10/01/86

Guideline S-26 - Enforcement of the Arsenic NESHAP for Glass Manufacturing Plants ( 40 CFR Part 60 Subpart N )


                              OCT - 1 1986

MEMORANDUM

SUBJECT: Guideline S-26 - Enforcement of the Arsenic NESHAP
         for Glass Manufacturing Plants ( 40 CFR Part 60 Subpart N )

FROM:    Director
         Stationary Source Compliance Division
         Office of Air Quality Planning and Standards

TO:      Air and Waste Management Division Director
         Region II

         Air Management Division Directors
         Regions I, III, V and IX

         Air, Pesticides, and Toxics Management Division Directors
         Regions IV and VI

         Air and Toxics Division Directors
         Regions VII

    The attached guideline is being forwarded to you to assist you in the
implementation and enforcement of the arsenic National Emission Standards
for Hazardous Air Pollutants ( NASHAPS ) for glass manufacturing plants
( Subpart N ).

    If you have any questions or comments on this guideline, please
contact Doreen Cantor in the Stationary Source Compliance Division at FTS
382-2874.

                             Edward E. Reich

Attachment

cc:  Michael Alushin                           Stan Cuffe
     Bob Ajax                                  Ron Myers
     George Walsh                              Jan Myers
                                               Jim Engel

               S-26 - GUIDELINE FOR THE ENFORCEMENT OF THE
                       ARSENIC NESHAP REGULATIONS
                     FOR GLASS MANUFACTURING PLANTS

    This guideline is being issued to assist the Regional Offices in the
enforcement of one of the NESHAP regulations for the control of arsenic
emissions.  Three types of facilities are required to be controlled by
these regulations:  (1) glass manufacturing plants, (2) primary copper
smelters, and (3) arsenic trioxide and metallic arsenic production
facilities.  This guideline addresses glass manufacturing plants only.

    Arsenic was declared a hazardous air pollutant on June 5, 1980.
Regulations were proposed for the control of arsenic under Section 112 of
the Clean Air Act, National Emission Standards for Hazardous Air
Pollutants ( NESHAPS ) 40 FR 59532, on August 22, 1983.  These regulations
were promulgated on August 4, 1986.  Since this regulatory framework has
been previously utilized for the control of asbestos, beryllium, mercury,
vinyl chloride, and benzene, additional helpful information is available
in Guidelines S-17 thru S-30, which offer some general guidance relative
to the procedural requirements of the NESHAP program.

Summary of Requirements

    The standard covers each glass melting furnace that uses commercial
arsenic as a raw material, except that pot furnaces ( refractory vessels in
which glass is melted by indirect heating and where the openings are
covered with refractory stoppers during melting ) are exempted.  Each owner
or operator must either 1) vent all inorganic arsenic emissions from each
glass melting furnace to a control device and reduce emissions by at least
85%, the level of reduction achievable by an electrostatic precipitator or
fabric filter ( Section 61.162(a)(2) and (b)(2) ), or 2) maintain
uncontrolled (i.e. preceeding an add-on control device) arsenic emissions
at 2.5 Mg/year (2.75 TPY) or less for existing plants ( Section
61.162(a)(1) ), or at 0.4 Mg/year (0.44 TPY) or less for new plants
( Section 61.162(b)(1) ).

    If the owner or operator intends to meet the standard by using a
control device, s/he is required to continuously monitor opacity and
temperature, and to submit semiannual reports of excess opacity.  An owner
or operator may bypass the control device for a limited period of time for
designated purposes such as maintenance of the control device, upon prior
approval from the Regional Office.

                                  - 2 -

    If the owner or operator intends to meet the standard by limiting
uncontrolled arsenic emissions, s/he is required to calculate the
uncontrolled arsenic emissions semiannually, and to report if the emission
rate is above the applicable limit.

                                  - 3 -

A.  Source Notification

    The Regional Offices should notify all potentially affected sources
immediately following the promulgation of the arsenic NESHAP regulations,
or immediately upon identification of affected sources anytime after
promulgation ( see guideline S-17 for example notification ).  A list of all
know glass manufacturing plants using arsenic as a raw material is
attached ( Table 1 ).  This list includes 75 furnaces at 27 plants, and
includes both controlled and uncontrolled furnaces, and furnaces emitting
arsenic at levels both above and below the threshold of 2.5 Mg/Yr.  This
regulation would require two of these furnaces to install additional
controls or reduce arsenic usage, and would require at least six furnaces
to maintain their present controls.  However, this list may not be
exhaustive, and it includes many emission estimates.  Since new plants may
have been constructed, additional plants may have begun using arsenic as a
raw material, and some plants may be unidentified as of yet, additional
investigation should be made to complete the list.  Also, a number of
companies are investigating the possibility of reducing or eliminating
arsenic in soda-lime batch formulations, which may reduce the number of
affected facilities.  Preferably, all glass plants should be notified of
the regulations, because they will become subject if they begin using
arsenic as a raw material in the future.  All affected sources should be
coded into CDS.

                                  - 4 -

B.  Determination of Compliance ( 40 CFR 61.164 )

1.  Initial Report ( 40 CFR 61.10 )

    The owner or operator of each existing source is required to submit an
initial report to the Administrator by November 2, 1986.  This initial
report should include:

    -name and address of the owner or operator,

    -source location,

    -brief description of the nature, size, design, and method of
      operation ( including capacity and emission points ),

    -the average weight of arsenic processed per month over the
      previous 12 months as determined by direct measure of
      materials balance,

    -a description of the existing control equipment ( including
      efficiency ), and


    -a statement of the feasibility of complying with the
      standard by November 2, 1986.

If the owner or operator is unable to comply with the standard within the
90-day period, s/he may apply for a waiver of compliance ( See Guideline
S-19).  Sources which need to install control equipment may be granted a
waiver for up to two years if the time is needed for purchase and
installation.  Reasonable compliance schedules for installing fabric
filters and ESP's are attached ( Table 2 ).

    For any source for which a performance test is required, the owner or
operator must notify EPA at least 30 days before the test and must submit
the results to EPA within 60 days of the test, as indicated in the next
section.

    For any source which can demonstrate compliance by means of an
emission calculation alone, the owner or operator must submit to EPA by
September 18, 1986 ( or within 45 days of start-up or modification ) a
written report of the calculated estimates of arsenic emissions.  ( NOTE:
In the proposal, this report was required to be submitted within 90,
rather than 45, days.  Sources may be unprepared for this change and may
require more time. )

                                  - 5 -

    For new and modified sources ( for which construction or modification
commenced after July 20, 1983, including any existing furnace which begins
to use arsenic - see following discussion ), the owner or operator must
apply for approval to construct or modify ( required by Section 61.07 ) and
provide process and emission data so that EPA may determine if the source
will be able to comply with the standard.  After approval, the owner or
operator is required to notify EPA of the anticipated and actual start-up
dates as indicated in Section 61.90.

                                  - 6 -

Determination of Whether a Source is "Modified" based on Arsenic Usage

    Background Information

    "Modification" is defined in the General Provisions, 40 CFR Section
61.02 as "any physical change or change in the method of operation . . .
which increases the amount of any hazardous air pollutant . . . or which
results in the emission of any hazardous air pollutant not previously
emitted, except that . . . an increase of the production rate, if such
increase does not exceed the operating design capacity . . . ( or ) an
increase in hours of operation . . . ( shall not be considered a change in
the method of operation )".

    "New source" is defined as "any stationary source, the construction or
modification of which is commenced after ( proposal )".

    The preamble to the promulgated standards ( Federal Register Vol. 51,
No. 149, August 4, 1986, p. 27997 ) states "(s)ince proposal, the use of
arsenic in some glass melting furnaces has been eliminated and the Agency
believes that this trend is likely to continue.  The companies that
operate these furnaces have indicated that they do not plan to resume
using arsenic.  The cutoff applied to new or modified glass melting
furnaces is based on consideration of cost and economic factors and has
been retained in the promulgated standard to discourage reintroduction of
arsenic in furnaces that have recently eliminated its use and to
discourage future use.  The Agency believes that this is appropriate to
prevent risks from increasing near those furnaces that have recently
eliminated arsenic use and because reasonable alternatives to exceeding
this cutoff level are available at these facilities.  These include the
use of low-arsenic glass recipes and the use of controlled furnaces for
production of those glass types which would result in uncontrolled
emissions of arsenic of more than 0.4 Mg ( 0.44 ton ) per year.

                                  - 7 -

    Discussion

    Many furnaces subject to the arsenic NESHAP will typically melt a
variety of glasses with different arsenic contents and emission factors.
It is necessary to determine whether furnaces will become modified
sources, and thus subject to the more stringent emission limit, on the
basis of these changes in the method of operation.

    The above information indicates that if a furnace has never used
arsenic and starts arsenic use any time after proposal, that furnace
should be considered a modified source.

    If a furnace has used arsenic in the past, but has ceased its use, it
becomes a modified source at any point after proposal that it resumes the
use of arsenic.  Because arsenic usage is to be calculated as a rolling
12-month average every 6 months, if a furnace does not use arsenic during
any such 12-month period, ( starting from the 12-month period immediately
preceeding proposal ) that furnace should be considered a non-arsenic
furnace, and any addition of arsenic in the future will cause this furnace
to become subject to the more stringent standard for new and modified
furnaces.

    If a furnace has continuously used arsenic since the 12-month period
before proposal, it would be a modified source if arsenic emissions
increase above previous levels.  Operating records should be reviewed to
determine if there has been any 12-month rolling average where arsenic
emissions were higher than a previous 12-month period.  If so, the source
should be considered modified.  If not, the semiannual rolling averages
calculated by the source should routinely be monitored to see that
emissions do not increase in the future.  If emissions do increase, the
source is modified and is required to either install controls or change
operation in some way so that uncontrolled emissions will be limited to
0.4 Mg/Yr arsenic.

    There are several exceptions to this:

    (1)  A source may argue that this period of lowest arsenic emissions
is not representative of the typical operation of that furnace.  These
claims should be evaluated on a case-by-case basis.  However, if the
reason for the low arsenic emissions was that the furnace was successfully
using a substitute for arsenic, then the lower emission rate should be
considered representative operation.

                                  - 8 -

    (2)  If the increase in arsenic emissions is due solely to an increase
in production, then the furnace should not be considered "modified".
However, this refers to the production rate and hours of operation of the
furnace, not for the individual glass types.  Therefore, if a furnace has
increased production of a big-arsenic glass but at the same time has
decreased production of a low-arsenic or non-arsenic glass such that
overall arsenic emissions increase but total production remains constant,
then the furnace should be considered modified.

    In summary, for all furnaces which choose to demonstrate compliance
with the 2.5 Mg/Yr uncontrolled arsenic emission standard for existing
sources, their operating records for the period from August 22, 1982 ( 12
months before proposal ) to the present, as well as all future semiannual
calculations of uncontrolled arsenic emissions, should be reviewed to
determine whether the furnace has been modified because of these changes
in operation.

                                  - 9 -

2.  Emission Test ( 40 CFR 61.164 )

    By November 2, 1986 (or within 90 days of startup for a new source),
the owner or operator must test emissions from the source unless a waiver
of emission testing is obtained under Section 61.13 ( See Guideline S-20 ).

    The owner of operator must provide the Regional Office at least 30
days prior notice of the emission test and demonstration of the opacity
monitoring system, if applicable.

    Emission tests are to be conducted while the source is operating under
conditions that are representative of those from which the maximum arsenic
emissions will result, as may be specified by the Regional Office.
Usually, this will be under conditions representative of the expected
maximum ( allowable ) production rate.  However, for sources melting more
than one type of glass, or for sources with multiple furnaces emitting to
a single control device, the emission test should be conducted while the
source is operating at the expected maximum production rates for the glass
types generating the greatest amounts of arsenic.  Furnaces producing non-
arsenic glass should also be operating during the emission test, as would
be representative of a source's usual operation.  Another test may be
required later if source operation changes so that the original testing
operating conditions are no longer representative of "worst case"
operation.

    The owner or operator must furnish the Regional Office with a written
report of the emission test results and associated calculations within 60
days of the test, and must retain records of emission test results and
other data needed to determine emissions for two years.

                                 - 10 -

    Furnaces with Uncontrolled Arsenic Emissions Above
      2.5 Mg/yr ( existing ) or 0.4 Mg/yr ( new or modified )
      ( Section 61.164(e) )

    Unless an alternative test method is approved ( refer to Guideline S-18
for procedure ), the owner or operator of each furnace must demonstrate
compliance with the 85% arsenic reduction requirement in Section
61.162(a)(2) or (b)(2) by using Method 108 to determine the concentration
of arsenic in the inlet and outlet has streams to the control device.
Each emission test is to consist of three 60-minute test runs, each
consisting of simultaneous testing of the inlet and outlet gas streams.
The gas streams must contain all of the gas exhausted from the gas melting
furnace.

    The percent reduction for each run will be computed as follows:

                        D = ( Cb - Ca ) x 100
                            _______________
                                  Cb

              D  = percent emission reduction

              Cb = arsenic concentration in stack gas entering the
                   control device, as measured by Method 108

              Ca = arsenic concentration in stack gas exiting the
                   control device, as measured by Method 108

The average percent reduction is equal to the arithmetic mean of the
results for the three runs, and must be equal to or greater than 85% for
the source to be in compliance.

                                 - 11 -

    Furnaces with Uncontrolled Arsenic Emissions Under
      2.5 Mg/yr ( existing ) or 0.4 Mg/yr ( new or modified )
      ( Section 61.164(c) and (d) )

    If less than 8.0 Mg arsenic/year ( 8.8 TPY ) is added to an existing
furnace, or less than 1.0 Mg arsenic/year ( 1.1 TPY ) is added to a new or
modified furnace, the owner or operator will usually be able to
demonstrate compliance with the uncontrolled emission limits by an
emission calculation only.  A theoretical arsenic emission factor should
be calculated for each type of glass produced ruing the 12-month period,
as follows:




                   Ti = ( Abi x Wbi ) + ( Aci x Wci ) - Agi

              Ti  = theoretical uncontrolled arsenic emission factor
                    (g/kg) for each glass type (i)

              Abi = fraction by weight of arsenic in fresh batch for
                    each glass type (i)

              Wbi = weight (g) of fresh batch melted per kg of glass
                    produced for each glass type (i)

              Aci = fraction by weight of arsenic in cullet for each
                    glass type (i)

              Wci = weight (g) of cullet melted per kg of glass produced
                    for each glass type (i)

              Agi = weight (g) of arsenic per kg glass produced for
                    each glass type (i)

The theoretical uncontrolled arsenic emissions for the 12-month period is
estimated as follows:

                        Yi = ( Ti x Gi )
                             _________
                                106

              Yi  = theoretical uncontrolled arsenic emission estimate
                    for the 12-month period for each glass type ( Mg/yr )

              Ti  = theoretical uncontrolled arsenic emission factor
                    for each type of glass produced during the 12-month
                    period (as calculated above)

              Gi  = kg of each arsenic-containing glass type (i)
                    produced during the 12-month period

                                 - 12 -

The total theoretical uncontrolled arsenic emissions for each 12-month
period is equal to the sum of these emission estimates ( Yi ) for each glass
type produced.  If this is equal to or less than 2.5 Mg for existing
plants, or 0.4 Mg for new plants, the source is in compliance and no
emission testing is required.  If the total is above these limits, then
the source is required to test as described below.
    The following procedure is required for existing sources using more
than 8.0 Mg arsenic/year, new sources using more than 1.0 Mg arsenic/year,
and for sources using less than these amounts but which are unable to
demonstrate compliance solely by the calculation procedure above.  The
theoretical uncontrolled arsenic emission factors ( Ti ) and estimates (Yi)
should again be calculated for each glass type produced during the 12-
month period as described above.  Emission testing, using Method 108, must
then be conducted during production of the glass type with the highest
theoretical uncontrolled arsenic emissions.  The actual uncontrolled
arsenic emission factor should be computed as follows:

                        Ra = Ea
                             __
                             P

              Ra  = actual uncontrolled arsenic emission factor ( g/kg )

              Ea  = actual uncontrolled arsenic emission rate, from
                    Method 108 ( g/h )

              P   = rate of glass production ( kg/h ), determined by
                    dividing the weight of glass pulled from the furnace
                    during the emission test by the number of hours taken
                    to perform the test

A furnace correction factor (F) to relate the theoretical and actual
uncontrolled arsenic emission factors should be calculated as follows ( Ra
and Ti should be the same glass type ):

                        F = Ra
                            __
                            Ti

                                 - 13 -

    The total uncontrolled arsenic emission rate for the 12-month period
should be computed by applying this furnace correction factor to all of
the theoretical emission factors, as follows:

U = ( Summation of n where i = 1 )*   ( T sub i x F x G sub i )/106th power

         U = total uncontrolled arsenic emission rate ( Mg/year )

         n = number of arsenic-containing glass types produced
             during the 12-month period

If the total uncontrolled arsenic emission rate is less than 2.5 Mg/yr for
an existing furnace, or 0.4 Mg/yr for a new furnace, the source is in
compliance.  If the total is above these values, then the source is in
violation and must install controls.  However, the source may opt to
conduct Method 108 tests on the remaining glass types compute type
specific correction factors, and attempt to demonstrate compliance in that
way.

Example 1:

    If the glass type produced during the Method 108 test is the only
    glass type to be produced for the initial 12-month period, then
    the actual arsenic emission factor can simply be multiplied by the
    amount of glass produced to calculate total yearly arsenic
    emissions.  ( If less than 8.0 Mg (or 1.0 Mg) arsenic/year were
    added to the furnace, a Method 108 test would be unnecessary. )

              Ea = .045 lb/hr ( from Method 108 )

              P  = 900 lb/hr

              Ra = Ea = .045 = .1 lb As/ton glass
                   __   ____
                   P     900

              Total yearly arsenic emissions = (Ra) (annual
                production) = (.1 lb/ton)(4000 ton/yr)= .2 TPY As

                                 - 14 -

Example 2:

    If two or more glass types are produced, a theoretical arsenic
emission factor, based on a materials balance, should be calculated for
every type of glass that will be produced.  This should be multiplied by
the correction factor to calculate an actual arsenic emission factor for
each type of glass.  Each actual arsenic emission factor should then be
multiplied by the amount of that glass that will be produced to calculate
yearly arsenic emissions for each glass type, and the results summed to
calculate total yearly arsenic emissions.

         Assume 3 types of glass (A,B,C) are produced in one
              furnace

         For Glass A, from above, Ra(A)= .1 lb As/ton glass

              Annual production of Glass A = 3000 TPY

              Theoretical arsenic emission factor (TA) =
                   .08 lb As/ton glass

              Correction factor = .1 = 1.25
                                  __
                                 .08

              For Glass B, TB = .075 lb/ton

                   Ra(B)= (.075)(1.25) = .09 lb/ton

                   Annual production of Glass B = 500 TPY

              For Glass C, TC = .4 lb/ton

                   Ra(C) = (.4)(1.25) = .5 lb/ton

                   Annual production of Glass C = 750 TPY

              Total yearly arsenic emissions

                   = (Ra(A))(A's annual production) + (RaB))(B's
                     annual production) + (Ra(C))(C's annual
                     production)

                   = (.1 lb/ton)(3000 TPY) + (.09 lb/ton)(500 TPY)
                     + (.5 lb/ton)(750 TPY)

                   = .15 TPY + .021 TPY + .19 TPY = .36 TPY

                                 - 15 -

    The Test Methods in Appendix B of Part 61 are to be used unless an
alternative method has been approved by the Director of the Emission
Standards and Engineering Division.  If the results obtained by an
alternative method are thought to be inaccurate, the Regional Office may
require the use of a reference method.  If the results obtained by the
reference method do not agree with those of the alternative method, the
results obtained by the reference method will prevail.

                                 - 16 -

C.  Emission Monitoring ( 40 CFR 61.163 )

    An owner or operator complying with Section 61.162(a)(2) or (b)(2)
must install, calibrate, maintain, and operate

    1) a continuous monitoring system for measuring opacity of
       the exhaust gas and

    2) a monitoring device for the continuous measurement of the
       temperature of the gas entering the control device.

These should be installed, and their operational status verified, prior to
the emissions test.  A report of the CEM evaluation should be furnished to
the Regional Office within 60 days of the evaluation.  The purpose of the
transmissometer will be to indicate when the control device may not be
operating properly and emissions may be exceeding the applicable limit.
The reference method used to demonstrate compliance with the emission
limitation remains Method 108.  As described in the following discussion,
a source specific opacity limit is to be derived for each individual
facility, which will be based on the opacity during an emissions test
demonstrating compliance.  This level would be viewed as indicative of a
properly operated and maintained control device.

    Opacity should be monitored during each of the three runs of the
emission test.  During the emission test, process and control equipment
should be operated so that opacity is minimized, as may be specified by
the Regional Office.  Monitoring results should be reduced to 6-minute
averages, and a source-specific opacity limit corresponding to the 97.5%
upper confidence level of a normal or lognormal ( whichever is more
representative ) distribution of the average opacity values should be
determined.  Temperature of the gas entering the control device should
also be monitored during each test run, and 15-minute temperature averages
should be determined.  An owner or operator may redetermine both these
values if this procedure is repeated during each test run of an emission
test demonstrating compliance.

    All continuous monitoring systems should be in continuous operation as
described in Section 61.163(f).  All opacity data should be reduced to 6-
minute averages, not including data from periods of breakdowns, repairs,
calibration checks, and zero and span adjustments.  Fifteen-minute
averages of temperature should also be calculated.

    The Regional Office may approve, after receipt and consideration of
written application, an alternative continuous monitoring system
( parameter-based, etc. ) to replace the CEM.

                                 - 17 -

D.  Recordkeeping ( 40 CFR 61.165 )

    All owners or operators of glass melting furnaces using arsenic as a
raw material are subject to recordkeeping and reporting requirements.
Each owner or operator must retain for a minimum of two years the
following information:

.   1) all measurements, including continuous monitoring for
       opacity and temperature,

.   2) all calculations used for emission estimates and all
       records of emission test data,

.   3) all monitoring system performance evaluations, including
       calibration checks and adjustment,

.   4) occurrence and duration of all startups, shutdowns or
       malfunctions of furnace,

.   5) all malfunctions of air pollution control system,

.   6) all periods when any continuous monitoring system or
       device is inoperative,

.   7) all maintenance and repairs made to each air pollution
       control system, continuous monitoring system, or
       monitoring device, and

.   8) if permission to bypass the control device is obtained,
       the dates the control device is bypassed and steps taken
       to minimize arsenic emissions during that period.

    Additionally, each owner or operator of a glass plant complying with
Section 61.162(a)(1) or (b)(1) must determine and record every six months:

    1) the uncontrolled arsenic emission rate for the preceeding
       12-month period ( or 6-month period, for the first determination )
       using measured or calculated arsenic emission factors ( as
       applicable ) multiplied by each respective glass production rate,
       and


                                 - 18 -

    2) an estimate of the uncontrolled arsenic emission rate
       for the forthcoming 12-month period, taking into
       consideration anticipated changes in production rates,
       glass types, and other factors.
For these semiannual determinations, it would not be necessary to conduct
a Method 108 test again.  The initial correction factor could be applied
again to calculate the measured arsenic emission factor for each glass
type.

                                 - 19 -

E.  Reporting ( 40 CFR 61.165 )

    Each owner or operator complying with Section 61.165(a)(2) or (b)(2)
must submit written reports to the Administrator semiannually if excess
opacity occurred during the proceeding six-month period.  An occurrence of
excess opacity is any 6-minute period where the average opacity exceeded
the source-specific opacity level.  Excess opacity reports would not be
used to cite a source in violation, but would alert enforcement personnel
that the control device may not be operated and maintained properly and to
indicate that an inspection and/or emission test may be appropriate.  All
semiannual reports should include:

    1) magnitude of excess opacity, conversion factors used,
       dates and times of each occurrence,

    2) specific identification of excess opacity occurring
       during start-ups, shutdowns, and malfunctions, and

    3) dates and times of each period when the continuous
       monitoring system was inoperative ( except for zero and
       span checks ) and the nature of repairs or adjustments.

These reports must be postmarked by the 30th day following the end of the
six-month period.

    An owner or operator may apply to the Regional Administrator for
approval to bypass the control device for limited periods, as described
previously.  This application must be submitted at least 60 days before
the bypass period is to begin, and should include:

.    1) name and address of owner or operator,

.   2) location of source,

.   3) description of nature, size, design, and operation of source,

.   4) the reason it is necessary to bypass the control device,

.   5) the length of time needed to bypass the control device,

                                 - 20 -

.   6) steps that will be taken to minimize arsenic emissions
       during the bypass,

.   7) the quantity of emissions that would be released if no
       steps were taken to reduce emissions,

.   8) the expected reduction in emissions due to steps taken
       during the bypass to minimize emissions, and

.   9) the type of glass to be produced during the bypass and an
       explanation of why non-arsenic or lower-arsenic glass
       could not be melted during the bypass period.

    If an owner or operator of a source complying with the 85% arsenic
reduction requirement wishes to reduce arsenic usage and comply with the
uncontrolled arsenic emission limitation instead, s/he should notify the
Regional Office of this change and include the necessary calculations and
emission test data to demonstrate that uncontrolled emissions will remain
below 2.5 (pr 0.4) Mg/year.

    Each owner or operator complying with Section 61.162 (a)(1) or (b)(1)
must report the uncontrolled arsenic emission rate if uncontrolled arsenic
emissions exceed 2.5 Mg/yr for existing plants, or 0.4 Mg/yr for new
plants.  If estimates show that arsenic emissions have exceeded 2.5 (or
0.4) Mg/yr for the preceeding 12-month period (or 6-month period, in a
first report following the compliance demonstration), this is a violation
and must be reported within 10 days of the end of the 6-month reporting
period.  If estimates show that arsenic emissions will exceed 2.5 (or 0.4)
Mg/yr, the owner or operator must comply with Section 61.162 (a)(2) or
(b)(2) and, within 10 days, notify the Regional Office of the anticipated
date of the emission test.

                                 - 21 -

    Table 1:  Emission Control for Arsenic Using Glass Plants

                                                               Expected
Plant No.            Name/Location               Number of    Compliance
                                                 Furnaces      Method a,b

  1      Corning, Martinsburg, WV                   1             PRc .
  2      Corning, Charleroi, PA                     1             PR
         Corning, Charleroi, PA                     1             CU
  3      Corning, Fall Brook, NY                    2             PR
         Corning, Fall Brook, NY                    3             UEL .
  4      Corning, State College, PA                 1             PR
  5      GTE-Sylvania, Central Falls, RI            1             PR
  6      North American Phillips, Danville, KY      1             PR
  7      Blenko Glass, Milton, WV                   1             UEL .
  8      Brooke Glass Co., Wellsburg, WV            2             UEL
  9      Corning, Corning, NY                       2             UEL
  10     Davis-Lynch Glass, Start City, WV          1             UEL
  11     Fenton Art Glass, Williamston, WV          4             UEL .
  12     Fostoria Glass, Moundsville, WV            1             UEL
  13     GTE, Versailles, KY                        1             UEL
  14     Indiana Glass, Dunkirk, IN                 9             UEL
  15     Jeanette Shade & Novelty, Jeanette, WV     3             UEL .
  16     Nourot Glass, Benica, CA                   2             UEL
  17     Owens-Illinois, Shreveport, LA             3             UEL
  18     Owens-Illinois, Mt. Pleasant, PA           1             UEL
  19     Owens-Illinois, Pittston, PA               2             UEL .
  20     Owens-Illinois, Toledo, OH                 9             UEL
  21     Paul Wissnach Glass, Paden City, WV        5             UEL
  22     Peltier Glass Co., Ottawa, IL              6             UEL
  23     RCA, Circleville, OH                       2             UEL .
  24     Scandia Glass Works, Kenava, WV            2             UEL
  25     Shott Optical, Duryea, PA                  3             UEL
  26     Vandermark Merritt Glass, Flemington, NJ   1             UEL
  27     Westmoreland Glass Co., Pittsburgh, PA     4             UEL .

a UEL = Uncontrolled Emission Limt  (2.5 Mg/yr)

   PR = Percent Reduction (85%)

   CU = Cease Arsenic Use



b Some of the furnaces emitting under 2.5 Mg arsenic/year also
   have control devices, and may comply using either method

c Needs to install controls

                                 - 22 -

Plants that are believed to have removed arsenic after proposal and which
would be subject to 0.4 Mg arsenic/year emission limit if arsenic is
re-introduced into glass:

    1.   American Stemware Corp.

    2.   Anchor-Hocking, Lancaster, OH

    3.   Anchor-Hocking, Baltimore, MD

    4.   Anchor-Hocking, Baltimore, MD

    5.   Corning, Charleroi, PA ( Soda-Lime furnace only )

    6.   Harvey Industries, Clarksburg, WV

    7.   Wheaton Industries, Millsville, NJ

Plants known to have used arsenic, but which were closed at last report:

    1.   Seneca Glass Company, Morgantown, WV

    2.   Sloan Glass, Inc., Culloden, WV

                                 - 23 -

                     TABLE 1:  Compliance Schedules

                                           Fabric Filter          ESP
     Time (days)

Contracts awarded or purchase
     orders issued                               60                60

Fabrication                                     270               360

Shipping                                         30                30

Installation                                    240               150 .

Start-up                                         40                40

Sampling, analysis, report                       90                90

Total                                           730               730 .



Return to Enforcement Policies