  Their comments did not
 specifically respond to the request pertaining to information on
 a section 112(c)(3)  health assessment.  These more general
 comments  are presented  in Section  2.6.1 of this  document, which
 discusses selection of  MACT for decorative chromium
 electroplaters using a  trivalent chromium electroplating process.
      Response;  The EPA has determined that the  delisting
 criteria  of  section 112(c)(9)  cannot be met at this time for
 trivalent chromium electroplating  processes and  that conducting a
 health assessment under section 112(c)(3) cannot be justified in
 light of  the limited data available for trivalent chromium.
 •Specifically,  the EPA is aware of  health concern issues
 associated with trivalent chromium exposure that would not allow
 the  statutory  criteria  of section  112(c)(9) nor
 section (112)(c)(3)  to  be met.  The testing of trivalent chromium
                               2-13

-------
compounds is significantly more limited than that available for
hexavalent chromium and is considered inconclusive at this time.
The currently available trivalent chromium toxicological studies
have not shown dose-'related increases in carcinogenic response;
however, trivalent chromium compounds exhibit geriotoxic
potential.  Of additional concern is the in vivo reduction of
hexavalent chromium to trivalent chromium which is believed to be
important in chromium's mechanism of carcinogenicity.  There are
also reports that under certain environmental conditions,
trivalent chromium can oxidize to the hexavalent state.
2.3.2  Selection of Source Categories to be Regulated
     Comment;  The EPA separated the decorative chromium
electroplating source category into two subcategories:
electroplating baths using a chromic acid electroplating process
and electroplating baths using a trivalent chromium
electroplating process.  The distinction between the
subcategories was based on process differences.  One commenter
(IV-D-22) disagreed with one of the differences cited by the EPA;
i.e., that additional posttreatment and rinsing tanks may be
needed with trivalent chromium processes.  The commenter pointed
out that trivalent chromium baths do not always have more rinse
tanks or have post dips.  According to this commenter, trivalent
solutions rinse better than hexavalent solutions,,
     Response;  In creating a subcategory, the EPA discusses
characteristics that generally distinguish one subcategory from
another.  It has been the EPA's experience that cidditional
posttreatment and rinsing tanks are required in some trivalent
chromium applications.  In an industry as large eind diverse as
chromium electroplating, however, the EPA is aware of the fact
that some trivalent chromium applications may not; require
additional rinses.  The description of the trivalent chromium
process in the proposed preamble was written generally so that
all known trivalent chromium processes would be covered by the
description.  In addition, the regulatory definition for the
trivalent chromium process, not the process description, is the
determining factor in deciding the applicable requirements in the
                               2-14

-------
final rule.   The definition used in the proposed rule defined the
trivalent chromium process as a process used for electro-
deposition of a thin layer of chromium onto a base material using
a trivalent chromium solution instead of a chromic acid solution.
In the definition, no mention is made of the other process tanks
used in the trivalent chromium electroplating line; therefore,
any process using a trivalent chromium solution to deposit
chromium onto a base metal would be subject to the requirements
in the final rule for the trivalent chromium process, regardless
of the existence of rinse or posttreatment tanks.
     Comment:  Six commenters  (IV-F-01  [Altmayer, McRae,
Sonntag], IV-D-51, IV-D-58, IV-D-59) said that maximum cumulative
potential rectifier capacity was an inappropriate parameter for "
determining facility size.  Commenters pointed out that sources
may have excess capacity for any of the following reasons:
     1.  Sources have purchased used equipment that  is larger
than actually necessary;
     2.  The rectifier is sized to accommodate either .the largest
piece plated, striking that is performed at the beginning of  the
process at high current densities, or some nonchromium
application that requires more capacity than chromium
electroplating;
     3.  The rectifier may be  oversized to prevent chromium burn
or DC power supply failure or because actual capacity needs are
unknown; or
     4.  In full-scale production, more rectifier  capacity may be
needed  than in test or pilot operations.
     One commenter  (IV-D-58) indicated  that using maximum
potential rectifier capacity would unfairly cause  small
facilities with multiple  tanks to be counted as  large,  even
though  it is not  customary to  use all of the tanks at once.   Two
other  commenters  (IV-D-17, IV-D-46) stated that  if maximum
potential rectifier capacity was used as the size  determinant,
,the  rule would have to be more clear that it is  only the
rectifiers associated with chromium electroplating that are  of
concern.  Finally, one  commenter  (IV-D-47) said  that the
                               2-15

-------
distinction between  small  and large  sources  was  not  sufficient
because MACT was proposed  for both categories, and the EPA has
made no determination as to what  level  of  chromium emissions
constitutes an adverse  effect on  human  health or the environment.
     Ten commenters  (IV-F-01  [Altmayer], IV-D-17, IV-D-21,
IV-D-28, IV-D-31,  IV-D-36, IV-D-46,  IV-D-51, IV-D-58, IV-D-59)
urged the EPA to explore alternatives to maximum potential
rectifier capacity for  determining facility  size.  Commenters
suggested using actual  annual ampere-hour  usage,  perhaps over the
most recent 12-month period,  which could be  demonstrated by using
nonresettable ampere-hour  meters  or  through  facility records.
Other commenters suggested raising the  maximum potential ampere-
hour limit for small sources  to 100  million  amp-hr/yr  (IV-D-46) ,"
or allowing sources  to  multiply the  maximum  potential rectifier
capacity by 0.75 to  account for oversizing (IV-D-51).  Two
commenters  (IV-D-21, IV-D-28)  suggested that, at a minimum, the
rule should clearly  indicate  how  sources could accept
Federally-enforceable limits  on their rectifier  capacity that
would allow them to  remain "small."
     Response:  The  distinction between small and large hard
chromium electroplating facilities in the  final  rule has not been
changed from the proposed  level of a maximum cumulative potential
rectifier capacity of 60 million  amp-hr/yr.  The EPA has,
however, included  provisions  in the  final  rule to allow sources
to use actual rectifier capacity  or  to  limit their potential
rectifier capacity.
     The final rule  allows hard chromium electroplating
facilities to determine their size by using  actual cumulative
rectifier capacity in lieu of  the  maximum  potential  capacity if
nonresettable ampere-hour meters are used  on the chromium
electroplating tanks.  The EPA believes that the use of such
meters will provide  an accurate accounting of the rectifier
capacity consumed  in a given year.  Also,  the final  rule has been
clarified to state that only  the rectifiers associated with hard
chromium electroplating should be  used  to determine maximum
cumulative potential rectifier capacity.
                               2-16

-------
      The  final  rule  also  allows  sources  to  establish
 Federally-enforceable  limits  on  their  rectifier capacity  through
 the  title V operating  permit  program.  For  example, a  source may
 exceed the  60 million  amp-hr/yr  cutoff based on maximum capacity,
 and  therefore be  considered large.   This source may, however,
 have excess capacity onsite for  any of the  reasons provided by
 the  commenters.   In  such  a case,  the source may agree  to  a
 Federally-enforceable  limit that would restrict the actual
 cumulative  capacity  used  to less than  60 million amp-hr/yr.  A
 Federally-enforceable  limit is obtained  through the title V .
 permit that is  required by §  63.340(e) of the final rule.
      To demonstrate  compliance with an actual cumulative
 rectifier capacity limit,  sources are  required to keep monthly
 records of  the  actual  cumulative rectifier  capacity, and  report
 these records,  in accordance  with §§ 63.346(b)  and  63.347(g) or,
 (h).  Once  a source  exceeds an actual  cumulative rectifier
 capacity of 60  million amp-hr/yr for any 12-month period  or less,
 the  source  is  considered  large and subject  to the existing source
 MACT standard for large sources.  A subsequent reduction  in
 rectifier capacity would  not  result in less stringent
 requirements;  the standard for large sources would  continue to
 apply.  Once a monthly record during the 12-month reporting
 period indicates  that  a source has exceeded the rectifier
 capacity of 60  million amp-hr/yr, the source has l  year from that
 month to come into compliance with the requirements for large
 existing sources.
      The EPA disagrees that the  size distinction between small
 and large hard chromium electroplating facilities "is  not
 sufficient."  The size distinction was made to more effectively
 evaluate the burden of the rule  on small sources, given the large
 number of small sources in this  source category.  The  EPA
 considers the issue  of applying  MACT to sources to be  different
 than the issue of size distinctions.  The rationale for applying
-MACT to these sources is  discussed in detail in Section 2.4.
      Comment:   Five commenters  addressed the applicability of the
 rule to the source category in general.   One commenter (IV-D-60)
                               2-17

-------
indicated .that the EPA was incorrect to establish subcategories
based on type of deposit  (hard chromium versus decorative
chromium electroplating).  Instead, this commenter stated that
subcategories should be based on the amperage used in a process,
similar to the California regulations.  Another commenter
 (IV-D-18) indicated that coating  (including chromium anodizing)
in the aerospace industry was to be covered under the control
techniques guideline  (CTG) and NESHAP that were currently under
development for the industry and, therefore, should not also be
covered by this rule.  One other commenter did not see his
process described in the proposed rule and was not sure how it
fit into the subcategories in the regulation.  This commenter
 (IV-D-35) described a process involving an alkaline solution used
to provide a chromate layer on copper foil.  According to the
commenter, the alkaline solution is made up in part by using a
small amount of chromic acid that is subsequently neutralized.
The commenter stated that there are no emissions to the
atmosphere from this process, nor are air pollution control
devices or fume suppressants appropriate for it.  As such it
should not be covered by the rule; this could be accomplished by
defining a chromic acid solution in terms of its pH.  The
commenter also urged the EPA not to require any recordkeeping to
document that such a process is being used because the facility
would then be subject to the NESHAP and to title V permitting.
     One commenter (IV-G-01)  cites the notice of the initial list
of source categories  (57 FR 31576), which states that only those
processes that use chromic acid in an electrolytic cell to
deposit chromium metal or to form an oxide film on a product are
covered.  This commenter requested that the EPA clarify that
chromic acid and/or sodium dichromate used in conjunction with
sulfuric acid for parts cleaning is not classified as chromium
anodizing even though the surface of the base metal may be
oxidized.
     Response:  The EPA disagrees that the type of deposit
(i.e.,  hard chromium versus decorative chromium)  is an
inappropriate basis for subcategories.  The EPA believes that
                              2-18

-------
subcategories based on the amperage  in a process would be
difficult to establish.  In finalizing this MACT standard,
however, the EPA recognized that the definitions provided for
,hard  chromium electroplating, decorative chromium electroplating,
and chromium anodizing may have been vague for some operations.
In the  final rule, the EPA has revised these definitions to
express the various chromium  electroplating and anodizing
operations in terms of process parameters such as plating
thickness achieved, current density, and plating or anodizing
time.   The definitions are clear that electroplating and
anodizing operations  are performed as part of a process; as  such,
cleaning activities that involve chromic acid would not be
included.  The  EPA believes that the revised definitions will
assist  affected sources and regulators in determining which
category is most appropriate  for a given process, regardless of
the name assigned  to  that process.   If the chromic acid
concentration and  other process parameters associated with the
alkaline solution  described by commenter IV-D-35 were
inconsistent with  the description of the chromium electroplating
or anodizing tanks, the source is not subject to the regulation.
If a  source is  not subject to the regulation, no recordkeeping
burden  is imposed  by  this rule.  However, State or local
permitting agencies may need  documentation to show that  the  rule
is not  applicable.
      There appears to be some confusion  surrounding  the
applicability of this rule to the aerospace  industry.  A GTG and
a NESHAP covering  coating operations in  the  aerospace  industry
are under development by the  Agency; the aerospace NESHAP is
•expected to be  promulgated by the  summer of  1995.  Neither the
aerospace NESHAP nor  the CTG  includes requirements  for chromium
electroplating  and anodizing  operations.  Those operations are
covered by  this standard.   It is not unusual for  large
^facilities,  such as aerospace facilities,  to be  covered by more
-than  one MACT  standard.  Often,  the  EPA develops  standards to
 control operations that  are  conducted across many industries,  and
are performed  in a relatively uniform fashion.   For example, the
                               2-19

-------
 EPA has promulgated standards that apply to industrial process
 cooling towers that could be located at many different types of
 manufacturing facilities.
      Comment;   Pour -commenters (IV-D-24,  IV-D-47,  IV-D-52,
 IV-D-58)  recommended that research and nonproduction operations
 be exempted from the rule as allowed by section 112(c)(7)  of the
 Act.
      Response:   The final rule exempts research or laboratory
 operations  as allowed by section 112(c)(7).   For the purposes of
 this  rule,  a research or laboratory source is defined as  any
 stationary  source whose primary purpose is to conduct research
 and development into new processes and products,  where such
 source  is operated under the close supervision of technically
 trained personnel and is not engaged in the manufacture of
 products for commercial sale in commerce,  except in a de  minimis
 manner.
      Comment:   Eleven commenters  (IV-D-23,  IV-D-24,  IV-D-27,
 IV-D-28,  IV-D-32,  IV-D-34,  IV-D-42,  IV-D-47,  IV-D-52,- IV-D-58,
 IV-G-01)  suggested that the EPA should consider applying  the rule
 differently to  area sources.   One commenter (IV-D-42)  suggested
 that  the EPA establish a lesser quantity emission rate to
 distinguish area and major  sources based on the potency of
 hexavalent  chromium.
      On  the other hand,  other commenters  stated that  the  EPA
 should establish a de minimis emission level,  below which sources
 would not be subject to the rule.   One commenter (IV-D-23)
 pointed  to  a previous decision that  demonstrates  that the EPA has
 the authority to establish, such cutoffs  (e.g.,  Alabama Power Co.
 v. Costle).  Another (IV-D-47)  stated  that  section 112(d)(l)
 allows the  EPA  to  establish such  exemptions.   Other commenters
 proposed  various  cutoff levels:   5  to  10 pounds (IV-D-47),
 1 million amp-hrs/yr (IV-D-47,  IV-D-52), 0.013  tons of  emissions
 (IV-D-28),  and  less  than 10  employees  (IV-D-32,  IV-D-34).
Another commenter  (IV-D-27)  proposed that sources  in  this
 subcategory that fall  below a de minimis level  (e.g.,  5 million
Amp-hr/yr)  be exempt  from the regulation.  One  commenter
                               2-20

-------
 (IV-D-47) .suggested that  the  EPA perform a risk assessment  to
 determine  the  level below which the health risk is  de  minimis,  or
•-..below which the  cost of compliance exceeds the benefit of
-emission reduction. '                                             .
      Response:   Although  this rule regulates both area and  major
 sources, the EPA has considered the potential burden on small
 sources (most  of which will also be area sources) by regulating
 small, hard chromium electroplating facilities differently  from
 large, hard chromium electroplating facilities.  Also, for  all
 source categories,  the monitoring, reporting, and recordkeeping
 requirements have been greatly reduced.   The EPA does  not  support
 a de minimis cutoff below which sources  would not be subject to
                                                                 «*
 this rule  because,  for this source category, the EPA has
 determined that emissions from area sources present, in the
 aggregate, a threat of an adverse effect to human health or the
 environment.  For this reason, area sources performing chromium
 electroplating and anodizing were placed on the list of area
 sources to be regulated  (57 FR 31588).
      On the other hand,  the establishment of a lesser quantity
 emission rate for the source category would"only decrease the
 population of area sources while  increasing the population of
 major sources.  Further,  because  of the risk potential of
 hexavalent chromium, the EPA has  concluded that it  is important
 to regulate all sources with MACT.   (See discussion in
 Section 2.4.)   The application of MACT to all  sources results  in
 equivalent control levels being applied to all sources.  At this
 time, the EPA is not proposing to establish,  for this rulemaking,
 a lesser quantity  emission rate for chromium emissions from this
 •source category.   The EPA is  still evaluating a lesser quantity
 .emission rate for  other  sources of chromium emissions, but no
 decisions have been made to date.  An Advance Notice  of Proposed
 Rulemaking will soon be  issued on this topic.
       Comment;  One commenter  (IV-D-18) states that  the rule
 should exempt facilities  that,are complying with a  similar State
 or  local  rule that requires  that  chromium emissions be controlled
 by  95 percent.
                                2-21

-------
      Response:   There are several State and local regulations
 that require control of chromium emissions from chromium
 electroplating  and chromium anodizing tanks.   Some of these
 regulations are expressed in terms of a control efficiency or in
 terms of mass emissions per amp-hr of operation (mg/amp-hr);  as
 such the requirements are not directly comparable to this rule.
 If a source is  subject to a State or local rule,  it is also
 subject to this NESHAP.  However,  States may  revise their rules
 to make them consistent with this rule.   Alternatively,  States
 may choose to have their regulations determined equivalent by the
 EPA.   If a source is currently controlling emissions from
 chromium electroplating and anodizing tanks to comply with a
 State or local  rule,  they may have sufficient control in place to
 meet  this standard as well.   Given the toxicity of chromium
 compounds,  the  EPA does not support the  exemption of any sources
 within the regulated source categories.   The  toxicity of chromium
 compounds is discussed in more detail in the  following section.
 2.4   SELECTION  OF MACT/GACT APPROACH
      Comment;   Four commenters (IV-D-07,  IV-D-40,  IV-D-41,
 IV-D-45)  agreed with the EPA that  it was appropriate to regulate
 area  sources with MACT standards.   One of these commenters
 (IV-D-07)  noted that  the EPA could not legally exempt sources
 emitting 9  tons of chromium without examining the  cumulative  HAP
 emissions of the entire facility,  not just the chromium
 electroplating  processes.   Ten commenters (IV-D-22,  IV-D-24,
 IV-D-28,  IV-D-32,  IV-D-34,  IV-D-43,  IV-D-47,  IV-D-58,  IV-D-59,
 IV-D-69)  questioned this approach,  however, for various  reasons.
 One commenter (IV-D-22)  did  not see how  the EPA could apply MACT
 to area  sources when  the source category as a whole  barely
 qualifies as  a  major  source.   Five  commenters questioned "...  the
 high  toxicity of chromium compounds  ..."  as the basis  for
 applying  MACT.   Two of  these commenters  (IV-D-24,  IV-D-58)  noted
 that  different  forms  of chromium (hexavalent  and trivalent) need
 to be considered separately,  especially when  determining whether
 to regulate all  sources under MACT.   A third  commenter (IV-D-60)
provided  information  to support his  claim that  research  to
                               2-22

-------
establish -the potential  risk associated with chromium  compounds
is  flawed and is not  intended  to establish real risk.  Two
..-commenters  (IV-D-32,  IV-D-34)  stated that the risk  of  hexavalent
chromium to humans has never been accurately assessed.
      Two commenters noted  that the  cost to area sources  subject
to  MACT is significant.  One  (IV-D-58) stated that  the EPA  failed
to  consider the costs for  area sources.  The same commenter
compared this industry to  the  dry cleaning industry in which area
sources were regulated by  GACT.  Another commenter  (IV-D-43)
noted that MACT could be an unnecessary burden on area sources
particularly if they  are required to obtain title V operating
permits.   (See related discussion in Section 2.14.)  One
commenter  (IV-D-24). questioned the  EPA's decision to apply  MACT
to  area sources on the grounds that the Act does not intend a
residual risk analysis for area sources.
      Three  commenters (IV-D-24,  IV-D-47, IV-D-58) suggested
specific changes  in the  way the EPA applies MACT.   One commenter
 (IV-D-58) suggested that the  EPA apply GACT standards  to small,
existing hard chromium electroplaters, decorative  chromium
electroplaters, and chromium anodizers to  allow the EPA to focus
its resources on  facilities posing  the greatest  impact.   Another
commenter  (IV-D-47) suggested that  the EPA establish a threshold
below which sources would  be subject  to  GACT.   Finally,  one
commenter  (IV-D-24) pointed out that  it  was  important  to have
separate  standards  for  area sources even if  GACT was as stringent
as  MACT to  preserve  the  intent of  section 112(d).
      Response:   In determining whether to apply MACT or GACT to
 the area sources  performing hard chromium electroplating,
 decorative  chromium electroplating, and chromium anodizing, the
 EPA considered the toxicity of chromium compounds emitted from
 such sources and the  availability of controls for sources in the
 source categories.
      In the final rule,  all area sources continue to be regulated
 by MACT standards.   The basis for applying MACT to area sources
 is that hexavalent chromium is emitted in significant quantities
 from most source categories and is a known human carcinogen.  In
                                2-23

-------
addition, .the EPA believes  all  chromium compound  emissions  should
be  considered toxic because adequate  data are not available to
determine otherwise.  At  least  three  epidemiological  studies have
shown a strong association  of lung  cancer and occupational
exposures to mixtures of  hexavalent and trivalent chromium.  Even
though hexavalent chromium  is a known human  carcinogen,  it
remains unclear whether any of  the  observed  association  is  due to
trivalent chromium.  Trivalent  chromium is a necessary dietary
nutrient; however, there'is some evidence to indicate that
hexavalent chromium is reduced  in part  to trivalent chromium in
vivo.  The link between hexavalent  and  trivalent  chromium in vivo
is  of concern.  Very limited data suggest that trivalent chromium
may have respiratory effects on humans.  No  data  on chronic or
subchronic effects of inhaled trivalent chromium  in animals is
currently available.  Due to the uncertainties associated with
the toxicity of trivalent chromium, the EPA  has chosen to
regulate sources using a  trivalent  chromium  process with MACT.
     The toxicity of hexavalent chromium is  well  established;
hexavalent chromium is a  highly potent,  known human carcinogen.
Even though total emissions of  hexavalent chromium from  the
source categories are low relative  to emissions of other
pollutants (such as solvents) from  other source categories, the
EPA decided that MACT rather than GACT  would be aipplied.  The EPA
believes that this decision is  justified because, given  the
cancer potency, the residual risk analysis may still  reveal a
risk greater than one in  a  million  for  facilities} using
hexavalent chromium.  The EPA did not think  it wets prudent  to
establish a threshold based on  risk below which sources were
subject to GACT given the potency of  hexavalent chromium and
uncertainties in the toxicity data  base discussed above.  As one
commenter points out, the EPA has applied GACT in other
rulemakings,  such as the  dry cleaning NESHAP.  However, the
pollutant of concern in that instance was perchloroethylene,
which is much less potent than  hexavalent chromium.
     In selecting MACT over GACT for  area sources, the EPA  also
evaluated the availability  of control technologies and the  cost
                              2-24

-------
 of compliance for area sources.  The control technologies that
 form the basis for MACT are widely available and, as such, would
 likely have been selected as GACT for most of the area source
 categories.  One exception is the large, hard chromium
 electroplating source category, for which packed-bed scrubbers
 may have been a possible alternative for GACT for existing
 sources.  In the EPA's analysis, however, a standard based on
 composite mesh-pad technology was found to be reasonable for
 large, hard chromium electroplaters.  The final rule does,
 however, allow existing small, hard chromium electroplaters to
 meet an emission limit that is based on packed-bed scrubbers;
 most of these small sources will also be area sources.  The EPA
 had determined that those existing area sources that fit the
 "small" designation could experience economic hardship if forced
 to meet a standard based on composite mesh-pad technology,
 especially if a packed-bed scrubber system was already in place.
      In light of comments received at proposal, the EPA has
 reduced the monitoring, reporting, and recordkeeping requirements
 for all sources  (see discussion in sections 2.11 and 2.13,
 respectively), which will further alleviate the burden on area
 sources.  Also, the compliance timeframe for decorative chromium
 electroplaters has been increased to 1 year, and the compliance
 timeframe for hard chromium electroplaters and chromium anodizers
 has been increased to 2 years.  These timeframes will allow
 existing area sources more time to obtain the capital necessary
 to purchase controls, train employees, and to implement
 monitoring, reporting, and recordkeeping systems.
      Many commenters point to section 112(d) of the Act to
 support their reasoning that all area sources should be regulated
 by GACT.  Section 112(d)(5) does indeed allow an alternative
 standard for area sources.  The EPA, however, interprets  this
 •paragraph as authorizing the Administrator to establish GACT
""standards for area sources when the imposition of MACT is
 determined to be unreasonable.  -The Agency considers the
 application of MACT to area sources in  this source category to be
 reasonable based on the risks of chromium compounds and the
                                2-25

-------
 results  of. the  cost  analysis,  as  discussed above.   The  EPA
 believes that its  interpretation  is  supported by section  112(f),
 which  discusses the  need for the  EPA to  investigate and report to
 Congress on residual" risk remaining  from sources regulated under
 section  112(d).  Section 112 (f) (5) excludes area sources
 regulated by GACT  under section 112(d)(5)  from this analysis.  If
 the  intent of the  Act was for all area sources to be regulated by
 GACT and not subject to a residual risk  analysis, it would not be
 necessary for section 112(f)(5) to exclude only those area
 sources  regulated  by GACT.
 2.5  SELECTION  OF  MACT  FOR HARD CHROMIUM ELECTROPLATING TANKS
 2.5.1  Selection of  the MACT Floor
     Comment;   Four  commenters (IV-D-10,  IV-D-41, IV-D-42,
 IV-D-43)  suggested that the  MACT  floor for new hard chromium
 electroplating  and chromium  anodizing tanks should  be based on
 the  use  of a fiber-bed  mist  eliminator  (FEME)  because this is the
 best technology in use.   All of these commenters disagreed with
 EPA's  rationale for  not requiring this technology as MACT for new
 sources;  they noted  that the Act  does not exclude government
 installations,  and in any event nongovernment installations
 exist.   One commenter (IV-D-41) knew of  two such installations in
 New  York State,   and  was told by a vendor of FEME that 60  units
 had  been installed by them nationwide.
     Response;   The  EPA has  further  investigated the prevalence
 of FEME  in response  to  comments received at proposal.   As
 discussed above, one commenter (IV-D-41)  indicated  that two FEME
 had  been installed in the State of New York and a total of
 60 units  had been  installed  nationwide.   The two installations in
 New  York are described  as "four-stage polypropylene pad
 de-misters," indicating that  there is some  confusion in control
 device terminology.   The description given  to the two
 installations in New York is  representative of  a composite
mesh-pad  system, not  a  FEME.   As  described  in the preamble to the
proposed  rule,  FEME  are typically installed downstream  of  an
 existing  control device and  consist  of hollow cylinders formed
 from two  concentric  screens.   The annular space between the
                               2-26

-------
cylinders -is filled with fibers,  which are typically glass,
ceramic,  plastic,  or metallic materials.  The EPA has also
contacted the vendor referenced by commenter IV-D-41 as having
60 fiber-bed units installed nationwide.  Again, it appears as if
the term fiber bed is being used to describe what is actually a
composite mesh-pad system.  The vendor confirmed with EPA that
they were discussing the installation of composite mesh-pad
systems with commenter IV-D-41, not fiber-bed systems.
     One other commenter also indicated that additional
facilities were using FEME to control chromium emissions from
hard chromium electroplating and chromium anodizing tanks
 (IV-D-10).  In addition, all four commenters stated that FEME
should be considered as the basis for MACT for new sources.
Based on  information gathered by EPA since proposal, a total of
five facilities are known to be using FEME to control chromium
emissions from affected hard chromium electroplating and chromium
anodizing tanks.  These five facilities represent different sizes
of hard chromium  electroplating and chromium anodizing
operations.
     Emission test data were obtained from four of the five
facilities using  FEME  (see Item No. IV-B-01  of  Docket A-88-02).
However,  after a  thorough evaluation, it  was determined that none
of the emission test data available were  adequate to  establish  an
emission  limit based on the use of fiber-bed units.
     The  test results  from one facility were incomplete to assess
 the performance of  fiber-bed units.  The  test  results from the
 other  facilities  were  adequate to evaluate  the performance of
 FEME;  however, the  limited data are not sufficient  to establish
 an emission limit which must be met on  a  continuous  long-term
 basis  based on the  use of FEME.   In one case,  at both the inlet
 and  outlet,  only  one  sample  port  was  installed.  At the  inlet,
 only a single point was  sampled.  At  the  outlet, only one
'traverse was made,  when two  traverses  should have been performed.
..'Based  on EPA's previous  experience  in sampling at electroplating
 facilities,  sampling  a single point  or a  single traverse is not
                                2-27

-------
 acceptable for providing accurate and representative emissions
 data.
      In order to use emissions test results for the purposes  of
 establishing emission limits,  the Agency also requires that the
 quantity of emissions captured by the sample catch exceed the
 detection limit of the analytical method by a factor of five.
 This requirements ensures that the measured data are accurate to
 within 10 percent.  Data from the remaining two facilities fall
 below this guideline.   Therefore,  the data must be treated as
 qualitative rather than quantitative results;  they may not be
 used to establish achievable emission limits.   Based on this
 qualitative assessment,  it appears that  FEME offer excellent
 control potential.
      In evaluating control technologies,  the Agency also must
 consider the sustainability of any performance level.   As
 discussed above,  the Agency does  not have data to establish
 emission limits associated with the FEME  technology.   In
 addition,  EPA is concerned with the long-term performance of
 these  systems because  of the tendency of  the fiber beds to plug.
 In other contexts,  most  vendors of FEME systems do not recommend
 their  use as primary pollution control systems.   Rather,  they
 recommend that coarse  prefiltering be provided upstream of the
 fiber  beds to prevent  plugging.   The prefiltering devices range
 from a series of mesh  pads to  a complete  packed-bed scrubber
 unit.   At  present,  there are no long-term data available to
 assess any actual deterioration or operational problems
 associated with FEME.  Fiber-bed mist eliminators to  control
 chromium electroplating  and anodizing tanks  have  only  recently
 been installed as a result of  local  air district  requirements;
 therefore,  it is  unlikely that  any long-term data are  available.
     Because  of  the uncertainties  in both the  measured FEME
performance data  and the potential  long-term variability of the
 system performance,  the  Administrator cannot at this time
determine  a more  stringent  emission  limit  could be achieved based
on the application  of  FEME  technology for  new  hard chromium
plating or chromium anodizing operations.  Therefore,  the MACT
                               2-28

-------
standard is 0.015 mg/dscm for new hard chromium electroplating
operations and is 0.01 mg/dscm for new chromium anodizing
operations.  [See discussion of emission limits in Section 2.8.]
As set forth in the proposal, these emission limits were based on
the performance level which could be achieved by composite
mesh-pad systems and fume suppressants, respectively.  These
emission limits correspond to an overall emission reduction of
99.8 percent for composite mesh-pad systems and 99.5 percent for
fume suppressants.  Based on available data, the Administrator
has been unable to conclusively determine that the performance
achieved in practice by the FEME is superior, considering
long-term performance, to that of either the composite mesh-pad
system or fume suppressants.  Therefore, the final MACT
performance level of new hard chromium electroplating and
chromium anodizing operations is -unchanged from the proposal.
However, the limited data do suggest that FEME systems can
achieve the emission limits established for composite mesh-pad
systems and fume suppressants.  Because this standard is a
performance standard, the use of a specific technology is not
mandatory; therefore, either system  (or any which meets or
exceeds this performance level) may be used for control of
emissions  from new hard chromium electroplating and  chromium
anodizing  operations.
     In order to  facilitate  the use of FEME to achieve compliance
with the standard, monitoring provisions have been  included in
the final  rule for use with  FEME.  The pressure drop across the
fiber-bed  unit is identified as the compliance monitoring
parameter  because of  the relationship between pressure drop and
the performance  of the unit.   Performance  of the  fiber-bed  unit
will decrease as  a result  of plugging  of the bed  if  the water
spray  system  is  not  activated  in  response  to an increase  in the
pressure  drop.   Therefore,  it  is  required  that the  pressure drop
be monitored  and recorded  once per day in  the  final  rule.   Any
.deviation outside the established parameter range would
constitute a  violation of  the  standard.  The  test methods in the
                               2-29

-------
 proposed rule are suitable for demonstrating compliance with the
 standard,  regardless of the control technology employed.
      Comment;  Two commenters (IV-D-07,  IV-D-43)  disagreed with
 the methodology used by the EPA in calculating the floor.   One
 commenter (IV-D-43)  pointed out that the calculation should be a
 straight arithmetic average,  not a median,  and account  for the
 five percent of the facilities using the composite mesh-pad
 system.   As  such,  the MACT floor would be 0.023 mg/dscm as
 presented in the preamble to the proposed rule.  The other
 commenter (IV-D-07)  also noted that in determining the  floor,  the
 EPA can  only exclude sources achieving low emission levels due to
 very recent  BACT and LAER determinations.   The EPA,  according  to
 this commenter,  may not exclude other sources  achieving these
 emission levels from the floor determination without positive
 information  that these levels came about because  of BACT or LAER
 determinations.   The commenter points to section  112(d) as the
 source of  this  information.
      Response;   The  EPA believes that the use  of  a median  for
 determining  the average emission limitation is warranted,  as
 discussed  in Section 2.1.
 2.5.2  Regulatory Alternatives Considered
      Comment;   Eight commenters  (IV-F-01 [Altmayer],  IV-D-12,
 IV-D-17, IV-D-21, IV-D-48,  IV-D-49,  IV-D-53, IV-D-58) suggested
 that  the EPA was too limiting in the  regulatory alternatives for
 hard  chromium electroplating  operations.  These commenters  said
 that  the EPA should  have accounted for sources  (especially  small
 sources) in  this subcategory  that  could  use  fume  suppressants  to
 comply with  the  standard,  instead  of  locking sources  into a
 control technology such as a  packed-bed  scrubber.   Four
 commenters (IV-D-17,  IV-D-48,  IV-D-49, IV-D-58) proposed that  the
 EPA allow new and existing hard  chromium electroplating
 operations the option of meeting the  same surface  tension limit
allowed for  decorative  chromium  electroplating  operations using a
wetting agent-type fume suppressant.  One of these commenters
 (IV-D-17) contended  that the  EPA did not present sufficient
evidence to  support  that fume  suppressants in hard chromium tanks
                               2-30

-------
are less e_ffactive than those in decorative chromium tanks.  Two
commenters  (IV-D-48, IV-D-58) further argued that there are fume
suppressants available that do not cause pitting in hard chromium
electroplating tanks (brochure provided); one of these fume
suppressants works by forming a foam blanket only near the anode
and cathode.  According to one commenter  (IV-D-48), the South
Coast Air Quality Management District  (SCAQMD) has approved the
use of such a fume suppressant in conjunction with poly balls for
controlling chromium emissions from hard chromium electroplating
tanks.
     Response:  The EPA has selected an emission limit format in
order to provide sources with the flexibility to choose the
emission control strategy best suited  to their facility.  The
regulation  only requires that any strategy selected meet the
emission limits set out in the rule.   As such, hard chromium
electroplating sources can use fume suppressants to achieve
compliance  with the standard, as long  as compliance testing and
ongoing compliance monitoring demonstrates that the emission
limit required by the standard is being achieved.
     The Agency allows decorative chromium electroplating and
chromium anodizing sources to achieve  a surface tension limit in
lieu  of conducting a performance test  if  the  criteria  of
§  63.343(b)(2) or  (3) are met.  This provision was  included in
the standard  because the EPA recognized that  some decorative
chromium electroplating and  chromium anodizing tanks do not have
ventilation stacks such that a source  test could be performed.
This  is not the case with hard chromium electroplating tanks.
Given the  quantity of chromium emitted from a hard  chromium
electroplating tank, it is the Agency's experience  that all such
tanks have ventilation systems to meet Occupational Safety and
Health Administration  (OSHA) requirements.  Thus,  testing  to
demonstrate initial  compliance with the standard  should not be  an
implementation problem for facilities  with hard  chromium
•electroplating tanks that wish to use  fume suppressants.
      The EPA  has  revised  the rule  to  clarify  that  monitoring
requirements  are  a  function  of the  control technique used,  and
                               2-31

-------
 are not source category-specific.   Thus,  the rule requires all
 sources using fume suppressants with wetting agents,  or foam
 blankets to monitor the surface tension or foam thickness,
 respectively,  to demonstrate continuous compliance with the
 standards regardless of the source category.   The value of the
 monitoring parameter used for ongoing compliance determinations
 is either (1)  specified in the final rule;  e.g.,  a source  may
 accept  a minimum foam blanket thickness of 1 inch as  compliant;
 or (2)  set during the initial compliance test;  e.g.,  if a  source
 using a foam blanket chooses to set the compliant foam thickness
 value.   The monitoring frequency,  and recordkeeping and reporting
 associated with monitoring,  that is identified in the proposed
 rule is also control technique-specific.
      Comment:   One commenter (IV-D-24)  suggested that the  section
 of the  proposed rule that allows alternate control technologies
 (§ 63.343(d))  also specify the timeframe within which the  EPA
 would approve the alternate technology.
      Response;   The installation of alternative control
 technologies do not have  to be approved under the final rule.  It
 is only necessary for the monitoring parameter that will be used
 to ensure continual compliance with the standard be approved, or
 if an alternative test method is used to demonstrate  initial
 compliance,  the alternative test method must  first be approved by
 the governing  agency.   As stated in section 63.343(d)  of the
 proposed (and  final)  rule,  sources  using air  pollution control
 devices  not  listed in the section are required to submit,  among
 other things,  verification of performance and appropriate
 operating parameters  to be monitored to establish compliance with
 the standards.   The testing procedures  used for verification
 would likely be consistent with those laid  out  in § 63.344(c) of
 the final rule.   If not,  the alternate  test procedures  must  be
 submitted for  approval  in accordance with § 63.7 (f) of  the
 General  Provisions.   Section 63.7(c)(3)  specifies  that  the
Administrator  will  approve or deny  the  alternate  test procedures
within 30 days  of  receipt  of the test plan  and  any supplementary
 information  requested by  the Administrator.   If silternate
                               2-32

-------
monitoring requirements are requested, these alternate
requirements  should be requested  in accordance with
.§  63.343(c)(8)  of  this rule and § 63 . 8 (f) (4) (i)  of subpart A.
;-The Administrator, as stated  in § 63.8 (f) (5) (i) ,  will approve  or
deny  the alternate monitoring methods within 30  days of  receipt
of the  alternate plan and any supplementary information
requested.
      Comment;   Several commenters responded to  the EPA's
solicitation  for comment  on the feasibility of  limiting  or
prohibiting wastewater discharges from sources  covered by the
proposed  rule.  Two  commenters  (IV-D-40,  IV-D-43) stated that
wastewater recycling should be  incorporated into the  rule.
However,  21  commenters did not  think that wastewater  should be
regulated.   Thirteen commenters   (IV-D-06, IV-D-14,  IV-D-15,
IV-D-17,  IV-D-22,  IV-D-26, IV-D-31,  IV-D-37,  IV-D-47,  IV-D-52,
IV-D-58,  IV-D-59,  IV-D-61) noted  that recirculated scrubber water
and  washdown water cannot always  be returned to the
electroplating bath because  of  the potential for contamination.
This is especially a concern when multiple process tanks
performing different operations  (decorative chromium
 electroplating, chromium anodizing,  nonchromium operations, etc.)
 are  vented to the same control  system.  In lieu of recirculation,
 scrubber water is routed to an on site wastewater treatment plant
 prior to being discharged to the POTW.  One commenter (IV-D-29)
 pointed out that to alleviate contamination concerns,
 recirculated water would have to be treated prior to reuse, but
 that such a practice would itself generate a significant quantity
 of waste.  Two commenters (IV-D-14,  IV-D-37) noted that the mass
.loading of chromium in the scrubber water  is typically  low and
 can be sent to directly  to a POTW.  Other  comments questioning
 the EPA's rationale for  regulating wastewater pointed out that:
      1.  The economic incentive  to recycle already exists
r'(IV-F-01 [Altmayer] , IV-D-09, IV-D-23, IV-D-61);
      2.  There may not be capacity in the  process to accept
 additional water  volume  (IV-D-58);
                                2-33

-------
      3.  The use of a scrubber for its secondary evaporative
 benefits is questionable because any benefit would likely be
 offset by increased energy consumption (IV-D-26, IV-D-59);
      4.  The EPA may not have statutory authority to regulate
 wastewater or any source that does not cause air emissions
 (IV-D-17,  IV-D-18,  IV-D-23);  and
      5.  There is limited benefit to regulating a source that is
 already regulated by an effluent standard (IV-D-17).
 Finally,  three commenters were concerned that a rule limiting or
 prohibiting wastewater discharge may be too  narrowly defined.
 For example,  one commenter (IV-D-36)  pointed out that a  rule
 covering wastewater discharge would have to  include  more options
 than the two proposed by the  Agency because  there are other
 potential  uses for recycled water.   Another  commenter (IV-D-51)
 noted that it was not prudent to regulate wastewater because the
 EPA is not familiar with all  existing or future technologies,  or
 the larger matter of total waste minimization for an entire
 industry.   The third commenter (IV-D-61)  suggested that  the  EPA
 should wait until technology  that eliminates the discharge of
 wastewater becomes  available  and affordable  to industry  before
 requiring  industry  to comply  with wastewater standards that  may
 lead to lower-quality products or price increases.
      Response:   In  the preamble to  the  proposed rule,  the EPA
 solicited  comments  on a  possible standard in which sources would
be  required to recirculate scrubber water until  it reaches a
 certain chromium concentration;  then, the water could be
discharged.   An even more  stringent requirement  presented by the
EPA for comment  was  one  in which sources  would not be allowed to
discharge any wastewater resulting  from the  use  of composite
mesh-pad and  packed-bed  scrubber systems.  The benefit of such
requirements  is  that  the quantity of  chromium-containing
wastewater would be  reduced.
     The comments received at proposal  offered many technical
reasons for why neither  of  these requirements would be feasible
for many sources.  In light of  these  comments, the EPA has
reevaluated the alternatives associated with wastewater.   There
                               2-34

-------
appear to _be many instances in which recirculation of wastewater
is not technically feasible because it could introduce
contamination into the plating baths.  Also, there are some
facilities that have' indicated that they do not have capacity to
accommodate all recycled water.  In soliciting comments on the
wastewater requirement, the EPA assumed that the requirement
would be  technically feasible for sources and therefore performed
at minimal or no costs.  In light of comments received, the
technical feasibility of the wastewater requirements is
uncertain, and the EPA has therefore chosen not to include such a
requirement in the final rule.
     Even though not regulated by this NESHAP, wastewater sources
from chromium electroplating and chromium anodizing tanks are not
unregulated.  The Agency recognizes  the environmental benefits of
reducing  wastewater containing chromium.  The Office of Water is
currently revising effluent guidelines that apply to the chromium
electroplating and chromium anodizing sources.  The Office of
Water  is  currently investigating  "end of pipe"  limits -that apply
to  chromium emission sources, as well as  in-plant methods  to
limit  the quantity of wastewater  that can be discharged.   These
guidelines are scheduled to be proposed  in  two  separate phases,
with Phase I  to be proposed  in the  Fall  of  1994,  and promulgated
by  May 1996.   Phase  I will cover  facilities that  sell  products  in
the following market  sectors:  aerospace, aircraft,  electronic
equipment, hardware, mobile  industrial  equipment,  ordnance,  and
stationary  industrial  equipment.   Phase  II  is  scheduled to be
proposed by December 1997  and promulgated by December 1999 and
will cover  sources  that sell  products  in the following markets:
bus and truck,  household equipment,  instruments,  motor vehicles,
 office machines,  precious  and nonprecious metals and instruments,
 railroad, and shipbuilding.   After completion of these two
phases, all  sources in the source categories subject to this
 NESHAP will  be covered by a revised wastewater effluent
..guideline.
                                2-35

-------
 2.5.3   Selection of
      Comment ;   Four commenters (IV-D-13,  IV-D-24,  IV-D-31,
 IV-D-58)  had questions on the EPA' s cost  and economic impact
 analysis  that was considered in selecting MACT for hard chromium
 electroplating operations.   One commenter (IV-D-13)  stated  that
 the cost  to retrofit existing control equipment was overstated.
 Two commenters (IV-D-24,  IV-D-58)  thought that the EPA's analysis
 was flawed for several reasons.   First, these commenters thought
 that costs were understated because the EPA did not consider  the
 cost of shutdown for a facility to  permit,  install,  and test  an
 air pollution control device.   One  of these commenters (IV-D-24)
 estimated that their cost would be  greater than $100,000/lb of
 chromium  reduction,  much  greater than the $l,850/lb estimated by
 the EPA.   The other (IV-D-58)  noted that  the costs did not
 account for the fact that multiple  control devices would be
 necessary to meet the standards.  These commenters also felt  that
 the cost  analysis was flawed because it assumed that all forms of
 chromium  have  the same toxicity.  Thus, the EPA has overestimated
 health benefits and underestimated  the cost per pound of
 pollutant reduced.   Finally,  one commenter (IV-D-31)  questioned
 the EPA's statement  that  the majority of  cost increases resulting
 from additional emission  controls can be  passed on to customers .
 The commenter  wanted to know the basis of this statement, and the
 framework for  the conclusion that product prices will increase by
 less than 1  percent.
     Response :   The  EPA considers the cost  estimates presented in
 the preamble to the  proposed rule to be generally  representative
 of  industry  costs to achieve compliance with the rule.   The EPA
 recognizes that in an industry of over 5,000  sources,  there is
 going to  be  variation in  the cost to achieve  compliance with  the
 standard.  The  costs developed by the EPA are not  intended  to
 represent every facility  within  the  source  category,  but  are
 intended  to  represent  the source category in  general.   Whereas
 commenter IV-D-24  feels that  costs will be much higher  than those
 estimated by the  EPA,  there  are  likely other  sources  whose  costs
will be lower.
                               2-36

-------
     Regarding  the  comment  that  the EPA underestimated  costs
because  it did  not  account  for the need for multiple  control
..devices,  the Agency did  account  for the fact  that medium and
large-sized facilities have more than  one  tank and would
therefore require more than one  control device per facility.  The
EPA assumed that at such facilities, multiple tanks could be
manifolded to one control device, thus saving on capital costs,
but did  not assume  only  one control device would be needed  when
large numbers of tanks are  present.   (For  more details  on costing
methodology, refer  to Chapter 5  of the proposal BID and to  the
New Technology  Document.)   Regarding the comment that the EPA
overestimated retrofit costs,  these costs  were supplied by
vendors  to the  industry. The commenter  is a  supplier of
composite mesh  pads who  sells the pads to  vendors  that actually
install  control systems. Although his costs  to vendors may be
less than those presented in  the proposal  preamble,  end users
purchase equipment  from  vendors, not  the original  equipment
component supplier, and  will  therefore incur  those costs charged
by the vendors.
      To  clarify the cost estimates presented, these costs are
 installed costs and thus do account  for  the  cost  of installation.
Shutdown costs  (i.e., downtime)  were  not included because,  based
 on information received  by the EPA,  the  majority of shops with
multiple tanks  have several control  system configurations for
 controlling  the tanks,  and one system can be operated while the
 other system is being upgraded or replaced.   In addition,
 facilities  have the option of installing the equipment during a
 normal  shutdown of the plant.  Data gathered on the industry
 indicate that facilities typically shutdown twice a year for a
 one week period.  A problem with shutdown may exist for small
 shops or shops that operate 365 days per year and have only one
 control system.  However, measures such as increased production
 prior to shutdown  could be used to minimize production losses.
..Although the costs of performance testing are not included as
 part of equipment  expenditures, the effect of these costs was
 considered by the  EPA in selecting the level of the standards.
                                2-37

-------
Also,  in response  to  commenters  that  stated  that  the EPA
overestimated  environmental benefits,  the EPA estimated the costs
per  quantity of hexavalent chromium removed  in developing impacts
for  the standard.  This was done because all chromium emitted
from chromic acid  electroplating and  anodizing tanks was assumed
to be hexavalent chromium.
      The EPA's analysis of the economic impacts of the standard
indicated that final  product prices will increase by less than
1 percent.  Because it would not be feasible to analyze the
impact of the  regulation on all  products that are chrome plated,
the  EPA's Economic Impact Analysis  (EIA) examined the impacts on
five different products.  Of these  five, three of the products
are  hard chromium  electroplated:  automobile parts, industrial
rolls, and hydraulic  cylinders.  For  each product analyzed, the
EIA  provided estimates of the original plating cost and both
the  percentage increase and the  actual dollar increase in
electroplating cost.  Then, if possible, the EIA  used the
increased cost of  electroplating, along with an estimated price
of the final product, to estimate a price increase for the final
product.  Using this  methodology, the EIA arrived at the
conclusion that the final product price would not exceed
1 percent for any  of  the products analyzed.
      The percentage increase in  electroplating costs estimated in
the  EIA did exceed one percent in some instances,,  However, the
EIA  indicated that these cost increases could be  passed on to the
consumer for the following reasons.   The primary  reason is
because the demand for most electroplated products is a derived
demand.  In other  words, the parts  that are electroplated are
intermediate goods in the production  of the final product.  As a.
result, the estimated increase in electroplating  cost is such a
small  percentage of the total cost  of producing the final product
that  the producer  of  the final product could raise the final
product price minimally (less than  one percent)  to recover the
cost  increase associated with purchasing the electroplated parts.
Another factor that should make  it  easier for hard chromium
electroplaters to pass cost increases is that there are limited
                              2-38

-------
 substitutes available for hard chromium electroplated parts.
 Thus,  customers are  less able to  switch to a different product
:when  faced with higher prices for hard chromium electroplating.
 (For  more details  on the EIA, refer  to Chapter 8 of  the proposal
 BID and  to the New Technology Document.)
      Comment:  Six commenters  (IV-D-17, IV-D-24, IV-D-28,
 IV-D-31, IV-D-36,  IV-D-58)  stated that the proposed  standard  for
 large, existing hard chromium electroplaters is inappropriate.
 Four  commenters  (IV-D-17, IV-D-24, IV-D-28, IV-D-58) did  not
 think that sources that had installed packed-bed scrubbers  prior
 to proposal  of the rule should have  to upgrade to  composite mesh-
 pad systems.  The  commenters did  not believe that  such an upgrade
 was cost justified;  one  (IV-D-17) stated  that large  facilities
 would have at least  as much difficulty in obtaining  capital as
 small facilities.  One of these commenters (IV-D-28)  also pointed
 to the language  of section  112(d)(3) to justify  that requiring
 this  upgrade is  inconsistent with the Act;  sources with  packed-
 bed scrubbers meet the requirement  that existing sources be
 subject  to a control requirement  no  less  stringent than  the MACT
 floor.
      Other  commenters  (IV-D-31,  IV-D-36)  also agreed that the
 standard should be set  at  the  MACT  floor  because:   (1)  the
 improvement  in  air emissions over the MACT floor is  minimal but
 cost  and energy impacts  are not;  (2) the  data on the number of
 control systems and their effectiveness is from vendors hoping to
 mandate their systems;  and  (3)  a level of 0.013  mg/dscm is not
 consistently achievable with composite mesh-pad systems under the
 normal range of operating conditions.  Another commenter
 (IV-D-58)  agreed with this last point and cited test data to show
 that a level of 0.013 mg/dscm could only be met if composite mesh
 pads are preceded by process controls such as fume suppressants
 or another air pollution control device.   This same commenter did
 not believe that the data provided by the EPA supported the EPA's
^position that the standard  is based on the best performing
 12 percent of sources.  One commenter (IV-D-24)  noted that the
 selection rationale for MACT was inappropriate because the EPA
                                2-39

-------
 established MACT based on lab test data or vendor values  rather
 than actual control  values,  which is contrary to the  CAA's
 language  that the emission control level be "achieved in
 practice."   This sairie commenter also felt that the EPA had not
 sufficiently investigated the parameters (emission reductions
 achievable,  cost of  control,  and economic impacts)  on which  the
 selection of MACT was based.
      Response;   Section 112(d)(3)  of the Act requires the EPA to
 set  emission standards for existing sources that are  no less
 stringent than,  but  may be more stringent than,  the average
 emission  limitation  achieved by the best performing 12  percent of
 existing  sources.  The EPA interprets this passage to clearly
 indicate  that sources that are meeting this floor level can
 potentially be subject to more stringent standards requiring
 additional  controls.   This same section of the Act states that
 this  "floor"  emission limitation shall be set by the
 Administrator based  on the sources for which the Administrator
 has emissions data.   For this source category, the EPA surveyed
 both  industry and vendors to  determine the types and  approximate
 numbers of  control systems in use.   The EPA did  not,  however, use
 vendor data  to establish the  emission limit.   The proposed
 emission  limit of  0.013  mg/dscm was based on tests that the EPA
 conducted on  composite mesh-pad systems operating under normal
 process conditions.   Lower limits  than the one selected were
 measured  from these  devices,  but the EPA based the emission limit
 on the highest measured  data  point and believes  that  this limit
 is consistently  achievable.   (The  emission limit in the final
 rule  has  been revised to 0.015  mg/dscm.   See  discussion in
 Section 2.8,  Selection of the Emission Limits.)   The  data cited
 by commenter  IV-D-58  are based  on  an older mesh-pad technology
 that  was  tested  prior to the  new composite mesh-pad designs and
 could not be  used  in  setting  a  standard based on the  use of
 composite mesh-pad technology.
      In setting  an emission standard,  the  Act  further directs the
Administrator to take  into account  the  cost,  nonaiir quality
health and environmental  impacts,  and  energy  requirements in
                               2-40

-------
 establishing  these  standards.  To  fulfill this  requirement, the
 EPA evaluated the cost,  impact, and benefit  of  requiring  large
/hard chromium electroplating  sources with existing packed-bed
 scrubbers  to  achieve' a  level  of control  corresponding  to  the use
 of  a composite mesh-pad system.  Requiring all  sources to achieve
 such a level  results in additional emission  control  of
 approximately 1.2 mg/yr,  with additional energy requirements of
 31,850 megawatt  hours per year (MWh/yr)  and  an  additional
 38  cubic meters  (m3)  of solid waste.   (These numbers are
 nationwide estimates.)   The incremental  cost effectiveness of
 requiring  all sources to meet a standard based  on composite mesh-
 pad systems compared to allowing sources with packed-bed
 scrubbers  to continue to use these devices  is approximately
 $4.2 million per Mg of  chromium controlled.   Although  this cost
 may seem high, the  EPA  believes the  toxicity of chromium
 justifies  these  costs.   The cost analysis  for composite mesh-pad
 technology accounted for the fact  that all  composite mesh-pad
 designs use either  a packed bed or other coarser mesh  pad prior
 to  the composite mesh pad to reduce  loading and prevent plugging.
 All composite mesh-pad  system are  single units  that combine both
 control strategies; thus, the cost analysis was performed on the
 whole unit.
      Based on the  EPA's economic  analysis,  the requirement that
 large, hard chromium electroplaters  meet a standard based on
 composite  mesh-pad technology would  not cause adverse  economic
 effects on large facilities that  currently use packed-bed
 scrubbers.  Due to economies of  scale, the relative impacts
 (e.g., the percentage increase in electroplating cost) on larger
 .facilities are consistently smaller that those on small
 facilities.  As a result, it will be easier for larger facilities
 to pass on control costs.  Financial institutions should
 recognize this when making decisions concerning whether to grant
 icapital to a larger chromium electroplating facility.   In
 ^addition,  the capital availability analysis that was  done for
 small electroplaters indicated that as  the revenue  of a facility
 increases, the ability of the facility  to obtain capital also
                               2-41

-------
 increases..  Because larger facilities tend to generate more
 revenue,  their ability to obtain the necessary capital should be
 enhanced.   Finally,  a larger facility may have more options
 available to raise the necessary capital (e.g.,  internal
 financing).   (See Chapter 5 of the New Technology Document  for a
 detailed  discussion of EPA's economic analysis for these
 systems.)
      It is  also important to keep in mind that the final  rule is
 expressed in terms of an emission limit,  not  a technology.  As
 discussed in the preamble to the proposed rule,  the EPA tested a
 system in which a foam blanket was used on a  hard chromium
 electroplating tank to reduce emissions going to the packed-bed
 scrubber.   Emissions from the outlet of the scrubber were
 measured  as  0.008 mg/dscm,  which would achieve compliance with
 the level of 0.015 mg/dscm required by the final rule for hard
 chromium  electroplating tanks.   Also,  a source may retrofit an
 existing packed-bed scrubber to add a composite mesh-pad
 component,  in lieu of replacing the system with a new composite
 mesh-pad  system.
      Comment:   Five  commenters  (IV-F-01 [Sonntag],  IV-D-24,
 IV-D-27,  IV-D-53,  IV-D-58)  stated that the proposed standard  for
 small, existing hard chromium electroplaters  is  too stringent.
 Three  commenters  (IV-F-01 [Sonntag] ,  IV-D-24,  IV--D-58)  disagreed
 that  small existing  sources  could meet the standeird even  if they
 were  equipped with packed-bed scrubbers.   These  commenters
 questioned the rationale  for basing the standard on ideal
 conditions,  especially when  the EPA's  own data indicate that
 control equipment not operating under  these conditions  would  not
 meet  the standard.   Another  commenter  (IV-D-53)  rioted that  the
 regulation required  even  very small sources to install  a  control
 device, which they believed  was an undue  burden.   In lieu of  such
 a requirement,  this  commenter proposed that the  EPA allow small
 sources to use  an approved mist suppressant.
     Response;  The  EPA did  test  some  packed-bed scrubber systems
 that were not  achieving the  level  of 0.03 mg/dscm required  by the
proposed standard.   However,  these devices  were  not  optimized to
                               2-42

-------
achieve the higher removal efficiencies.  Specifically, when
scrubbers were operated with periodic or continuous washdown in
which fresh water was supplied as makeup to the top of the bed, a
limit of at least 0/03 mg/dscm was achieved.  The EPA also notes
that the selection of the emission limit for packed-bed scrubbers
was based on the highest emission level obtained during any of
the tests.  The final rule requires the use of fresh water added
to the top of the packed bed whenever.makeup additions occur as
one of the work practice standards identified in Table 2 of the
final rule.  Thus, packed-bed scrubbers that are operated in
accordance with the requirements of the rule should be able to
achieve a limit of 0.03 mg/dscm.
     As mentioned in a previous response, the emission limit of
0.03 mg/dscm for small, hard chromium electroplaters does not
require these sources to install any particular type of control
equipment.  A fume, or mist, suppressant can be used if it meets
the 0.03 mg/dscm emission limit, as could any other control
technique that meets the limit.
     In addition, the Agency has considered potential  adverse
impacts to small sources by 'requiring a less stringent standard
for small, hard chromium electroplaters than the larger ones,  and
by reducing the monitoring, reporting,  and  recordkeeping
requirements for all sources.   (See Sections 2.11  and  2.13,
respectively.)
     Comment:  Five  commenters  (IV-F-01 [Schott],  IV-D-07,
IV-D-13,  IV-D-30, IV-D-39) stated  that  the  proposed standard  for
small, existing hard chromium electroplaters was too  lenient.
One commenter  (IV-D-07) noted that the  proper  interpretation  of
the MACT  floor is a  straight average, and  as such  is
0.023  mg/dscm (see discussion in Section 2.5.1).   Therefore,  the
EPA could not legally establish an emission limit  of
0.03 mg/dscm; the appropriate standard  is  0.013  mg/dscm (more
Astringent than the floor  level  of  0.023 mg/dscm).
     Two  other commenters  (IV-F-01 [Schott],  IV-D-39)  indicated
that the  risk remaining after the proposed MACT was  applied to
small  sources warranted more  stringent  controls, and that the
                                2-43

-------
 residual risk analysis to be conducted 8 years after the standard
 would necessitate a more stringent standard.   Two commenters
 (IV-D-13,  IV-D-30)  thought that the EPA's rationale for selecting
 MACT for small,  hard chromium electroplaters  was flawed from both
 a technical  feasibility and cost basis.   One  of these commenters
 (IV-D-30)  stated that packed-bed scrubbers are an inappropriate
 design improperly evolving from vapor/liquid  mass transfer  and
 are not good controls for mist elimination.   The other commenter
 (IV-D-13)  stated that the EPA's cost-effectiveness rationale is
 flawed because it overstates the cost  of retrofitting packed-bed
 scrubbers, as well  as the efficiency of  these devices;  the
 efficiency is actually 97 percent,  not 99 percent.
      Response;   As  discussed in Section  2.1,  NESHAP Decision
 Process,  the Agency believes that the  MACT floor is properly
 based on the median level of control (use of  packed-bed
 scrubbers) for the  hard chromium electroplating source  category.
      The Agency  considers any discussion of the residual  risk
 from small,  hard chromium electroplaters  to be premature  at this
 time.   Under the Act,  standards are first set based on  control
 technology;  risk is  considered as a factor when deciding  to
 regulate area sources  with a MACT or a GACT standard.   Following
 the  establishment of  technology-based  standards,  the  Act  has
 provisions to set risk-based standards if the residual  risk
 remaining after  implementation of the  standard is  determined  to
 be significant.   In developing the  technology-based standard  for
 sources  that  emit hexavalent  chromium  (a  known csircinogen)  and
 trivalent chromium, the EPA regulated  sources  with  MACT to  ensure
 that a  residual  risk analysis  will  be  conducted.  Currently,  the
 EPA has  insufficient information  to determine  if the  residual
 risk remaining after the  implementation of the  standard will
warrant development of a  risk-based standard.
     The EPA also considers  its cost analysis  for small, hard
 chromium electroplaters to be  sound.  As  stated in a  response to
a previous comment, the EPA based its  retrofit  costs  on
information from vendors who supply  the equipment to  the
industry.  Commenter IV-D-13 may be  correct in stating  that
                               2-44

-------
vendors receive equipment for a lower cost than indicated in the
proposal preamble, but the costs incurred by affected sources are
the costs charged by vendors.
     The EPA disagrees with the assertion that packed-bed
scrubbers are not good controls for mist elimination.  While
composite mesh-pad systems may be superior technology, packed-bed
scrubbers have been tested by the EPA and can consistently
achieve emission limits of 0.03 mg/dscm, which represents a
99 percent reduction from uncontrolled emissions in a typical
situation.  The EPA recognizes the shortcomings in expressing the
control efficiency as an indication of a technology's efficiency,
which is why percent reduction was not chosen as the format of
the standard.  The control efficiency of any particular device
varies according to the inlet loading such that any one device
may only achieve an efficiency of 97 percent or less; however,  if
the scrubber is properly operated and maintained, the emission
level of 0.03 mg/dscm can still be achieved at the outlet.
     Comment:  Two commenters  (IV-D-41, IV-D-61) asked for
clarification on various aspects of the rule.  One commenter
 (IV-D-41) wanted to know if MA.CT was at least equivalent  to BACT
established by the California Air Resources Board  (GARB), which
requires 95 percent control  (or a maximum emission rate of
0.15 mg/amp-hr) for small facilities; a 99.8 percent  efficiency
 (or a maximum emission rate  of 0.006 mg/amp-hr) for  sources
emitting greater than 10 Ib/yr with controls; and a  99 percent
control  (or a maximum emission rate of  0.03 mg/amp-hr) for all
other sources.  This commenter felt that  if the proposed  standard
was not equivalent to 99 percent  control, it  could not be MACT.
This same  commenter noted that the proposed standard does not
meet health  risk-based ambient guidelines.  According to  this
commenter, a hard  chromium  electroplating source  emitting
0.03 mg/dscm of hexavalent  chromium exceeds New York's ambient
"-guideline  concentration  of  2 x 10"5 ug/m3 by  two  orders  of
^magnitude.   Finally, this commenter,requested clarification  on
the  control  technology  that forms  the  basis  for MACT.  This
 commenter  questioned whether the  proposed standard  of
                               2-45

-------
 0.013  mg/dscm for hard chromium electroplating represents the use
 of mist suppressant controls plus scrubbers  with high-efficiency
 mist eliminators,  or can it be achieved using a four-stage
 high-efficiency fiber mist pad air cleaning  system.   A second
 commenter (IV-D-61)  noted that the proposed  rule was  unclear  as
 to whether the air pollution control  device  must operate  24 hr/d,
 or only during tank operation.
     Response:   The EPA cannot state  whether the final NESHAP for
 this source category is equivalent to the  GARB standards.  The
 format of those standards is mg/amp-hr,  which is not  directly
 comparable to the  concentration-based format of the this  rule.
 (See discussion in Section 2.7,  Selection  of the Format of the
 Standard.)   A source that operates at a high amperage  may achieve
 a  limit expressed  in terms of mg/amp-hr, yet may be emitting  high
 levels of chromium in terms of the concentration (mg/dscm).
     The basis  for MACT for small and large  hard chromium
 electroplaters  is  packed-bed scrubbers and composite mesh-pad
 systems,  respectively.   When typical  inlet conditions  are
 assumed,  packed-bed scrubbers can control  99 percent of emissions
 from hard chromium electroplating sources  while composite mesh-
 pad  systems  can control up to 99.8 percent.   However,  as  noted
 above,   control  efficiencies can vary  from  system to system.   Also
 as noted above,  even though MACT  was  based on the use  of  specific
 controls,  any control  technique can be used  to meet the standard
 as long as  it is demonstrated through performance testing.
     In terms of meeting health risk-based standeirds,  MACT
 standards  developed  in  accordance with section 112(d)  of  the  Act
 are  technology-based, not  risk-based.   The EPA considered  risk
 for  this  source  category by regulating all area sources covered
by these  standards with MACT.  Thus,  an analysis  will  be
 conducted  in accordance with section  112 (f)  to  determine the
 residual risk from these sources.
     In response to  the second commenter,  the  final rule has been
clarified to require control  of chromium emissions only during
tank operation; tank operation has also been defined.
                               2-46

-------
     Comment:  One commenter  (IV-D-39) suggested that the
proposed rule should specify ventilation system design to prevent
excess fugitive emissions or dilution, or the EPA should issue
guidance concerning this to the States.
     Response;  Based on EPA's experience, hard chromium
electroplating sources have to adhere to strict OSHA standards to
protect workers at these facilities.  Hard chromium
electroplating tanks must be properly ventilated in order to meet
OSHA and,  as such, the EPA does not think it is necessary to
include ventilation guidelines in  the MACT standards as well.
The  OSHA requirements specify that the exposure level to workers
not  exceed 0.05 mg/m3 over a normal 8-hour workday and a 40-hour
work week  for water soluble hexavalent chromium.  As explained  in
the  preamble to the proposed  rule, one concern with a
concentration standard is that dilution  of the exhaust gases  can
occur.  Possible  dilution of  the exhaust can be checked by
regulators by comparing  the air  flow  measured during testing  with
the  design air  flow specified for  the system in the  source's
construction or operating permit,  or  in  manufacturer's
specifications.   Such a  circumvention of the standard  is also
expressly  prohibited  by  §  63.4(b)  of  the General  Provisions;
dilution of the air stream would cause a source  owner  or operator
to be  subject  to  enforcement  action.
2.6   SELECTION  OF MACT  FOR DECORATIVE 'CHROMIUM ELECTROPLATING AND
      CHROMIUM ANODIZING  TANKS
2.6.1   Decorative Chromium Electroplating--Trivalent Chromium
        Electroplating Process
      Comments:   Two commenters (IV-D-10, IV-D-43)  stated that the
preamble to the proposed rule should have a more detailed
 discussion of how the MACT floor for this subcategory was
 determined.
      Eleven commenters (IV-F-01 [Altmayer,  Sonntag],  IV-D-01,
 IV-D-02,  IV-D-03, IV-D-22,  IV-D-25,  IV-D-53,  IV-D-58,  IV-D-59)
• .questioned why the EPA was including those decorative chromium
 tanks that use a trivalent chromium process in the proposed rule;
 these commenters do not believe this process should be regulated.
                               2-47

-------
      Many_of the commenters (IV-D-01,  IV-D-03,  IV-D-25,  IV-D-53,
 IV-D-58,  IV-D-59)  felt that the EPA's  decision to regulate
 triyalent chromium based on its risk was flawed,  and that the EPA
 had insufficient data to reach this conclusion.   Two commenters
 (IV-F-01  [Altmayer],  iv-D-58)  pointed  out that trivalent chromium
 baths meet OSHA requirements without ventilation.   One of these
 commenters (IV-D-58)  provided an epidemiological  study discussing
 the toxicity of chromium.   This same commenter stated that the
 toxicity  of trivalent and hexavalent chromium should be
 considered separately when addressing  health effects and
 determining whether to regulate all sources  under MACT.   Four
 commenters  (IV-F-Ol [Altmayer],  IV-D-02,  IV-D-22,  IV-D-58)
 questioned why  the EPA once considered requiring  trivalent
 chromium  baths  for new sources and are now regulating them
 equally with hexavalent chromium baths.   Finally,,  three
 commenters  (IV-D-01,  IV-D-03,  IV-D-53)  stated that use of the
 trivalent chromium process should be encouraged by the EPA,  not
 discouraged through regulation,  because  trivalent  processes
 result in less  total  chromium in process  wastewater and  less
 sludge generation.
      Four commenters  (IV-F-01  [Altmayer],  IV-D-22,  IV-D-25,
 IV-D-58)  found  fault  with  the  data supporting the  EPA's  reasoning
 for regulating  trivalent chromium baths.   The commenters noted
 that  the  level  of  hexavalent  chromium  identified by the  EPA  from
 a trivalent  chromium  bath  (0.004  mg/dscm)  would require  that
 hexavalent  chromium be  present  in the  bath at a level  of
 8 percent.   The  commenters  point  out that  the presence of only a
 few parts per million (ppm) of hexavalent  chromium in  a  trivalent
bath will destroy  the bath; this  suggests  that the  presence  of
hexavalent  chromium was due to analytical  integrity.
     There were  ten comments  (IV-F-01  [Bortoli, Sonntag],
IV-D-01,  IV-D-02,  IV-D-03,  IV-D-22, IV-D-25,   IV-D-26,  IV-D-58,
IV-D-59)  concerning specific requirements  of  the proposed
regulation.  These commenters pointed  out  that the  reporting,
recordkeeping, and monitoring requirements eliminate the
incentive for facilities to switch to a trivalent chromium
                               2-48

-------
electroplating process; according to the preamble,  the "no
action" alternative proposed by the EPA will result in
$1.6 million in reporting and recordkeeping.  One commenter
(IV-D-22)  stated that the EPA recognizes that adding just wetting
agents alone to a trivalent chromium bath can jeopardize the
operation; this commenter questioned why the monitoring and
reporting is necessary.  Another commenter  (IV-D-59) questioned
how this burden can be justified in light of the fact that
sources using a trivalent chromium electroplating process
comprise only 10 percent of the decorative chromium
electroplating subcategory, and they already have significant
regulatory burden due to Clean Water Act requirements.  Two
commenters  (IV-D-01, IV-D-03) noted that facilities using the
trivalent chromium electroplating process are already using fume
suppressants and should therefore not be regulated further.  One
commenter (IV-F-01  [Sonntag]) suggested regulating trivalent
chromium electroplating processes under GACT to eliminate some, of
the burden.
     One commenter  (IV-D-02) asked for clarification  on the
proposed regulation.  This  commenter asked whether sources using
the trivalent chromium electroplating process have to meet both
the compliance monitoring and the operation and maintenance
requirements in the proposed rule if they employ only fume
suppressants and not an air pollution control device.  The same
commenter also stated  that  although the EPA had previously
indicated that this was not the  intent, the proposed  rule is
written such that sources have to meet both the surface tension
limit  and an emission  standard.
     Twelve commenters responded to the EPA's request for comment
on whether  the trivalent chromium electroplating process should
be required for new sources.  Two commenters  (IV-D-01,  IV-D-03)
thought that this would be  reasonable in  situations where the
process would be technically feasible.  However, the  remaining
-ten  commenters  (IV-F-01  [Sonntag],  IV-D-09,  IV-D-10,  IV-D-22,
IV-D-26,  IV-D-29,  IV-D-40,  IV-D-43, IV-D-57,  IV-D-59) did not
think  that  this should be  a requirement.   Four  commenters
                               2-49

-------
 (IV-D-26, .IV-D-29,  IV-D-57,  IV-D-59)  did not think that the
 process was technically feasible for the full range of decorative
 chromium electroplating.   One of these commenters pointed out
 that decorative chromium platers using a trivalent chromium bath
 need stringent process controls and possible proprietary
 processes to achieve a finish comparable to a chromic  acid bath.
 Three commenters (IV-D-22,  IV-D-26,  IV-D-29)  stated that
 requiring a trivalent chromium electroplating process  for new
 sources would force manufacturing overseas,  and that this problem
 is  accentuated by a narrow source definition in which  each tank
 is  considered a source.   Two commenters (IV-D-10,  IV-D-59)
 pointed out inconsistencies  in the EPA's reasoning;  the EPA can
 only require trivalent chromium baths if it recognizes the
 difference  in toxicity between hexavalent and trivalent chromium.
 In  the  preamble to  the proposed rule,  the EPA discusses the
 adverse environmental impacts of trivalent chromium, but then
 suggests requiring  this process for new sources.   Finally,  one
 commenter  (IV-D-40)  suggested that it is not necessary to require
 the  trivalent chromium process for new sources as  long as other
 requirements are met,  and the source  demonstrates  that the
 residual risk is negligible.
      Response:   The  EPA has  reconsidered the requirements
 associated  with regulation of decorative chromium  electroplating
 tanks using a trivalent chromium electroplating process.
 Specifically,  the EPA has examined the  technical basis for
 including these tanks  in  the  rule  and the technical  feasibility
 of requiring such a  process  for new sources.
     During development of the proposed  standards,  the EPA
 evaluated the  trivalent chromium electroplating process  and
 considered  requiring this process  for all  new  decorative  chromium
 electroplating tanks.  The trivalent  chromium  electroplating
process  is  considered  to be a pollution prevention alternative.
Chromic  acid is  not  present in the plating solution  in the
trivalent chromium processes,  and hexavalent chromium  is  regarded
as a bath contaminant  in these processes.  In  addition, all of
the trivalent  chromium plating  solutions with which EPA  is
                               2-50

-------
familiar contain a wetting agent as an inherent bath component.
That is, the wetting agent is part of the plating solution
purchased from the vendor; it is not added separately by the end
user.
     With a trivalent chromium electroplating process, the
potential emissions of chromium in any form are much lower
because the concentration of total chromium in trivalent chromium
baths is approximately four times lower than the total chromium
concentration in chromic acid baths.  Trivalent chromium  .
processes greatly reduce emissions of the most potent form of
chromium  (hexavalent), and significantly lower emissions of
chromium in other forms.  In addition to reduction of air
emissions, the use of trivalent chromium processes results in
lower chromium concentrations in process wastewaters and,
consequently, reduces the amount of  sludge generated.  Based on a
source  test conducted by  the EPA,'hexavalent chromium emissions
were present at a concentration of 0.004 mg/dscm, and total
chromium emissions were present at an average  concentration of
0.023 mg/dscm.  Total chromium  emissions  from  a  trivalent
chromium bath are approximately 99 percent less  than those  from a
traditional, uncontrolled decorative hexavalent  chromium bath.
Hexavalent  chromium emissions are  approximately  equivalent  to
those  emitted  from  a hexavalent chromium bath  using a wetting
agent.
     Although  chromium  emissions  from the trivalent chromium
process were low,  the EPA had not anticipated  the presence  of
hexavalent  chromium in  emissions  from the trivalent
 electroplating process  nor the  level of total  chromium emissions.
.Given that  the Act  lists all forms of chromium on the HAP list,
 the EPA considered the  trivalent chromium electroplating process
 as a source of chromium emissions as well as an emission control
 alternative for the chromic acid electroplating process.  Based
 on the emission test results,  a decorative hexavalent chromium
 bath controlled by adding a wetting.agent had equivalent
 hexavalent chromium emissions and less total chromium emissions
 than a trivalent chromium plating bath.  (As previously stated,
                                2-51

-------

 for trivalent chromium baths, the wetting agent is inherent to
 the solution; it does not need to be added by the user.)   In
 addition,  the trivalent chromium process may not be technically
 feasible for all decorative chromium applications.  Therefore,
 the final  rule does not require the use of a trivalent chromium
 electroplating process for either existing or new decorative
 chromium electroplating tanks.
      Commenters indicated that  the levels of hexavalent chromium
 emissions  identified in the EPA test results must be erroneous
 because levels such as these would have contaminated the
 trivalent  chromium plating bath.   Therefore,  it is not clear
 whether the  EPA data accurately reflect emissions from the
 trivalent  chromium electroplating process or if the analytical
 integrity  of the data is suspect.   In addition,  no additional
 control measures beyond the use of the wetting agent in the bath
 were  identified for trivalent chromium processes.   Therefore,  in
 light of the ambiguity of the air emissions  data from trivalent
 chromium process and the other  environmental  benefits-of  the
 trivalent  chromium process (less  wastewater  and sludge
 generated),  the EPA has  decided to regulate  these baths
 differently  from hexavalent chromium electroplating baths.   The
 EPA has revised the final rule  so  that users  of  trivalent
 chromium baths  are required only  to submit initial  notifications
 certifying that a trivalent chromium bath is  being  used, and to
 keep  records  of bath chemicals  purchased.  Subsequent
 notifications are required only if  the process  is  changed,  or if
 a new trivalent chromium process is  introduced.
      The final  rule does  make a distinction between types of
 trivalent chromium electroplating baths.  The minimal
 notification and recordkeeping  requirements apply only to those
 trivalent chromium baths  that incorporate a wetting  agent.  The
 EPA has evaluated baths with this characteristic and found  them
 to have the environmental benefits discussed above.  However, the
EPA cannot be certain that  a trivalent  chromium bath will not be
developed that does not contain a wetting agent as a bath
ingredient.  Therefore, the final rule  regulates trivalent
                              2-52

-------
chromium baths that do not incorporate a wetting agent in the
same manner as decorative chromium baths using a chromic acid
solution.  The EPA believes that this will discourage the use of
a trivalent chromium bath that does not have a wetting agent as
an inherent bath component.
2.6.2  Decorative Chromium Electroplaters--Chromic Acid
       Electroplating Process
     Comment:   Three commenters questioned the EPA's rationale
for setting the MACT floor for this subcategory.  Two commenters
(IV-D-10, IV-D-43) thought more rationale was required to explain
why the use of fume suppressants in conjunction with a packed-bed
scrubber was not a more stringent control option than fume
suppressants alone, and thus, the MACT floor for new sources.
Another commenter  (IV-D-50) stated that the EPA's standard
development process ignored the 12 percent criteria set out in
section 112(d)(2) of the Act in setting a standard for
2,800 decorative chromium electroplaters.
     Response:  The EPA received surveys from 63 decorative
chromium electroplating operations, which had a total tank
population of  102 decorative chromium electroplating tanks.  .In
.addition, 11 decorative chromium electroplating operations were
visited.  Section 112(d)(3) instructs the EPA to establish the
minimum emission limitation  (i.e., the MACT floor) based on the
average emission limitation achieved by the best performing
12 percent of  the existing sources for which the Administrator
has emissions  information  (emphasis added).  The EPA recognizes
that the decorative chromium electroplating source category  is
large, but  is  directed to use the information available in
establishing the MACT floor for the source category.
     Through analysis of this background  information, the EPA
concluded that approximately 15 percent of decorative chromium
electroplating operations were uncontrolled, 40 percent use  fume
suppressants,  40 percent use fume suppressants  in conjunction
with packed-bed  scrubbers, and 5-percent  use only packed-bed
scrubbers.  Source testing was conducted  to establish the
effectiveness  of  the  fume  suppressants used by  sources  in this
                               2-53

-------
 source category.   Wetting agent-type fume suppressants were found
 to result in outlet concentrations ranging from 0.001 mg/dscm to
 0.003 mg/dscm.   Foam blanket fume suppressants were found to
 result in outlet  concentrations ranging from 0.003  mg/dscm to
 0.007 mg/dscm.  Although tests were not conducted on sources
 using packed-bed  scrubbers in conjunction with fume suppressants,
 the EPA does not  think that in this application the packed-bed
 scrubbers provide any significant additional environmental
 benefit.   When  packed-bed scrubbers are installed on decorative
 chromium electroplating tanks,  they are typically used as
 evaporators  to  reduce the wastewater treatment burden.   Rinse
 waters (contaminated with chromium)  are recirculated through the
 scrubber until  their volume is reduced.   The inlet  concentration
 to a scrubber used for a tank that is already controlled with
 fume suppressants would be very small.   Likewise, the particle
 size of  any  remaining pollutants is small.   These factors combine
 to make  any  additional contaminant removal  in the scrubber
 difficult.   There are insufficient data available to quantify any
 improvement  in  overall emissions;  therefore,  the  EPA based the
 MACT floor on the use of fume suppressants  alone.
      Comment:   Two commenters asked for clarification of the
 proposed  standard.   One commenter (IV-D-41)  noted that  the
 standard  does not meet health risk-based ambient  guidelines.   For
 example,  a source emitting 0.003  mg/dscm of hexavalent  chromium
 exceeds New  York's ambient guideline concentration  of 2  x 10"5
     *3
 ug/m by  an  order of magnitude.   This  same  commenter requested an
 explanation  on  how the proposed rule compares  to  BACT established
 by GARB,  which  requires a  95  percent control  efficiency  or a
 0.15 mg/amp-hr  emission limit.   The  second  commenter (IV-D-45)
 suggested that  §  63.342(b)-(d)  be  clarified;  as it  is  currently
 written,  the rule assumes  that  either an air pollution control
 device or fume  suppressant will be used,  and uncontrolled tanks
 can remain uncontrolled.
     Response:  As  explained  in a previous  response  in
 Section 2.5,   Selection of MACT  for Hard  Chromium Electroplating
Operations, MACT  standards developed in  accordance with
                               2-54

-------
section 112(d) of the Act are technology-based, not risk-based.
The EPA considered risk for this source category by regulating
•all area sources with MACT standards.  Thus, an analysis will be
'conducted in accordance with section 112 (f) to determine the
residual risk that remains from area sources.  If considerable
risk is found to remain, the source category can be subjected to
residual risk standards.
     The EPA cannot state whether the.final NESHAP for this
source category is equivalent to the BACT  standards established
by GARB.  The format of the CARS standards is mg/amp-hr, which is
not directly comparable to the concentration-based format  of this
rule.  A source that operates at a high  amperage may  achieve a
limit expressed in terms of mg/amp-hr, yet may be emitting high
levels of chromium in  terms of the concentration  (mg/dscm).  The
basis for MACT for decorative chromium electroplating tanks  is
the use of  fume suppressants.  Data  indicate that when these fume
suppressants are used  according  to manufacturers recommendations,
control efficiencies of greater  than 99.5  percent are-achieved.
     Sections 63.342(c) and  (d)  of the final rule express
standards in  terms of  an emission  limit  and a  surface tension
limit.  The rule has been  clarified  to state that the emission
limit applies to all sources, regardless of the  control  method
used  (if any).  Sources using wetting agent-type  fume
suppressants  to meet the  standard  are allowed  to  demonstrate
initial  compliance by  measuring  surface  tension if  the  criteria
of  §  63.343(b)(2) are  met.   This compliance method  has  been
allowed  because  it  is  the  EPA's  experience that sources
performing  decorative  chromium electroplating  and chromium
anodizing may not have ventilation stacks to the atmosphere,
making  source testing  impractical.
      Comment;   Two  commenters suggested changes to the proposed
 rule.   One  commenter  (IV-F-01 [Sonntag]) proposed that decorative
 chromium electroplaters using either a chromic acid or trivalent
•chromium bath be regulated by GAGT.   Another commenter  (IV-D-41)
 suggested that  the  rule contain specific ventilation design
 criteria because the efficiency of mist suppressants is highly
                               2-55

-------
 variable.  	This commenter proposed that the rule require
 high-efficiency slot exhaust designed in accordance with the
 specifications  in the American Conference of Governmental
 Industrial Hygienists industrial ventilation manual,  with the  use
 of high efficiency air cleaning equipment and high efficiency
 de-misters capable of 99  percent control.
      Response:   The EPA does not believe that it is appropriate
 to regulate new or existing area sources performing decorative
 chromium electroplating operations with GACT due to the  toxicity
 of chromium compounds and the availability of control techniques.
 (See  discussion in Section 2.4.)   As  discussed in Section 2.6.1,
 regulation of sources using the trivalent chromium electroplating
 process has been revised  such that the regulatory burden for
 sources using a trivalent chromium bath that incorporates a
 wetting agent has been substantially  reduced.   However,  as
 discussed  in Section 2.4,  regulation  by MACT is appropriate  given
 the toxicity of chromium  compounds in general.   Also,  the
 technologies that form the basis  for  MACT are widely available
 and will not cause undue  burden on any source.
     Regarding  the suggestion that ventilation specifications  be
 added to the rule,  the EPA believes that this is unnecessary
 because sources covered by this standard are also covered by OSHA
 guidelines.  If a fume suppressant, or any control  technique,  is
 not properly operated such that workers  are being exposed to
 chromic acid fumes,  the source  is  in  violation of OSHA.   In
 addition,  the EPA has  established  monitoring techniques  that
 would ensure that the  chromium  electroplating or anodizing
 processes  are operated properly at all  times  with appropriate
 fume suppressants.   Therefore,  it  is  the EPA's  opinion that  to
 specify ventilation  requirements in this rule  would be redundant.
     Comment;   Three  comments pertained  specificailly  to  the
proposed emission limit of  0.003 mg/dscm.   One  commenter
 (IV-D-58) proposed that sources that  choose  to  use an air
pollution  control device  in lieu of fume suppressants be allowed
to meet a  limit of 0.03 mg/dscm.   Two commenters  (IV-D-50,
IV-D-58) did not  think  that a source using  either a fume
                               2-56

-------
suppressant or a. fume suppressant in conjunction with a
packed-bed scrubber could consistently meet a limit of
0.003 mg/dscm.  One of these commenters  (IV-D-50) supplied test
results to support this claim.  One commenter  (IV-D-09) requested
further justification for the 0.003 mg/dscm proposed standard;
this commenter noted that the EPA had previously proposed a level
of  0.01 mg/dscm.
     Response:  As a result of comments  received at proposal, the
EPA has reconsidered the emission limit  of 0.003 mg/dscm for
decorative chromium electroplating and chromium anodizing tanks.
As  stated in  the preamble to the proposed rule, this emission
limit was based on tests of a decorative chromium electroplating
                                                                ^
tank in which a combination wetting agent/foam blanket was used
to  control emissions.  Tests had also been conducted on a
decorative chromium electroplating tank  using  only a foam blanket
for control.  The analytical data for all types of fume
suppressants  ranged from 0.001 to 0.007  mg/dscm, with  the wetting
agent/foam blanket data ranging  from 0.001 to  0.003 mg/dscm  and
the foam blanket data  ranging  from 0.003 to  0.007 mg/dscm.    At
one time, the EPA had  considered (but had never proposed) an
emission limit  of 0.01 mg/dscm which was based on applying a
safety  factor to the highest measured emission limit
 (0.007  mg/dscm)  in  the data base.  After further  consideration,
the EPA did  not include the  foam blanket data  when  setting  the
proposed emission limit because  it was  thought that  these data
may indicate that  foam blankets  are  less effective  than wetting
agents  in  reducing  chromium emissions.
      In evaluating  whether the proposed emission limit of
•0.003 mg/dscm should  be revised  in  the  final rule,  the EPA
 reassessed the  effect  the  test methods  may have had on the
 emission data obtained.  The analytical method used for the fume
                                                       /
 suppressant  test was  colorimetric  spectroscopy.  As more
,: efficient  control  technologies (such as composite mesh-pad
 systems)  were developed,  a more-sensitive analytical method was
 needed to measure the lower concentrations of chromium being
 emitted.   Therefore,  the  more sensitive ion chromatography method
                               2-57

-------
 was  used  in the  later phases  of  emission  testing  for  these
 standards involving  add-on control  devices.   By using the less
 sensitive colorimetric analytical method,  it  is unclear whether
 the  variation  found  between the  two types  of  fume suppressants
 was  due to a performance  difference in  the fume suppressants or
 was  an artifact  of the analytical method used.  The fact that
 there is  overlap between  the  foam blanket  and wetting agent/foam
 blanket data (that is,  one foam  blanket test  resulted in the same
 emission  limit as one of  the  wetting agent/foam blanket tests
 [0.003 mg/dscm]) further  indicates  that this  could be the case.
 Therefore,  the EPA concluded  that the emission  limit  in the final
 rule should be based on the performance of both foam  blankets and
 wetting agents.  Accordingly,  the emission limit  selected for
 decorative chromium  electroplating  and  chromium anodizing tanks
 in the final rule is 0.01 mg/dscm.   This emission limit was
 selected  by applying a safety factor to the highest measured data
 point (0.007 mg/dscm)  to  account for variation  in the sampling
 and  analytical procedures.  The  selection  of  this emission limit
 is consistent  with the methodology  used to select emission limits
 based on  other control  techniques.   (See discussion in
 Section 2.8, Selection of Emission  Limits.)
 2.6.3  Chromium  Anodizing Operations
     Comment:  Three  commenters (IV-D-14, IV-D-47,  IV-D-58)
 questioned the MACT  floor established by the  EPA  for  sources
 performing chromium  anodizing.   Two commenters  (IV-D-47, IV-D-58)
 stated that  it did not  appear that  the  EPA had  sufficient data to
 perform a MACT floor analysis  for these sources,  and  that the
 final rule should not  include a  MACT standard for chromium
 anodizing sources unless  additional  data are  gathered or higher
 emission  rates are allowed.   The third  commenter  (IV-D-14) noted
 that the  emission limit for the  top  12  percent  is  surprisingly
 low and should be reevaluated.   Alternatively,  this commenter
 suggested that anodizing  tanks in the aerospace industry be
 considered a separate  subcategory because  fume  suppressants are
not always feasible.   Another commenter (IV-G-01)  also stated
 that chromium anodizers,  as well as  decorative  chromium
                               2-58

-------
 electroplaters,  that  cannot  use  fume  suppressants  should be
 considered separately.   This commenter states  that the MACT  floor
 for such sources should be based on packed-bed scrubbers.
      Response;   The TVIACT floor for existing chromium anodizing
 sources  was based on  information available  to  EPA  on the source
 category.   Information on the industry was  obtained through
 survey questionnaires to both industry representatives and
 control  system vendors,  site visit reports,  and available
 emission data.   Although information  was not available from  all
 sources  in the category, EPA believes the information was
 sufficient to satisfy the requirements of section  112(d)(3)  of
 the Act.  The industry survey, which  included  some aerospace
 facilities, indicated that fume  suppressants were  the control
 technique predominantly used in  the industry.   Data based on
 tests at decorative chromium electroplating tanks  were used  to
 establish the efficiency of  fume suppressants; in  that sense,
 data was transferred from one source  category  to another. As
 stated in the previous response, the  emission  limit associated
 with fume suppressants has been  raised from 0.003  mg/dscm to
 0.01 mg/dscm.  No technical  reason was provided by industry, nor
 is one known by the EPA, for creating a separate subcategory of
 sources  for which fume suppressants are not technically feasible.
 Thus, all new and existing sources performing  chromium anodizing
 must meet an emission limit of 0.01 mg/dscm (or maintain the
 surface tension specified in the rule).
      Comment:  Six commenters (IV-D-14, IV-D-17, IV-D-36,
 IV-D-44, IV-D-47, IV-D-58)  did not think that  the  standard for
 chromium anodizing tanks was achievable in all situations,
 especially when an add-on control device was used in lieu of fume
 suppressants.  Two commenters (IV-D-47, IV-D-58) cited  test data
 to support this claim.  One commenter stated that  tests of
.sources in the SCAQMD indicate that a level of 0.003 mg/dscm is
-not consistently achieved by using fume suppressants  (IV-D-58).
 -Test data were provided by commenter IV-D-47 for sources
 performing chromium anodizing and using add-on controls.  One
 commenter  (IV-D-58) suggested that the standard for chromium
                               2-59

-------
anodizing tanks  controlled with add-on  control devices be set at
0.03 mg/dscm.  Otherwise, this commenter believes that sources
will have to use an air pollution  control device followed by a
state-of-the-art fiber-bed mist eliminator.
     Response:   The MACT  floor for existing and new chromium
anodizing sources  is based on the  use of fume suppressants.  The
EPA selected this  technology as the basis for MACT for all
existing and new sources.  The emission limit associated with the
use of fume suppressants  has been  established as 0.01 mg/dscm in
the final rule.  Section  112(d)(3)  of the Act prohibits the EPA
from establishing  a standard that  is any less stringent than the
MACT floor for a category or subcategory of sources.  The EPA has
explored the possibility  that sources performing decorative
chromium electroplating and chromium anodizing operations that
cannot use fume  suppressants may constitute a separate
subcategory.  Thus, a separate MACT floor would be established
for the subcategory, and  MACT selected  based on ci cost-benefit
analysis of selecting a standard more stringent than the floor
for this subcategory.  However, as  discussed in Section 2.6.2 for
decorative chromium electroplaters, the EPA could not establish
that a distinct  subcategory based  on the technical infeasibility
of fume suppressants existed within the chromium anodizing source
category.  The EPA was informed by  commenters thait fume
suppressants were  not feasible for  all  operations, but
information was  not provided on the technical recisons for this.
For instance, two  commenters alluded to the fact that fume
suppressants may not be feasible for adhesive bonding
applications because corrosion protection failure* may result from
the change in bath chemistry.  However, in a follow-up letter,
one of these commenters conceded that other factors may have
caused corrosion failures.  Due to  the  lack of data, the EPA
could not conclusively state that there is a separate subcategory
for which fume suppressants are not technically feasible.
     The data submitted by commenter IV-D-47 were evaluated by
EPA to determine the performance level  of scrubber systems
controlling emissions from chromium anodizing tanks.  Post-
                              2-60

-------
control emissions from the four scrubbers for which there were
data range between 0.0026 and 0.033 mg/dscm.  One of the scrubber
systems met the proposed limit of 0.003 .mg/dscm; two of the four
scrubber systems would meet the emission limit of 0.01 mg/dscm in
the final rule.  The operation and maintenance practices
associated with the systems tested are not known.  Implementation
of the work practice standards in § 63.342(f) may result in all
systems achieving a level of 0.01 mg/ciscm.
     The EPA believes that the revised emission limit of
0.01 mg/dscm in the final rule makes it easier for sources to
comply using add-on control technology.  Also, commenters
indicated that they would like to use fume  suppressants but would
need time to test the feasibility of the fume suppressants for
their application, and in the case of government contractors,
obtain permission for their use.  The EPA believes that the
extended compliance time of 2 years for the chromium anodizing
source category will allow sources more time to explore the
possibility of using fume suppressants.  Also,  this timeframe
will allow sources that prefer to use air pollution control
devices time to optimize their systems.
     Comment:  There were four comments related to the  efficiency
of  fume suppressants in  chromium anodizing  tanks.  Three
commenters  (IV-D-01, IV-D-03, IV-D-43)  stated that allowing  fume
suppressants for  the control  of  large  chromium  anodizing  tanks
should be  reevaluated; some tanks  are very  large and may  require
more control.  One of  these commenters  supplied source  test  data
to  support this  statement.  Another commenter  (IV-D-47)  indicated
that the  EPA had overstated the  control efficiency of  fume
suppressants;  the commenter felt that  the control  efficiency was
closer to 95 percent.  Based  on  an uncontrolled emission rate of
2.59 x 10~4  gr/dscf, a control  efficiency of 99.5  percent would
be  needed to meet the  proposed standard.
     Response;   The  EPA  did not  propose a standard expressed in
terms  of  control efficiency  for the very reason cited by
commenters;  the efficiency of fume suppressants vary according to
 the electroplating or  anodizing tank operation.  Based on source
                               2-61

-------
 tests conducted by the EPA,  fume suppressants used on "typical"
 decorative chromium electroplating tanks  in accordance with
 manufacturers specifications can reduce over 99.5  percent  of
 emissions  when compared to an uncontrolled tank.   The outlet
 concentrations measured during these  tests ranged  between
 0.001 and  0.007 mg/dscm.   If the uncontrolled tank had lower
 emissions,  the same controlled outlet concentration could  result
 but  the overall efficiency would be lower.   Thus,  this rule
 requires sources to measure an outlet concentration,  not a
 control efficiency.
      The emission limit for chromium  anodizing tanks  was based on
 tests of decorative chromium electroplating tanks  in  which fume
 suppressants  were used.   However,  the EPA believes that the final
 emission limit of 0.01 mg/dscm is appropriate for  both decorative
 chromium electroplating and chromium  anodizing operations.  In
 response to the commenters that were  concerned that the use of
 fume  suppressants was  not sufficiently stringent for  large
 chromium anodizing tanks,  data available  to the EPA indicate that
 an uncontrolled chromium  anodizing tank has total  chromium
 emissions  typically ranging  from 0.54 to  1.6  mg/dscm,  with an
 average  of 0.97 mg/dscm.   These uncontrolled  emission rates are
 similar  to those seen  for decorative  chromium electroplating (see
 Chapter  3  of  the Proposal BID).   Therefore, the EPA believes that
 fume  suppressants  will be equally effective for both  chromium
 anodizing and decorative  chromium electroplating tanks.
      Comment;   One commenter  (IV-D-14)  stated that  sources  should
 have  the option of using  reducing  agents  in wet scrubbers  to
 convert hexavalent chromium to  trivalent  chromium.
      Response:   Both the  proposed  and final rules  regulate
 chromium compounds, not just hexavalent chromium.  Therefore,
using a reducing agent to change  the  form of  the chromium  from
hexavalent to  trivalent is not  allowed by the  final rule.
     Comment;   There were  two comments related  to the  impact of
the proposed regulation.   One commenter (IV-D-14) stated that in
the impact analysis the EPA erroneously assumed that all sources
can use fume suppressants  to meet  the standard.  Because this is
                               2-62

-------
not the case, the analysis should include costs for emission
control equipment.  Another commenter (IV-D-47) was concerned
that the EPA had underestimated the impact of the rule on small
sources performing chromium anodizing to support the aerospace
industry.  To minimize the impact, this commenter suggested a
reduction in monitoring and recordkeeping for small sources.
     Response;  In calculating impacts for the proposed rule, the
EPA did assume that all sources performing decorative chromium
electroplating and chromium anodizing operations would use fume
suppressants to comply with the rule.  The EPA still thinks that
the vast majority of sources will opt to use fume suppressants;
all of the commenters that stated that they could not use fume
suppressants had a control device in place.  The industry did not
supply information on the number and types of  sources that could
not use  fume suppressants.  The EPA thinks that there is only a
small portion of the industry that would be unable  to use fume
suppressants.  Also, the extended compliance time may allow
sources  that thought they could not use fume suppressants time to
test additional fume suppressants that may be  viable.  For these
reasons, the EPA believes that capital expenditures for  the
decorative chromium electroplating and chromium anodizing source
categories will be minimal.
     Of  course, all source categories will experience costs
associated with testing, monitoring, reporting, and record-
keeping.  The EPA has  revised these  requirements  to minimize the
burden on area sources.   (See discussion  of  Selection of
Monitoring and Reporting  and Recordkeeping Requirements  in
Sections 2.11 and 2.13,  respectively.)
2.7  SELECTION OF THE  FORMAT OF  THE  STANDARD
     Comment:  Seven  commenters  (IV-D-07,  IV-D-10,  IV-D-41,
IV-D-42, IV-D-47,  IV-D-50,  IV-D-53)  stated that  the format  of the
standard should be  expressed as  milligrams  of  chromium emitted
per amp-hour of  operation (mg/amp-hr),  not  mg/dscm. According to
•-•the commenters,  concentration-based standards  are flawed because
 they can be  circumvented by dilution (IV-D-10, IV-D-41,  IV-D-53);
 concentration can vary from system to system depending on the
                               2-63

-------
 design flow rate (IV-D-io,  IV-D-41);  and source test data
 indicate that  outlet concentrations  do vary widely over various
 inlet conditions (IV-D-07).
      Five commenters" (IV-D-10,  IV-D-14,  IV-D-41,  IV-D-47,
 IV-D-53)  pointed out that California's rule is  expressed
 as  mg/amp-hr and requested that EPA  maintain consistency with the
 California rule.  According to  one commenter (IV-D-10),  this
 request is reasonable because,  in general,  process emission rate
 standards result in lower emissions  than the proposed concen-
 tration-based  standards.   Another commenter (IV-D-53)  provided
 calculations to  demonstrate  that sources that meet the
 concentration  limit will  also meet California's process  emission
 rate.   Two commenters (IV-D-10,  IV-D-14)  were concerned  that  some
 well-controlled  sources that currently meet the process  emission
 rate  standard  would have  to  install  additional  controls  to meet
 the concentration limit.   Finally, one commenter (IV-D-41)
 pointed out  that  the California standards address  the problem of
 residual  risk, and  the EPA should therefore relate their standard
 to  the  California standard.
      Two  commenters (IV-D-10, IV-D-50)  stated that emissions
 should  be correlated to production rates because chromium
 emissions increase  proportionately with increased  current;
 whereas ventilation air flow is  not  related to  electroplating
 bath  operation.
      One  commenter  (IV-D-42)  stated  that a  mg/amp-hr format is
 the most  appropriate because that  is  the one permitting  agencies
 will  use.  Another  commenter (IV-D-53)  stated that a process
 emission  rate  is  readily  enforceable  and can be verified without
 requiring a  source  test;  the commenter did  not  expand on how this
 would be  possible.
     Two  commenters  (IV-D-10, IV-D-42)  suggested that, at a
minimum,  the final  rule should recognize  process emission rates
 as acceptable,  and  identify  acceptable alternatives  to avoid an
 equivalency  evaluation.
     Response;   The majority of  commenters  felt that  the format
 of the standard should be changed  from a  concentration format to
                              2-64

-------
a process emission rate format.  However, available test data
indicate that a process emission rate format will not ensure that
the control level required by the standard will be achieved
consistently.  A comparison of the composite mesh-pad systems and
packed-bed scrubber data sets reveals that there is no overlap in
the concentration data sets; however, there is overlap in the
process emission rate data sets.  For example, the most stringent
limit  for existing hard chromium electroplaters is 0.015 mg/dscm
and is based on the use of composite mesh-pad systems.  The
concentration data on which this emission limit is based ranges
between 0.003 and 0.013 mg/dscm; whereas, the process emission
rates  for this same data set varies from 0.002 to
0.090  mg/Amp-hr.  The data set associated with the use of
packed-bed scrubbers with periodic or continuous washdown shows a
variation in concentration from  0.021 to 0.028 mg/dscm;  the
process emission rates  for the same data set vary from 0.068  to
0.219  mg/Amp-hr.  A concentration  format ensures that the
appropriate control technology is  employed because  the
concentration data sets are mutually  exclusive between the  two
technologies.  Therefore, a  concentration-based  format was
selected.  This  data  analysis  also refutes  one  commenter's
statement  that process  emission  rate  standards  result  in lower
emissions  than  the proposed  concentration-based standards.   As
evident  from  the data presented  above,  both packed-bed  scrubbers
and composite mesh-pad systems can achieve  similar  process
 emission rates;  however,  composite mesh-pad systems are  a
 superior technology and result in lower emissions  than packed-bed
 scrubber systems.   Because  of the data overlaps in the process
•emission rates,  the EPA also cannot establish in the final  rule
 process emission rate limits that would be considered equivalent
 to the concentration limits presently in the rule.
      Some commenters were concerned that a concentration-based
^standard would be easily circumvented through dilution of the
-.emission stream.  The EPA examined concerns over dilution of the
 exhaust gases as a means of circumventing the standards prior to
 selecting concentration as the format for the standard.   In
                               2-65

-------
 considering the effects of dilution, it was determined that the
 dilution of the gas stream at the outlet of the control system is
 more of a concern than any dilution taking place at the inlet to
 the control systems "because the control systems operate as
 relatively constant outlet devices.  (That is,  the concentration
 at the outlet is relatively independent of the inlet loading.)
 Dilution of the gas stream at the outlet of the control system
 can be addressed by a review of the test data arid permit data
 from a given facility.   The air flow rate measured during testing
 should approximate the design air flow rate for the control
 system reported on the permit application.   If  the two values
 differ significantly,  then an inspection of the control system
 can be made to determine if dilution air is being introduced.   it
 is also possible for a facility to dilute the inlet gas stream to
 the control device by designing a system to ventilate the
 electroplating tanks at air flow rates  substantially above those
 required for adequate ventilation.   However,  the increased costs
 associated with heating the room air in the electroplating shop,
 supplying additional makeup air,  oversizing the control system
 for the dilution air,  and the increased maintenance cost of such
 a  system would outweigh the costs of complying  with the standard
 without dilution.   In addition,  as  stated previously,  the control
 systems function as constant outlet control devices;  therefore,
 the introduction of dilution air at the inlet to the  control
 system would not substantially bias the outlet  emission level
 except in cases  of  large  volumes  of dilution air being added.
 Further,  §  63.4(b)  of the General Provisions  expressly prohibits
 dilution as  a means to  comply with  an emission  limit.   Violations
 of  this provision would cause the EPA to bring  an enforcement
 action against the  owner  or operator of  the source.  Therefore,
 concerns  of  dilution of the air stream  were not  considered  to be
 significant  enough  to warrant changing  the  format  of the
 standard.
     Other commenters stated that emissions should be  correlated
 to production  rates  because chromium emissions  increase
proportionately with increased  current;  whereas ventilation air
                               2-66

-------
flow is not related to electroplating bath operation.  The
.commenters are correct in stating that the production rate, as
measured by ampere-hours, is related to .emissions.  However, the
amount of current supplied to the tank is an indicator of the
amount of uncontrolled emissions from the tank, not the
controlled emission level from  the tank.  As shown previously,
the controlled mass emission rate is not directly related to the
amount of current supplied to the tank.  As stated in the
proposal preamble, the process'emission rate is based on the
outlet mass emission  rate and the current supplied to the plating
tank.  The outlet mass emission rate varies depending on the
chromium concentration and the  exhaust gas flow rate.  Because
                                                                ^
the outlet chromium concentration levels do not vary
significantly with the current  loading to the  plating tank,
facilities that  operate  at high current loadings would have a
much  lower process emission rate that facilities that operate at
low current loadings  even though both facilities have applied the
•same  control  technology  and may be  controlled  to the same
emissions level.  Conversely,  facilities that  have a low process
emission rate could have higher emissions than intended by  the
standards because of  a high inlet  loading  (high current loading)
to the  control  device.   Therefore,  the EPA does  not  consider  the
use of  process  emission  rates  as an appropriate format  for  the
standard.
      One commenter pointed  out  that because  the California
standards address the problem of  residual  risk,  the  standard
should be related to  the California standard.   As  mentioned in
previous comment responses, MACT  standards  are based on the
 control technologies  used in  the  industry  and risk is  not
 considered  in selecting the level  of the  standard.   However,  risk
may be considered in determining  whether or not an area source  is
 regulated with MACT or GACT.   Residual  risk impacts  are addressed
,,when  a MACT standard is reviewed as required by section 112 (f) .
      Other  comments  received suggest that  a mg/Amp-hr format is
 the  one permitting agencies will  use and that a process emission
 rate  is readily enforceable and can be verified without requiring
                                2-67

-------
 a source test.   Regardless of the format of the standard,  all
 sources subject to an emission limit would be required to  perform
 an initial source test to demonstrate compliance.   It is unclear
 how a process emission rate standard would eliminate the need for
 compliance testing.   Sources would need to demonstrate compliance
 with a standard expressed in terms of mg/Amp-hr;  the selection of
 the format of the standard does not affect the need for
 compliance testing.   Further,  in all source tests,  the
 concentration of emissions at the outlet is measured.   Other
 formats,  such as the process emission rate format,  build on these
 data.   Therefore,  no additional testing burden is  imposed  by the
 concentration format.
      Based on the data collected on this industry,  the EPA
 believes  that a concentration format is the most appropriate
 format for this standard.   Therefore,  no change to  the format of
 the  standard  has been made in the final rule.
      Comment:   Eight commenters (IV-D-10,  IV-D-14,  IV-D-24,
 IV-D-32,  IV-D-34,  IV-D-47,  IV-D-53,  IV-D-58)  disagreed with the
 EPA's  decision  to base the standard on emissions of total
 chromium  rather than on emissions of hexavalent chromium.   Five
 commenters  (IV-D-10,  IV-D-14,  IV-D-24,  IV-D-53, IV-D-58)
 contended that  there are limited data to support the toxicity of
 trivalent chromium,  and thus  to include it in the total  chromium
 number.   Two  other commenters  (IV-D-32,  IV-D-34) stated  that  data
 show that an  average of 30  percent  of  the  total chromium will  be
 trivalent chromium emissions.   Two  other commenters  (IV-D-47,
 IV-D-53)  agreed,  stating that  the rule  should  focus  on hexavalent
 chromium because hexavalent chromium emissions  are  lower than
 total  chromium  emissions, and  hexavalent  chromium is truly  the
 compound  of concern.   One commenter  (IV-D-53) pointed  out that
 the EPA Report  68-D-90155 gives  the  ratio  of hexavalent  chromium
 emissions to total chromium emissions as 85 percent.   One
 commenter  (IV-D-58)  explained  that hexavalent chromium emissions
are lower than  total chromium  emissions because hexavalent
chromic acid mist  is very reactive and unstable.
                               2-68

-------
     Two commenters suggested alternate compliance methods to
account for the difference between hexavalent and total chromium
'emissions.  One commenter  (IV-D-47) stated that sources should be
allowed to demonstrate compliance by testing for hexavalent
chromium in lieu of total chromium.  Another commenter  (IV-D-24)
suggested that if a source exceeds the standards based on its
total  chromium emissions, the source should then be allowed to
test for hexavalent chromium to demonstrate compliance.
     Response;  Several  commenters questioned the validity of
basing the standard on total chromium rather than hexavalent
chromium.  The EPA selected total chromium because the HAP list
identifies all chromium  compounds, not just hexavalent chromium
                                                                •»
compounds.  In addition, based on testing conducted by the EPA
for these source categories, there is little difference between
total  and hexavalent  chromium emitted from the processes.  The
compound emitted from these electroplating and anodizing baths  is
chromic acid, a hexavalent compound of chromium.  For tests where
samples were  analyzed for both hexavalent and total chromium, it
was shown that, considering the precision of the  sampling and
analytical methods used, the hexavalent  and total chromium levels
were essentially the  same  for chromic acid baths  (varying within
±10 percent in most  instances).   In some cases, the hexavalent
chromium amount was more than the  total  chromium  amount as a
result of the higher  sensitivity  of the  hexavalent  chromium
analytical  technique. Because  the EPA data base  is mainly
comprised of  data measured as hexavalent chromium,  the  final  rule
does allow  sources  to demonstrate  compliance with the total
chromium emission  limits through  the measurement  of either
hexavalent  or total  chromium  for  all sources.   The  final  rule
also  incorporates  acceptable  analytical  methods  for hexavalent
chromium  into Methods 306  and 306A.
      Comment:  One commenter  (IV-D-42)  suggested that the format
should be  flexible because States will  also be  requiring  sources
'to meet health risk-based  standards.
      Response;  A concentration-based  format  is flexible  and
 should not  interfere with  any health risk-based standards set by
                               2-69

-------
 States.  During  source tests,  the  concentration of  the  emission
 stream is  determined as the building  block  for  other  formats.
 Therefore,  the concentration at  the outlet  will be  known prior to
 calculating other  formats  such as  mass  rate or  process  emission
 rates.
     Comment:  One comtnenter (IV-D-16)  questioned how compliance
 would  be determined if tank utilization is  low;  that  is, would
 facilities  be penalized by inspectors suggesting that sources are
 diluting their airflow.
     Response: Initial compliance  tests must be conducted while
 the electroplating tanks are operated at or near maximum
 capacity.   If a.  source can meet  the standard under  maximum
 operating  conditions,  then the standard would continue  to be met
 under  normal or  low tank utilization  periods.   In addition, the
 ventilation rate for any given tank or  series of tanks  is
 determined  based on the dimensions and  surface  area of  the
 electroplating tank.   The  ventilation rate  remains  constant
 regardless  of the  number of parts  placed in the tank  or tank
 utilization.  Operating a  tank at  its ventilation rate  under low
 utilization periods  should not be  considered as  dilution by an
 inspector.
 2.8  SELECTION OF  THE  EMISSION LIMITS
     Comment;  Two commenters  (IV-D-24,  IV-D-50) stated that an
 insufficient number of production  operations were tested for the
 EPA to establish an emission limit.   One commenter  (IV-D-24)
pointed out  that out of  5,000  facilities, only  13 hard  chromium,
 3 decorative chromium,  and no  chromium  anodizing sources were
 tested.  This commenter suggests that additional tests  be
 conducted.   The  other  commenter  (IV-D-50) concurred that the
EPA's data base  is  insufficient to support  either the numeric
standards or the surface tension limits.  This  commenter
suggested that the proposed standard  be  issued as an  interim rule
until more data  are gathered and a justified standard set.  Two
commenters  (IV-D-11, IV-D-33)  stated  that the steindards are based
on false assumptions,  not  data.
                               2-70

-------
     Response;  The EPA believes that the data base of testing
information is suitable for setting the emission limits and
.surface tension limit in the final rule.  Section 112(d)(3) of
the Act directs the "Administrator to establish an emission
limitation using the emissions information available.  To
establish the level of control appropriate for the source
categories covered by this rule, the EPA surveyed over
120 operations that represented 215 hard chromium electroplating
tanks, 102 decorative chromium electroplating tanks,  and
31 chromium anodizing tanks.  Additional information  was obtained
from site visits to numerous hard and decorative chromium
electroplating and chromium anodizing facilities.  The EPA also
                                                                *
conducted performance tests at 10 facilities representing 13 hard
chromium and  3 decorative chromium electroplating tanks  to
establish the performance level of controls used in the  industry.
These  test results, not assumptions, were used to establish the
emission limits.  Finally, the rule  is  proposed for public
comment before it is finalized by the EPA and before  sources are
required to comply.  In response to  comments received during this
public comment period, the EPA has further  evaluated  the emission
limits, and revised them when appropriate as described  in
Section 2.6.2 and below.
     Comment;  Four commenters  (IV-F-01 [Schott],  IV-D-13,
IV-D-30, IV-D-50) representing three companies stated that  the
emission limit should be  changed for hard chromium plating
operations.   Three  commenters suggested changing  the  emission
limit  that  is based on the use of  composite mesh-pad  systems.
One  commenter (IV-F-01  [Schott]) stated that  the  control level
should be based  on  optimum design, as well  as  cost, energy
consumption,  and environmental  impact,  and  suggested  a  level  of
0.006  mg/dscm.   In  a  subsequent  letter  (IV-D-13),  this  same
commenter  stated that  the EPA did  not test  the best  systems
..-•available  and suggested  a level  of  0.009 mg/dscm,  which in the
-EPA's  data  base  is  the worst result  achieved by  the best system.
Another  commenter (IV-D-30)  agreed that the best  systems were  not
tested and that  the standard as  written penalizes companies that
                               2-71

-------
have  installed high-efficiency units.  Therefore,  this  coinmenter
suggested a level  of  0.001 mg/dscm to  correspond with the use of
composite mesh-pad systems.   Finally,  one commenter  (IV-D-50)
suggested that numeric  limits be based on 95 percent removal of
uncontrolled emissions.
      Response:  The EPA does  not think it is appropriate to lower
the emission limit that is based on the use of composite mesh-pad
systems.  The commenters did  not provide any data  to support
their claim that the  emission limit consistently achievable with
a composite mesh-pad  system is 0.001 mg/dscm, 0.006 mg/dscm, or
0.009  mg/dscm.  After meeting with the commenters  (IV-D-13,
IV-D-30), the EPA  agreed to contact a  State agency that the
commenters suggested  would have data to support such a  limit.
The data were requested, received,  and evaluated,  but did not
support a lower limit.
     As was discussed in Section 2.6.2, the emission limit for
all decorative chromium electroplating tanks and chromium
anodizing tanks at all  existing and new sources has been changed
to 0.01 mg/dscm in the  final  rule.  This limit is  based on
applying a safety  factor to the highest data point
(0.007 mg/dscm) in the  fume suppressant data base.  Likewise, as
described in the preamble to  the proposed rule, the emission
limit  that is based on  packed-bed  scrubbers is based on rounding
the highest value  (0.028 mg/dscm)  in the packed-bed scrubber data
base to 0.03 mg/dscm  to incorporate a  safety factor.  To be
consistent in setting the emission limit that is based on the use
of composite mesh-pad systems,  the emission limit  of
0.013 mg/dscm has  been  changed to  0.015 mg/dscm in the final
rule.  As with the other emission  limits, this emission limit is
based on applying  a safety factor  to the highest value
(0.013 mg/dscm) in the  data base of information on composite
mesh-pad systems to ensure that the limit is consistently
achievable.
     Regarding the  suggestion that  the numeric limit be based on
a 95 percent control  efficiency, for the reasons described in the
preamble to the proposed rule,  the  EPA does not believe that
                               2-72

-------
 using a percent efficiency format is the best method for
 evaluating the performance of control techniques for these source
 categories.  For the devices expected to be used by sources
 covered by these standards,  a relatively constant outlet
 concentration results.  Therefore,  percent reduction is not a
 good indicator of performance because it depends on the inlet
 conditions as well as the outlet conditions.  A numeric limit
 based on percent reduction could penalize facilities with low
 inlet loading, while facilities with high inlet loadings could
 meet the 95 percent efficiency but be emitting larger quantities
 of chromium.  For these reasons, it would also be impossible to
 select a numeric limit that is based on a 95 percent control
 efficiency.
 2.9  SELECTION OF DEFINITION OF SOURCE
      Comment;  Three commenters (IV-D-06, IV-D-17, IV-D-24)
 thought that the proposed source definition did not adequately
 address the situation in which multiple tanks are tied to a
 single control device.  One commenter (IV-D-06) thought that the
 manner in which this situation was addressed in the preamble to
 the proposed rule was confusing.  In a situation with multiple
 tanks tied to a common control device, this commenter asked if
 compliance would be achieved if emissions from the new tank in
 combination with emissions from existing tanks do not exceed the
 required emission levels at the exit of the emissions control
 system.  Two commenters (IV-D-17,  IV-D-24) suggested alleviating
 the potential confusion by selecting a broader definition of
 source for facilities controlling multiple tanks with one control
 device.  The source could be defined as the population of tanks
 connected to a single control system, or a group of tanks
 associated with a single electroplating system in a designated
 subcategory (i.e., each electroplating line).  Alternatively,
 when multiple tanks are tied to one control system, new source
. MACT could be triggered if greater than 50 percent of the tanks
 existing at the time of proposal have been replaced or added
 (IV-D-17).  One commenter (IV-D-07) thought the rule should
 clarify that.new tanks at existing sources are subject to new
                                2-73

-------
source MACT even if they are tied into an existing control
system.
     Five commenters stated that the source definition is too
narrow; three of these commenters (IV-D-22, IV-D-26, IV-D-59)
suggested that the source be defined as the plant.  Commenters
pointed out that a broad definition of source red.uces the
recordkeeping burden (IV-D-26), whereas the narrow definition
results in burdensome compliance testing and monitoring
(IV-D-57).  One commenter  (IV-G-01)  pointed out that a narrow
definition of source will subject a facility to the
preconstruction review requirements of § 63.5, even in the case
of a single tank being added at a major source.  One commenter
(IV-D-59) was also concerned that, should the EPA require the use
of trivalent chromium baths as new source MACT, the narrow
definition of source could result in an existing hexavalent
chromium bath having to be replaced by a trivalent chromium bath
once it has been consumed or contaminated.
     Response:  In the final rule, the affected source continues
to be defined as each chromium electroplating or chromium
anodizing tank.  If an additional tank is tied into an existing
control system and the reconstruction criteria are met, that new
tank would have to meet the requirements for new source MACT.  As
discussed in Section 2.12.3, the final rule contains procedures
for demonstrating compliance when multiple tanks are controlled
by one control system.  Alternately, a source may choose to
control the new or reconstructed tank separately.  The EPA
believes that identifying affected sources in terms of the
control strategy used to comply with the standard will make
enforcement of the rule difficult.
     The final rule also clarifies the reporting and
recordkeeping requirements for affected sources.  In general, the
reporting and recordkeeping requirements-in the final rule would
not vary according to the affected source definition.  Regarding
the preconstruction review requirements of § 63.5 of subpart A,
the final rule overrides these requirements in § 63.345.
Section 63.345 of the final rule requires notification of
                               2-74

-------
construction or reconstruction of an affected source (i.e., the
tank).   The EPA believes that this requirement is reasonable
because many tanks are controlled separately, and the permitting
authority will need "information on the control strategy.  If the
new or reconstructed tank is to be tied into an existing control
scheme, the requirements of § 63.345 should not be difficult to
fulfill.  Also, this section does not require approval for
construction or reconstruction.  These activities may begin once
a complete notification of construction or reconstruction  is
submitted.
     Cojsment:  Two  commenters  (IV-D-07, IV-D-28) suggested that
the discussion of definition of  source be clarified.  One
commenter (IV-D-07)  thought the  rule should  clarify that the
narrow new source definition applies to new  and modified area
sources as well  as  major  sources.   The  second commenter requested
an  explanation of  the control  requirements  if a facility has  one
tank  controlled with a packed-bed scrubber  and then adds another
tank    The commenter stated the new tank would obviously  require
a composite mesh-pad system for control and questioned whether
 controls for the existing tank would also have to be upgraded.
      Response:  Section 63.340(a)  of the final rule states that
 the affected source is each hard chromium electroplating,
 decorative chromium electroplating, and chromium anodizing tank.
 The affected source definition does not differentiate between new
 and existing sources, or between major and area sources;  the same
 affected source definition applies  to all sources.
       If an existing facility has one tank that  is  tied to a
 packed-bed scrubber and adds another tank to be controlled by a
 separate  control system, only the  new tank  would have  to  be
 controlled by a composite mesh-pad system  (or  some control
 technique that  achieves  an equivalent  control  level).  If the new
 tank  is  to be tied into  the existing control device,  the  new tank
 would still  need  to achieve the higher control level.  The
  compliance method to be  followed in this situation is  outlined in
  §  63.344(e)  of  the final rule.
                                 2-75

-------
  2.10  SELECTION OF COMPLIANCE DATES
       Comment;:   Several commenters stated that the proposed
  compliance timeframe of 1 year for hard chromium electroplating
  operations is  insufficient.   The following reasons were given:
       1.   It is insufficient  time to inform and educate
  responsible parties,  install equipment,  and train operators
  (IV-F-01  [Altmayer],  IV-D-58);
       2.   More  time is needed to account  for design,  bidding,
  installation,  and  testing (IV-D-17,  IV-D-21,  IV-D-27,  IV-D-44
  IV-D-47,  IV-D-52);
       3.   At  government  installations, more  time  is needed  to
 procure funding, submit a control plan,  award a  contract,
 install,  and test  (IV-D-37,  IV-G-01);
       4.   Sources need time to build up inventories to account for
 downtime  during retrofit  (IV-D-24);
      5.   Large sources will have large expenditures  for the
 equipment and personnel necessary to do  testing, reporting, and
 recordkeeping  (IV-D-57);
      6.  Budgeting, test planning, testing, and results
 evaluation will take more than 1 year (IV-D-16);  and
      7.  There  may not be a sufficient supply of control devices
 for all of the  existing sources nationwide  (IV-D-47,  IV-D-49).
      Eight commenters recommended 2  years for compliance
 (IV-D-21,  IV-D-27,  IV-D-37, IV-D-44,  IV-D-47,  IV-D-49,  IV-D-52,
 IV-D-57).   Five commenters recommended 3  years for compliance
 (IV-F-01  [Altmayer],  IV-D-17,  IV-D-24,  IV-D-57, IV-D-58).   One
 commenter  (IV-D-37)  suggested having a, tiered compliance
 approach,  whereby uncontrolled sources would have to  use a  foam
 blanket or poly balls  within  6 months and install MACT within
 2 years.   Also, this commenter felt  that  facilities that have
 installed  controls  in  the  absence of a Federal standard  should be
allowed 2  years to meet MACT requirements.
     .Comment:  Two  commenters pointed to  previous regulatory
actions as reasons why the proposed timeframe  of  l year  for hard
chromium electroplaters is inappropriate.  One  commenter
 (IV-D-23)  suggested that sources will need at  least 24 months to
                               2-76

-------
estimate emissions; identify appropriate technology; order,
install, and test equipment; and comply with the General
Provisions and title V.  This same commenter pointed out that the
EPA has established 'a 24-month compliance period in the
halogenated solvents NESHAP and does not think that the EPA
adequately explains why a shorter timeframe was proposed for this
rule.  Another commenter  (IV-D-17) also pointed out that other,
higher priority source categories such as those covered by the
HON have been given 3 years to comply.
     Comment:  One commenter  (IV-D-43) stated that  the proposed
timeframe of 1 year for hard  chromium electroplaters does  not
give States sufficient time to submit requests for  equivalency. ^
This same commenter suggested that no notification  or compliance
dates  should be prior  to  1  year  after promulgation  so that
sources are not confused  as to who is the  implementing  agency
 (Federal, State,  or local).
     Comment:  Several commenters stated that  the proposed
compliance  timeframe of 3 months for decorative  chromium
electroplaters and chromium anodizers was  insufficient  for the
following reasons:
       1.  Trade  associations  need time  to  educate  their members
 (IV-F-01  [Altmayer,  Sonntag], IV-D-19,  IV-D-58);
       2.  Equipment may have  to be purchased,  the  lead time to
purchase  equipment is  greater than 120  days, and employees must
be trained  to use a stalagmometer (IV-F-01  [Altmayer],  IV-D-22,
 IV-D-58);
       3.   Facilities will have to research the appropriate
 wetting agent,  test,  and develop a startup/shutdown/malfunction
 plan  (IV-D-26,  IV-D-35,  IV-D-59);
       4.   The compliance timeframe does not consider decorative
 chromium platers and chromium anodizers that may have to  install,
 test, and possibly upgrade add-on controls  (IV-F-01  [Altmayer],
.IV-D-09, IV-D-14, IV-D-25, IV-D-40, IV-D-43, IV-D-44,  IV-D-47,
 IV-D-58, IV-D-59, IV-D-61);
                                2-77

-------
        5.   Large sources will have large expenditures for
  equipment  and personnel to do testing,  reporting,  and
  recordkeeping (IV-D-57);
        6.   Emissions "from anodizing tanks are  low,  and there
  should therefore be no  urgency (IV-D-14);
        7.   Most of the facilities  are area sources  and will need
  more time  to  comply  (IV-D-42);
        8.   In  the aerospace industry, additional time will be
  needed because fume suppressants  must be tested to verify that
  they do not interfere with other  steps  or  affect the mechanical
  and fatigue properties necessary  when chromium anodizing is used
  as a pretreatment  for adhesive bonding,  and government
  contractors must get government approval prior to  changing a
 process (IV-D-47,  IV-D-52,  IV-D-58);
       9.  The use  of fume  suppressants would have  to be approved
 by European customers (IV-D-54); and
      10.  A 3-month timeframe does not give States time to submit
 requests for equivalency  (IV-D-43).
      Compliance timeframes ranging from 6 months to 3 years were
 suggested  by commenters  (IV-F-01  [Altmayer],  IV-D-14, IV-D-19,
 IV-D-24, IV-D-22, IV-D-26, IV-D-35, IV-D-40,  IV-D-45, IV-D-47,'
 IV-D-52, IV-D-54, IV-D-57, IV-D-58, IV-D-59).   Two commenters'
 (IV-D-58,  IV-D-61)  suggested that  sources that have to install
 add-on  controls be given the same  compliance  timeframe as hard
 chromium electroplating  operations.
     Response;   The Agency agrees  with  the commenters that  the
 compliance  timeframes for hard chromium electroplating,
 decorative  chromium electroplating,  and chromium anodizing
 sources should be increased.   The  EPA recognizes that some of the
 facilities  within all of  the source categories will have  to
 investigate the technical  feasibility of installing control
 devices  at  their facility  to meet  the standards.  Also, some
 sources  performing  decorative chromium electroplating and
 chromium anodizing  will have  to investigate the feasibility of
using fume suppressants.    Likewise, some  hard chromium
electroplating  sources that, have control  systems that may not
                               2-78

-------
 quite meet the standard may want to investigate using fume
 suppressants in conjunction with a control device as a way to
 meet the standards.   Also,  many area sources are not yet aware
•that a rule is to be promulgated for their industry, and the EPA
 and various trade associations/industry groups need an
 opportunity to notify these facilities of the new requirements.
      Based on these and the other reasons provided by the
 commenters, the EPA has extended the compliance date to 1 year
 after the promulgation date for existing decorative chromium
 electroplaters and 2 years after the promulgation date for
 existing hard chromium electroplaters and existing chromium
 anodizers.  The EPA believes that the 1 year timeframe for
 decorative chromium electroplaters is sufficient rather than
 2 years because, based on EPA's survey of the  industry,
 80 percent of these sources are already using  fume  suppressants.
 The EPA believes that very few sources will need to install
 add-on control devices because fume suppressants are widely
 applicable to this source category.   In addition, should  a source
 have  to install an add-on air pollution control device, there are
 procedures  in place for requesting an extension of  compliance,  if
 necessary [§  63.6(i) and 63.343(a)(6) of  subpart A].   A larger
 percentage  of hard chromium electroplating  and chromium anodizing
 source  categories, on  the  other  hand, will  need  to  evaluate  and
 implement control techniques  to  meet  the  emission  limitations.
 Therefore,  a  2 year  compliance  timeframe  is identified for
 existing  sources  in  these  source categories.   The  EPA thinks that
 the timeframes  in the  final rule will address commenters
 concerns, and still  ensure implementation of controls in a  timely
 fashion.
       Although the Act  allows  up to 3  years for an existing  source
 to comply with a standard,  the EPA does not think that a 3-year
 compliance time is  necessary  or prudent in this case.  A longer
 •compliance time has not been  allowed due to the toxicity of
 -•chromium compounds  and the importance of controlling chromium
  emissions to protect human health and the environment.  Also, the
  types of control techniques that are expected to be used to
                                 2-79

-------
  comply with the standards are readily available within the 1-  or
  2-year compliance  timeframes.
       There  may be  some  special situations  in  which sources
  require additional time to comply with the rule.   For  example,
  government  installations covered  by  this rule have indicated that
  they  will need extra  time to procure funding  and to implement
  changes.  Also,  for defense contractors, it is  expected that a
  new procedure  or process (e.g., inclusion  of  a  fume suppressant)
  for electroplating or anodizing operations will require
  government  approval prior to its  use.  Based  on industry
  comments, a minimum of  18 months  is  required  to verify a new
 process change such as  the use of  fume suppressants  in addition
 to the time necessary for implementation.
      The EPA recognizes  that unique  compliance considerations
 such as these may exist.  Therefore,  as previously stated,
 § 63.6 (i) of the General Provisions discusses mechanisms by which
 a source can be granted an extension of compliance with the
 emission standards.  This section allows a source to request a
 1-year extension for compliance.  In the General Provisions, the
 request must be submitted 12 months in advance of the compliance
 date identified in  the regulation.  In § 63.343(a)(6) of the
 final  rule,  the extension must be  submitted no later than
 6 months before the compliance date for the source.  If granted,
 this extension  combined  with the compliance timeframes  in the
 proposed rule provides a total  of  2 years for  compliance for
 decorative chromium electroplaters and 3 years for  compliance for
 hard chromium electroplaters and chromium anodizers.  This  same
 section of the  General Provisions  allows an extension to sources
 that have complied  with  BACT or LAER  requirements specifically
 related to chromium compounds prior to  promulgation of  this  rule.
 Such sources may  be granted an  extension of the  compliance  date
 for up  to 5 years after  the date on which such installation was
achieved if the requirements of §  63.6 (i) (2) (ii)  of  subpart A are
met.
     Some commenters, specifically military installations, have
indicated that 5 years would be the minimum timeframe needed for
                               2-80

-------
them to comply with the rule.  The EPA believes that the
extensions described above that are allowed by the General
Provisions may offer some relief.  At best, the EPA could extend
the compliance time "up to 3 years which is the maximum allowable.
However, the EPA did not think that the compliance timeframe
should be increased beyond 2 years to accommodate these
facilities due to the toxicity of chromium compounds and the
provisions for obtaining extensions to compliance time under the
General Provisions.
     Comment:  Four commenters  (IV-D-27, IV-D-36, IV-D-47,
IV-D-52) found the proposed rule confusing in its description of
the relationship between the compliance timeframe and the
timeframe for completing a performance test.  One commenter
 (IV-D-27) found confusing the statement that the compliance
timeframe is 1 year, but that sources have 120 days following
that to conduct a performance test.  This  commenter had
understood that the test would have to be  conducted and
monitoring parameters set within the l year timeframe.  Another
commenter  (IV-D-36) stated that  the General Provisions should
allow  testing before the effective date.   Two  commenters
 (IV-D-47, IV-D-52) pointed out  that the compliance time specified
 in § 63.342(a) presents a dual  requirement that  sources must
 comply within 1 year or at the  end of  the  compliance  test,
whichever is shorter.  These commenters point  out that such a
 clause is a disincentive  for sources to comply early  in the
 compliance period,  especially  if sources have  to shut down their
 operations if they fail the  compliance test.   A  single  compliance
 time is suggested.
     Response;  As previously  discussed,  the  compliance time
 specified in  the  final  rule  has been changed  to  1 year  from the
 effective date  (i.e., promulgation date)  for  existing decorative
 chromium electroplaters and 2  years  from  the  effective  date for
 existing hard chromium electroplaters  and chromium anodizers.
•.The compliance  time for new sources  is still  immediately  upon
 startup.   Since proposal  of this rule, the General  Provisions to
 part 63 have also been promulgated (59 FR 12408).
                                2-81

-------
 Section 63.7(a)(2) requires that source testing be conducted
 within 180 days after the compliance date.  A source must be in
 compliance by the compliance date (e.g.,  install control
 equipment, set up recordkeeping system, etc.)  but does not have
 to demonstrate compliance through a performance test,  nor conduct
 compliance monitoring,  until 180 days after the compliance date.
 This point is clarified in the final rule.
      In addition,  neither this NESHAP nor the General  Provisions
 forbid performance testing prior to the effective date of this
 standard.   Typically,  sources wait until  promulgation  of the rule
 to be certain that all  changes made from proposal to the final
 rule have  been accounted for in a facility's compliance efforts.
 Some facilities may have a control system in plcice prior to
 promulgation,  and  may also have tested it prior to the
 promulgation date  to fulfill State permitting requirements.   The
 EPA recognizes that it  would be costly to repeat; a performance
 test after promulgation of a Federal rule if a representative
 test had recently  been  conducted.   Section 63.344(b) of the final
 rule allows a  source to use the results of a performance test
 conducted  by the source at startup to obtain a State operating
 permit,  as long as the  following conditions are met:
      1.  The test  methods  and procedures  identified in
 §  63.344(c)  of the final rule were used during the performance
 test;
      2.  The performance test was  conducted under representative
 operating  conditions for the source;
      3.  the performance test  report  contains  the elements
 required by §  63.344(a); and
   •   4.  The owner or operator of  the affected  source  for which
 the performance test was conducted has  sufficient  data  to
 establish  the  operating parameter values that corresponds to
 compliance with the standards,  as required  for  continuous
 compliance monitoring under  §  63.343(c) of  subpart N.
     Regarding the dual compliance date, it was not the EPA's
intent to penalize sources that conduct a performance test prior
to the deadlines specified in this rule and in  the General
                               2-82

-------
 Provisions.   Section 63.342(a)  of the final  rule has been
 clarified;  emission limits must not  be exceeded on or after the
.compliance  date  specified in §  63.343(a)  for the affected source.
 A facility  may conduct  its performance test  at  any time prior to
 the  compliance date,  but  the facility would  not be in violation
 due  to exceedances  until  the actual  compliance  date itself.
      Comment:  One  commenter (IV-D-47)  suggested that if a system
 fails to meet the standard upon initial testing,  the rule should
 allow additional time to  investigate and correct the problem.
 The  commenter points to the  treatment,  storage,  and disposal
 facilities  (TSDF) rule  (56 FR 33490,  section 265.1081)  as an
 example.
      Response:   The final rule  allows a 2-year  compliance
 timeframe,  and the  General Provisions allow  an  additional
 180  days after the  compliance date for a source to demonstrate
 compliance  through  a performance test.   If a test is conducted
 prior to the  compliance deadline and fails,  the source can retest
 up until the  compliance deadline without violating the standard.
 The  EPA believes that the timeframe  for compliance specified in
 the  final rule is sufficient, and no additional time is necessary
 to allow sources to investigate control device  problems and
 retest.
 2.11  SELECTION  OF  MONITORING REQUIREMENTS
 2.11.1  Enhanced Compliance  Monitoring for Packed-Bed Scrubber
         and Composite Mesh-pad  Systems
      Comment;  There were 11 (IV-D-06,  IV-D-10,  IV-D-14, IV-D-23,
 IV-D-24,  IV-D-27, IV-D-36, IV-D-37,  IV-D-43, IV-D-47, IV-D-53)
 comments on the  suitability  of  measuring gas velocity to
 demonstrate ongoing compliance.
      One commenter  (IV-D-24)  stated that, in general, the EPA has
 not  demonstrated the correlation between the monitoring
 parameters  and compliance.   Four commenters  (IV-D-14, IV-D-23,
..IV-D-27,  IV-D-36) stated that measurement of both gas velocity
.and  pressure  drop is redundant  because they  are proportional; one
 of these commenters (IV-D-27) indicated that gas velocity
 measurement should  not  be required because it is a more
                                2-83

-------
complicated measurement than pressure drop.  Three commenters
 (IV-D-06, IV-D-10, IV-D-43) stated that, in their opinion, gas
velocity is not related to scrubber performance.  According to
one commenter  (IV-D-36), only a major malfunction of the scrubber
would cause a change in gas velocity.
     One commenter  (IV-D-23) suggested daily prejssure drop
measurements be required in lieu of daily gas velocity
measurements.  Two commenters  (IV-D-47, IV-D-53) suggested
installing an air pressure differential gauge across the scrubber
to monitor air flow, which can be recorded once per day.  One of
these commenters  (IV-D-53) suggested that the operating permit
should state the maximum and minimum for the velocity or pressure
drop.
     Nine commenters (IV-D-06, IV-D-16, IV-D-20, IV-D-28,
IV-D-36, IV-D-37, IV-D-38, IV-D-43, IV-D-53) stated that there
would be difficulties in implementing the gas velocity monitoring
requirements.  Two commenters  (IV-D-06, IV-D-16) stated that the
measurement method for gas velocity is laborious?.  Two commenters
(IV-D-16, IV-D-36) pointed out that a source may not be able to
access an area for measurement and a permanent measurement device
may be fouled by chromic acid.  One of these commenters  (IV-D-16)
stated that it was unreasonable to require testing for all
combinations of tank operation and duct flows that can occur;
testing at maximum amperage and air flow should be sufficient.
This commenter also indicated that air flows at each tank can
change daily due to the addition of makeup air cind seasonal
changes of air pressure.  This commenter and one other commenter
(IV-D-53) each suggested that gas velocity should be measured at
the entrance to the control device, not at each tank.  One
commenter (IV-D-38) stated that air velocity should be monitored
in the stack after discharge of the fan to .allow the pitot tube
to stay clean, and not be measured prior to the control device.
     Two commenters (IV-D-28, IV-D-37) suggested that the rule
clarify how the acceptable gas velocity ranges cire to be
established; development of a protocol is needed.  These
commenters also stated that no compliance test can fully examine
                               2-84

-------
the range of gas velocities and evaluate their impact on control
performance, especially if the system is designed to operate at a
constant velocity (IV-D-37).  This same .commenter (IV-D-37)
suggested that velocity measurements can vary based on ambient
conditions, operator error, differences in pitot tubes, etc.;
therefore, the rule should specify an allowable variance from
that measured during the performance test, e.g., ± 5 percent.
Likewise, one commenter  (IV-D-49) recommended that,the regulation
specify an acceptable pressure drop variance of ± 33'percent.
Another commenter  (IV-D-27) indicated that during performance
testing,  the accuracy of  the measurement  device should be
accounted for when  establishing  ranges  for the monitoring
parameters.
      One  commenter  (IV-D-38)  suggested  that  for  facilities that
 cannot afford an automatic monitoring and recording  system,  a
 simple alternative  should be  considered.   For example,  because
 pressure drop  is the primary  indicator  of proper operation,  a
 facility could  measure pressure drop with a photohelic gauge
 which contains  a set of relays to perform a function,  i.e.,  an
 alarm or a system shutdown.'  The commenter suggests that an
 automatic or manual washdown for composite mesh-pad systems and a
 conductivity control system for packed-bed scrubber systems could
 be used in conjunction with this.
      Six commenters made remarks regarding the measurement of
 chromium concentration in the scrubber water.  Four commenters
  (IV-D-10, IV-D-24, IV-D-36, IV-D-43) stated that there is no
 obvious  relationship between scrubber water chromium
 concentration and  scrubber performance.  Another commenter
  (IV-D-53)  said  further research is necessary to establish such a
 relationship.   This commenter suggested  that the scrubber water
 chromium concentration  limitation be determined on  a  case-by-case
 basis and also  suggested that  in lieu  of scrubber water  chromium
  concentration,  the regulation should require a minimum scrubbing
  recirculation  rate to the packed-bed  scrubber.  One commenter
  (IV-D-47)  indicated that it  would be more appropriate to monitor
  the pH of the  scrubber water and to record this information once
                                 2-85

-------
 per day.  One  commenter (IV-D-24)  requested  that  the  requirement
 for measuring  scrubber  water chromium concentration be  eliminated
 because the hydrometer  is  not accurate and the process  parameters
 do not change  on a daily basis.
      Ten commenters  (IV-D-06,  IV-D-10, IV-D-24, IV-D-27, IV-D-28,
 IV-D-38, IV-D-43, IV-D-44) provided comments on the difficulty of
 measuring the  chromium  concentration  in the scrubber water.
 Three commenters (iv-D-06, IV-D-24, IV-D-27)  indicated  that
 measurement of the chromium  concentration in the scrubber water
 with a hydrometer is not accurate.  Three commenters  (IV-D-10,
 IV-D-28, IV-D-43)  requested  that the rule identify a simpler
 measurement method.   One commenter (IV-D-44)  indicated that use
 of a hydrometer to measure scrubber water concentration is
 impractical when the scrubber is also used for control of
 phosphoric acid or pickling baths.  Another commenter (IV-D-38)
 suggested that a conductivity control system should be identified
 as an alternate method of measurement.
      Two commenters  (IV-D-10, IV-D-43) stated that small,  hard
 chromium electroplaters  should be exempt  from the  scrubber water
 chromium concentration measurement requirement  unless  these
 requirements are made more  flexible;  one  of these  commenters
 (IV-D-10)  suggested a threshold level  of  less than
 60,000,000 amp-hr/yr.  One  commenter  (IV-D-45)  indicated that  a
 default compliant scrubber  water chromium concentration  should
 not be specified and  that sources  should  set  their own value.
     Response:   The proposed  rule  included compliance  monitoring
 provisions in § 63.343(b) and (c).  The intent of  compliance
 monitoring is to monitor a  process  parameter  that  is directly
 linked to control device performance.   In  the proposed rule,
 sources using a composite mesh-pad  system  to comply with the
 emission limits would monitor gas velocity to determine
 compliance with these limits.  For  sources using a packed-bed
 scrubber, it was proposed that gas velocity and concentration of
 chromium in the scrubber water be monitored to determine
compliance with the emission  limits.  The proposed rule also
contained operation and maintenance (O&M)  provisions in
                               2-86

-------
§ 63.345(b)  and (c)  that were not intended to determine
compliance with the emission limits,  but to ensure that the
control device was properly maintained.   The EPA added the O&M
provisions because poor maintenance could result in system
degradation over time, and eventually, system breakdown that
would not be immediately obvious through compliance monitoring.
     The final rule continues to require compliance monitoring
 [§ 63.343(c)], but the O&M provisions have been replaced with
work practice standards  [§ 63.342 (f)], which address O&M
practices.  Requirements for the compliance monitoring and the
work practice standards are summarized in Table 2-2.  As
indicated in Table 2-2, there have been changes to these
requirements from proposal.  Changes  have been made to address
commenters' concerns  while still providing  the EPA with the
ability to  determine  the compliance  of a  source at any time.   The
parameters  identified for  compliance monitoring have  changed
based  on  the EPA's review  of  comments received  on the proposed
 rule and  further  investigation  of  which process parameters  relate
 best to proper  performance of the control systems.
     For  sources  complying with the  emission limits  for hard
 chromium  electroplating, decorative  chromium electroplating,  or
 chromium  anodizing by using a composite mesh-pad system,  the
 final  rule requires  that the pressure drop across the control
 system be monitored daily to determine compliance with the
 emission limits.   During the initial performance test,  sources
 using composite mesh-pad systems will determine the outlet
 chromium concentration, and will also determine the average value
 of pressure drop across the device that corresponds to a
 compliant emission limit.   Subsequent operation outside of
 ± 1 inch of H20 of this pressure drop range constitutes
 noncompliance with the emission limits.  In lieu of accepting
 this default compliant range, owners or operators may conduct
 multiple performance  tests to establish a  range  of compliant
 pressure drop values.
       in developing the proposed rule, the  EPA chose  gas velocity
 as the process parameter^to monitor for  compliance when a
                                2-87

-------
S,
3
u
«s
c
£
en
,s

•g
§

•2 u,

a$ 3
jQ u





<
2














3
2














^
2

>>
.0
"3
>
2
0.
Oi
I
                                                           ^(Cg
OJ
I
If
I

I Parameteri
1 monitoring
liliil compliance tett



1*
•5
S
1

if
ft
li
§ § §
111
B9

•a

-------
composite mesh-pad system was used because the control system is
designed to operate within a specific gas velocity range.
.Operation of the control system outside of this velocity range
can  lead to performance problems.  In revising the proposed rule,
the  EPA recognizes that the measurement of gas velocity could be
burdensome and  that other control system parameters' could
potentially be  used to determine ongoing compliance.   In the
final  rule, sources using composite mesh-pad  systems are required
to monitor pressure drop across the device for compliance
purposes.  Based on information gathered by the EPA, the pressure
drop is directly related to  composite mesh-pad system
performance.  Also, based on comments received, measurement of
pressure drop is straightforward, and is,  in,fact,  currently
being monitored by some users  of  composite mesh-pad systems.   The
EPA believes  that  this  change  makes  the  rule  more flexible for
regulated  sources, while still ensuring  that  the  EPA has  a
mechanism  for determining  compliance with  the emission limits at
any given  time.
      The  compliance monitoring for  packed-bed scrubbers in the
 final rule is slightly different  than  that for composite  mesh-pad
 systems.   As  indicated in  Table 2-2, a  source using a  paeked-bed
 scrubber to meet  an emission limit  must  measure the velocity
 pressure at the inlet to the control system as well as the
 pressure drop across  the device.   The  relationship between
 pressure drop and packed-bed scrubber performance is less
 reliable than the relationship between pressure drop and
 composite mesh-pad system performance because of the lower
 pressure drop in packed-bed scrubbers.   Therefore, the EPA
 requires sources using packed-bed scrubbers  to also monitor the
 velocity pressure at the inlet to the control device.   This
 requirement will ensure that the gas velocity through the control
 system is being maintained in accordance with vendor
 recommendations and,  along with the pressure drop monitoring,
 will  ensure that the source is in compliance with the standard.
       The requirement that sources using packed-bed  scrubbers
 monitor the chromium concentration in the scrubber  water  has been
                                2-89

-------
 eliminated.  Upon further review of the proposed monitoring
 requirements, the EPA concluded that monitoring of the velocity
 pressure at the control device inlet and the pressure drop across
 the device was sufficient to demonstrate compliance with the
 emission limits when packed-bed scrubbers are used.  It is
 unlikely that the chromium concentration in the scrubber water
 would increase to the level identified in the proposed rule
 (45 g/L [6 oz/gal])  because of the industry practice to recycle
 the scrubber water as plating bath make-up water.   Also,  most
 sources using packed-bed scrubbers will be small sources; in
 revising the regulation, the EPA worked towards reducing the
 burden on small sources wherever feasible.
      Compliance monitoring requirements for fiber-bed mist
 eliminators have been added in the final rule.   For sources
 complying with the  emission limits for hard chromium
 electroplating,  decorative chromium electroplating,  or chromium
 anodizing by using  a fiber-bed mist eliminator,  the rule  requires
 that the pressure drop across both the fiber-bed mist-eliminator
 and the upstream control system used to prevent plugging  be
 monitored daily to determine compliance with the emission limits.
 During the initial performance test,  sources using  fiber-bed mist
 eliminators will  determine the outlet  chromium  concentration,  and
 will also  determine  the  average value  of pressure drop  across the
 fiber-bed  mist  eliminator  and upstream control  device that
 corresponds to  a  compliant  emission limit.   Subsequent  operation
 outside  of  ±  1  inch  of water column of  this  pressure  drop range
 constitutes noncompliance with the  emission  limits.   In lieu  of
 accepting  this default compliant range,  owners  or operators may
 conduct multiple performance  tests  to establish  a range of
 compliant pressure drop values.
     Manycommenters requested  guidance  on how to measure the
process parameters to determine compliance.  The final rule
 includes a method that sources must use  for  establishing  the
velocity pressure.  The rule also contains procedures for
establishing the pressure drop across the unit that sources may
use if they wish.  The methods identified are straightforward and
                               2-90

-------
can be performed at a low cost to the plants.  For example, the
velocity pressure at the inlet to the control device can be
monitored by permanently mounting a plastic type-S pitot tube
.•that is connected to" a magnahelic gauge that records the velocity
pressure.  To measure the pressure drop, pressure taps can be
located on both sides of the packed bed, mesh pads, or fiber
beds, as appropriate.  Tubing connected to these two taps can
then be sent to another magnahelic gauge.  The EPA has installed
such a system and has found that corrosion problems associated
with chromic acid are alleviated by using a plastic pitot tube
and plastic pressure taps  instead of metal.  Once the monitoring
devices are installed, the only labor required is the recording^
of the magnahelic readings.  Alternately, a  strip chart  recorder
could be used with  the gauges, and high/low  alarms used  to
..indicate when values were  close to exceeding the  compliant
ranges.  Once the pitot  tube  is in place, the magnahelic gauge
connected  to the pitot  tube  can then be "zeroed."  The magnahelic
gauge would then be monitored once a day to  ensure no  change in
the velocity pressure  has  occurred.  The final  rule  (Table 2)
also  specifies  work practice standards  for pitot  tubes.
      One  requirement  that  is consistent between the  proposed and
 final  rules is  the  requirement that  sources  set the  values for
 the  operating  parameters to be monitored for compliance.  The EPA
 does  not  think it  is  appropriate  to identify the actual values
 for the compliant  operating parameters in the rule.   Each control
 device is different depending on the manufacturer,  the age, the
 type of process operation being controlled,  maintenance
 procedures, etc.   The EPA does believe, however,  that it is
 reasonable to specify an allowable variance because it would be
 very difficult for an owner or operator to establish a variance
 during one, three-run performance test.  In the final rule, an
 .owner or operator is out of compliance with the standard if the
-pressure drop varies by ± 1 inch of water column from the average
^operating parameter value established during the initial
 performance test.  Likewise, an owner or operator is out of
 compliance with the standard if the velocity pressure varies by
                                2-91

-------
 ± 10 percent from the average operating parameter value
 established during the initial performance test.  In lieu of
 accepting these default compliant ranges, an owner or operator
 can set the compliant range of operating parameter values by
 conducting multiple performance tests.
      Comment:  Two commenters (IV-D-20, IV-D-43) suggested that
 § 63.343(b), which requires determination of outlet chromium
 concentration in an initial performance test and monitoring of
 gas velocity and scrubber concentration,  should apply to all
 tanks,  not just hard chromium electroplating tanks.   Otherwise,
 sections 63.345(b)  and (c),  which refer to maintenance practices
 for composite mesh-pad and packed-bed scrubber systems,
 respectively,  are not enforceable.
      Response:   The rule has been revised to clarify that
 monitoring requirements are  based on the  type of control
 technique  used,  not on the type  of  tank operated.   Therefore,  the
 same monitoring applies regardless  of whether the  control device
 is being used to control  chromium emissions  from a hard  chromium
 electroplating  tank,  decorative  chromium  electroplating  tank,  or
 a chromium anodizing  tank.
      Comment;   One  commenter (IV-D-20)  noted  that  the preamble
 states  that  pressure  drop  measurement will be used as an
 indication of noncompliance  but will not be a noncompliance
 trigger; this exclusion is not evident in the rule.
     Response;  As stated  above, the compliance  monitoring
 requirements have been revised so that pressure  drop monitoring
 is required  in lieu of gas velocity  monitoring.  Table 2  in the
 final rule specifies work practice standards, and  § 63.343(c)
 specifies monitoring requirements to determine continuous
 compliance with the emission limits.
 2 •1:L'2  Selection of Proposed Operation and Maintenanr.A
        Requirements for Packed-Bed  Scrubber and Composite Mesh-
        Pad Systems
     2.11.2.1  Maintenance Requirements for Packed-Bed Scrubbers.
One commenter (IV-D-59) indicated that wastewater requirements
for packed-bed type washers are unreasonable,  but did not
                              2-92

-------
 elaborate on the specific "wastewater requirements"  to which he
 was referring.   Three commenters (IV-D-14,  IV-D-37,  IV-D-38)
 suggested the EPA revise the requirement for adding  makeup water
'at the top of the scrubber because makeup is typically added to
 the scrubber basin or at the front of the scrubber.   One of these
 commenters (IV-D-37)  stated that some scrubber systems are
 vertical and some are horizontal and suggested that  the term top
 be further clarified; this commenter questioned whether the term
 top refers to the upstream or downstream side.  Another commenter
 (IV-D-24) indicated that the location for addition of make-up
 water should not be specified in the rule.   One commenter
 (IV-D-58) noted that several scrubber manufacturers  have
 indicated that the requirement to supply fresh makeup water at
 the top of the scrubber unit is either impossible or not
 recommended for their equipment.
      Three commenters (IV-D-21, IV-D-37, IV-D-51)  questioned
 whether washdown water for the composite mesh pads and makeup
 water additions for scrubbers can be done with electroplating
 rinse instead of fresh water.  One of these commenters (IV-D-21)
 pointed out that wastewater would be increased by using fresh
 water.  This commenter also questioned whether shutdown of the
 control system is desirable because OSHA regulations may be
 exceeded during this time.  Another of these commenters (IV-D-37)
 stated that use of fresh water deters pollution prevention
 practices; according to another (IV-D-51),  the concentration of
 chromium in the rinse water will be much less than that on the
 pad and should therefore be acceptable.  This commenter also
 stated that for composite mesh-pad systems that contain three
 pads, the fluid from the second stage is recirculated and used to
 clean the front pad;  the concentration of this recirculated fluid
 is usually below 100 mg/L.
      One commenter (IV-D-29)  requested that section  63.345(b)(3)
 and (c)(3) be revised to state, "When makeup additions occur,
Censure that all makeup water is fresh or lower in total dissolved
 solids (TDS)  (2,000 ppm or less) and total chromium  (400 ppm or
 less) and supplied either to the top of the packed bed or its
                                2-93

-------
reservoir."  One commenter  (IV-D-37) asked that the term fresh
water be defined in the rule.
     Response:  Section 63.345(c)(3) of the proposed rule
erroneously required" that sources using composite mesh-pad
systems ensure that all makeup water is fresh and supplied to the
unit at the top of the packed bed whenever makeup additions
occur.  This requirement was intended for packed-bed scrubbers
only and has been corrected in the final rule.
     In light of comments received on the proposed O&M monitoring
requirements, the EPA has included the O&M requirements as work
practice standards in § 63.342(f) the final rule, and has reduced
the burden to the extent possible while still ensuring that units
will be properly maintained.  In the final rule,, the inspections
to ensure proper drainage, no chromic acid buildup, and no
chemical attack on the structural integrity, and the inspection
of the chevron-blade mist eliminator remains, but the frequency
of inspections has been reduced from daily or monthly to .
quarterly.  An additional quarterly inspection has been added,
however; owners or operators must inspect ductwork from the
affected source (s) to the control device to ensiire that there are
no leaks.  These requirements apply to all add-on air pollution
control devices.
     The final rule continues to require sources using packed-bed
scrubbers to meet an emission limitation ensure that all makeup
water is fresh and supplied to the unit at the top of the packed
bed whenever makeup additions occur.  The EPA considers this
requirement essential to meeting the prescribed emission limit.
During source testing conducted by the EPA to esstablish the
performance level of packed-bed scrubbers, it was noted that a
system equipped with an overhead spray system that periodically
cleaned the packing with fresh water performed much better than a
system without such cleaning.  Therefore, the EPA conducted a
test to determine the affect of periodic and continuous washdown
of the packed bed on system performance.  In this test, emissions
were measured when no overhead washdown occurred, when periodic
washdown occurred, and when continuous washdown occurred.  Clean
                               2-94

-------
water was used  in  the washdowns.  The tests showed  that the
emissions from  the scrubber were  reduced when periodic washdown
..occurred, yet continuous washdown did not offer  substantial
additional  benefit over periodic  washdown.   (Test results are
described in detail in Chapter  4  of  the Proposal BID.)
Packed-bed  scrubber systems that  did not incorporate  this
washdown technique could not meet the emission limit  in the  rule.
     Based  on the  above results,  the EPA decided that the rule
should  require  periodic washdown  of  packed-bed scrubbers with
fresh water whenever makeup additions occur because such systems
offer additional environmental  benefit.   (The final rule defines
fresh water.)   Makeup water is  added to replace  water that has
been evaporated in the scrubber system, to replace  scrubber  water
recycled to the plating tanks,  or to replace water  discharged to
the wastewater  treatment system.  Therefore, there  is no increase
in the  wastewater  burden on the facility.  Water from the
scrubber will continue to  be  recycled to the electroplating  bath
to make up  for  electroplating bath  evaporative  losses.  The  only
difference  is that the water  that would then be  added to the
scrubber as makeup must be fresh  and added to the  top.  If
electroplating  rinse water meets  the definition of  fresh provided
in the  final rule, it  can  be  used.
     To further clarify  the makeup  water requirement, top has
been defined in the rule  for  horizontal-flow scrubbers  as the
section of  the  unit directly  above  the packing  media  such that
the makeup  water would flow perpendicular  to the air  flow through
 the packing.  For  vertical-flow units,  the  top  would  refer  to  the
area downstream of the packing material  such that  the makeup
water  would flow countercurrent to  the  air flow through the unit.
The EPA believes that  if  the  requirement  that makeup  water  be
 fresh  and  added to the top of the packed bed is not included,
 scrubbers  will  not continuously meet the  required emission  limit,
 even if the scrubber met the  limit  during  the  initial performance
 test and the pressure  drop and velocity pressure measurements  are
 within the appropriate ranges.   Therefore,  the  washdown
                                2-95

-------
 requirement was added as a work practice standard that serves as
 another means of ensuring compliance with the standard.
      Based on comments received, there seems to be some confusion
 related to implementing this periodic washdown with fresh water.
 Contrary to commenters concerns, the scrubber system does not
 have to be turned off in order to wash down the packed bed.
 Also,  the EPA does not think that substantial retrofit would be
 required to incorporate periodic washdown of the packed bed.
 During the source test to determine the effect of periodic
 washdown,  the EPA retrofitted an existing scrubber using readily
 available equipment at minimal cost.
      2.11.2.2  Maintenance Requirements for Composite Mesh-Pad
 Systems.
      Comment:  Eleven commenters (IV-F-01 [Schott,  Kimre],
 IV-D-16,  IV-D-17,  IV-D-23,  IV-D-24,  IV-D-37,  IV-D-38,  IV-D-49,
 IV-D-51,  IV-D-58)  provided comments on the washdown requirements
 for  composite mesh-pad systems.
     Three commenters (IV-D-38,  IV-D-51,  IV-D-58)  indicated  that
 a 10-minute wash is not appropriate for every unit.   Two of  these
 commenters (IV-D-38,  IV-D-58)  also noted that although the rule
 requires  shutdown of the fan for washdown,  some units  are cleaned
 better with the exhaust fan on.   One  cornmenter (IV-D-51)  noted
 that tests performed on one composite mesh-pad unit  indicate  that
 there is no further benefit to washdown after 1 1/2  minutes.
 Pour commenters (IV-D-16,  IV-D-17,  IV-D-37,  IV-D-58)  said daily
 washdown is not feasible if electroplating time exceeds 24 hours.
 Another commenter  (IV-D-17)  indicated that if multiple tanks  are
 connected  to a  control  device, it  is  likely that at  least  one
 will always be  operating, and  shutting down the control system
 for washdown would  ruin work in  progress.   This commenter
 recommended deleting  the washdown  requirements.  Seven commenters
 (IV-F-01  [Schott, Kimre], IV-D-23, IV-D-24,  IV-D-37, IV-D-38,
 IV-D-58) suggested  washdown requirements  for  composite mesh-pad
 systems be  site-specific, as recommended  by vendors, or only  if
pressure drop determinations indicate  the potential presence  of  '
 chromic acid buildup  (IV-D-23).
                               2-96

-------
      Two commenters (IV-D-16,  IV-D-37)  indicated that some
 composite mesh-pad systems have automatic daily washdown and
•suggested that such systems be excluded.from the requirements.
 Another commenter (IV-D-38) recommended that daily recordkeeping
 of washdown times and durations not be required if the source
 demonstrates that a well-designed and functioning automatic
 washdown control panel is being used.  One commenter (IV-D-24)
 stated that systems with automatic washdown are difficult to
 inspect on a monthly basis.  Another commenter  (IV-D-49) noted
 that, regarding the monthly inspection, chromic acid should never
 be present in the back portion of the pad but moisture will be
 present when washdown occurs from the back of the pad.
      Two commenters  (IV-D-23, IV-D-51) suggested that the
 washdown water will likely exceed the quantity  of water that can
 be recycled, thus resulting in a wastewater stream that needs to
 be treated.  One of these  commenters  (IV-D-51)  stated that  the
 volume  of fluid from washdown and reclaimed rinse waters  cannot
 exceed  the volume of water evaporated by  the  electroplating tank
  (in  an  attempt to eliminate wastewater).  This  commenter  also
 noted it  is desirable to  use  rinsewater  as makeup water in  the
 electroplating tank because fresh water  can then be  added to the
 rinse tank, thus  reducing chromium concentration and exposure.
 In addition to the requirement  for daily spray cleaning of
 composite mesh-pad systems, commenter IV-D-51 stated that the
 first mesh pad should be  immersed  in a rinse  tank at a  frequency
 of once per week to  once  per  month, depending on the system.
 According to  this commenter,  mesh  pads should always be removable
  to  facilitate immersion cleaning and to facilitate visual
  inspection and repair or replacement.
       Response;   In the final  rule,  the EPA has revised the
  requirement that sources complying with an emission limit by
  using a composite mesh-pad system perform washdown of the pads on
 ,a daily basis.   Instead,  the final rule requires that washdown of
 ^the pads in a composite mesh-pad system occur in accordance with
  manufacturer's recommendations as part of a facility's O&M plan
  required by the work practice standards  [§ 63.342 (f)] .  The EPA
                                 2-97

-------
 believes that washdown is an essential part of composite mesh-pad
 system operation; if proper system maintenance such as washdown
 does not occur, there will be a decline in system performance.
 The EPA recognizes that vendor designs for these systems vary
 significantly and the requirements for washdown are based on the
 design of the unit and the operation of the plating tanks.
 Therefore, a specific washdown frequency is not included in the
 final rule.  The frequency of washdown is dependent upon the
 position of the pad in the control unit.  Pads located in the
 front portions of the unit are exposed to higher chromium
 concentrations and,  therefore,  require washdowns more frequently
 than those in the back of the unit.   Washdown practices
 recommended by manufacturers vary from continuous in some cases *
 to a maximum of once every one to two weeks.   Also,  as noted in
 the previous response,  the monthly inspections of the composite
 mesh-pad systems have been changed to quarterly inspections.
      As  previously stated,  in the final rule  monitoring and work
 practice requirements for fiber-bed  mist eliminators have also
 been added.   A washdown requirement,  similar  to that for
 composite mesh-pad systems,"is  included for fiber-bed units.
 Fiber-bed units  installed downstream of more  efficient
 prefiltering systems,  such as packed-bed scrubbers,  will  require
 less  frequent washdown  than those  using a  less effective
 prefiltering device because  of  the lower inlet loading to the
 unit.  Most  vendors of  fiber-bed units  recommend monitoring  of
 the pressure drop  as  a means of gauging when  the unit  needs  to  be
 washed down.   If an increase in pressure drop  is observed, then
 the unit will be washed down to remove  any  chromium built up on
 the fiber elements.
     2 ' -11 *2 '3  Operation and Maintenance Requirements- -General
     Comment;  Eight  commenters (IV-D-01, IV-D-03, IV-D-09,
 IV-D-17,  IV-D-24, IV-D-45, IV-D-47, IV-D-49) provided  comments  on
 the appropriateness of O&M requirements in general.  Two
 commenters  (IV-D-01, IV-D-03) suggested that the O&M plan
requirements be made more general to make them applicable to
sources complying with either a control device, a fume
                              2-98

-------
suppressant, or both.  One commenter (IV-0-09)  questioned whether
the EPA has justification for setting standards for O&M
requirements of scrubbers used at decorative chromium facilities
because none of the decorative chromium facilities tested use
scrubbers.
     One commenter  (IV-D-47) stated the maintenance practices are
burdensome and would require reconfiguration of most systems.
This commenter also suggested that only vendor-recommended
operating parameters should be monitored on a routine basis.  In
addition to these comments, this commenter also suggested that
more appropriate monitoring could include the following
measurements that would be  recorded once each day:   (1)  airflow,^
 (2) recirculating spray water and make-up water flow rates,  and
 (3) pH of  scrubber  water.
     One  commenter  (IV-D-17)  indicated that  requiring  a  source  to
notify supervisory  personnel  immediately upon  identifying a
malfunction is  unnecessary  because  the General  Provisions require
sources to affect a repair  of a malfunction as  soon as is
practicable.   Another commenter (IV-D-45)  requested that,  in
addition to the maintenance practices specified,  the rule also
 specify procedures  to follow if something is found to be wrong.
      Response:   The final rule  has  more detailed requirements for
 the O&M plan required by the work practice standards,  some of
 which refer to all  sources regardless of control technique, and
 others that are control technique-specific.  The requirements for
 the O&M plan in the final rule have been made more detailed than
 the requirements in the proposed rule, because these requirements
-are overriding § 63.6(e)(3) of the General Provisions.  This
 section of the General Provisions requires a startup, shutdown,
 malfunction plan, which in many respects is not applicable  to
 this  source category.  Therefore, the O&M plan incorporates those
 parts that are applicable, and has some additional requirements
..specific to this source category.  The  requirements of  the  O&M
 plan  are outlined  in  § 63.342(f) (3)  of  the  final  rule.
       The EPA disagrees that the work practice standards are
 burdensome and will  require substantial reconfiguration of
                                2-99

-------
 control systems.  The final  rule has reduced the frequency of
 maintenance inspections and  has revised some of the other work
 practice standards to make them less burdensome.  As stated
 above, the EPA believes that the retrofit that will be required
 to incorporate work practice standards can be accomplished at
 minimal cost to the facility.  The EPA also believes that the
 work practice standards specified for the various control devices
 are justified, even if the devices evaluated and tested by the
 EPA were not used to control decorative chromium electroplating
 baths.  Proper maintenance of the control devices that are likely
 to be used by this source category is not dependent on the type
 of electroplating or anodizing bath being controlled.
      Cpmment;   One commenter (IV-D-24)  suggested several changes
 to § 63.345(a) (1)- (3),  including use of the term velocity
 pressure traverse in place of velocity traverse and use of static
 pressure in place of velocity pressure.   Other changes that this
 commenter suggested for the gas velocity measurement include
 (1)  addition of  instructions  for ducts  with larger  diameters;
 (2)  addition of  barometric pressure and duct temperature
 measurements;  (3)  removal  of  the temperature requirement of 70°F
 because  tank temperature would have to  be  decreased, and the
 measurement  would  not reflect actual  operating  conditions;  and
 (4) use of  the term velocity  points in place of velocity pressure
 readings and use of  the  term  traverse test point to  obtain the
 traverse point of  average  velocity  pressure.  Another  commenter
 (IV-D-28)recommended deleting  the  requirement that  stack
 temperature must be approximately 70 °F during initial performance
 testing of the velocity  traverse because this temperature cannot
 be controlled or adjusted.  One commenter  (IV-D-37)  asked that
 alternate velocity measurements be allowed,  such as the use of
manometers that measure static pressure to indicate changes in
ventilation.  Another commenter  (IV-D-38) suggested that  use of
an air velocity magnahelic gauge is appropriate.  This commenter
also suggested that pressure  drop could be measured using a
pressure differential magnahelic gauge located inside the
building at a convenient area for visual monitoring.
                              2-100

-------
     Response;   As discussed previously,  gas velocity is no
longer considered as a continuous monitoring requirement for
control systems.  However, the final rule does require that the
velocity pressure at the inlet to the control system be measured
once per day to ensure compliance with the emission limit for
users of packed-bed scrubbers.  The comments above are based on
the need to do a velocity traverse every day and to calculate
velocities; this is not required in the final rule.  In addition,
the stack temperature specification of 70°F was selected because
the gas temperature would reach that temperature value at some
time of the day every day of the year.  However, the final rule
does not contain this temperature requirement.
     Comment:  Two commenters  (IV-D-17, IV-D-28) requested that
the preparation of the O&M plan be linked to the compliance date.
As the rule  currently reads, the O&M plan must be  implemented
within 90 days of the effective date while  hard chromium  sources
have 1 year  in which to comply; this suggests that sources may
have to develop a plan for a control device prior  to its  being
installed.
     Response;  The final rule has been  clarified  so that
implementation  of the O&M plan is tied to the compliance  date,
not  the effective date, of  the standards.
2.11.3  Frequency of Monitoring
      Comment;   Fourteen commenters  indicated that  the  required
 frequency of monitoring is  inappropriate.   One  commenter
 (IV-D-59)  explained that  daily monitoring of gas velocity is
unreasonable because  it takes considerable  time,  equipment,  and
 training.   Two other commenters  (IV-D-14,  IV-D-61) said the rule
 should state that daily monitoring is only required when the tank
 is in operation.   One commenter (IV-D-36)  suggested velocity
 measurements be done quarterly because pressure drop would be
 measured daily.  Another commenter (IV-D-06)  suggested that the
.ventilation systems should be checked annually as is consistent
 with CAL/OSHA.
      Two commenters (IV-D-06, IV-D-36) stated that the
 monitoring, recordkeeping,  and reporting requirements are
                               2-101

-------
 excessive and daily monitoring  is not necessary  for nonvariable
 control systems such as packed-bed scrubbers and composite mesh-
 pad systems; one of these commenters  (iv-D-36) stated that, in
 general, quarterly monitoring is sufficient.  Another commenter
 (IV-D-28) indicated that weekly monitoring would be sufficient.
 Another commenter (IV-D-09) agreed, stating that daily monitoring
 of steady state control systems is too frequent.
      One commenter (IV-D-14) suggested that the maintenance
 schedule for the control equipment should be consistent with the
 manufacturer's recommendations; for example, some manufacturers
 recommend monthly pressure drop readings.  This  commenter also
 recommended that monitoring should be part of normal preventive
 maintenance that occurs annually,  monthly, or weekly depending on
 the parameter.   One commenter (IV-D-16)  stated that the
 monitoring requirements in the rule would require a full time
 position to measure  and record air flow information.   One
 commenter (IV-D-57)  indicated that daily monitoring is excessive
 for small  sources  and suggested that  this requirement  be reduced
 if the source can  show that  compliance can be maintained with
 less  frequent monitoring.
      One  commenter (IV-D-10)  suggested monitoring be tied to
 production  rate  or throughput (amp-hr/yr).   Three commenters
 (IV-D-23, IV-D-24, IV-D-28)  suggested monitoring  should  only  be
 conducted on the days when electroplating operations are taking
 place,  and  one of  these commenters  (IV-D-24)  requested that the
 tank be considered in operation  only  when the  current  is actually
 applied to  the bath.  One commenter  (IV-D-14)  suggested
 monitoring  frequency be established through  the O&M plan,  which
 is reviewed  by the local air district.  One  commenter  (IV-D-44)
 indicated that monitoring frequency should be reduced after
 compliance has been demonstrated for  6 months as  is allowed in
 40 CFR part  61 subpart  V, and another commenter  (IV-D-27)
 suggested this reduction in monitoring frequency  specifically for
monitoring of the scrubber water concentration.
     .Comment:  Six commenters indicated that the  required
 frequency of monitoring chromium concentration in the scrubber
                              2-102

-------
water is inappropriate.  One  commenter .(IV-D-09) noted  that  for
scrubber systems, NPDES  only  requires weekly monitoring of
..wastewater discharge.  Another  commenter  (IV-D-27)  also suggested
that the specified  frequency  for measuring  scrubber water
concentration  should be  weekly.  This commenter  indicated that
daily monitoring  of scrubber  water is excessive  and expensive,
costing $200 per  day per scrubber  unit.   One  commenter  (IV-D-37)
pointed out that  it would  take  most scrubber  systems  a  very  long
time to reach  a level  of 45 g/L, and therefore,  daily monitoring
is overly  conservative.  This commenter also  suggested  that
monitoring occur  monthly or before any  complete  turnover of
scrubber water volume,  whichever  is earlier.   One  commenter
 (IV-D-36)  stated  that  daily measurement of  scrubber water
concentration  is  not necessary if  a system is automated such that
waters  with a  chromium concentration greater than that  specified
in the  proposed rule are not  discharged.   Another commenter
 (IV-D-14)  suggested that daily measurement of scrubber water
 concentration is  inappropriate for a continuous discharge system
and should only be required for batch discharge.  Another
 commenter (IV-D-17) stated that daily measurement is  not
 necessary for systems that only have one pass of water through
 the scrubber because the chromium concentration would not be near
 the concentration limit.
      Comment:   One commenter (IV-F-01  [Sonntag]) stated that the
 paperwork, testing, and sampling burden is excessive, especially
 the OSM plan and the 10-minute washdown requirement for the
 composite mesh-pad system.  One commenter  (IV-D-61) suggested the
 rule clarify that daily washdown needs to  occur only when the
 tanks are operating.  One  commenter  (IV-D-24) indicated that
 measuring pressure drop daily  is excessive and should be measured
 only weekly.  Another commenter (IV-D-57)  stated that  daily
 monitoring is excessive for  small  sources  and should be reduced
 ,if a source can  show that  compliance can be maintained with less
..frequent monitoring.  Two  commenters  (IV-D-17,  IV-D-58) suggested
 O&M requirements should be consistent  with the manufacturer's
                                2-103

-------
 recommendations and should not require hourly or daily
 inspections, measurements, or recordkeeping.
      Comment:  One commenter (IV-D-58) suggested inspections
 should only occur if other factors indicate the equipment is
 malfunctioning.  Two commenters (IV-D-17, IV-D-58)  asked that
 inspection of packed-bed scrubber, chevron-blade mist eliminator,
 and composite mesh-pad systems be changed to semiannual and
 pointed out that daily monitoring results in increased exposure
 hazards to persons conducting the inspections.   Three commenters
 (IV-D-24,  IV-D-47, IV-D-49)  pointed out that some control systems
 (i.e.,  packed-bed scrubber systems)  are inaccessible and that
 inspections will result in down-time;  therefore,  the inspection
 frequency should be an annual requirement or should be dropped.
 One of  these commenters (IV-D-24)  explained that baffle-type
 chevron-blade mist eliminators  would have to be removed
 (down-time is 2 to 3  days)  and  suggested that this  requirement be
 eliminated.   Another  (IV-D-49)  stated that visual inspection of
 the pad, which is required by §63.345(c)(2),  should occur on a
 monthly, not a daily,  basis.
      Response;   As stated  above,  the compliance monitoring and
 work practice standards (which  were  O&M requirements in the
 proposed rule)  have been reduced  to  minimize the  burden on
 regulated  sources.  The final rule continues to require daily
 monitoring of  pressure drop and velocity pressure for compliance,
 but  the monitoring procedures specified in the  rule are minimal.
 Once the monitoring devices are in place,  the labor required is
 only that  needed to read the gauges.   Compliance  monitoring for
 sources using packed-bed scrubbers has  been reduced in that these
 sources are  not  required by the final  rule to measure the
 chromium concentration in the scrubber  water  (see discussion in
 Section 2.11.1).
     The frequency  of  inspections  associated  with the work
practice standards has  also been reduced or revised.   The
proposed rule required  that inspection  of  the control  device  for
drainage,   chromic acid  buildup and system  integrity occur once
per month.  Also, the back portion of the mesh pads  (for
                              2-104

-------
composite mesh-pad systems)  or the chevron-blade mist eliminator
(for packed-bed scrubbers)  was to occur once per day.  These
requirements remain in the final rule, but the frequency of both
has been reduced to "once per quarter  (every 3 months).  (A
similar inspection has also been added for users of fiber-bed
mist eliminators.)  The EPA believes that the inspections are
still necessary to ensure that system degradation is not
occurring over time, because gradual degradation may not be
indicated by the compliance monitoring.  The EPA also believes
that the reduction in frequency will minimize the burden on
affected sources.  Some commenters noted that their systems were
not accessible for inspection, or that the  inspection would
result in extended downtime.  The compliance timeframe  in the
final rule  should allow sources sufficient  time to retrofit their
systems, if necessary, so that inspections  can occur.   The
negative effects  of any downtime are  minimized by inspection
frequency being reduced to  once per quarter.
     Other  aspects of the work practice  standards have  been
revised.  For  example, daily  washdown of composite mesh pads  is
no longer required;  instead,  manufacturers  recommendations  are to
be followed.   In  the proposed rule, pressure drop measurement was
specified as an O&M monitoring  requirement  in addition  to  the
compliance  monitoring.   In  the  final  rule,  pressure  drop  is
required for compliance monitoring  purposes only.
      Several commenters  noted that  in terms of monitoring and
 compliance  with  the emission limit,  the proposed rule was unclear
 as to whether  a  facility must comply with the rule at all times
 or only when the  tank is operated.   The final rule has been
 clarified so that requirements  apply only during tank operation.
 Further, tank operation has been defined as any time that current
 and/or voltage is being applied to a chromium electroplating or
 chromium anodizing tank or any one affected tank in a series of
 chromium electroplating or anodizing tanks ventilated to a common
 control system.                                          .
                               2-105

-------
 2-11-4  Alternate Compliance Monitoring for sources Using
         Suppressants
      2.11.4.1  Appropriateness of Surface Tension Value for Tvr
 Baths.              •"
      Comment:  Five commenters suggested that the requirement for
 maintaining surface tension below 55 dynes/cm for trivalent
 chromium baths is inappropriate.  One commenter  (IV-D-59)
 indicated that the surface tension limits should, be eliminated.
 This commenter and one other (IV-F-01 [Altmayer]) pointed out
 that, typically,  surface tension is approximately 40 dynes/cm.
 These commenters indicated that the electroplating process would
 not operate properly unless the surface tension was maintained
 properly,  and therefore the surface tension should not be
 regulated.   One commenter (IV-F-01 [Sonntag]}  questioned why
 surface tension is regulated if it is an inherent bath property.
 One commenter (IV-D-22)  stated that a 55 dynes/cm limit may not
 be practical for  optimum operation;  this commenter indicated that
 the EPA data in this  regard may be coincidental and adequate
 research has not  been done to demonstrate that all commercially
 available TVC processes  operate best  at  this  surface tension.
 Two other commenters  (IV-D-22,  IV-D-26)  also  noted that surface
 tensions above 40 dynes/cm result  in  unacceptable electroplating
 quality.  These commenters stated  that monitoring the  surface
 tension every  4 hours  to show that it remains  below 55 dynes/cm
 is  unnecessary.
      Response:  AS discussed  in Section  2.6.1,  the regulatory
 burden  on owners  or operators of decorative chromium
 electroplating tanks that  use a trivalent chromium electroplating
 process  has been  greatly reduced if the  trivalent  bath
 incorporates a wetting agent.   There  are no monitoring
 requirements associated  with such baths  in the final rule.  For
 those trivalent chromium processes that do not incorporate a
wetting agent  as part of the process, the requirements  for
decorative chromium electroplaters using a chromic acid bath
apply.  That is, monitoring of the bath surface tension is
necessary to maintain a level of 45 dynes/cm,  or monitoring must
                              2-106

-------
 be  conducted to  demonstrate an emission control  limit  of
 0.01  mg/dscm if  an add-on air pollution control  device is  used.
      2.11.4.2 Appropriateness of  Surface  Tension for  Chromic
 Acid  Baths.
      Comment: Seven commenters (IV-D-09,  IV-D-10,  IV-D-24,
 IV-D-43,  IV-D-45,  IV-D-47,  IV-D-52)  indicated that the
 requirement  for  maintaining surface tension below 40 dynes/cm  for
 chromic acid baths is inappropriate.   Two  commenters (IV-D-10,
 IV-D-43)  indicated that a surface  tension  standard may not be
 prudent to demonstrate compliance.  One commenter (IV-D-24)
 stated that  a direct correlation between exceedance of parameters
 and the emission limit has not been established.  One  commenter
 (IV-D-45)  suggested that a default compliant surface tension
 should not be specified by the rule and that sources should set
 their own compliance value.
      Two commenters (IV-D-09, IV-G-01) stated that the
 40  dynes/cm limit is too low.  Two other commenters (IV-D-47,
 IV-D-52)  stated that vendors recommend a 40-45 dynes/cm range and
 that the rule should state that the surface tension be consistent
 with manufacturers' recommendations.
      Response:  There are two types of fume suppressants that are
 expected to be used to comply with the final rule:  wetting
 agents and foam blankets.  Sources can use one or the other of
 these to inhibit chromic acid emissions, or may use them in
 combination.  Wetting agents inhibit misting by lowering the
 surface tension of the electroplating or anodizing bath; thus,
 there is a direct relationship between surface tension and
 wetting agent performance.  Foam blankets, on the other hand,
 limit.mists by trapping the mist under a layer of foam; the
 surface tension is not reduced.  Section 63.342(b)(2)  of the
 proposed rule allowed decorative chromium electroplating sources
 .using fume suppressants to comply with the standard to meet a
 .surface tension of 40 dynes/cm in lieu of meeting the emission
:i.limit.  This section should have allowed the surface tension
 limit only for those sources using wetting agents, or a
                               2-107

-------
combination wetting agent/foam blanket.  The final rule is clear
in this requirement.
     The EPA believes that it is necessary and appropriate to set
a default value for surface tension in the rule.  Based on the
EPA's experience, many decorative chromium electroplating tanks
are not ventilated, making source testing impossible without
considerable retrofitting.  The EPA had proposed a default
surface tension limit of 40 dynes/cm, but has increased this
value to 45 dynes/cm based on information submitted at proposal
regarding the surface tension ranges recommended by vendors of
wetting agents.  However, if a facility believes that a different
surface tension value is appropriate, the rule allows a source to
conduct a performance test concurrently with surface tension
monitoring to establish the maximum surface tension that
corresponds to compliance with the emission limits.  The source
would subsequently monitor surface tension, with an exceedance
occurring if the surface tension value of the bath was greater
than that measured during the performance test.
     Comment:  Two commenters (IV-D-09, IV-D-59) noted that the
rule appears to penalize owners using fume suppressants in
conjunction with a control device in that both the surface
tension and the control device must be monitored.  Two commenters
(IV-D-09, IV-G-01) stated that monitoring requirements should be
selected by these sources and should be less frequent than for
those with only one control method.  Another corrmenter (IV-D-59)
stated that it is unclear which monitoring method applies if a
source measures surface tension in association with fume
suppressant use and also has an air pollution control device.
     Response;  The final rule is clear in that the monitoring
requirements of the source are associated with the control
technique that is being used to meet the emission limits or
surface tension limits.  For example,  a source may be using both
a wetting agent and a packed-bed scrubber on a decorative
chromium electroplating bath.   If the wetting agent alone meets
the standards (i.e.,  lowers the surface tension to below
45 dynes/cm),  testing and monitoring of the control device is not
                              2-108

-------
necessary.  If however, a source is using a foam blanket and a
packed-bed scrubber to control emissions from a hard chromium
electroplating tank, and both control techniques are necessary to
meet the prescribed emission limit, the monitoring and testing
associated with both the foam blanket and the scrubber must be
performed.
     2.11.4.3  Frequency of Monitoring Surface Tension.
     Comment:  Twenty-one commenters '(IV-F-01  [Altmayer, Wenske],
IV-D-02,  IV-D-05,  IV-D-09, IV-D-10, IV-D-17, IV-D-18, IV-D-22,
IV-D-24,  IV-D-25,  IV-D-26, IV-D-35, IV-D-47, IV-D-48, IV-D-53,
IV-D-57,  IV-D-58,  IV-D-59, IV-D-61, IV-G-01) provided comments
related'to  the  frequency of monitoring surface  tension.   Several
of these commenters made recommendations for more appropriate
monitoring  schedules,  ranging from daily to monthly monitoring,
 in place of the 4-hour schedule.
      One commenter (IV-D-02)  asked the  basis for the 4-hour
monitoring  schedule.   One  commenter (IV-D-18)  stated that
monitoring  every 4 hours  is  burdensome and suggested once per
 month as an alternative.   Two commenters (IV-F-01 [Altmayer],
 IV-D-09) indicated that monitoring surface tension every 4 hours
 is onerous  because it is time-consuming to perform the
 measurement.  These commenters suggested the frequency be
 decreased if a continuous feed system is used or if there has
 been compliance over the 4-hour period.  Three commenters
  (IV-D-24, IV-D-53, IV-D-57)  recommended once per operational day
 as an appropriate  schedule; one of these commenters indicated
 that testing requires 45 minutes  to complete  (IV-D-53) and
 another indicated  0.5 to 1.5 hours  (IV-D-59) is  needed for
 testing.   One commenter  (IV-D-05) stated that  calibration and
 analysis of one  sample with  a  stalagmometer will take 1.5 to
  1.75 hours, and  monitoring every  4 hours could conceivably  occupy
  someone full-time if  there are multiple tanks.
      Three commenters (IV-D-01 [Wenske], IV-D-05,  IV-D-17)
 -suggested  that  a statistically .significant increase or  change  in
  surface tension should be the trigger  for  increasing the
  measurement frequency;  one  commenter (IV-D-47)  stated that
                                2-109

-------
  surface  tension does  not change day-to-day or week-to-week.   Two
  commenters  (IV-D-10,  IV-D-43)  stated that  the monitoring schedule
  for  fume suppressants or alternate  technologies  (i.e.,  foam)
  should be set  throug-h permit  review and be based  on  amp-hrs  of
  throughput, not on operating  time.   Another commenter  (iv-D-58)
  stated that the manufacturer's  recommendations should be  followed
  and  that, generally,  daily or weekly testing is appropriate.  One
  commenter (IV-D-10) suggested that monitoring be  based on
  rectifier throughput.  One commenter (IV-D-22) noted that
  frequent  monitoring results in  increased exposure of workers to
 hexavalent chromium.
      Several commenters noted specific situations in which
 monitoring every 4 hours is excessive.  Three commenters
  (IV-D-26, IV-D-48, IV-D-59)  stated that sampling every 4 hours is
 not necessary when chromium additions are performed
 automatically;  weekly sampling is sufficient in this situation
 (IV-D-26, IV-D-59).   One commenter (IV-D-25) indicated that if a
 source uses  a fluorocarbon chemical and provides evaporative
 recovery of  chromium dragout,  surface tension changes occur very
 slowly.   The wetting agent is  returned to the bath and is
 virtually indestructible; in this situation, daily measurement is
 sufficient.   One commenter (IV-D-35)  stated that  their operation
 is  continuous with constant  operation characteristics,  and if the
 alkaline  baths  are also covered by the rule, it would be  more
 appropriate  to  monitor once per day  or when a new  production is
 introduced into the system.
      One  commenter (IV-D-22) indicated that because decorative
 platers and anodizers  make up  only 10 percent of the  emissions,
 the proposed paperwork that applies  to them is  overkill;  for
 these sources,  monitoring of baths once per week is sufficient.
     Two  commenters  (IV-D-05, IV-D-17)  stated that sampling
 frequency  should be dependent upon the surface tension a source
 is attempting to maintain.  Sampling  every  4  hours is appropriate
when surface tension is maintained between  38-40 dynes/cm,  if
the surface tension is between 36-38  dynes/cm, daily sampling is
sufficient, and if it is less than 36  dynes/cm, then weekly
                              2-110

-------
 sampling is  sufficient.   According to one  commenter,  the language
 in section 63.345(d)  should be modified to reflect this
 .dependency (IV-D-05).
      Two commenters '{IV-D-18,  IV-D-61)  requested that the rule
 clarify whether surface tension measurement occurs only during
 hours of operation or whenever the tank is available for use; the
 former of these would be preferred.
      Response:  The final rule specifies that monitoring of
 surface tension should initially be done once every 4 hours, but
 that the frequency may be decreased if no exceedances occur
 during 40 hours of tank operation.  The EPA had originally
 proposed monitoring every 4 hours based on testing that indicated
' that the surface tension tended to increase at that point unless
 additional wetting agent was added.  In response to comments
 received, the EPA recognizes that this frequency may be
 burdensome, and in some cases, unnecessary.  However,  the EPA^has
 insufficient data to establish the monitoring frequency that  is
 appropriate for each mode of bath  operation.  Instead,  the  final
 rule allows a decrease  in monitoring frequency  if no exceedances
 occur.   Section 63.343 (c) (5)'(.ii)  specifies  that  if no  exceedances
 occur  during  40 hours  of tank operation,  monitoring  can occur
 once every  8  hours of  tank  operation.   If there are  again  no
 exceedances during 40  hours of tank  operation,  monitoring  can
 then occur  once every  40 hours of tank operation.   The minimum
 frequency allowed by the rule is  once every 40  hours of tank
 operation.  Also,  once an  exceedance occurs,  the original
 monitoring  schedule  of once every 4  hours must  be resumed.   A
 subsequent  decrease  in monitoring frequency can occur according
 to the criteria identified above.  Likewise,  if the bath solution
 is changed out,  the  schedule  must begin again at once every
 4 hour.  Documentation must be maintained on site to verify the
 number of exceedances and that the appropriate monitoring
 .-schedule is being followed.
       As stated in a previous.response, the final rule is clear
  that the emission limitations only apply during tank operation.
                                2-111

-------
 Also, surface tension monitoring would only be required during
 those times that the tank is operating.
      2.11.4.4  Surface Tension Measurement Technique.
      Comment;  Four commenters provided comments on the
 measurement technique for surface tension of tank baths.  One
 commenter  (IV-D-09) suggestedthat stalagmometers, which are used
 to measure surface tension, are not easy to use.  Another
 commenter  (IV-D-22) stated that stalagmometers are time consuming
 to set up, calibrate,  and make a determination, and that
 tensiometers,  an alternative measurement device,  are expensive.
 This commenter and one other (IV-F-01 [Altmayer])  indicated that
 alternative surface tension measurement methods that have
 equivalent accuracy and repeatability should be allowed.   One . .",.
 commenter (IV-D-05) suggested that Method. 306B reflect different
 sampling intervals  for different ranges of surface tension.
      Response:  Measurement of  the surface tension of  tank baths
 with a stalagmometer has  been performed during several source
 tests.   For those commenters  who believe that  the  stalagmometer
 is  not easy to use, section 3.1  of Method 306B states  "...The
 surface  tension of  the  tank bath may be  measured by using  a
 Preciser tensiometer, a stalagmometer,  or any  other device
 suitable for measuring  surface tension  in dynes per
 centimeter..."
      The time interval for measuring  surface tension has been
 changed  from once every 4 hour of  tank operation to a  progressive
 system that minimizes measurements if the proper surface tension
 is maintained.  This system is described  in section 3.2 of
Method 306B, and the maximum measurement  interval is once every
40 hours.
      For proper tank operation,  the EPA believes that  the surface
tension of the liquid in the tank must be maintained at 45 dynes
per centimeter or less.  The surface tension of the tank solution
will fall into one of two categories:  45 dynes per centimeter or
less and greater than 45 dynes per centimeter.   Different
sampling intervals for different ranges of surface tension would
                              2-112

-------
mean little and unnecessarily complicate the measurement
procedure.
     2.11.4.5  Testing and Monitoring T?«=-gui remits for Sources
Using Foam Blankets.'
     Comment:  Thirteen commenters  (IV-F-01  [Altmayer], IV-D-02,
IV-D-10, IV-D-20, IV-D-22, IV-D-25, IV-D-26, IV-D-37, IV-D-43,
IV-D-52, IV-D-57, IV-D-59, IV-G-01) provided remarks  regarding
the burden of hourly testing for sources using foam blankets.
Commenters also questioned the testing method and recommended
alternative testing schedules.
     Two commenters  (IV-D-37, IV-D-43) stated that hourly
monitoring is excessive.  Foam blankets that are used according
to operator's instructions are designed to  last 24 hour, as  long
as air  agitation does not surface  near the  anodes and freeboard
height  is adequate.  These commenters said  visual observation is
adequate  for determining  foam blanket effectiveness.  One  of
these commenters  (IV-D-37) indicated that the rule could specify
the volume and  frequency  for additions of foam blankets based on
the surface area of  the tank or  the rule  could specify that
manufacturer's  recommendations be  followed.
     One  commenter  (IV-D-22) stated that  hourly monitoring of
thickness  is an undue burden and can be  substituted  with the use
of automated .feeders and  ampere-hour monitoring.  Another
commenter (IV-D-57)  also  stated  that  this is burdensome and noted
that other monitoring within  the rule  is proposed on a daily
basis.   One  commenter  (IV-D-02)  questioned the  basis for
requiring hourly monitoring  for  foam blanket thickness.   One
 commenter (IV-D-52)  indicated that the same level of
 environmental  protection could be achieved through weekly
monitoring.
      Two commenters (IV-D-26,  IV-D-59)  suggested that it  is
 obvious that the EPA is trying to discourage the use of foam
 blankets; these commenters note that several States recommend or
-require foam blankets with less  testing and recordkeeping than
 that proposed by the EPA.  These commenters stated that the EPA
 should consider eliminating the performance test requirement and
                               2-113

-------
 should encourage foam blankets when wetting agents are not
 feasible.  Another commenter  (IV-D-37) also indicated that foam
 blankets are discouraged because of the burdensome monitoring
 requirements; this is inappropriate because foam blankets are
 effective pollution prevention techniques.  One commenter
 (IV-F-01 [Altmayer]) stated that the rule should not distinguish
 between foam blankets and wetting agents.  Two other commenters
 (IV-D-10, IV-D-43)  suggested the EPA increase the rule's
 flexibility by allowing suppressant foams and plastic balls as
 equivalent methods.
      One commenter (IV-D-37) recommended that the rule not
          ii-'!!;:,        ••            "       „ ' '' ''   'l|'"'
 require hourly monitoring of foam blankets when the tank is
 unattended^ (i.e.,  overnight).   Two other commenters (IV-D-25,
 IV-D-37)  also requested that exemptions be made for foam blanket
 measurement for those times when the tank is idle,  and time
 should be allowed for the foam blanket to build up after the tank
 has been left idle.
      Two commenters (IV-D-02,  IV-G-01)  requested that the EPA
 describe the  method for measuring foam blanket thickness on
 trivalent or  hexavalent chromium baths.   Two commenters  (IV-D-20,
 IV-D-43)  requested  that the rule clarify how sources  using foam
 blankets  should establish the  outlet  chromium concentration,
 especially  if  the tanks  are not  ventilated.   Another  commenter
 (IV-D-59) stated that  the stack  testing requirement is
 unreasonable as  it  can cost up to $10,000.   Two commenters
 (IV-D-20, IV-D-43)  suggested section  63.343(b)  specify that
 sources complying by use of a foam blanket should establish a
 foam blanket thickness as  a site-specific operating parameter as
 opposed to establishing  a  single  default limit.  These commenters
 note that if section 63.343(d), which requests  information on
 alternate air pollution  control devices that a  source may  be
 using to achieve compliance, applies to foam blankets, then foam
 blankets must be classified as air pollution control devices.
     Respc-nse;  The final rule is unchanged in  that sources using
 a foam blanket must conduct a performance test, and the initial
monitoring frequency is  once per hour.  However, as with wetting
                              2-114

-------
 agents,  if  there are no exceedances  of  foam blanket  thickness  for
 40  hours of tank operation,  monitoring  can occur once every
:4 hours.  If there are no exceedances for 40 more hours  of  tank
 operation,  monitoring can occur once every 8 hours.   The minimum
 frequency specified for foam blankets is once per 8  hour of tank
 operation.   As with wetting agents,  if  there is an exceedance,
 the original monitoring frequency of once per hour must  be
 resumed.  A subsequent decrease in monitoring frequency  can occur
 in accordance with the criteria outlined above.  If the  bath
 solution is changed out, the initial monitoring schedule of once
 per hour must be resumed.  The maximum frequency of 1 hour  is
 based on the EPA's experience that foam blankets can deplete
 quickly and must be closely monitored.
      The EPA believes that wetting agents are safer than foam
 ,,blankets because foam blankets present a potential safety hazard.
 The foam traps the hydrogen gas and chromic acid mist in the foam
 layer;  if these gases build up and a spark is generated, a
 hydrogen explosion will result.  As a means of encouraging
 wetting agent use over  foam blankets,  sources using wetting
 agents  do not have to conduct a performance test unless they want
 to set  a surface tension limit other than the default value of
 45 dynes/cm.  The EPA believes that the  increased compliance
 timeframes will allow sources that  currently use foam blankets
 the opportunity to explore the use  of wetting agents.   Sources
 that wish  to  continue using  foam blankets will be required to
 conduct a performance test,  and perform the necessary retrofit  to
 accommodate such a  test.   It  is also important  to note  that
  sources using a  foam blanket  in combination with a  wetting agent
  can demonstrate  compliance through  the  surface  tension  limit.
       Several  commenters were concerned  about  the technique for
 measuring  foam  blanket  thickness, and  the potential hazards
  associated with this measurement.   The  EPA does not believe that
 .it is necessary to  specify a procedure  because it  is simply a
  depth measurement.   Specifying a  technique may also hinder the
  development of  site-specific techniques to reduce worker
  exposure.   For  example, a plastic ball  of l-in.  radius  may be
                                2-115

-------
 marked at the center and allowed to float on the liquid in the
 tank.   The foam blanket measurement could then consist of
 ensuring that the foam is at the same level as the mark on the
 plastic ball.
                             •      ' '  '                 '        '•'  "-I
      The final rule specifies that sources using foam blankets
 are to establish foam thickness as the site-specific operating
 parameter,  and set the value of this parameter during the initial
 performance test.   Alternately, the default thickness of 1 inch
 could be used as the compliant thickness.   As such,  § 63.343(d),
 which specifies procedures to establish compliance monitoring for
 control methods not identified in the rule,  does* not apply.
 2.11.5  Monitoring Requirements--General
      Comment:  One commenter (IV-D-02)  questioned how the
 $14 million and $1.6 million cost estimates were reached by the
 EPA in predicting  the rule's economic impact on industry to meet
 the monitoring,  recordkeeping,  and reporting requirements  for
 compliance  with chromic acid emission standards for hexavalent or
 with surface tension levels for trivalent  chromium.   One
 commenter (IV-D-28)  found it unreasonable  that in the EPA's cost
 estimates,  monitoring and recordkeeping constitute 28 percent of
 the total annual cost.   A commenter  from the aerospace industry
 (IV-D-47) is concerned that the small businesses  that supply
 products  to the aerospace industry will not  be able  to afford the
 technology  to meet  the standards.  According to this  commenter,
 the impact  of the  regulation will  be  greater than the EPA's
 estimate  unless  the  standard is adjusted and the  monitoring and
 recordkeeping requirements  are  reduced.  This commenter also
 stated that the  burden of monitoring, reporting,  and
 recordkeeping is underestimated and it  will  likely require
 600  hours per year per source.   Another commenter (IV-D-58)
 suggested that  the complicated  monitoring, recordkeeping, and
 stack measurement methods place an unreasonable economic burden
 on  small  sources.
     Response;   The burden  of monitoring, reporting,  and
 recordkeeping are estimated by  the EPA  based  on the number of
measurements and records an owner would have  to take, and the
                              2-116

-------
 time  to  file  reports  to  the permitting agency.  The  estimates
 presented in  the  proposal  preamble  also  include the  burden of  the
"initial  performance test,  and for a certain  number of  facilities,
'repeat tests  in the event  of  failure.
      As  described in  this  section and  in Section  2.13,  the
 monitoring, reporting, and recordkeeping requirements  in the
 final rule are less than what were  proposed.  The EPA  reduced  the
 burden specifically due  to the many area sources  that  will be
 subject  to these  standards.   The ongoing yearly cost of the final
 monitoring,  reporting, and recordkeeping is  approximately
 159,100  hours (or $3.5 million) for hard chromium electroplaters,
 289,300  hours (or $6.4 million) for decorative chromium
 electroplaters, >and 70,200 hours (or $1.6 million) for chromium
 anodizers.  The EPA believes that the monitoring, reporting, and
 recordkeeping has been reduced to the extent possible  while still
 providing the EPA with the ability to determine a source's
 compliance status.
      Comment:  One commenter  (IV-G-01) requested that the EPA
 clarify the monitoring to be performed when hard chromium
 electroplating and chromium anodizing are performed in the same
 tank.  This  commenter stated that in such situations the tank
 should be considered in compliance if the surface tension limit
 is met.
      Response;  As clarified in  the final rule,  the monitoring
 requirements are  specific to the type of control  technique used,
 not  the  type of  tank operation.  If hard chromium electroplating
 and  chromium anodizing are performed in the same tank, and a
 wetting  agent  (or combination  foam blanket/wetting  agent) are
 used to  control  emissions from the bath during both modes of
 operation, surface tension monitoring is appropriate.  However,
 in accordance  with the  final  rule, an initial performance test
 -would have to  be conducted to demonstrate that the  emission limit
 vis met  when  hard chromium electroplating  is taking  place.
 :.      Comment;  Three commenters (IV-D-10, IV-D-42,  IV-D-43)
 suggested that the permitting agencies  be allowed to  identify
 suitable monitoring  for affected sources  in their jurisdiction.
                                2-117

-------
 One of these commenters (iv-p-io)  recoinmended the frequency of
 monitoring be based on permit review or a subcategory-specific
 basis and also suggested that the requirements could be subject
 to one-time public and Federal comment.
      One commenter (IV-D-58)  recommended the EPA develop a
 two-tiered approach to monitoring requirements:   one tier would
 be low-burden for those sources that do initial  performance
 testing,  O&M plans,  and maintain compliance,  and the second tier
 would be more rigorous for sources that need moire scrutiny or
 could be used as  an enforcement option for the EPA.   This
 commenter also suggested facility certifications for small
 facilities through industry-developed recommended-best-practices
 and automated monitoring or recordkeeping could  be employed in
 lieu of  the proposed requirements.   This commenter recommended
 that technical and compliance assistance for small businesses  be
 utilized under section 507 of the  Act.
      Response;  The  EPA recognizes that  many State and  local air
          jn, !             ,                  • sir'•!            '
 pollution control  agencies have established monitoring,
 reporting,  and recordkeeping  requirements for the sources covered
 by the proposed standards.  However,  the EPA is  required to
 establish the  minimum monitoring,  reporting,  and recordkeeping
 requirements  in the  Federal rule that are necessary  to  determine
 a  source's  compliance status  at any given time.   State  and local
 agencies  reserve the  right  to require other types of monitoring,
 reporting,  and recordkeeping  in addition to the  requirements of
 the  Federal rule.
     , The  final  rule contains  fewer  monitoring, reporting,  and
 recordkeeping  requirements  than the proposed  rule and should
 therefore alleviate some of the burden on small  sources.   The  EPA
believes  that any  tiered compliance requirements  would  further
 complicate  the  rule, making compliance by small  sources more
difficult.  In  regard  to the  EPA actions  for  small businesses
under section 507, the EPA is developing  an enabling document  to
accompany the rule, and the Small Business Administration  is
developing a pamphlet  to assist  small businesses.  Both of these
                              2-118

-------
documents are intended to help sources understand the
requirements of the rule and implement the rule.
     Comment;  Two commenters  (IV-D-20, IV-D-43) suggested that
all performance test" requirements establishing site-specific
operating parameters should be in one section of the rule, such
as section 63.343.  One commenter  (IV-D-24) suggested a specific
set of monitoring requirements be developed to address systems
using multiple technologies.
     Response:  The final rule has been reorganized so that
testing and monitoring requirements have been made more clear,
and also addresses sources that are using multiple technologies
to comply with the rule.  All  compliance monitoring requirements
are contained in  § 63.343.  All performance testing requirements
are in § 63.344.
     Comment;  One commenter  (IV-D-61) pointed  out that although
the term "operation of the tank"  is referred to throughout  the
monitoring  requirements,  the  term is  not  defined.  This term
could have  several meanings,  such as  when the tank  is  heated to a
specific temperature, when it is  charged,  or when it  contains
electroplating material.  The commenter  recommended that  tank
operation be defined  as  when  the  contents of  the tank reaches a
temperature greater than room temperature.
     Response:   In the  final  rule,  tank  operation is defined as
any  time that current and/or  voltage  is  being applied to  a single
 chromium electroplating  or  chromium anodizing tank  or to  any one
 of a series of  such tanks that are ventilated to a  common control
 device.
 2.12  SELECTION OF TEST METHODS
 2.12.1   Test Methods  306 and 306A
      Comment;  One commenter  (IV-D-17) noted that Method 306
 presents difficulties when the sample contains significant
 quantities of dissolved salts.  To alleviate this problem, the
 commenter suggested the EPA modify Method 306 to allow use of a
rnitric acid/hydrogen peroxide matrix in lieu of the
 sodium hydroxide/bicarbonate matrix.  Another commenter  (IV-D-53)
 indicated that sodium hydroxide causes analytical difficulties
                               2-119

-------
 and recommended using a 0.1 N solution.  This commenter also
 recommended that if a nitric acid solution is used,  it should
 also be used for calibration standards to minimize matrix
 effects.  The test methods in 40 CFR part 266,  appendix IX and
 SW846 Method 6010 should be allowed by the EPA as an equivalent
 method (IV-D-17).
      Comment;   Two commenters (IV-D-10, IV-D-24)  indicated that
 the methods proposed in the rule for determining chromium
 emissions, do not have detection limits low enough to determine
 compliance with the standard.  Both of these commenters indicated
 that Method 306 is not appropriate for determining emissions at
 very low levels and that the EPA provides no guidance on
 extending the  duration of the test run (increasing the sample
 volume of air  collected)  or limiting the volume of probe wash
 (less solvent)  to decrease the level of detection for the method.
 The other commenter (IV-D-24)  pointed out that  the proposed
 emission standards for hard chromium sources are  below the
 detection limits  of the proposed ICP analytical method.   In fact,
 using ICP with a  concentration of 50 ug/L as a  minimum
 requirement, the  method detects  chromium concentrations  at values
 of  8  times greater than the standard.   This  commenter also
 pointed out that  the  minimum detection limits for ICP and GFAAS
 are not within the linear ranges  for these methods.
      Another commenter (IV-D-53)  suggested use  of an  ICP/MS
 system as  an alternative  method  for  detecting low concentrations.
      Response:  One commenter (IV-D-17)  referred  to the
 difficulty of  analysis  when the  sample contains significant
 quantities  of  dissolved salts, and suggested using nitric
 acid/hydrogen peroxide  matrix in  lieu  of  the sodium
 hydroxide/bicarbonate matrix.  If the  dissolved salts make
 analysis difficult, it  has  not made  it  difficult  enough  that the
 labs  that have done our analyses  (and  written the  analysis
methods) have felt  it necessary to point  this out  as a problem.
 If a nitric acid/hydrogen peroxide matrix is  superior to the
 sodium hydroxide/bicarbonate matrix  for analysis purposes, the
 commenter may verify this possibility by performing a Method 301
                              2-120

-------
evaluation.  The comment on the 0.1 N solution is difficult to
respond to because we use 0.1 N sodium hydroxide already and the
reference in the letter  (IV-D-53)  was not specific.  Other
commenters recommending nitric acid in the analysis may also use
Method 301 to verify the validity of their suggestions.  The
analytical method in 40 CFR part 266 (ICPCR) will be allowed for
hexavalent chromium analysis.
     Other commenters stated that the "methods proposed for
determining emissions do not have detection limits low enough to
determine compliance with the standard.  This comment doesn't
make any sense because EPA used these analytical methods to set
the standard  (see Docket No: A-88-02 for an explanation of why
the colorimetric method was dropped in lieu of atomic absorption
graphite furnace  (AAGF)  or ion chromatography using a post column
reactor  (ICPCR)).  The reasons for using AAGF and  ICPCR was to
increase the  accuracy of the analytical technique  and still
maintain a reasonable sample time.  The suitability of the
detection  limits  can be  illustrated with the  following example
when using AAGF  on an outlet location  (ICPCR  is  even more
sensitive  than AAGF).
     A 2 hour sample would collect about 90 cubic  feet of  sample.
This 90  cubic feet is equal  to 90  cubic  feet  times 0.02832  cubic
meters per cubic foot which  equals 2.5488  cubic  meters.   If  the
concentration of total  chromium  in this  2.5488  cubic meters  of
sample were  at  the limit of  the  standard for  large hard  chromium
platers  (0.013  mg/dscm), the sample  would  contain:
      2.5488 m3 (0.013 mg/dscm)  =
33.1 micrograms
of total chromium
      The sensitivity of the AAGF method is 1 microgram per
 1,000 milliliters.  A sample containing 5 to 10 micrograms would
 be quantifiable if the total sample collected was contained in
 1 liter or less of solution.  The maximum quantity of sample
 solution would probably not exceed 800 milliliters, so 4 to
 8 micrograms collected would be sufficient for a reliable
                               2-121

-------
 analysis at the level of the standard, the 33.1 micrograms of
 sample collected is 4 to 8 times the minimum catch for
 quantifiability.
      Perhaps the commenters were referring to the
 diphenylcarbazide colorimetric method, or the ICP method.  The
 colorimetric method was replaced in 1991 because it was not
 sensitive enough for some of the more recently developed control
 systems.   The ICP method which is mentioned in Method 3Q6 is good
 for large concentrations of chromium, such as that collected at
 the inlet to the  control system.  It is not suitable for
 measuring outlet  concentration that are very close to the
 standard.  The point about the length of the test is a valid one,
 and we will specify a 2 hour test time at a sampling rate of
 about 0.75 dscfm.
      A commenter  also stated that the minimum detection limits
 for ICP and AAGF  analysis methods were not within the linear
 ranges for those  methods.   The detectable limits do not have to
 be  within the  linear range.   The quantifiable limits,- however,
 are within the linear range.   As mentioned before,  the ICP method
 is  acceptable  only  for high concentration samples.   The AAGF
 analysis  can detect  l microgram per liter of  total  chromium.   A
 concentration  of  5  to 10 micrograms per liter is quantifiable.
 It  is  not likely  that the  impinger  solutions  and sample rinse
 collected following  sampling  would  exceed 1 liter.   The combined
 impinger  solutions and sample rinse would probably  be  between
 500  and 800 milliliters.   It  is  also not  likely  that less  than
 4 to 8 micrograms of  chromium would be  caught  in a  2 hour  sample
 that was  taken at an  extremely well  controlled cutlet.   If  the
 sample catch is less  than 4 to 8 micrograms of chromium (per
 800  milliliters of sample solution),  or if the chromium cannot be
 detected  in a sample,  the source would be well below the emission
 limits established in  the standard.
     Comment:  Five commenters  (IV-D-10, IV-D-24, iv-D-37,
IV-D-49, IV-D-53)  suggested changes or additions for Methods 306,
306A, and 306C. [Note;  There is no method 306C  in the proposed
                              2-122

-------
rule.  The commenter appears to be referring to
section 5.1.1.2--Velocity Traverse section in Method 306A.]
     One commenter (IV-D-24) requested that the EPA provide a
margin of error to reflect the precision of the sampling methods
and analytical test methods.  One commenter  (IV-D-10) suggested
the addition of a colorimetric analysis to Method 306 to provide
hexavalent chromium in addition to total chromium.  This would
prevent a source that is subject to a State emission standard
based on hexavalent chromium from having to perform two source
tests.  One commenter  (IV-D-37) made the following suggestions
for Method 306:   (1) the method should state that probe heat is
not necessary;  (2) the method should state that use of probe and
nozzle brushes  during recovery is unnecessary; and  (3) the method
should suggest  specific recovery procedures  that minimize sample
volume.
     One  commenter  (IV-D-53) made several  remarks concerning
Methods  306 and 306A.  For section  5 of Method 306,  the  commenter
suggested that  trip blanks be used  rather  than reagent blanks
because  the trip blanks provide more  indication  of  contamination
problems.  The  results of  blank runs  should be reported  if  the
method does not specify  a  maximum acceptable blank  level.   The
 commenter also  suggested using more than 10 ml of acid to acidify
 300 ml of 0.1 N sodium hydroxide.   Also,  the amount of dilution
 allowed in bringing a sample into calibration range should be
 limited.  For section 7  of Method 306,  the commenter recommends a
 standard check every ten injections,  or every set,  whichever is
 greater.  According to the commenter,  a quality control check
 standard that is closer to the limit of detection would detect
 nonlinearities on the lower end of the calibration curve; a
 factor of 20 times the limit of detection may be more appropriate
 than 100 times the limit of detection for a quality control check
 standard.  For Method 306A, the commenter recommends that
 flexible tubing containing phthalate plasticizers not be used for
 chromium sampling trains as it is mechanically sticky and
 possibly reactive.  One commenter  (IV-D-10) indicated that
 Method 306A would seem to  result in only minimal cost savings
                               2-123

-------
 compared to Method 306 and should be Devaluated before being
 approved.
      Another commenter  (iv-D-49) suggested that Method 306C
 should contain provisions for a 33 percent increase in pressure
 in addition to the provisions for a 33 percent decrease in
 pressure that the method currently contains.  This commenter also
 indicated that some newer control devices are basically
 extensions of the hood ducting and may not have one diameter
 between the hood and control device for insertion of the pitot
 tube; an alternative velocity measurement point for Method 306C
 should be specified.
      Response:   One commenter requested that the EPA provide a
 margin of error to reflect the precision of the sampling methods
 and analytical  test methods.   It is the EPA's opinion that
 providing a margin of error to reflect  the precision of the
 sampling methods and  analytical test methods would cost money,
 take time,  and  would  be nonproductive in view of. the fact that
 future testing  and analysis will be performed in a manner
 consistent  with that  used to  gather the data to establish the
 standard.   Another commenter  also suggested adding a colorimetric
 method to the analysis section of Method 306.   The EPA has found
 that  the diphenylcarbazide colorimetric method the commenter is
 probably referring to does not have  the sensitivity required to
 measure the  low chromium  concentrations found  in  the outlet  stack
 gases.  However,  the  EPA  has  added the  use of  the  ion
 chromatography with a post-column reactor  as an acceptable
 analytical method for determining the hexavalent chromium
 concentration in  the  exhaust  stack.
      One commenter made the following suggestions for Method  306:
 (1) the method should state that probe  heat is not necessary;
 (2) the method should  state that use of probe and nozzle brushes
during recovery is unnecessary; and  (3)  the method should  suggest
specific recovery procedures  that minimize sample volume.  In
reviewing these comments, the EPA agreed that any mention of
probe heat and probe and nozzle brushes for sample recovery will
be deleted from the method.  Regarding specific recovery
                              2-124

-------
procedures to minimize sample volume, the EPA believes that the
analytical methods have sufficient sensitivity to detect small
quantities of chromic acid collected and no procedures are
necessary to minimize sample volumes.  On the contrary, the EPA
believes such procedures could increase the risk of missing part
of the sample collected.  The EPA checked the sensitivity of the
methods against 1 liter of collected sample solution in order to
determine if longer sample times were necessary to ensure
sufficient sample collection.  The results of this sensitivity
indicated that a 2-hour sample collects a sufficient quantity of
chromic acid mist for accurate determination of concentration in
mg/dscm.
      Another commenter  indicated  that Method 306A would result  in
minimal cost savings  to the  industry.   For the  response to  this
comment, please refer to  section  2.12.3.1.   In  addition,  another
commenter  (IV-D-49)  suggested that Method 306C  should contain
provisions  for a  33  percent  increase in pressure  in addition to
the provisions for  a 33 percent  decrease in  pressure that the
method currently  contains.   Please  refer to  the response  to
 comments  on Method  306A in section  2.11.2.3  for the response to
 this comment.   The  monitoring equipment (magnahelic gauges and
 pitot tubes)  are  used to assess  system operation,  and accuracy is
 not as important  as reproducibility.  A magnahelic gauge is easy
 to adjust by securely mounting the gauge and adjusting the gauge
 to "zero."
      Comment;   Three commenters  (IV-D-10, IV-D-14, IV-D-47)
 requested that GARB Method 425 be evaluated for equivalency, and
 if determined to be equivalent,  be identified as such in the
 rule.  These commenters also stated that sources that have
 performed this test should not have to retest.  Two of these
 commenters  (IV-D-14, IV-D-47) and one other  (IV-D-15) pointed  out
 that ion chromatography  is the preferred analytical method  and
 should be cited as equivalent in the rule.  One of  these
 commenters  (IV-D-10) indicated that Method 306A is  not acceptable
 for  demonstrating compliance with California chromium
 electroplating emission  limits.
                               2-125

-------
      Four commenters  (iv-D-06, IV-D-io, IV-D-14, IV-D-27) asked
 whether retesting will be required if sources have done previous
 performance testing using 306, 306A, or an equivalent test
 method.
      Response;  The GARB Method 425 train is almost identical to
 the Method 306 train except that a filter is used between the
 third and fourth impingers.  The catch on these filters is
 usually nondetectable.  There is no need to evaluate the GARB
 sampling train for equivalency.  The analytical method used after
 the sample is collected needs tobe considered.  In the early
 days of the NESHAP development, the analysis for hexavalent
 chromium using either the Method 306 train or the GARB 425 train
 was the diphenylcarbazide colorimetric method.   As control device
 efficiency improved,  the colorimetric method was not sensitive
 enough toensure accuracy without going to excessively long
 sampling times.   The  acceptability of previously performed GARB
 Method 425 tests will depend upon the analysis  rather than the
 sampling train or the sampling procedure.   For  GARB tests using
 the colorimetric method for analysis,  the  sample catch will have
 to  be  at least 5 times the minimum detection limit  to be
 acceptable.   The GARB method is sensitive  to 0.33 micrograms per
 100 milliliters.   To  ensure that  the  data  are accurate to within
 ± 10 percent  accuracy,  the concentration at  the detector would
 have to  be 5  times  0.33 micrpgrams  per  100 milliliters or
 1.65 micrograms  per 100 milliliters.  Because the sample is
 diluted  to 1/2 the  original  concentration prior to  analysis,  the
 sample solution  recovered  from the  train must contain
 3.30 micrograms per 100 milliliters, or 0.033 micrograms  per
 milliliter.  If  the sample meets these concentration
 requirements, the colorimetric method of analysis for  hexavalent
 chromium, and consequently,  the results, may  be accepted.  If  the
 sample does not meet these criteria, it is not possible to assure
 ±10 Percent accuracy.   If another type of analysis  is used,  such
as atomic absorption graphite furnace for total chromium, or ion
chromatography with a post column reactor for hexavalent
chromium, there should be no problem using GARB Method 425.
                              2-126

-------
     One commenter said that ion chromatography was the preferred
analytical method, and this is true if this means ion
chromatography with a post column reactor (ICPCR)•   I°n
chromatography with a post column reactor is the preferred
method, but finding a laboratory with this analytical capability
is difficult.  Nonetheless, the final rule will be revised to
include ICPCR as an acceptable analytical method.
     The commenter who stated that Method 306A was unsuitable for
demonstrating compliance with the California standard was
incorrect.  The commenter suggested that Method  306A should be
thoroughly compared with Method 306 before it is officially
approved.  Method 306A has been carefully compared against
Method 306 and is, in fact, suitable  for showing compliance with
the  California standard.
     Four commenters asked whether retesting will be required if
sources have done previous performance  testing with Method 306,
306A or an equivalent test method.  Tests using  other  equivalent
sampling methods  for chromium emissions should be  reviewed by the
Agency prior to determining whether or  not another test  should  be
performed in accordance with  §  63.344 (c) (4) of subpart N and
§  63.7 (f) of subpart A.   Facilities that use Method  306A prior  to
December  1991 will have to retest.
2.12.2 Demonstrating  Compliance  When Using Fume Suppressants
     Comment;   Two commenters (IV-D-22, IV-D-59) questioned
whether facilities will have  to measure emissions  to support
meeting the  emission standard for fume  suppressants.   Language  in
 the preamble to the  proposed  rule on pages 65772 and 65774
 implies that facilities using fume suppressants  are  expected to
 achieve a specific outlet concentration.
      Response;   The  final rule includes a dual standard for
 sources performing decorative chromium electroplating and
 chromium anodizing operations.   These sources must either:
 (1) discharge no greater than 0.01 mg/dscm of total chromium to
 the atmosphere, or (2)  operate the electroplating or anodizing
 tank such that the surface tension does not exceed 45 dynes/cm.
 With the first alternative, a source must conduct a source test
                               2-127

-------
 in accordance with the methods identified in the rule to
 demonstrate that the emission limit of 0.01 mg/dscm is not being
 exceeded.  It is assumed that a source complying with the first
 alternative is using" an add-on control device or a foam blanket.
 With the second alternative, a source must measure surface
 tension po demonstrate .initial compliance; ongoing compliance is
 also demonstrated through surface tension measurement in
 accordance with the monitoring procedures in § 63.343 (c)  of the
 final rule.   In this latter case,  an initial performance test is
 not required if the criteria of § 63.343(b)(2)  are met.   That is,
 the source is a decorative chromium electroplating or chromium
 anodizing tank,  a fume suppressant containing wetting agents is
 used to limit chromium emissions,  and the owner or operator of
 the source accepts 45  dynes/cm as the surface tension value that
 demonstrates compliance with the standard.  The rule  is  written
 as  an either/or alternative; that is,  only the emission  limit or
 the surface  tension limit must be met.   It was not  the EPA's
 intent to require a source test if the  criteria of  §  63.343(b)(2)
 are met.   The EPA recognizes that  for some decorative chromium
 electroplaters and chromium anodizers  source testing  is  not
 feasible  because ventilation stacks do  not exist.   Therefore,  the
 surface tension alternative is provided.
      One  exception to  the above conditions is the case in which
 sources using wetting  agent-type  fume suppressants  want  to
 establish a  surface tension limit  other than the 45 dynes/cm
 limit  in  the  final  rule.   In such  a case,  a  source  would  have  to
 conduct a performance  test  to establish the  alternate surface
 tension value  that  corresponds  to  the emission  limit of
 0.01 mg/dscm.
     Sources performing hard chromium electroplciting that  use a
 fume suppressant  (either  a  wetting  agent,  a  foam blanket,  or a
 combination) to comply with the standard will have  to conduct an
 initial performance test  to demonstrate compliance  with the
standard.  The EPA did not  fully evaluate  the use of fume
suppressants to reduce emissions from hard chromium
electroplating tanks because  only a limited number  of operations
                              2-128

-------
used this technique.  Also, there were safety concerns associated
with their use, as well as the fact that fume suppressants are
not universally applicable to all types of hard chromium plating
operations.  Therefore, all hard chromium electroplating tanks,
regardless of the type of control technique employed  (e.g., APCD,
fume suppressant, or combination) must conduct an initial
compliance test.  All hard chromium electroplating tanks will
have ventilation stacks to comply with OSHA requirements.  Thus,
source testing will not be as burdensome as it would be for
decorative chromium electroplating and chromium anodizing  tanks
that may not have ventilation systems.  Ongoing compliance
monitoring will be the same for  all tanks using fume
suppressants,  requiring sources  to monitor bath surface tension
if  a wetting agent  is used and foam blanket thickness  if a foam
blanket  is used.
2.12.3   Demonstrating  Compliance When Using Add-On  Controls
     2.12.3.1   Impacts of  Stack  Testing.
     Comment:   Five commenters addressed  the  burdens  associated
with stack testing.  Two  commenters  (IV-D-06,  IV-D-40)  stated
that stack testing is  burdensome and  can  cost up  to $8,000
 (IV-D-06)  or more than $10,000  (IV-D-40).   Onecommenter
 (IV-D-59)  stated that  stack testing costs could close some
decorative chromium electroplating sources.   Another commenter
 (IV-D-47)  suggested that  stack testing be minimized because not
 all control systems were  designed to allow easy measurement of
 outlet concentration,  and alterations that may be necessary to
 enable the performance of the testing (e.g.,  stack may need to be
 extended)  could be significant.   Onecommenter (IV-D-53)   stated
 that the proposed method requires burdensome, yearly testing and
 indicated that the process emission rate (mg/amp-hr)  can  be used
 to demonstrate the correlation between compliance and the
 applicable emission limit.  This commenter states that with a
 mg/amp-hr limit, 'compliance can be easily verified without a
 source test.  Another commenter  (IV-D-42) stated that the EPA
 needs to specify the initial performance test to demonstrate
 compliance and set enforceable  operating parameters and to
                               2-129

-------
 evaluate the economic impact of such requirements because these
 requirements may significantly hinder small sources.
      Response;  Based on the comments received at proposal, there
 appears to be some confusion surrounding the stack testing
 requirements.  First, an annual test is not required.  All
 sources are required to conduct an initial performance test
 within 18.0 days after the compliance date in accordance with
 §63.7 °| the General Provisions to part 63, unless the criteria
 of § 63.343(b)(2)  are met.  After the initial test, ongoing
 compliance is demonstrated by performing compliance monitoring in
 accordance with §  63.343(c)  of the final rule.   In terms of
 modifications necessary to conduct a performance test,  the method
 in the rule requires that the stack height be 2.5 times the stack
 diameter,  not 8  times as thought by commenter IV-D-47.   The EPA
 has estimated the  cost to conduct performance tests at  all sizes
 of sources and this cost was considered in selecting the
 standard.   The EPA recognizes that testing may be a burden for
 some sources,  and  has pursued methods to reduce this burden.   For
 example,  in developing the proposed rule,  the EI?A found that  the
 analytical  methods  for measuring chromium emissions were
 expensive,  and the  number of laboratories  qualified to  do the
 testing  limited.  Therefore,  the EPA developed  and proposed
 Method 306A,  which  is simpler and less  expensive  that other
 methods.  Based on  a  review  of stack testing costs,  the EPA has
 determined  that the cost for stack testing is approximately
 $4,500 per  stack.   The  EPA is aware of  firms that  may charge  as
 much as  $8,000 to $10,000 per stack.  However,  it  is  up to  the
 owner of the  electroplating  facility  to  ensure  they are paying a
 fair competitive price  for the stack  test.   In  addition,  the  EPA
 is aware of the fact  that even $4,500 per  stack could be
 burdensome to some  small plating shops.  Furthermore, if  the
 initial test reveals  that the control device is inadequate, and
modifications must be made to the system,  the cost of an
additional test is  incurred.  To respond to  these  concerns, the
EPA developed the alternative Method 306A.  The cost  of
fabricating a Method 306A sampling train is about  $600.   A pitot
                              2-130

-------
tube (which can later be used for enhanced monitoring),  a
manometer, and chemicals should run another $50.  Three sampling
runs will take about 15 hour, which at $12/hour would cost $180.
The analysis of three samples and a blank would cost about
$60/sample for a total chromium analysis and would cost about
$80/sample for a hexavalent chromium analysis.  The cost of
having the company perform its own test is:
     Sampling train:               $600.00
     Pitot, manometer, chemicals:   $50.00
     Labor                         $180.00
     Analysis                      $320.00
     Total                        $1150.00
     The  company also gets to keep the equipment.  Additional
testing then costs $500 for labor and analysis plus about $10 for
chemicals, or $510.
     If the company does its own testing and performs only one
test, it  saves $3,350 over having a testing firm do the work.   If
two tests are needed, the company saves the original $3,350  plus
$3,990 or $7,340.  If the company has more than one stack, it
would definitely pay the company to do its own  testing.  A major
airline company used Method  306A to test  its plating tanks and
saved over $33,000 in source testing  costs.
     When the final rule is  in place, there will probably be some
companies that will rent Method 306A  equipment.  Stack  testing
firms may also use Method 306A for sampling sources and still use
their isokinetic  equipment.  This should  be less expensive for
the plater because Method 306A is less  labor  intensive  and easier
to perform that Method  306.  There may  also be  equipment
manufacturing companies that will sell  assembled trains ready  for
use.
     Another way  to  alleviate  the expense of  testing  so many
platers  would be  for one of  the  trade associations to set up its
 own source sampling group,  dedicated to sampling the
 electroplating and anodizing industry.   A competent individual
 well versed in chromium sampling and knowledgeable of the methods
 could  head up such a group  and save the industry considerable
                               2-131

-------
 money.   To further assist sources in using the Method 306A
 sampling train,  the EPA has developed a video entitled
 "Construction and Operation of the EPA Method 306A Sampling Train
 and Practical Suggestions for Monitoring of Electroplating and
 Anodizing Facilities."   This tape will be available in
 January 1995  for a nominal fee through North Carolina State
 University, Registrar,  Environmental Programs,  Box 7513,
 Raleigh,  NC  27695-7513.   The telephone number is  (919)  515-4659;
 the fax number is (919)  515-4386.
     Also,  the EPA believes that  extending the compliance  time  to
 1  year  for decorative chromium electroplaters and  2 years  for
 hard chromium electroplaters and  chromium anodizers will provide
 sources with  more time  to obtain  capital for both  control  device
 purchase and  performance  testing.   Finally,  § 63. 344 (b)  of
 subpart N allows sources  to fulfill the initial compliance test
 requirements  for this standard by submitting compliance  test
 results from  tests  conducted at startup in accordance with State
 operating permits.   The test must  have been conducted -using the
 EPA- approved  test methods and procedures in subpart N.  This
 provision may offer some  relief to sources that have recently
 conducted performance tests.
     One  commenter  (IV-D-53)  also  suggests that an annual  source
 test is nqcessary to demonstrate  compliance with a
 concentration-based standard,  whereas  compliance with a mg/amp-hr
 standard  can  be  demqnstrated without a performance test.   As
 explained above,  an annual  test is  not necessary.   Also, the
 format  of  the standard is  irrelevant to performance test
 requirements.  For  the reasons  described in Section 2.7,
 Selection of  the  Format of  the  Standard,  the  EPA has  promulgated
 a rule  expressed  in mg/dscm.  However,  even  if  the  rule had been
 expressed in mg/amp-hr,  the  EPA would  have  required a  source test
 to demonstrate initial compliance and  parametric monitoring to
demonstrate ongoing  compliance.  Thus,  the  format  of  the standard
has no bearing on the source  test requirements.
     In regard to the comment that  the  EPA should specify the
initial performance test and  set enforceable operating
                              2-132

-------
parameters, the final rule explicitly requires an initial
performance test except as allowed by § 63.343(b)(2) and (3).
The EPA has also specified which operating parameters correlate
to compliance with the emission limit.  For example, the pressure
drop is directly related to the performance of a composite mesh-
pad system.  The EPA does not, however, think it is appropriate
to identify the operating parameter value.  The final rule allows
sources to set the operating parametef value during the initial
performance test because each control system is different
depending on its design, age, operation, and condition.  Rather
than hindering small sources, the EPA believes that this
condition  lends flexibility to all sources.  The final rule  does
specify default compliant ranges.  For example,  a  source is  in
compliance if the measured pressure drop value is  within ± 1 inch
of water column from the average value established during the
initial performance  test, and a source is  in  compliance  if the
measured velocity pressure value is within ±  10  percent  of the
average value established during the  initial  performance test.
These  ranges have been specified because  EPA  recognizes  that it
would  be difficult  to  set a  range  of  compliant  operating
parameter  values  during one  performance  test.   For those owners
 or operators  that wish to  establish their own range of compliant
 operating  parameter values,  the final rule allows them to  do this
 by conducting multiple performance tests.   The following comment
 and response  discuss the importance of compliance testing for all
 sources,  regardless of size.
      Comment:   One commenter (IV-D-57)  suggested that at large
 sources where many tanks have identical design,  a compliance test
 protocol that requires a source to test a percentage of the
 comparable tanks may be more practical.  Two commenters (IV-D-27,
 IV-D-36)  requested that the EPA allow a source to test one
 representative unit to demonstrate compliance for several similar
 units.  Another commenter (IV-D-40) suggested that the EPA
 certify commonly used control devices based on the EPA's test
 methods to avoid compliance  testing by small businesses.
                               2-133

-------
      Response;  The EPA does not believe that representative
 testing is adequate to demonstrate that each source is in
 compliance with the standard.  Therefore, the final rule requires
 that each source conduct a performance test to demonstrate
 compliance with the emission limit, unless the source is
 performing decorative chromium electroplating or chromium
 anodizing and is complying with the surface tension limit.   Based
         :    ,             -      ,,        " •  .in	    „         , ,   "  ii.,,,1' '
 on the EPA's experience testing control devices in use in this
 industry,  a device's performance will vary based on its age,  its
 condition, and the manner in which it is maintained and operated.
 A performance test is also necessary for sources to establish the
 Val!3es f3* the °Peratin£r parameters that will be monitored to
 establish continuous compliance.   For example,  a source operating
 a composite mesh-pad system will  have to establish the pressure
 drop value that corresponds to the emission limit.
      2.12.3.2  Control Systems Controlling Multiple Tanks.
      Comment;  Two commenters (IV-D-36, IV-D-44)  requested that
 the rule provide guidance on how  to verify compliance  when both
 chromium anodizing and hard chromium electroplating tanks are
 vented toa common control  device.   Three commenters (IV-D-17,
 IV-D-47, IV-D-61)  pointed out that  the regulation does not
 account for the situation in which  chromium electroplating
 sources share a ventilation system  with nonchromium sources that
 could introduce dilution  air;  these commenters  questioned where
 the emissions should be measured.   Three commenters (IV-D-47,
 IV-D-52, IV-D-61)  noted that it is  extremely difficult to
 reconfigure the existing  system such  that  only  the  air from
 chromium electroplating or  anodizing  will be tested.   One of
 these commenters  (IV-D-61)  indicated  that  tanks  cannot be
 selectively turned  off to accommodate individual  testing; some
 air flow would  still come from each tank.
      Response;  There are basically two situations  involving
multiple tanks manifolded to one control system:   (1)  the
multiple tanks  include a chromium electroplating  or chromium
anodizing tank among other  tanks not  affected by  the rule;  or
 (2) the multiple tanks include chromium tanks performing
                              2-134

-------
different operations (e.g., electroplating and anodizing)  or the
same type of tanks subject to different emission limits (e.g.,  a
new hard chromium electroplating tank with an existing small,
hard chromium electroplating tank).   The final rule includes
compliance provisions for either of these situations.
     In the first situation, a chromium electroplating or
anodizing tank is tied to a control system that is also
controlling tanks that are not covered by this rule.  In such a
case, the nonaffected tanks may be turned off, but air flow will
still result from each tank thus diluting the chromium measured
at  the outlet from  the control device.  The final rule contains a
new procedure for estimating emissions for these special cases. ^
     When multiple  affected sources performing the same type of
operation and subject to  the same emission limitation are
controlled with  a common  add-on air pollution control device that
is  also  controlling emissions  from sources not affected by these
standards,  the following  procedures should be followed to
determine compliance with the  applicable  emission  limitation:
      1.  Calculate  the  cross-sectional area  of  each  inlet  duct
 (i.e., uptakes from each  hood)  including  those  not affected by
the standard.
      2.  Determine  the  total  sample  time  per test  run by dividing
 the total  inlet  area from all  tanks  connected to the control
 system by  the total inlet area for all ducts associated with an
 affected source, and then multiply this number by 2 hours.  The
 calculated time  is the minimum sample time required per test run.
      3.   Perform Method 306 testing and calculate an outlet mass
 emission rate.
      4.   Determine the total ventilation rate from the affected
 sources by using equation 1:
                               IDA-
                               IATOTAL
                                                               (1)
 where VRtot is the average total ventilation rate  in dscm/min  for
 the three  test runs as determined at  the outlet by means  of  the
                                2-135

-------
 Method 306 testing; IDA± is the total inlet area for all ducts
 associated with affected sources;  lAp^^ is the sum of all
 inlet duct areas from both affected and nonaffected sources; and
 winlet is the totar ventilation rate from all ducts associated
 with the affected sources.
      5.  Establish the allowable mass emission rate of the system
 (AMRgyg)  in milligrams of total chromium per hour (mg/hr)  using
 equation 2:
      EVRinlet x EL x 60 minutes/hour = AMR.
                                           sys
(2)
 where 2VRinlet is the total ventilation rate in dscm/min from the
 affected sources,  and EL is the'applicable emission limitation. *
 The allowable mass emission rate calculated from equation 2
 should be equal to or less than the outlet three-run average mass
 emission rate determined from Method 306 testing in order for the
 source to be  in compliance with the standard.
      The second situation likely to be  encountered in industry is
 one in which  affected sources in two different  source categories,
 or  £!ource.s. in the  same source category  (i.e., hard chromium
 plating)  but  subject  to different emission limits  (i.e.,  new
 versus  existing small tanks),  are tied  into the common control
 device.   For  example,  a control  system  may be controlling
 emissions  from a hard chromium electroplating ta.nk,  a chromium
 anodizing  tank,  and other nonaffected sources.   A  similar
procedure  as  outlined above  is followed,  except that  the
 allowable  mass  rate is  the sum of the chromium  mass  emission
 rates calculated by equation 1 for each source  category or group
 of affected sources subject  to the same emission limitation.   For
 example, assume  that  a  control device controls  emissions  from  a
 chromium anodizing tank with a ventilation  rate  of 140 dscm/min
and a hard chromium electroplating tank at  a large facility with
a ventilation rate of 75 dscm/min.  The  allowable chromium mass
emission rate is calculated  as:
                              2-136

-------
     [ (140 dscm/min) (0.01 mg/dscm)  + (75 dscm/min)
     (0.015 mg/dscm)]  * 60 min/hr = 15'2 mg/hr

The. calculation is based on a 0.01 mg/dscm limit for the chromium
anodizing tank and a 0.015 mg/dscm limit for the hard chromium
electroplating tank.   Thus, the allowable chromium mass emission
rate is 152 mg/hr.
     The allowable chromium mass emission rate and operating
parameter values established for a control system that controls
multiple tanks described by either of the above situations are
valid only for that specific control system and tank
configuration.  Therefore, if a tank is added to or removed from
the control system, a performance test must again be conducted to
establish the allowable mass rate and the operating parameter
values.  This must be done whether or not the tank added or
removed is an affected tank; although nonaffected tanks will not
affect the allowable chromium mass emission rate, the  operating
parameter value would be  different.  Retesting would not be
required  if a nonaffected tank  is  replaced with a tank that  is
the same  size and  is connected  to  the  control system through the
same  size  inlet duct.
2.13   SELECTION OF REPORTING AND RECORDKEEPING REQUIREMENTS
2.13.1 Reportina/Recordkeepina Associated With Fume Suppressants
      Comment;  Seven commenters (IV-D-17,  IV-D-24,  IV-D-26,
IV-D-37,  IV-D-45,  IV-D-58,  IV-D-59)  stated that the types  of
records  that  the  rule  requires  sources to maintain  are
inappropriate.
      Five commenters  (IV-D-17,  IV-D-24,  IV-D-26,  IV-D-45,
IV-D-59)  pointed out  that requiring facilities  to maintain
records  of the amount  of chemicals used and purchased does not
 indicate compliance and should not be required.   Another
 commenter (IV-D-37)  suggested that this requirement be made more
 flexible by allowing sources to retain copies of  purchase orders,
packing slips,  etc.,  because in some large procurement systems,
 the purchaser does not receive copies of the invoices.
                               2-137

-------
      Fourcommenters  (IV-D-24, IV-D-26, IV-D-58, IV-D-59)
 indicatedthat requiring facilities to maintain records of the
 frequency of maintenance additions and the amount of fume
 suppressant material' added is not related to compliance and
 should not be required.
      Two commenters (IV-D-26, IV-D-59 ) stated that recordkeeping
 requirements be limited to only surface tension measurements
 because this measurement is the basis of compliance.
      ResP°nse:  In- finalizing the rule, the EPA attempted to
 reduce the recordkeeping burden to the extent possible, while
 still requiring sources to maintain sufficient records to
 demonstrate compliance.  The recordkeeping requirements are
 contained in § 63.346  of the final rule.   Recordkeeping
 associated with fume suppressants requires that sources maintain
 records of the date and time of surface tension or foam blanket
 thickness measurements, as appropriate,  and the value measured,
 and the date and time  of additions of fume suppressants to the
 bath.   The EPA concurs that the records required to be kept
 should correspond specifically to that which is required to
 demonstrate compliance.  As such,  records  of fume suppressant
 purchases and the amount added to the bath are no longer
 required.
 2.13.2   Reporting Frequency
     Cornme.nt:   SJ-X companies  (IV-D-06,  IV-D-07,  IV-D-10,  IV-D-18,
 IV-D-27,  IV-G-01)  provided  comments on the reporting frequency.
     Five commenters suggested that the frequency of reporting
 outlined in the proposed rule  was  burdensome.   One  of  these
 commenters  (IV-D-06) indicated that neither  quarterly  nor
 semiannual  reporting is necessary; two commenters  (IV-D-18,
 IV-G-01)  suggested that this reporting schedule be  replaced with
 a requirement  that the  source  submit an annual  certification  that
 necessary control  parameters have  been met,  consistent  with the
annual certification requirements  of title V.  One of  the
 commenters  (IV-D-27) suggested semiannual  compliance reporting
would be appropriate because it is consistent with the  title V
operating permit program.  Another of  these  four commenters
                               2-138

-------
(IV-D-10)  indicated that sources should not have to submit
compliance reports if the source's permitting agency inspects the
onsite records on an annual basis.
     One commenter CIV-D-07) suggested that the rule require
quarterly reporting until 2 years pass without exceedances of any
state or federal emission standards; after 2 years pass without
an exceedance, the source would be allowed to report semiannually
until a subsequent exceedance.
     Response:  The final rule requires submission of ongoing
compliance status reports to document whether a source has been
in continuous compliance with the standards.  The final rule
contains different reporting schedules for major and area
sources.  Major sources are required to submit ongoing compliance
status reports semiannually, unless an exceedance occurs.  If  an
exceedance occurs, quarterly reports are  required.  This  change
reflects changes made  to the General Provisions, which were  only
a proposal when this rule was proposed.   For major  sources,  the
reporting requirements in the final rule  are analogous to those
in the final  General Provisions.
      In an effort  to reduce the burden on area sources,  the  final
rule allows area  sources  to complete an annual compliance report,
and  allows the  source  to maintain the  report on site,  to be  made
available  to  the  Administrator  or permitting authority upon
request.   The EPA recognizes that many permitting authorities are
not  equipped  to handle reports  from area  sources,  and that these
 sources may  not be the sources  of primary concern to the
 authority.   However,  the requirements  in the final rule  do not
 alleviate affected area sources from complying with the  reporting
 requirements of State or Federal operating permit programs under
 title V.   The rule does require that area sources submit reports
 semiannually if exceedances occur, or if required by the
 Administrator or permitting authority.
      The rule also contains provisions for a reduction in
 reporting frequency for those major and area sources that are
 required to submit quarterly and semiannual reports,
 respectively.  These  sources must have 1 full year of compliance
                               2-139

-------
 with the standards, and request to have reporting frequency
 decreased.  The Administrator  (or permitting authority) may grant
 or deny the request.
 2•13•3  Reporting and Recordkeepina Requirements--General
      Comment;  Seven commenters (IV-D-10, IV-D-22, IV-D-23,
 IV-D-36, IV-D-47, IV-D-52, IV-D-58) indicated that the reporting
 and recordkeeping requirements of the proposed rule are
 burdensome for both small and large facilities and suggested the
 EPA consider the impact of these requirements.
      Two commenters (IV-D-10, IV-D-43) explained that small
 businesses do not have the resources to keep extensive records
 and to submit reports.   According to these commenters, extensive
 data may be necessary to maintain federal enforceability,  but the
 data is not needed for small businesses because the EPA is
 unlikely to seek enforceable action against such a small firm.
 One commenter (IV-D-58)  said that  the complicated monitoring,
 recordkeeping,  and stack measurement methods place an
 unreasonable economic burden on small sources.   One commenter
 (IV-D-22)  indicated that the EPA's  reporting and recordkeeping
 cost  estimate  of  $14 million does not account for the negative
 financial  impact  on small  shops that do not have personnel
 availableto perform testing every  four hour.
     Another commenter  (IV-D-47) suggested that  the  EPA's
 estimate q,f  the burden of  monitoring,  reporting,  and
 recordkeeping has  been underestimated and  provided an industry
 estimate of  600 hours per  year  per  source  for these  requirements.
 This commenter pointed out that, based on  the EPA's  estimate of
 recordingand recordkeeping  costs,  for chromium  anodizing  sources
 alone,  the reporting and recordkeeping requirements will cost
 almost  $1 million per ton  of  reduction.  The  commenter also
 indicated that the aerospace  industry  is concerned that the small
businesses that supply products to the  aerospace  industry will
not be able to afford the  technology to meet the  standards.  The
impact of the regulation will be greater than the EPA's estimate
unless the standard  is adjusted and the monitoring and
recordkeeping requirements are reduced.  This coimmenter stated
                              2-140

-------
that the EPA should consider eliminating most of the reporting
and recordkeeping requirements for area sources.
     One commenter (IV-D-23) pointed out that the EPA has
recognized differences in large and small facilities in selecting
MACT emission standards.  The EPA should also recognize
differences between large and small facilities in selecting
reporting, recordkeeping, and permitting requirements.  The
commenter suggested the following for small, hard chromium
platers:  (1) exemption  (especially for sources using batch
operations)  from startup/shutdown requirements of the proposed
General Provisions;  (2) exemption from extensive preconstruction
review requirements  (even though the proposed General Provisions
include area sources, the intent of the Act was to include
construction and reconstruction of major sources only); and
 (3) exemption from obtaining a title V permit.
     One  commenter  (IV-D-36) explained that reporting and
recordkeeping will also be  substantial for large companies where
multiple  sets of records will be maintained at multiple
facilities across the nation.  One commenter  (IV-D-52) indicated
there is  no environmental benefit to keeping records of gas
velocities, pressure drops, washdown conditions, and scrubber
water chromium  concentrations.
     Response:  The  EPA recognizes that  the reporting and
recordkeeping contained in  the proposed  rule could be burdensome
for many  sources, especially  area sources.  Given the large
number  of area  sources  in these  source  categories, the final  rule
contains  reporting  and  recordkeeping requirements that are
substantially reduced  from  those required at proposal.
Specifically, the amount of records that a  source is  required to
maintain  has been reduced,  the  frequency of  reporting has been
 reduced (especially for area  sources),  and  the requirements  of
 the General  Provisions  have been further evaluated  to assess
 their applicability to these  source  categories.  Where
 requirements  of the General Provisions  were considered  redundant
 or confusing, subpart  N has overridden the  requirement  as
 specified in Table  l of the final  rule.   For example,  the
                               2-141

-------
 requirements of § 63.6(e)  of the General Provisions associated
 with the startup, shutdown,  and malfunction plan have been
 overridden.   Sources subject to subpart N are required instead to
 complete an O&M plan in accordance with § 63.342(f)(3),  with
         ji, iiinj                  ,       ,        ,             . .         i'.in
 fewer elements than that required by the startup,  shutdown,
 malfunction plan.  Also, the preconstruction review requirements
 of the General Provisions  have been overridden,  and instead  more
 straightforward preconstruction review requirements are contained
 in subpart  N.   Finally,  as previously described  in Section 2.11,
         ii"  (   ••     _  |Y '   :        ,. •  •  "i.. .        •      ,     >• •,	•	i,
 the monitorxng requirements  have been reduced where possible.
      The EPA estimates that,  on an ongoing basis,  the impacts  of
 the revised reporting  and  recordkeeping requirements are
 $3.5 million and 159,100 hours per year for hard chromium
 electroplating operations, $6.4 million and 289,300  hours per
 year for decorative chromium electroplating operations, and
 $1.6 million and 70,200  hours per year for chromium anodizing
 operations.   This is compared to the impacts of  the  proposed rule
 of $8.6  million  and 388,400  hours per year for hard  chromium
 electroplating operations, $15.6 million and 706,200  hour per
 year for decorative chromium electroplating operations, and
 $3.8  million and 171,500 hour per year for chromium  anodizing
 operations.
      Comment;  Two  commenters (IV-D-10,  IV-D-43)  stated that
 state and local  permitting agencies  should be allowed to
 determine appropriate  reporting and  recordkeeping  requirements,
 especially for area sources because  enforcement  actions against
 area  sources are the responsibility  of  local  agencies.  These
 commenters also  recommended that no  reports be submitted  for a
 set period of  time  following  promulgation;  one commenter
 suggested 6 months  (IV-D-10)   and the  other suggested  l year
 (IV-D-43) as an  appropriate timeframe.   This  would allow  State
and local agencies  an opportunity to  request  and assume
delegation of  the air toxics  program  and to minimize  the
confusion on the part of sources  as to which  agency has
authority.
                              2-142

-------
     Response;   The EPA is required to identify the minimum
reporting and recordkeeping requirements that are necessary to
determine compliance with subpart N.  State and local agencies
may choose to require additional reporting and recordkeeping if
they feel it is necessary to determine the compliance status of a
source.  States that currently have existing regulations for
chromium electroplating and anodizing tanks may submit their
programs to the EPA for approval as equivalent in accordance with
subpart E of part 63.
     In the final rule, the first report that must be submitted
is the initial notification report required by § 63.347(c), which
is due to the Administrator 180 days after the effective date.
The report may be submitted to the permitting authority,  if the
authority has been granted approval to implement and enforce this
rule.  The compliance date for existing sources subject to this
rule is 1 year from the effective date for decorative chromium
electroplaters (i.e., November 1995) and 2 years from the
effective date for hard chromium electroplaters and chromium
anodizers (i.e.,  November 1996).  The EPA believes that this
timeframe is sufficient for States to apply for implementation
and enforcement responsibilities for this rule.
     Comment:  One commenter  (IV-D-10) suggested the length of
time a facility is required to keep records should depend on its
inspection frequency; however, records should be maintained for a
minimum of 2 years.  A tiered approach to reporting frequency
based on the facility's emission level could also be adopted.
Two commenters (IV-D-24, IV-D-58) stated that maintaining records
at a facility for 5 years is excessive; a more appropriate length
of time would be 3 years.
     Response:  The final rule requires that owners or operators
of affected sources maintain records for a period of 5 years
following each occurrence, measurement, maintenance, corrective
action, report, or record.  This requirement is unchanged from
the proposed rule because it is consistent with the General
Provisions and with the title V permit program.  As stated in the
preamble to the final General Provisions, the EPA believes
                              2-143

-------
retention of records for 5 years allows the EPA to establish a
source's history and pattern of compliance for purposes of
determining the appropriate level of enforcement action.
Allowing the destruction of potential evidence of violations may
prevent the EPA from pursuing the worst violators.
     Comment;  One commenter  (IV-D-20) asked that section 63.9(b)
specify that all sources are allowed to report potential
rectifier capacity or some other parameter in lieu of chromium
emission estimates.
     Response:  Section 63.9(b) of subpart A requires an initial
notification, but does not require that the emissions from the
affected source be estimated.  In any event, § 63.9(b) has been
overridden by § 63.347(c) of the final rule.  This section
requires identification of facility name and address, a statement
that subpart N is the basis for the notification, a description
                                                           " '     "" t
of affected sources and the applicable emission limitation for   '
the source, and a statement of whether the source is located at  a
major or area source as defined in § 63.2 of subpart A.  For
owners or operators performing hard chromium electroplating, the
maximum potential rectifier capacity associated with these
operations must also be reported so the Administrator  (or
permitting authority) knows if the source(s) is located at a
small or large, hard chromium electroplating facility.  The
initial notification does not require an emissions estimate
because the affected sources in these source categories are
relatively uniform in terms of emissions potential.
     Comment;  One commenter  (IV-D-38) explained that
automatic/continuous monitoring of conductivity, washdown timer,
magnahelic gauges, etc.  may be equipped with a transmitter and
programmable logic controller system to provide printing and
storage of parameters that can be maintained via computer hard
drive.  If the EPA approves the use of such a monitoring system,
software can be written to accomplish the storage of data on the
computer hard drive.
     Response:  Section 63.10(b)(1) of subpart A, which applies
to sources covered by subpart N, specifies the manner in which
                              2-144

-------
records should be kept.  In accordance with this provision, files
must be maintained for at least 5 years following each
occurrence, measurement, maintenance, corrective action, report,
or record.  The most recent 2 years of data must be retained on
site; the remaining 3 years, however, may be retained off  site.
Such files may be maintained on microfilm, computer, floppy disk,
magnetic tape disks, or microfiche.
     Comment;  Six commenters  (IV-D-df, IV-D-03, IV-D-08,
IV-D-19, IV-D-20, IV-D-22)  suggested  the EPA develop
user-friendly reporting forms  to  facilitate the  reporting
requirements of  the rule.   One of these commenters  (IV-D-08)  also
suggested  that because the  formal rule is  difficult to  understand
as  written,  the  EPA should  produce an easy to  follow guidance
document  to  accompany  the rule.   Another  commenter (IV-D-20)
indicated that  the rule (or preamble or  Fact  Sheet)  should
contain a chronological listing  of compliance and reporting
requirements that incorporates requirements of both the chromium
electroplating  NESHAP  and the General Provisions.
      Response:   The  EPA is  developing an enabling document to
accompany the rule,  and the Small Business Administration is
 developing a pamphlet  assist small businesses.  Both of these
 documents are intended to help sources understand the
 requirements of the rule and implement the rule.
      Comment:  One commenter  (IV-D-20)  requested that  the
 reporting dates for the initial  notification of compliance status
 and for the initial performance  test referenced in § 63.9(h)(4)
 of the proposed General Provisions be revised.  Notification  of
 compliance status is  due within  45 days of compliance  date,
 however the performance test  is  not  due until 120 days after  the
 compliance date; the  emission data gathered during the
 performance test will be necessary  for submission of compliance
 status.
      Response:   In the General  Provisions, the  notification of
 compliance  status is  required to be submitted within  60 days
 following the  completion of the required performance  testing, as
 is stated in §  63.9(h)(2)(ii),  not 60 days from the  compliance
                               2-145

-------
 date.   Subpart N overrides  this  requirement  to  allow sources
 additional  time  to  submit the  notification of compliance  status.
 Additional  time  may be  required  for  the many sources in these
 source  categories to  obtain performance test results.  A  shorter
 timeframe may result  in owners or  operators  having  to pay extra
 to obtain analytical  results within  60 days.
 Section 63.347(e)(3)  of the final  rule requires the  notification
 of compliance status  be submitted  90 days following  completion of
 the performance  test  required  by § 63.7 of subpart A and
 § 63-343fb) of the  final rule  if a performance test  is required.
 The results of performance  testing are also  due no later  than
 90 days after completion of the test.  Thus,  the timeframes are
 consistent.  If a performance  test is not required  (e.g.,  sources
 meet the requirements of §  63.343 (b)  (2)  or (3))/the compliance
 status report must be submitted no later than 30 days after the
 compliance date.
 2•13•4  Requirements of the General Provisions
      Comment;  One commenter (IV-D-23)  requested that the  rule
 clearly state  which sections of the General  Provisions apply to
 chromium electroplating sources and which do  not.
     Response:  Table  1 has  been added to the final  rule.   This
 table  identifies  which sections of  the General  Provisions  apply
 to sources  subject to  subpart N,  and  which do not.   If a
 requirement  in the General Provisions is  overridden  by a
 requirement  in subpart N, the applicable  section in  subpart  N  is
 cited.
     Comme:nt:  Three commenters (IV-D-20,  IV-D-28, IV-D-43)  noted
 that the proposed rule does  not contain any requirements for
 continuous monitoring  systems but subsequently requires a
 continuous monitoring  system performance report  in §  63.347(c).
 Two of these commenters  (IV-D-20, IV-D-43) asked whether the EPA
 intended to  refer to section 504(a) of the Act or part 70  of the
General Provisions.  One of  the three commenters (IV-D-28)
suggested deleting this  reference from the rule.
     Response;   in the General  Provisions, continuous monitoring
systems  (CMS) are intended to include continuous emission
                              2-146

-------
monitors, continuous opacity monitors, continuous parameter
monitoring systems, or other manual or automatic monitoring that
is used for demonstrating compliance with an applicable
regulation on a continuous basis.  In accordance with this
definition, monitoring of surface tension, foam blanket
thickness, pressure drop, and velocity pressure would be
considered CMS.  The EPA recognizes that the term CMS could cause
some  confusion among owners or operators of affected sources.
Therefore,.subpart N has overridden the requirements in the
General  Provisions that deal with CMS and replaced  them with
requirements  in subpart N that are more appropriate to these
subcategories.  As previously stated, Table 1  specifies which  ^
requirements  of the General Provisions apply to  sources subject
to subpart N, which do not, and  which have been  overridden by a
specific requirement  in  subpart  N.
      Comment:  One commenter  (IV-D-10)  suggested that  the initial
notification requirement of  section 63.7  of  the  General
 Provisions be waived  for sources that hold operating permits with
 the State or local agency to reduce reporting burdens for the
 sources.  In lieu of  this requirement,  the State or local agency
 would submit a list of affected sources to the EPA.
      Response;   The requirements for the initial notification
 that are contained in § 63.347(c)  of the final rule are minimal.
 The EPA does not  think it will be burdensome for sources to
 submit this report, especially those sources that  hold operating
 permits.  The information required in the initial  notification
 has probably been compiled previously by such sources.
      Comment;  One commenter (IV-D-10) indicated that the
 timeframe for review  of test plans that is provided in subpart A
 of part  63 of the General Provisions is inappropriate and should
 be overridden by  the  chromium electroplating  rule.
      Response:  The  final rule  overrides the  requirement that
 sources  required  to  conduct a performance test  prepare a
 site-specific  test plan that meets  the requirements  of
 § 63.7(b)(l).  instead,  § 63.344(a)  identifies  the minimum
 elements to be contained in  test  reports  documenting the initial
                                2-147

-------
  performance  test,  which shall be made available to the
  Administrator upon request.
       Comment;   Given the implications of the  General  Provisions
  for  this  rule,  one^commenter (IV-D-23)  statedthat he reserved
  the  right to submit  additional comments on  the proposed chromium
  electroplating NESHAP following final promulgation of the General
  Provisions.
       Response;  The  preamble to the proposed  subpart  N and the
  proposed  rule  clearly stated that;the General  Provisions were
  proposed  rules  at  the time,  but that  sources  subject  to subpart N
  would eventually have to comply with  the General Provisions as
  promulgated.  As such, comments  regarding the General  Provisions
  were  to fce made during the comment period for subpart A.
 Alternatively, owners or operators of sources affected by
  subpart N could have provided comments on the requirements of the
 proposed General Provisions during the comment period for
 subpart N.
 2.14   OPERATING PERMIT PROGRAM
      .Comment;  Eleven commenters stated that area sources  should
 not be required to obtain title V operating permits.  Four
 commenters (IV-D-14,  IV-D-23, IV-D-24, IV-D-47)  suggested  that
 the rule should exempt nonmajor sources from title V.   One
 commenter (IV-D-15) suggested that the NESHAP  be written so  that
 it  requires all sources to comply with the emission standards  but
 does  not require title V permits for area sources.   Another
 commenter  (iv-D-10)  stated that a title V permit is not necessary
 because  existing requirements are enforceable  through  State and
 local  permits.   This  commenter and one other (IV-D-43)  pointed
 out that because area sources are not  likely to  be  subject to
 multiple MACT standards or to employ emissions  averaging and
 complex  alternate operating scenarios,  title V permits do not
 benefit  the area sources.  This  commenter also indicated that
part 70  requires periodic review of the permit,  and  for area
 sources, this review would not result  in significant changes from
the original operating permit; periodic review is most  useful for
large, complex operations.  Two  commenters, (IV-D-10, IV-D-22)
         ::;  '              '    2-148

-------
said it will be a cost burden for area sources to obtain title V
permits, and the emissions from these sources may be
insignificant  (IV-D-22).
     Four commenters-  (IV-D-10, IV-D-20, IV-D-23, IV-D-43)
suggested that there  should be an applicability threshold, or
cut-off, for area sources required to obtain permits.  One of
these commenters  (IV-D-10) suggested that the threshold be
risk-based.  Another  of these commenters  (IV-D-20) suggested that
the threshold  determination for title V permit applicability
could be set at a later date after the EPA has received and
reviewed the initial  notifications of  compliance status.
     Two commenters  (IV-D-14, IV-D-47) stated that in preparing^
their  title V  permit  programs, States  did not anticipate  a need
for emission-unit specific permits at  nonmajor  sources,  and
inclusion  of nonmajor sources under  title V  will  require that
many local agencies  revise their  permit programs.  Two  other
commenters (IV-D-08,  IV-D-10) stated that States  will not have
the resources  for completing  title V permits for  area sources;
some states have exempted nonmajor  sources  from their permitting
programs until the  nonmajor  source  permitting rule is promulgated
 in the late 1990's  (IV-D-10).
      Finally,  two commenters  (IV-D-40, IV-D-43)  asked that  a
 general permit be included in the final rule to reduce  the  burden
 for small  facilities.
      Response:  The EPA believes that requiring all sources that
 are subject to the standards, including area sources,  to obtain
 title V operating permits is important because of the toxicity of
 chromium compounds and the close proximity of many of these
 sources to residential areas.  The EPA believes that permitting
 area sources  will not be overly burdensome to permitting
 authorities and affected sources for  the reasons given below.
      First, many States are already permitting these sources
 under  their State permit programs.  The  preamble to the  final
 part  70 rule  states  that  "some nonmajor  sources would already be
 permitted at  the State level, and therefore would have  some
 experience with  the  permitting process and  completing permit
                               2-149

-------
 applications."  Therefore,  a  State would have  little  reason to
 defer title V permitting  of the  sources that already  have State
 operating permits.  Second, the  burden may be  reduced
 significantly by issuing  general permits to these sources.
 According^to the preamble to  the final part 70 rule,  general
 permits  «...provide an alternative means for permitting sources
 for which the procedures  of the  normal permitting process would
 be overly burdensome, such as area sources under section 112...".
 Under this option, States would  develop a single general permit
 for this source category and issue it to individual sources; or
 alternatively,  a letter or certification may be used.   The burden
 would also be reduced by using general permits because public
 participation and EPA and affected State review is only necessary
 when the initial general permit is drafted and issued.  When
 subsequent general permits are issued to individual sources,
 these activities are not required.   Finally,  States are
 developing small business assistance programs (SBAP's) to assist
 these types  of  sources with the permitting process that will be
 funded using  the annual fees collected from permitted  sources.
 Small businesses may also be eligible for  reduced permitting
 fees.     i          '.     .   ,      .     f    '' .    ,.        .
      Cgmnjgnt:   Three commenters (IV-D-14,  IV-D-23,  IV-D-47)
 suggested that  in  the  event  that  nonmajor  sources are  included
 under title V permitting,  the  rule should  explicitly state that  a
 permit isrequired  only for  applicable emissions  units at
 nonmajor  sources.
      Responge;   Under  title  V,  sources must include  information
 on all emission  points  (except  those  considered insignificant
 under the State  or  local permit program) in their permit
 application.  However, applicable emission limitations  only need
 to be identified for those emission points that are  subject to
 regulation.
 2.15  WORDING OF THE REGULATION
     .Comment;   Several commenters requested changes or additions
 to the definitions of terms provided in section (53.341 of the
regulation.   One commenter (IV-D-45)  suggested changes regarding
                              2-150
          iiii

-------
the definitions of the terms decorative chromium electroplating,
hard chromium electroplating, and stalagmometer.  According to
this commenter, the text "...to provide .-a bright surface with
wear and tarnish resistance" for decorative chromium
electroplating and the text  "...to provide a surface with wear
resistance, a low coefficient of friction, hardness, and
corrosion resistance" for hard chromium electroplating that are
currently used in the definitions for/these terms limit the
applicability of the regulation; the regulation should apply
regardless of the purpose of the electroplating.  Also, this
commenter stated that the term tensiometer should be defined
separately from the term stalagmometer.   Two other  commenters   ^
 (IV-D-17, IV-D-58) also  suggested changes to the definitions  for
the terms hard and decorative chromium electroplating;  they
believed that  these terms should emphasize the purpose  of  the
electroplating operation.   Decorative  electroplating  is done
primarily  to provide  cosmetic effects  and hard electroplating is
done primarily to provide  functional  effects.   According  to  the
 commenters,  the  definition of hard  electroplating should  also be
written so that  the  applicability  of  only one  of the  listed
 functional effects would qualify the  electroplating operation as
 hard chromium electroplating.   These  commenters also suggested
 that the definition for the term chromium electroplating be
 amended to include the electrodeposition of chromium onto
 undercoatings, in addition  to the electrodeposition of chromium
 onto base metals or plastics as the definition currently reads.
      Two commenters  (IV-D-20,  IV-D-43) requested clarification of
 the definitions for the terms air pollution control device and
 foam blanket and specifically questioned whether foam blankets
 are considered to be air pollution control devices according  to
 the current definitions.   In the instance that foam blankets  are
 an air  pollution control device, the  final regulation  should
 clearly identify them as such.  Also,  in this instance, the
„monitoring, recordkeeping,  and reporting requirements  for control
 devices should apply.   However, nonapplicable requirements such
 as pressure drop monitoring should be excluded.  Another
                               2-151

-------
 commenter  (IV-D-24) asked that a definition for the term add-on
 air pollution control device be added.
      Finally, one commenter  (IV-D-41) pointed out that the proper
 terminology for referring to chromium electroplating emissions is
 "mist emissions," as is used by California and other agencies.
 According to this commenter, chromium emissions should not be
 referred to as "fumes," as the proposed rule currently is worded.
      ResRonse:  As mentioned in previous sections of this
 document, the definitions for hard chromium electroplating,
 decorative chromium electroplating, and chromium anodizing have
 been revised.  In addition to describing the function of each
 type of electroplating and anodizing, the definitions in the
 final rule express these electroplating and anodizing operations
 in terms of process parameters such as typical electroplating
 thickness achieved,  current density,  and electroplating time.
 The term tensiometer has also been added to the list of
 definitions.   Another new term,  "air pollution control
 technique," which includes methods such as fume suppressants that
 reduce  or prevent chromium emissions, has also been  added.   A
 "foam blanket"  is a type of fume  suppressant and is  considered to
 be an air pollution control technique.   As previously discussed,
 the monitoring, reporting,  and recordkeeping that  applies  to
 sources  controlled with  an air pollution control  technique  has
 been  clarified  in the  final  rule;  requirements  are control
 technique specific,  not  source  category  specific.  The  term air
 pollution control  device has  been  changed to add-on  air pollution
 control  device, and  refers  to add-on  equipment  siuch  as  packed-bed
 scrubbers,  composite mesh-pad systems, and fiber-bed mist
 eliminators.  A definition  for fiber-bed mist eliminator has  been
 added in the final rule.
     The  final rule  also  contains a definition  for fume
 suppressant; this definition  clarifies that fume suppressants  are
 considered the same  as mist suppressants.   It is the EPA's
 experience that both terms are used in the  industry,  and
 therefore a definition of fume suppressant was added to eliminate
any confusion.
                              2-152

-------
     Comment;   One commenter (IV-G-01)  requested that the term
facility in the proposed rule be consistent with the definition
of major source under 40 CFR part 70.  Otherwise, large
industrial complexes" that consist of multiple, distinct major
sources under title V are treated as one source under Title III
(of the CAA Amendments).  One problem that would arise under the
proposed rule is that maximum cumulative rectifier capacity would
have to be calculated for the entire industrial complex and not
just each major source.
     Response:  The definition of facility in subpart N is
consistent with the definition of major source in subpart A.
     Comment;  Four commenters  (IV-D-01, IV-D-03, IV-D-40,
IV-D-43) suggested that allowable emission limits, alternative
compliance procedures,  compliance dates, monitoring  requirements,
recordkeeping requirements,  and  reporting requirements should be
summarized  in table format  in the regulation  in  a manner  similar
to  the  summary tables  in  the preamble.  The table format  is
easier  to follow  than  the current written version.
     Response:  The Agency  agrees that  providing information in
table  format whenever  possible  in the  rule  is a  convenient method
for summarizing the requirements of the rule.  However,
requirements  that are  written in text  format  in  the rule cannot
be  repeated in a  table format because  of potential
inconsistencies between the text and table.   Two tables  have been
included in the final  rule  that present requirements not found
elsewhere in text.  Table 1 specifies  which requirements of the
General Provisions apply, and Table 2  specifies the work practice
 standards.   In addition,  summary tables of the standards and the
monitoring and reporting requirements are provided in the
 preamble.
      Comment;  Two commenters (IV-D-16 and IV-D-24) indicated
 that § 63.345(c)  should refer to the O&M plan in
 section 63.342(e) rather than section 63.342(d), which refers  to
 chromium anodizing tanks.  Another  commenter  (IV-D-49) indicated
 that § 63.345(c)(3), which  refers to the addition of makeup  water
 for a packed-bed scrubber,  is not applicable to the
                               2-153

-------
 composite	mesh-pad system that is the subject of section 63.345(c)
 One commenter  (IV-D-24) suggested that section 63.343, which
 references compliance and performance testing, should reference
 pressure drop measurement.  This commenter also suggested that
 the wording of section 63.345(d)  be consistent with the wording
 of section 63.345 (e).
      Response:  The commenters are correct in that § 63.345(c) of
 the proposed rule should have referred to § 63.342 (e), not
 § 63.342(d).   The final rule is organized differently so all
 operation and maintenance requirements are included as work
 practices in § 63.342(f).   Section 63.343 did not reference
 pressure drop measurement because,  at proposal,  pressure drop
 measurement was not being conducted for purposes of compliance. *
 In the final  rule, pressure drop  measurement is being done for
 purposes of compliance and is required as such in § 63.343(c)(1)
 through (4).   in a previous response in Section 2.11,  the EPA
 acknowledged  that § 63.345(c)  erroneously required the addition
 of makeup water to the packed bed for composite mesh-pad systems.
 The addition  of makeup water to the  packed bed is a work practice
 standard only for owners or operators of  packed-bed scrubbers, as
 is clear in §  63.342(f) and Table 2  of the final  rule.   Finally,
 consistent wording is  used,  as applicable,  in the final  rule.
     gommeilt:   One commenter (IV-D-33)  requested  that  the EPA
 review the wording of  the  regulation.   The commenter indicated
 that the wording  of the regulation will cost  the  company a  lot of
 money.
          lit                .         '       '•• •         .•          r
     Response:  The final version of  the  regulation differs  from
 the proposed version.  The EPA made changes to clarify
 applicability,  and compliance, monitoring, recordkeeping, and
 reporting  requirements.  The EPA  believes  that the  final
 regulationis clearer, and will be easier  for sources to
understandand  implement.
     .Comment:   One commenter  (IV-D-21) suggested that the rule
for hard chromium  electroplating  should clearly indicate that  a
source needs to meet the designated emission limit and not a
specified control  system design.
                              2-154

-------
     Response;  The standards in the rule require a specific
emission limitation (expressed as mg/dscm),  and any control
system may be used to meet these standards.   Exceptions are the
surface tension limits that are included as standards; surface
tension limits are included for the specific situation in which a
wetting agent-type fume suppressant is used to control chromium
emissions from decorative chromium electroplating or chromium
anodizing tanks.  The preamble to the proposed rule described
only those control technologies that the EPA assumed would be
used by industry in meeting the emission limits.  The fact that
other control methods were not described in the proposal preamble
does not preclude their use as long as the applicable emission
limit is achieved.
2.16  MISCELLANEOUS
     Comment:  One commenter suggested that an emissions
averaging scheme be incorporated into the chromium electroplating
NESHAP.
     Response;  The EPA does not think that there are
opportunities for emissions averaging in this NESHAP.  Typically,
there are multiple emission points within a source category, and
emissions averaging allows a source to average across these
various emission points.  In the chromium electroplating and
chromium anodizing source categories, there is really only one
emission point, the tank.  One might suggest that averaging could
occur by controlling  one tank to a level higher  than  the standard
to  compensate for another tank that is controlled to  a lesser
degree.  However, the EPA does not support leaving a  tank
uncontrolled even if  another tank at the facility is  controlled
to  a greater extent.
     Comment;  One commenter  (IV-D-23) noted that the proposed
rule differs significantly from  the information  that  was
presented at the National Air Pollution  Control  Techniques
Advisory Committee  (NAPCTAC) meeting.  Specifically,  at this
meeting, the EPA stated  that area and major sources would  be
regulated separately, with different technology requirements
 (i.e.,  MACT v.  GACT). Also, according to  this  commenter,  the  EPA
                              2-155

-------
 stated that  the compliance timeframe would be 3  years after
 promulgation for existing sources.   At NAPCTAC,  composite mesh
 pads  were the basis  of MACT for new hard chromium electroplaters
 only,  not new and existing.   Finally,  the trivalent chromium
 electroplating process was identified as MACT for new decorative
 chromium electroplaters.
      Response;   The  EPA presented the status  of  the chromium
 electroplating and chromium anodizing NESHAP  at  a briefing for
 the NAPCTAC  on January 29 through 31,  1991.   The purpose  of the
         if '•!,"•       ' •     ,      ,         , ;        |         ,'•.,,
 NAPCTAC briefing is  for the EPA to  present the current status of
 their regulatory actions  to industry,  regulatory agencies,
 environmental groups,  and the public.   The information that is
 presented at NAPCTAC briefings is subject to  chcinge based on
 public comments,  comments received  from the NAPCTAC panel,  and
 additional information that may be  gathered by the EPA in
 response to  these comments.   One of the major changes since
          i. 'i          '                   ''                        ' • '
 NAPCTAC is that composite mesh-pad  systems became much more
 widely used  in the chromium electroplating industry.   In  the
 NAPCTAC briefing,  the  EPA stated that  emerging sidd-on control
 technology (e.g.,  composite mesh pads)  were being considered as a
 regulatory alternative that was more stringent than the MACT
 floor,  and that such technologies may  become  demonstrated during
 the rulemaking  process.   The  most recent  cost-benefit analysis
 conducted by the  EPA indicated that these devices are
 cost-effective  for tanks  at existing large, hard chromium
 electroplating  facilities and all new  hard chromium
 electroplating  facilities.  Therefore,  they were selected as MACT
 for those  categories.   Regarding the regulation  of  decorative
 chromium electroplating facilities,  for the reasons  discussed in
 Sectipn 2.6.1,  the EPA concluded that  there was  insufficient
 information  to  select  the trivalent  chromium  electroplating
process  as new  source  MACT for decorative  chromium  electroplating
sources.
     During  the NAPCTAC briefing, GACT was presented  as a
possible alternative for  regulating  area  sources.  Regulation of
area sources by GACT was  considered  for the source categories
          :• :                  ,,,,'2-156.

-------
covered by this rule, and, for the reasons discussed in
Section 2.4, was not selected.  -At the time of the NAPCTAC
briefing, these decisions had not been finalized.  Likewise,
section 112(i)(3) of the Act states that the EPA shall establish
a compliance time for existing sources, which provides for
compliance as expeditiously as practicable, but in no event is
later than 3 years after the effective date of the standard.  At
the time of the NAPCTAC briefing, this information may have been
presented because the compliance timeframe for the source
categories  covered by this NESHAP had not yet been selected.
When the rule was to be proposed, the EPA selected the compliance
dates.  For the  reasons discussed in Section 2.10, the compliance
date in the final rule is 1 year from the effective date for
existing decorative  chromium  electroplaters and 2 years from  the
effective date for existing hard chromium electroplaters and
chromium anodizers covered by the standard.
     Comment:  Two commenters (IV-D-52,  IV-D-58)  suggested that
the EPA educate  small businesses on the  requirements  of this  and
other NESHAP  through public outreach programs.   One commenter
 (IV-D-58) pointed out that the provisions  of  section  507(b)(2)
could be  implemented to  inform small businesses  of  the  technical
requirements  of  this rule.
     Response:   In  order to aid small  business  in the
requirements  for stack testing,  the EPA has prepared  a  videotape
on how  to build the  Method  306A sampling train,  how to  operate
the  train,  and how  to make  and install the enhanced monitoring
equipment.   The videotape is  entitled "Construction and Operation
of the  EPA Method 306A Sampling Train and Practical Suggestions
 for  Monitoring of Electroplating and  Anodizing Facilities," and
will be available in January 1995  for a nominal fee.   The
videotape is available through North  Carolina State University,
 Registrar,  Environmental Programs,  Box 7513,  Raleigh,
 NC 27695-7513.  The telephone number  is (919) 515-4659.  The fax
 number is  (919)  515-4386.  In addition, articles on how to make
 and operate the testing and enhanced monitoring equipment will be
 written and published in trade journals of the industry such as
                               2-157

-------
 the Plating and Surface Finishing and Metal Finishing.  Also,  as
 previously stated,  the EPA is developing an Enabling Document  and
 the Small Business  Administration is developing a pamphlet to
 help owners or operators understand and implement the rule.
      Comment:   One  commenter (IV-D-19)  stated that the EPA should
 promulgate the final rule by November 15,  1994 as scheduled.
      Response:  The final rule will be promulgated by
 November 23,  1994 in accordance with a consent decree reached
 between the Sierra  Club and the EPA.
        ,„ Ijl1	           '       '   '   |   , .   '  ,   ,,,",'  .    ,      . ;
      Comment;   One  commenter (IV-D-04)  questioned whether they
 were on the mailing list because they did  not receive a copy of
 the background information document (BID)  for the proposed
 standards.   Other commenters (IV-D-32,  IV-D-34)  requested a copy
 of  an organization's public comments.
      Response:   Two documents,  the  BID  for the proposed standards
 and the New Technology Document, were prepared to identify the
 information used in developing  the  proposed standard.   The BID
 consists of two  volumes,  and is too lengthy to distribute to
 everyone on the  mailing list.   The  BID  and the new technology
 document can be  obtained from the U.  S.  EPA Library  (MD-35),
 Research Triangle Park,  North Carolina  27711,  telephone number
 (919)  541-2777.  These documents can  also  be retrieved from one
         Ml   '          .               .   ", '      .      '     .    .
 of  the  EPA's electronic bulletin boards, the Technology Transfer
 Network (TIN).   The TTN provides information and technology
 exchange in various areas  of  air pollution control.   The service
 is  free, except  for the cost  of the phone  call.   Dial
 (919) 541-5742 for  up  to a 14,400 bps modem.   If  more  information
 on  TTN  is" needed, call  the HELP line at  (919)  541-5384.
     Copies of public  comment letters are  available for  public
 inspection  and copying  between  8:00 a.m. and 5:30 p.m.,  Monday
through Friday, at  the  EPA's Air and Radiation Docket  and
Information Center^  Waterside Mall,  Room 1500, 1st Floor, 401 M
Street, S.W., Washington,  DC 20460.   The telephone number is
                              2-158

-------
(202)  260-7548.  The fax number is (202) 260-4400.  When
requesting information on this standard, refer to Docket
No. A-88-02.  Public comment letters are identified according to
the nomenclature in "Table 2-1.  A reasonable fee may be charged
for copying.
                               2-159

-------
•i	If

-------
                                  TECHNICAL REPORT DATA
                             (Please read Instructions on reverse before completing)
1. REPORT NO.
  EPA 453/R-94-082b
                             2.
4. TITLE AND SUBTITLE
  Chromium Emissions from Chromium Electroplating
and Chromic Acid Anodizing Operations-Background
Information for Promulgated Standards
7. AUTHOR(S)
   Lalit Banker, BSD/Organic Chemicals Group
9. PERFORMING ORGANIZATION NAME AND ADDRESS
   Emission Standards Division
   Office of Air Quality Planning and Standards
   U. S. Environmental Protection Agency
   Research Triangle Park, NC  27711
 12. SPONSORING AGENCY NAME AND ADDRESS
  DAA for Air Quality Planning and Standards
  Office of Air and Radiation
  U. S. Environmental Protection Agency
  Research Triangle Park, NC  27711
 15. SUPPLEMENTARY NOTES
                                                                 3. RECIPIENT'S ACCESSION NO.
                5. REPORT DATE
                  November 1994
                6. PERFORMING ORGANIZATION CODE
                                                                 8. PERFORMING ORGANIZATION REPORT NO.
                                                                 10. PROGRAM ELEMENT NO.
                11. CONTRACT/GRANT NO.
                13. TYPE OF REPORT AND PERIOD COVERED
                Final
                14. SPONSORING AGENCY CODE
                EPA/200/04
     National emission standards to control emissions of chromium compounds from new and existing
 chromium electroplating and chromium anodizing tanks are being promulgated under section 112 of the
 Clean Air Act.  This document summarizes public comments received during proposal and EPA s
 responses to these comments.		.	•
 17.
                                     KEY WORDS AND DOCUMENT ANALYSIS
                   DESCRIPTORS
    Air pollution control
    Chromium
    Environmental protection
    National emission standards
    Hazardous air pollutants
    Electroplating
    Anodizing	
  18. DISTRIBUTION STATEMENT
    Release Unlimited
                                                 b. IDENTIFIERS/OPEN ENDED TERMS
 Air pollution control
 Chromium
 Stationary sources
19. SECURITY CLASS (Report)
   Unclassified
                                                 20. SECURITY CLASS (Page)
                                                    Unclassified
                                                                                     c. COSATI Field/Group
21. NO. OF PAGES

     170
                                                                                     22. PRICE
EPA Form 2220-1 (Rev. 4-77)   PREVIOUS EDITION IS OBSOLETE

-------

-------