National Emission Standards for Hazardous Air Pollutants: Proposed Standards for Hazardous Air Pollutants for Hazardous Waste Combustors (Phase I Final Replacement Standards and Phase II) [[pp. 21297-21346]]

 [Federal Register: April 20, 2004 (Volume 69, Number 76)]
[Proposed Rules]
[Page 21297-21346]
From the Federal Register Online via GPO Access [wais.access.gpo.gov]
[DOCID:fr20ap04-27]
 
[[pp. 21297-21346]]
National Emission Standards for Hazardous Air Pollutants: 
Proposed Standards for Hazardous Air Pollutants for Hazardous Waste 
Combustors (Phase I Final Replacement Standards and Phase II)

[[Continued from page 21296]]

[[Page 21297]]

standard for existing sources, we conclude that a beyond-the-floor DRE 
standard for new sources is not warranted. Consequently, after 
considering nonair quality health and environmental impacts and energy 
requirements, we are proposing the floor DRE standard for new sources.

XIII. What Is the Rationale for Proposing an Alternative Risk-Based 
Standard for Total Chlorine in Lieu of the MACT Standard?

    Under authority of CAA section 112(d)(4), we propose standard 
procedures to allow you to establish a risk-based emission limit for 
total chlorine in lieu of compliance with the section 112(d)(2) MACT 
emission standard. See proposed Sec.  63.1215. The risk-based approach 
would be applicable to all hazardous waste combustors except 
hydrochloric acid production furnaces. Because we are proposing to use 
the MACT standard for total chlorine as a surrogate to control metal 
HAP for the hydrogen chloride production furnace source category, we 
cannot allow any variance from the standard. For the other hazardous 
waste combustor source categories, we are proposing the section 
112(d)(4) standard as an alternative to the MACT standard. Sources 
could choose which of these two standards they would prefer to apply.
    The alternative risk-based emission limit for total chlorine would 
be based on national exposure standards established by EPA that ensure 
protection of public health with an ample margin of safety. The 
standard would consist of a nationally-applicable, uniform algorithm 
that would be used to establish site-specific emission limitations 
based on site-specific input from each source choosing to use this 
approach. Thus, these standards would provide a uniform level of risk 
reduction, consistent with the requirement of section 112(d)(4) that 
EPA establish ``emission standards'', i.e., a requirement established 
by EPA which limits quantity, rate or concentration of air emissions 
(see CAA section 302(k)).
    We also request comment on an alternative approach to implement 
section 112(d)(4) for cement kilns in which we establish a national 
risk-based emission standard for total chlorine that would be 
applicable to all cement kilns. Under this approach, EPA would issue a 
single total chlorine emission standard using an emission level that 
meets our national exposure standards if each cement kiln were to emit 
at that level.
    We believe that most hazardous waste combustors are likely to 
consider establishing risk-based standards for total chlorine because 
the MACT standards proposed today are more stringent, and in some cases 
substantially more stringent, than currently applicable standards 
(e.g., the total chlorine standard for incinerators is currently 77 
ppmv while we propose today a MACT standard of 1.4 ppmv).

A. What Is the Legal Authority To Establish Risk-Based Standards?

    Under the authority of section 112(d)(4), the Administrator may 
establish emission standards based on risk, in lieu of the technology-
based MACT standards, when regulating HAP for which health threshold 
levels have been established. Under section 112(d)(4), Congress gave 
EPA the discretion to consider the health threshold of any HAP and to 
use that health threshold, with an ample margin of safety, to set 
emission standards for the source category or subcategory. In the 
legislative history accompanying this provision, the Senate Report stated,

    ``To avoid expenditures by regulated entities that secure no 
public health or environmental benefit, the Administrator is given 
discretionary authority to consider the evidence for a health 
threshold higher than MACT at the time the standard is under review. 
The Administrator is not required to take such factors into account; 
that would jeopardize the standard-setting schedule imposed under 
this section with the kind of lengthy study and debate that has 
crippled the current program. But where health thresholds are well 
established, for instance in the case of ammonia, and the pollutant 
presents no risk of other adverse health effects, the Administrator 
may use the threshold with an ample margin of safety (and not 
considering cost) to set emissions limitations for sources in the 
category or subcategory.'' (S. Rep. No. 228, 101st Cong. 1st Sess. 
at 171 (1989); see also id. at 175-176 (1989).)

    EPA has previously used section 112(d)(4) authority in the 
Industrial Boiler and Process Heater MACT Final Rule signed Feb. 26, 
2004, the Pulp and Paper MACT Phase II (66 FR 3180, January 12, 2001) 
and the Lime Manufacturing MACT (69 FR 394, January 5, 2004), and has 
proposed to use it in a different manner in several other MACT 
rulemakings (e.g., the Reciprocating Internal Combustion Engine MACT 
(67 FR 77830, December 19, 2002).\164\ The approach we propose today is 
nearly identical to the approach EPA recently adopted for the 
Industrial Boiler and Process Heater MACT source category, which allows 
a source to establish a site-specific risk-based emission limit for 
threshold HAP using prescribed procedures. This approach differs from 
the previous MACT rules where EPA simply determined, on a national 
basis, what level of exposure from each source in the category would be 
protective of public health with an ample margin of safety, and did not 
pose significant adverse environmental impacts. This previous approach 
resulted in a determination that no standard was necessary because no 
source in the category could exceed such a risk-based standard. Today's 
proposal varies in that the level of protection afforded by the 
standard is uniform, but the limits for individual sources differ due 
to site-specific factors. As explained later in this section of the 
preamble, EPA is, however, also considering for cement kilns applying 
the single national standard approach adopted in earlier rules.
---------------------------------------------------------------------------

    \164\ The Agency also proposed to use Section 112(d)(4) 
authority in two other MACT rulemakings--the Combustion Turbine MACT 
(68 FR 1888, January 14, 2003), and the Chlorine Production MACT (67 
FR 44671)--but determined that MACT standards for those source 
categories are not warranted and delisted the source categories from 
the section 112(c) list of major sources pursuant to the authority 
in section 112(c)(9).
---------------------------------------------------------------------------

B. What Is the Rationale for the National Exposure Standards?

    We identify as national exposure standards threshold levels that 
are protective of human health from both chronic and acute exposure. In 
addition, because EPA has discretion whether or not to promulgate risk-
based standards pursuant to section 112(d)(4), we would not allow an 
alternative standard where emission levels may result in adverse 
environmental effects that would otherwise be reduced or eliminated. We 
would not issue the alternative standard even though it may be shown 
that emissions do not approach or exceed levels requisite to protect 
public health with an ample margin of safety because we believe the 
statute requires that we consider effects on terrestrial animals, 
plants, and aquatic ecosystems in addition to public health in 
establishing a standard pursuant to section 112(d)(4). See S. Rep. 228 
at 176: ``Employing a health threshold or safety level rather than the 
MACT criteria to set standards shall not result in adverse 
environmental effects which would otherwise be reduced or eliminated.''
1. What Are the Human Health Threshold Levels?
    a. Chronic Exposure. Hydrogen chloride is corrosive to the eyes, 
skin, and mucous membranes. Chronic exposure may cause gastritis, 
bronchitis, dermatitis, and dental discoloration and erosion. Chronic 
exposure to chlorine gas can cause respiratory effects

[[Page 21298]]

including eye and throat irritation and airflow obstruction. See 
discussion in Part One, Section I.E of this preamble.
    Given that neither hydrogen chloride nor chlorine gas is known to 
produce a carcinogenic response,\165\ we use reference air 
concentrations (RfC) to assess the likelihood of non-cancer health 
effects in humans. The RfC is an estimate of a continuous inhalation 
exposure to the human population, including sensitive subgroups, that 
is likely to be without an appreciable risk of deleterious effects over 
a lifetime. We use an RfC for hydrogen chloride of 20 [mu]g/m\3\, as 
presented in EPA's Integrated Risk Information System (IRIS). We 
propose to use an RfC for chlorine gas of 0.2 [mu]g/m\3\ based on a 
provisional assessment prepared by EPA on inhalation hazards from 
chlorine.\166\ This is the same as the value for chlorine used by the 
State of California's Office of Environmental Health Hazard Assessment, 
which they refer to as a chronic ``Reference Exposure Level'' (REL). 
Because RfCs can change over time based on new information, the rule 
would require you to use the current RfC value found at 
http://epa.gov/ttn/atw/toxsource/summary.html.

---------------------------------------------------------------------------

    \165\ EPA conducted an assessment of the carcinogenicity of 
chlorine gas and concluded that it is not likely to be a human 
carcinogen (see EPA's June 22, 1999 Risk Assessment Issue Paper for 
Derivation of a Provisional Chronic Inhalation RfC for Chlorine, 
p.12). The International Agency for Research on Cancer (IARC) 
concluded that hydrochloric acid is not classifiable as to its 
carcinogenicity to humans (see IARC Monographs, Vol. 54: 
Occupational Exposures to Mists and Vapours from Strong Inorganic 
Acids; and Other Industrial Chemicals (1992) p.189).
    \166\ See EPA's externally peer-reviewed ``Risk Assessment Issue 
Paper for Derivation of a Provisional Chronic Inhalation RfC for 
Chlorine'' (June 22, 1999) that can be found in the docket for 
today's proposal.
---------------------------------------------------------------------------

    We considered how to account for the fact that chlorine gas 
photolyzes in the atmosphere in bright sunlight to chlorine ions and 
then quickly reacts with hydrogen or methane to form hydrogen chloride. 
The half-life of chlorine due to photolysis in bright sunlight is 
estimated to be 10 minutes.\167\ Nonetheless, this is generally 
sufficient time for the plume to reach nearby ground-level receptors 
without significant transformation. In addition, such transformation is 
possible only a portion of the time. Photolysis does not occur at night 
and is reduced on overcast or cloudy days. Generally speaking, the rate 
of photolysis depends on the particular wavelength and intensity of 
solar radiation reaching the earth's surface which varies greatly 
depending on the solar angle which changes with the time of day, the 
season of the year, and the latitude at a given location. While the 
ideal approach would be explicit modeling of photolysis rates as a 
function of solar insolation, sky conditions, absorption cross-section, 
quantum yield, and subsequent transformation to hydrogen chloride, to 
our knowledge no such regulatory air dispersion model currently exists.
---------------------------------------------------------------------------

    \167\ As determined by a modeling analysis done by the Air 
Pollution Research Center at the University of California at 
Riverside, as reported in a California Air Resources Board fact 
sheet, ``Toxic Air Contaminant Identification List Summaries--ARB/
SSD/SES,'' p. 231, September 1997. See also 
http://www.arb.ca.gov/toxics/tac/factshts/chlorine.pdf. 

---------------------------------------------------------------------------

    Because it is reasonable to believe that receptors will be exposed 
to chlorine gas before appreciable transformation occurs due to the 
variability and complexity of the transformation and the fact that 
chlorine gas is considerably more toxic than hydrogen chloride, we 
conclude that, for the purpose of protection of public health, it is 
prudent to assume that chlorine gas is not transformed to hydrogen chloride.
    b. Acute Threshold Levels. Short-term exposure to hydrogen chloride 
may cause eye, nose, and respiratory tract irritation and inflamation 
and pulmonary edema. Short-term exposure to high levels of chlorine gas 
can result in chest pain, vomiting, toxic pneumonitis, and pulmonary 
edema. At lower levels, chlorine gas is a potent irritant to the eyes, 
the upper respiratory tract, and lungs. See Part One, Section I.E. 
Please note that, although we discuss here how we would consider acute 
exposure, we conclude below that you need not assess acute exposure to 
establish an emission limit for total chlorine. See discussion in 
Section B.2.e.
    To assess effects from acute exposure, we would use the acute 
exposure guideline level (AEGL). AEGL toxicity values are estimates of 
adverse health effects due to a single exposure lasting 8 hours or 
less. Consensus toxicity values for effects of acute exposures have 
been developed by several different organizations. EPA, in conjunction 
with the National Research Council and National Academy of Sciences, is 
in the process of setting acute exposure guideline levels. A national 
advisory committee organized by EPA has developed AEGLs for priority 
chemicals for 10-minute, 30-minute, 1-hour, 4-hour, and 8-hour airborne 
exposures. They have also determined for each exposure duration the 
levels of these chemicals that will protect against notable discomfort 
(AEGL-1), serious effects (AEGL-2), and life-threatening effects or 
death (AEGL-3).\168\ To be protective of public health, we propose to 
use the AEGL-1 values to assess acute exposure: 2.7 mg/m\3\ (1.8 ppm) 
for hydrogen chloride, and 1.4 mg/m\3\ (0.5 ppm) for chlorine gas.\169\ 
Airborne concentrations of a substance above the AEGL-1 could cause 
notable discomfort, irritation, or certain asymptomatic nonsensory 
effects in the general population, including susceptible individuals. 
Please note, however, that airborne concentrations below the AEGL-1 
could produce mild odor, taste, or other sensory irritations. Effects 
above the AEGL-1 (but below the AEGL-2) are not disabling and are 
transient and reversible upon cessation of exposure.
---------------------------------------------------------------------------

    \168\ The full definitions of the AEGL values are more nuanced. 
AEGL 1: The airborne concentration of a substance above which it is 
predicted that the general population, including susceptible 
individuals, could experience notable discomfort, irritation, or 
certain asymptomatic nonsensory effects. However, the effects are 
not disabling and are transient and reversible upon cessation of 
exposure. AEGL 2: The airborne concentration of a substance above 
which it is predicted that the general population, including 
susceptible individuals, could experience irreversible or other 
serious, long-lasting adverse health effects or an impaired ability 
to escape. AEGL 3: The airborne concentration of a substance above 
which it is predicted that the general population, including 
susceptible individuals, could experience life-threatening health 
effects or death.
    \169\ For hydrogen chloride and chlorine gas (individually), the 
AEGL-1 values for 10-minute, 30-minute, 1-hour, and 8-hour exposures 
are the same. Therefore, when comparing predicted ambient levels of 
exposure to the AEGL-1 value, we believe it is reasonable to 
evaluate maximum 1-hour ground level concentrations.
---------------------------------------------------------------------------

2. What Exposures Would You Be Required to Assess?
    We discuss below the following issues: (1) Use of the Hazard Index 
to assess exposure to both hydrogen chloride and chlorine gas; (2) 
exposure to emissions of respiratory irritant HAP other than hydrogen 
chloride and chlorine gas; (3) exposure to emissions of respiratory 
irritant HAP from collocated sources; (4) exposure to ambient 
background levels of respiratory irritant HAP; and (5) our conclusion 
that acute exposure need not be assessed to establish emission limits 
because the Hazard Index for chronic exposure is expected to be higher 
in all situations.
    a. Hazard Index. Noncancer risk assessments typically use a metric 
called the Hazard Quotient (HQ) to assess risks of exposures to 
noncarcinogens. The HQ is the ratio of a receptor's potential exposure 
(or modeled concentration) to the health reference value or threshold 
level (e.g., RfC or AEGL) for an individual pollutant. HQ values less 
than 1.0 indicate that exposures are below the

[[Page 21299]]

health reference value or threshold level and, therefore, that such 
exposures are without appreciable risk of adverse effects in the 
exposed population. HQ values above 1 do not necessarily imply that 
adverse effects will occur, but that the likelihood of such effects in 
a given population increases as HQ values exceed 1.0.\170\
---------------------------------------------------------------------------

    \170\ See US EPA Glossary of Key Terms for National Air Toxics 
Assessment, at http://www.epa.gov/ttn/atw/nata/gloss1.html.

---------------------------------------------------------------------------

    When the risk of noncancer effects from exposure to more than one 
pollutant to the same target organ must be assessed, the effects are 
generally considered to be additive and the HQ values for each 
pollutant are summed to form an analogous metric called the Hazard 
Index (HI). Assuming additivity, HI values less than 1.0 indicate that 
exposures to the mixtures are likely to be without appreciable risk of 
adverse effects in the exposed population. HI values above 1.0 do not 
necessarily imply that adverse effects from exposure to the mixture 
will occur, but that the likelihood of such effects in a given 
population increases as HI values exceed 1.0.
    For purposes of establishing risk-based emission limits for total 
chlorine, we propose to allow a maximum HI value of not greater than 
1.0.
    b. Exposure to Emissions of HAP other than Hydrogen Chloride and 
Chlorine Gas that Have a Common Mechanism of Action. We have identified 
in the table below 40 HAP that are respiratory irritants, including 
hydrogen chloride and chlorine gas. Because these HAP have a common 
mechanism of action, we must determine whether exposure to these HAP 
must be considered when determining that the HI is less than or equal 
to 1.0.
Respiratory Irritant HAP
1,2-Epoxybutane
1,3-dichloropropene
2,4-Toluene diisocyanate
2-Chloroacetophenone
Acetaldehyde
Acrolein
Acrylic acid
Acrylonitrile
Antimony
Beryllium
Bis(2-ethylhexyl)phthalate
Chlorine
Chloroprene
Chromium
Cobalt
Diethanolamine
Epichlorohydrin
Ethylene glycol
Formaldehyde
Hexachlorocyclopentadiene
Hexamethylene 1,6-diisocyanate
Hydrochloric acid
Maleic anhydride
Methyl bromide
Methyl isocyanate
Methyl methacrylate
Methylene diphenyl diisocyanate
N-hexane
Naphthalene
Nickel
Nitrobenzene
Phosgene
Phthalic anhydride
Propylene dichloride
Propylene oxide
Styrene oxide
Titanium tetrachloride
Toluene
Triethylamine
Vinyl acetate

    In making this determination, we would consider only those 
respiratory irritants that are HAP (as opposed to also considering 
respiratory irritants that are criteria pollutants) not only because 
section 112 deals with control of emissions of HAP, but also because 
ambient levels of criteria pollutants that have a common mechanism of 
action with hydrogen chloride and chlorine gas (e.g., SO_X, 
NO_X, PM, ozone) are controlled through the applicable State 
Implementation Plans demonstrating compliance with the National Ambient 
Air Quality Standards for these pollutants.
    In addition to hydrogen chloride and chlorine gas, several of the 
respiratory irritant HAP listed in the table above may be emitted by 
hazardous waste combustors, including the metals antimony trioxide, 
beryllium, chromium (VI), cobalt, and nickel, and the organic compounds 
Bis(2-ethylhexyl)phthalate, formaldehyde, napthalene, and toluene.\171\ 
We do not believe, however, that these respiratory irritant HAP would 
be emitted by hazardous waste combustors at levels that would result in 
significant Hazard Quotient values. Beryllium and chromium would be 
controlled by emission standards for low volatile metals and the 
remaining metal HAP would be controlled by a particulate matter 
standard. Emissions of the respiratory irritant organic HAP would be 
controlled to trace levels by the MACT standards for carbon monoxide or 
hydrocarbons and destruction and removal efficiency (DRE). Accordingly, 
we propose to require you to quantify and assess emissions from the 
hazardous waste combustor of hydrogen chloride and chlorine gas only; 
you would not be required to account for these other respiratory 
irritant HAP because they would not contribute substantially to the 
Hazard Index.
---------------------------------------------------------------------------

    \171\ Betty Willis, et al., Agency for Toxic Substances and 
Disease Registry, U.S. Department of Health and Human Services, 
``Public Health Reviews of Hazardous Waste Thermal Treatment 
Technologies: A Guidance Manual for Public Health Assessors,'' March 
2002, Table 4.
---------------------------------------------------------------------------

    c. Exposure to Emissions of Respiratory Irritant HAP from 
Collocated Sources. You would be required to account for exposure to 
emissions of hydrogen chloride and chlorine gas from all on-site 
hazardous waste combustors subject to subpart EEE, part 63. EPA will 
address exposure to emissions of respiratory irritant HAP from other 
sources that may be collocated with a hazardous waste combustor--for 
example, process vents and fossil fuel boilers--under the residual risk 
requirements of section 112(f) for both hazardous waste combustors and 
(potentially) other MACT source categories. See A Legislative History 
of the Clean Air Act Amendments of 1990 (Senate Print 103-38, 103d 
Cong. 1st sess.) vol. 1 at 868-69 (floor statement of Sen. Durenberger 
(Senate floor manager for section 112) during debate on the Conference 
Report, indicating that EPA is obligated to consider ``combined risks 
of all sources that are collocated with such sources within the same 
major source'' but going on to state that the determination of ample 
margin of safety from emissions from all collocated sources need not 
occur at the same time, but rather can be spread out over the course of 
the residual risk determination process for all major sources.
    d. Exposure to Ambient Background Levels of Respiratory Irritant 
HAP. Background levels of respiratory irritant HAP attributable to 
emissions from off-site sources would not be considered when 
establishing risk-based limits for total chlorine under section 
112(d)(4). Rather, these background levels will be addressed (as may be 
necessary) through other CAA programs such as the urban air toxics 
program.
    e. Acute Exposure Need Not Be Assessed. We have determined that you 
need not assess acute exposure to establish an emission limit for total 
chlorine. You would not be required to model maximum 1-hour average 
off-site ground level concentrations to calculate a Hazard Index (HI) 
based on acute exposure for purposes of establishing an emission limit 
for total chlorine. We conclude that the chronic exposure Hazard Index 
(HI) for the hazardous waste combustor(s) would always exceed the acute 
exposure HI. Thus, the emission limit for total chlorine based on 
chronic exposure would always be more stringent than the limit based on

[[Page 21300]]

acute exposure. As an example, the Cement Kiln Recycling Coalition 
evaluated both chronic and acute exposure to hydrogen chloride and 
chlorine gas for the 14 cement facilities that burn hazardous 
waste.\172\ In all cases, the chronic HI exceeded the acute HI. In 
addition, we determined that the Hazard Quotient (HQ) for chronic 
exposure was always higher than the HQ for acute exposure for the HAP 
we evaluated in the risk assessment we used to support the 1999 Final 
MACT Rule for hazardous waste combustors.\173\
---------------------------------------------------------------------------

    \172\ See Trinity Consultants, ``Analysis of HCl/Cl2 Emissions 
from Cement Kilns for 112(d)(4) Consideration in the HWC MACT 
Replacement Standards,'' September 17, 2003.
    \173\ See USEPA, ``Human Health and Ecological Risk Assessment 
Support to the Development of Technical Standards for Emissions from 
Combustion Units Burning Hazardous Wastes: Background Document,'' 
July 1999.
---------------------------------------------------------------------------

    Not requiring an acute exposure analysis reduces the burden on both 
the regulated community and regulatory officials to develop and review 
an analysis that would be superseded by the chronic exposure analysis 
when establishing an emission limit for total chlorine.
    Please note that this discussion relates to evaluating acute 
exposure in establishing an emission limit for total chlorine. Although 
we conclude that the chronic exposure Hazard Index would always be 
higher than the acute exposure Hazard Index, and thus would be the 
basis for the total chlorine emission rate limit, this relates to acute 
versus chronic exposure to a constant, maximum average (e.g., a maximum 
annual average) emission rate of total chlorine from a hazardous waste 
combustor. Acute exposure must be considered, however, when 
establishing operating requirements (e.g., feedrate limit for total 
chlorine and chloride) to ensure that short-term emissions do not 
result in an acute exposure Hazard Index of 1.0 or greater even though 
long-term (e.g., annual average) emissions do not exceed the limit. See 
discussion in Section G.1 below.
3. Does the Proposed Approach Ensure an Ample Margin of Safety?
    Section 112(d)(4) allows EPA to develop risk-based standards for 
HAP ``for which a health threshold has been established'', and the 
resulting standard is to provide an ``ample margin of safety.'' The 
``ample margin of safety'' standard, at least as applied to 
nonthreshold pollutants, typically connotes a two-step process (based 
on the standard first announced in the so-called Vinyl Chloride 
decision (NRDC v. EPA, 824 F. 2d at 1146 (D.C. Cir. 1987)), whereby EPA 
``first [determines] * * * a `safe' or `acceptable' level of risk 
considering only health factors, followed by a second step to set a 
standard that provides an `ample margin of safety', in which costs, 
feasibility, and other relevant factors in addition to health may be 
considered.'' 54 FR at 38045. It is not clear that Congress intended 
this analysis to apply to section 112(d)(4) standards, since the 
principal legislative history to the provision indicates that costs are 
not to be considered in setting standards under section 112(d)(4) (S. 
Rep. 228 at 173), whereas cost normally is a relevant consideration in 
the second part of the ample margin of safety process, as described 
above. Further, if issues of feasibility, cost, and other non-health 
factors are to be taken into account in establishing section 112(d)(4) 
standards, it would be exceedingly difficult, if not practically 
impossible, to do so on a site-specific basis, undermining the approach 
we are proposing here. Nor is it clear that the two-step approach is 
necessarily warranted when considering threshold pollutants, since 
there is greater certainty regarding levels at which adverse health 
effects occur. See Vinyl Chloride, 824 F. 2d at 1165 n. 11.\174\
---------------------------------------------------------------------------

    \174\ Indeed, using the classic two-step approach to ``ample 
margin of safety'' could result in the same standards we are 
proposing as MACT for HCl and Cl2 for all of the affected source 
categories (if one assumes that all of the standards would be below 
protective risk-based levels for all sources), since we believe that 
the proposed technology-based standards would be justifiable based 
on considerations of technical feasibility and cost, and so would 
provide a reasonable margin of safety beyond the risk-based level 
considered protective.
---------------------------------------------------------------------------

    We specifically request comment on how to ensure that the emission 
limits calculated using the health threshold values (e.g., RfCs and 
AEGL-1 values), and after considering emissions of respiratory irritant 
HAP from collocated hazardous waste combustors, achieve an ample margin 
of safety.
4. How Are Effects on Terrestrial Animals Addressed?
    We believe the RfC values for hydrogen chloride and chlorine gas 
should be generally protective for chronic effects in most, if not all, 
fauna. We note that the RfC values are based on animal studies. 
Although the AEGL-1 values for acute exposure are based on human data, 
we nonetheless expect that they too would be generally protective of 
most fauna, absent information to the contrary.
5. How Are Effects on Plants Addressed?
    EPA has not established ecotoxicity values that are protective of 
vegetation. Nonetheless, for the reasons discussed below we do not 
believe that ambient concentrations of hydrogen chloride and chlorine 
gas that meet the human health threshold values discussed above will 
pose adverse effects on plants.
    As discussed in the preamble to the Lime Manufacturing NESHAP 
proposed rule (67 FR 78056),\175\ chronic exposure to about 600 [mu]g/
m3 can be expected to result in discernible effects, 
depending on the plant species. Effects of acute, 20-minute exposures 
of 6,500 to 27,000 [mu]g/m3 include leaf injury and decrease 
in chlorophyll levels in various species. The hydrogen chloride RfC of 
20 [mu]g/m3 is well below the 600 [mu]g/m3 effect 
level, and the AEGL-1 value for hydrogen chloride of 2,700 [mu]g/
m3 is far below the 6500 [mu]g/m3 acute effect 
level. Therefore, no adverse exposure effects are anticipated.
---------------------------------------------------------------------------

    \175\ EPA published the final rule at 69 FR 394, January 5, 
2004.
---------------------------------------------------------------------------

    We specifically request additional information on ecotoxicity for 
both acute and chronic exposure of vegetation to hydrogen chloride and 
chlorine gas.

C. How Would You Determine if Your Total Chlorine Emission Rate Meets 
the Eligibility Requirements Defined by the National Exposure Standards?

    Under the risk-based approach to establish an alternative to the 
MACT standard for your total chlorine emission limit, you would have to 
demonstrate that emissions of total chlorine from on-site hazardous 
waste combustors result in exposure to the actual most-exposed 
individual residing off site of a Hazard Index of less than or equal to 
1.0. (Put another way, we are proposing to establish this level of risk 
as the national emission limitation, with the rule further establishing 
the mechanisms by which this demonstration can be made, such 
demonstrations yielding a site-specific limit for total chlorine.) 
\176\ The rule would also establish two ways by which you could make 
this demonstration: by a look-up table analysis or by a site-specific 
compliance demonstration (as explained below). The look-up table is 
much simpler to use, but establishes emission rates that are quite 
conservative because there are few site-specific parameters considered and

[[Page 21301]]

therefore the model's default assumptions are conservative. If you 
elect not to comply with those conservative emission rates, you may 
perform a site-specific compliance demonstration.
---------------------------------------------------------------------------

    \176\ Rather than establishing emission rate limits for hydrogen 
chloride and chlorine gas, or for total chlorine, for each 
combustor, you would actually establish an HCl-equivalent emission 
rate limit for each combustor, as discussed below in the text.
---------------------------------------------------------------------------

    The look-up table identifies the total chlorine emission limit in 
terms of a toxicity-weighted HCl-equivalent emission rate. Under the 
site-specific compliance demonstration alternative, the total chlorine 
limit would also be expressed as a toxicity weighted HCl-equivalent 
emission rate even though you would model emissions of hydrogen 
chloride and chlorine gas from each on-site hazardous waste combustor. 
We define the toxicity-weighted HCl-equivalent emission rate below.
1. Toxicity-Weighted HCl-Equivalent Emission Rates
    Although the MACT emission standards for total chlorine are 
expressed as a stack gas emission concentration--ppmv--we must use an 
emission rate (e.g., lb/hr) format for risk-based standards. This is 
because health and environmental risk is related to the mass rate of 
emissions over time.
    In addition, we propose to use a toxicity-weighted HCl-equivalent 
emission rate (HCl-equivalents) as the metric for the combined 
emissions of hydrogen chloride and chlorine gas. The HCl-equivalent 
emission rate considers the RfCs of hydrogen chloride and chlorine gas 
when calculating the combined emission rate according to this equation:
    ER_dtw = [Sigma](ER_i x (RfC_HC1/
RfC_i))

where:

    ER_tw is the HC1-equivalent emission rate, lb/hr
    ER_i is the emission rate of HAP i in lbs/hr
    RfC_i is the reference concentration of HAP i
    RfC_HC1 is the reference concentration of HCl
    Expressing the risk-based emission limit as HCl-equivalents enables 
you to use the equation to apportion the emission rate limit between 
hydrogen chloride and chlorine gas as you choose. Thus, you need to be 
concerned with ensuring compliance with the HCl-equivalent emission 
rate only, rather than with emission rates for hydrogen chloride and 
chlorine gas individually.
    Under the look-up table analysis discussed below, you would use the 
hydrogen chloride and chlorine gas emission rates you choose for each 
on-site hazardous waste combustor to calculate the HCl-equivalent 
emission rate for the combustor. You would sum the HCl-equivalent 
emission rates for your hazardous waste combustors. If you elect to use 
the site-specific compliance demonstration to document eligibility, you 
would model emission rates of hydrogen chloride and chlorine gas that 
you choose for each on-site hazardous waste combustor to document that 
the facility Hazard Index is less than or equal to 1.0. You would then 
use the hydrogen chloride and chlorine gas emission rates you model to 
establish an HCl-equivalent emission rate limit for each combustor.
2. How Would You Conduct a Look-Up Table Analysis?
    You would sum the HCl-equivalent rates for all combustors, and 
compare the sum to the appropriate allowable emission rate in Table 1 
of proposed Sec.  63.1215. Emission rates are provided as a function of 
stack height and distance to the nearest property boundary. If you have 
more than one hazardous waste combustor at your facility, you would use 
the average value for stack height (i.e., the averaged stack heights of 
the different hazardous waste combustors at your facility), and the 
minimum distance between any hazardous waste combustor stack and the 
property boundary.\177\
---------------------------------------------------------------------------

    \177\ HCl production furnaces are not eligible for the risk-
based total chlorine emission limits because we are proposing that 
the MACT standard for total chlorine would be used as a surrogate to 
control metal HAP. Nonetheless, if you operate an HCl production 
furnace at a facility where you would establish risk-based emission 
limits for total chlorine for other hazardous waste combustors, you 
would account for total chlorine emissions from the HCl production 
furnace in your risk-based eligibility demonstration for the other 
combustors. If, for example, you use the look-up table to 
demonstrate eligibility, you would include the stack height of the 
HCl production furnace in the calculation of average stack height 
for your combustors, and you would consider whether the HCl 
production furnace stack is the closest hazardous waste combustor 
stack to the property boundary.
---------------------------------------------------------------------------

    If one or both of these values for stack height and distance to 
nearest property boundary do not match the exact values in the look-up 
table, you would use the next lowest table value. This would ensure 
that the HCl-equivalent emission rate limits are protective.
    You would not be eligible for the look-up table analysis if your 
facility is located in complex terrain because the plume dispersion 
models used to calculate the emission rates are not applicable to 
sources in complex terrain.
    You would be eligible to comply with the risk-based alternative 
HCl-equivalent emission rate limits you calculate for each combustor if 
the facility HCl-equivalent emission rate limit (i.e., the sum of the 
HCl-equivalent emission rates for all hazardous waste combustors) does 
not exceed the appropriate value specified in the look-up table. Please 
note, however, that we also propose to cap the HCl-equivalent emission 
rate limits for incinerators, cement kilns, and lightweight aggregate 
kilns at a level that ensures that the current total chlorine emission 
standards are not exceeded. See discussion below in Section D.
    Please note that the emission rates provided in Table 1 are 
different from those provided for industrial boilers in the Industrial 
Boiler and Process Heater MACT rule recently promulgated. This is 
because the key parameters used by the SCREEN3 atmospheric dispersion 
model to predict the normalized air concentrations that EPA used to 
establish HCl-equivalent emission rates as a function of stack height 
and distance to property boundary for industrial boilers--stack 
diameter, stack exit gas velocity, and stack exit gas temperature--are 
substantially different for hazardous waste burning incinerators, 
cement kilns, and lightweight aggregate kilns. Thus, the maximum HCl-
equivalent emission rates for hazardous waste combustors would 
generally be lower than those EPA established for industrial boilers.
    To ensure that the HCl-equivalent emission rate limits in a look-up 
table analysis for hazardous waste combustors would not result in a 
Hazard Index of more than 1.0, we propose to establish limits based on 
the maximum annual average normalized air concentrations in U.S. EPA, 
``A Tiered Modeling Approach for Assessing the Risk Due to Sources of 
Hazardous Air Pollutants,'' March 1992, Table 1. Those normalized air 
concentrations are based on conservative simulations of toxic pollutant 
sources with Gaussian plume dispersion models. The simulations are 
conservative regarding factors such as meteorology, building downwash, 
plume rise, etc.
    We specifically request comment on whether the HCl-equivalent 
emission rates in Table 1 are too conservative and thus have limited 
utility because they apply to all hazardous waste combustors 
generically. Alternatively, we could establish less conservative 
emission rates in look-up tables specific to various classes of 
hazardous waste combustors (e.g., cement kilns, incinerators) that have 
similar stack properties that affect predicted emissions. We request 
comment on whether industry stakeholders would be likely to use the 
proposed look-up table eligibility demonstration or revised

[[Page 21302]]

look-up tables tailored to specific classes of hazardous waste 
combustors, in lieu of the site-specific compliance eligibility 
demonstration.
3. How Would You Conduct a Site-Specific Compliance Demonstration?
    If you fail to demonstrate that your facility is able to comply 
with the alternative risk-based emission limit using the look-up table 
approach, you may choose to perform a site-specific compliance 
demonstration. We are proposing that you may use any scientifically-
accepted peer-reviewed risk assessment methodology for your site-
specific compliance demonstration. An example of one approach for 
performing the demonstration for air toxics can be found in the EPA's 
``Air Toxics Risk Assessment Reference Library, Volume 2, Site-Specific 
Risk Assessment Technical Resource Document,'', which may be obtained 
through the EPA's Air Toxics Web site at http://www.epa.gov/ttn/atw.
    Your facility would be eligible for the alternative risk-based 
total chlorine emission limit if your site-specific compliance 
demonstration shows that the maximum Hazard Index for hydrogen chloride 
and chlorine gas emissions from all on-site hazardous waste combustors 
at a location where people live (i.e., the maximum actual most exposed 
individual) is less than or equal to 1.0, rounded to the nearest tenths 
decimal place (0.1).\178\ You would estimate long-term inhalation 
exposures for this individual most exposed to the facility's emissions 
through the estimation of annual or multi-year average ambient 
concentrations. You would use site-specific, quality-assured data 
wherever possible, and health-protective default assumptions wherever 
site-specific data are not available. You would document the data and 
methods used for the assessment so that it is transparent and can be 
reproduced by an experienced risk assessor and emissions measurement 
expert.
---------------------------------------------------------------------------

    \178\ When calculating Hazard Index values, the final HI value 
should be rounded to one decimal place given the uncertainties in 
the analyses. For example, an HI calculated to be 0.94 would be 
presented as 0.9, while an HI calculated to be 0.96 would be 
presented as 1.0 (which would pass the eligibility demonstration). 
Intermediate calculations should use as many significant figures as 
appropriate.
---------------------------------------------------------------------------

    Your site-specific compliance demonstration need not assume any 
attenuation of exposure concentrations due to the penetration of 
outdoor pollutants into indoor exposure areas. In addition, we are 
proposing that the demonstration need not assume any reaction or 
deposition of hydrogen chloride and chlorine gas from the emission 
point to the point of exposure. In particular, you would assume that 
chlorine gas is not photolyzed to hydrogen chloride, as discussed in 
Section B.1 above.
    If your site-specific compliance demonstration documents that the 
maximum Hazard Index from your hazardous waste combustors is less than 
or equal to 1.0, you would establish a maximum HCl-equivalent emission 
rate limit for each combustor using the hydrogen chloride and chlorine 
gas emission rates you modeled in the site-specific compliance 
demonstration. Please note, however, that we also propose to cap the 
HCl-equivalent emission rate limits for incinerators, cement kilns, and 
lightweight aggregate kilns at a level that ensures that the current 
total chlorine emission standards are not exceeded. See discussion 
below in Section D.

D. What Is the Rationale for Caps on the Risk-Based Emission Limits?

    The HCl-equivalent emission rate limits would be capped for 
incinerators, cement kilns, and lightweight aggregate kilns at a level 
that ensures total chlorine emissions do not exceed the interim 
standards provided by Sec. Sec.  63.1203, 63.1204, and 63.1205. These 
caps on the risk-based emission limits would ensure that emission 
levels do not increase above the emission levels that sources are 
currently required to achieve, thus precluding ``back-sliding.'' Given 
the discretionary nature of section 112(d)(4), and the general purpose 
of the section 112(d) standard-setting process to lock-in performance 
of current emission control technology, we think it appropriate to 
invoke the provision in a manner that does not result in emission 
increases over current regulatory levels.
    We considered whether to propose emission caps for boilers at the 
levels allowed by the RCRA emission standards under Sec.  266.107 but 
conclude that this would be inappropriate. This is because the RCRA 
emission standards are also risk-based standards but are based on risk 
criteria that we considered appropriate in 1987 when we proposed those 
rules. The risk criteria we propose today are substantially different 
from those used to implement Sec.  266.107. For example, the RfC for 
hydrogen chloride is higher now while the RfC for chlorine gas is 
lower. In addition, we considered a Hazard Index of 0.25 acceptable 
under the RCRA rule, while we propose today a Hazard Index limit of 
less than or equal to 1.0. Because the risk criteria for the current 
RCRA rules are substantially different from the risk criteria we 
propose today for invoking Section 112(d)(4), we do not believe it is 
appropriate to use the RCRA standards as a cap for establishing risk-
based standards under Section 112(d)(4).
    Capping risk-based emission limits for incinerators, cement kilns, 
and lightweight aggregate kilns at an HCl-equivalent emission rate 
corresponding to the MACT interim standards would not increase 
compliance costs (by definition). Thus, the cap would help ensure that 
emissions are protective of public health with an ample margin of 
safety, and that there are no significant adverse environmental 
impacts.
    To implement the cap, you would ensure that the hydrogen chloride 
and chlorine gas emission rates you use to calculate the HCl-equivalent 
emission rate for incinerators, cement kilns, and lightweight aggregate 
kilns would not result in total chlorine emission concentrations 
exceeding the standards provided by Sec. Sec.  63.1203, 63.1204, and 
63.1205.

E. What Would Your Risk-Based Eligibility Demonstration Contain?

    To enable regulatory officials to review and approve the results of 
your risk-based demonstration, you would include the following 
information, at a minimum: (1) Identification of each hazardous waste 
combustor combustion gas emission point (e.g., generally, the flue gas 
stack); (2) the maximum capacity at which each combustor will operate, 
and the maximum rated capacity for each combustor, using the metric of 
stack gas volume emitted per unit of time, as well as any other metric 
that is appropriate for the combustor (e.g., million Btu/hr heat input 
for boilers; tons of dry raw material feed/hour for cement kilns); (3) 
stack parameters for each combustor, including, but not limited to 
stack height, stack area, stack gas temperature, and stack gas exit 
velocity; (4) plot plan showing all stack emission points, nearby 
residences, and property boundary line; (5) identification of any stack 
gas control devices used to reduce emissions from each combustor; (6) 
identification of the RfC values used to calculate the HCl-equivalent 
emissions rate; (7) calculations used to determine the HCl-equivalent 
emission rate as prescribed above; (8) for incinerators, cement kilns, 
and lightweight aggregate kilns, calculations used to determine that 
the HCl-equivalent emission rate limit for each combustor does not 
exceed the standards for total chlorine at Sec. Sec.  63.1203, 63.1204, 
and 63.1205; and (9) the HCl-equivalent emission rate limit for each 
hazardous waste

[[Page 21303]]

combustor that you will certify in the Documentation of Compliance 
required under Sec.  63.1211(d) that you will not exceed, and the 
limits on the operating parameters specified under Sec.  63.1209(o) 
that you will establish in the Documentation of Compliance.
    If you use the look-up table analysis to demonstrate that your 
facility is eligible for the risk-based alternative for the total 
chlorine emission limit, your eligibility demonstration would also 
contain, at a minimum, the following: (1) Calculations used to 
determine the average stack height of on-site hazardous waste 
combustors; (2) identification of the combustor stack with the minimum 
distance to the property boundary of the facility; (3) comparison of 
the values in the look-up table to your maximum HCl-equivalent emission 
rate.
    If you use a site-specific compliance demonstration to demonstrate 
that your facility is eligible for the risk-based alternative for the 
total chlorine emission limit, your eligibility demonstration would 
also contain, at a minimum, the following: (1) Identification of the 
risk assessment methodology used; (2) documentation of the fate and 
transport model used; and (3) documentation of the fate and transport 
model inputs, including the stack parameters listed above converted to 
the dimensions required for the model. In addition, you would include 
all of the following that apply: (1) Meteorological data; (2) building, 
land use, and terrain data; (3) receptor locations and population data; 
and (4) other facility-specific parameters input into the model. Your 
demonstration would also include: (1) Documentation of the fate and 
transport model outputs; (2) documentation of any exposure assessment 
and risk characterization calculations; and (3) documentation of the 
predicted Hazard Index for HCl-equivalents and comparison to the limit 
of less than or equal to 1.0.

F. When Would You Complete and Submit Your Eligibility Demonstration?

    You would be required to submit your eligibility demonstration to 
the permitting authority for review and approval.\179\ In addition you 
would submit an electronic copy of the demonstration to reag@epa.gov 
(preferably) or a hard copy to: U.S. EPA, Risk and Exposure Assessment 
Group, Emission Standards Division (C404-01), Attn: Group Leader, 
Research Triangle Park, North Carolina 27711.
---------------------------------------------------------------------------

    \179\ Since the Title V permitting authority is delegated to 
States in virtually all instances, the permit limit would thus be 
issued as a matter of State authority (generally in parallel with a 
delegation of section 112 authority pursuant to CAA section 112(l)), 
and be reviewable only in State courts.
---------------------------------------------------------------------------

    Requiring prior approval of these eligibility demonstrations is 
warranted because hazardous waste combustor may feed chlorine at high 
feedrates which may result in emissions of hydrogen chloride and 
chlorine gas that approach or exceed the RfCs (i.e., absent compliance 
with either the MACT standards or the section 112(d)(4) risk-based 
standards). Thus, prior approval of alternative HCl-equivalent emission 
rate limits is warranted to ensure that emissions are protective with 
an ample margin of safety.
1. Existing Sources
    If you operate an existing source, you must be in compliance with 
the emission standards on the compliance date. Consequently, if you 
elect to comply with the alternative risk-based emission rate limit for 
total chlorine, you must have completed the eligibility demonstration 
and received approval from your delegated permitting authority by the 
compliance date.
    You would submit documentation supporting your eligibility 
demonstration not later than 12 months prior to the compliance date.
    Your permitting officials will notify you of approval or intent to 
disapprove your eligibility demonstration within 6 months after receipt 
of the original demonstration, and within 3 months after receipt of any 
supplemental information that you submit. A notice of intent to 
disapprove your eligibility demonstration will identify incomplete or 
inaccurate information or noncompliance with prescribed procedures and 
specify how much time you will have to submit additional information. 
If your permitting authority has not approved your eligibility 
demonstration to comply with a risk-based HCl-equivalent emission 
rate(s) by the compliance date, you must comply with the MACT emission 
standards for total chlorine gas under Sec. Sec.  63.1216, 63.1217, 
63.1219, 63.1220, and 63.1221.\180\
---------------------------------------------------------------------------

    \180\ Please note that, if your eligibility demonstration is not 
approved prior to the compliance date, a request to extend the 
compliance date to enable you to undertake measures to comply with 
the MACT standards for total chlorine will not be approved unless 
you made a good faith effort to submit a complete, accurate, and 
timely eligibility demonstration and to respond to concerns raised 
by the permitting authority or U.S. EPA.
---------------------------------------------------------------------------

2. New Sources
    If you operate a source that is not an existing source and that 
becomes subject to Subpart EEE, you must comply with the MACT emission 
standards for total chlorine unless and until your eligibility 
demonstration has been approved by the permitting authority.
    If you operate a new or reconstructed source that starts up before 
the effective date of the emission standards proposed today, or a solid 
fuel-fired boiler or liquid fuel-fired boiler that is an area source 
that increases its emissions or its potential to emit such that it 
becomes a major source of HAP before the effective date of the emission 
standards proposed today (and thus becomes subject to emission 
standards applicable to major sources, including the standard for total 
chlorine), you would be required to comply with the emission standards 
under Sec. Sec.  63.1216 and 63.1217 until your eligibility 
demonstration is completed, submitted, and approved by your permitting 
authority.
    If you operate a new or reconstructed source that starts up after 
the effective date of the emission standards proposed today, or a solid 
fuel-fired boiler or liquid fuel-fired boiler that is an area source 
that increases its emissions or its potential to emit such that it 
becomes a major source of HAP after the effective date of the emission 
standards proposed today (and thus becomes subject to emission 
standards applicable to major sources including the standard for total 
chlorine), you would be required to comply with the emission standards 
under Sec. Sec.  63.1216 and 63.1217 until your eligibility 
demonstration is completed, submitted, and approved by your permitting 
authority.

G. How Would the Risk-Based HCl-Equivalent Emission Rate Limit Be 
Implemented?

    Upon approval by the permitting authority of your eligibility 
demonstration, the HCl-equivalent emission rate limit established in 
the demonstration for your hazardous waste combustor(s) becomes the 
applicable emission limit for total chlorine in lieu of the MACT 
standard for total chlorine.
1. What Are the Testing and Monitoring Requirements?
    To ensure compliance with the alternative HCl-equivalent emission 
rate limit for your combustor(s), you would conduct performance testing 
as required for the MACT standards and establish limits on the same 
operating parameters that apply to sources complying with the MACT 
standards for total chlorine under Sec.  63.1209(o). You would 
establish and comply with these operating parameter limits just as you 
would establish and comply with the limits for the MACT emission 
standard for total chlorine, with the exception of the

[[Page 21304]]

chlorine feedrate limit, as discussed below. For example, existing 
sources would establish these limits in the Documentation of Compliance 
required under Sec.  63.1211(c) and begin complying with them not later 
than the compliance date. Existing sources would also revise the 
operating limits as necessary based on the initial comprehensive 
performance test and begin complying with the revised operating limits 
not later than when the Notification of Compliance is postmarked, as 
required under Sec. Sec.  63.1207(j) and 63.1210(b).
    The limit on chlorine feedrate required under Sec.  63.1209(o)(1) 
would be established differently to ensure compliance with the HCl-
equivalent emission rate limit rather than the total chlorine emission 
standard. To ensure that facility-wide hazardous waste combustor 
emissions of HCl-equivalents result in exposures equivalent to a Hazard 
Index of less than or equal to 1.0, the feedrate limit for chlorine 
would be established as the average of the test run averages and the 
averaging period for compliance would be one year. A yearly rolling 
average is appropriate for risk-based emission limits rather than the 
12-hour rolling average applicable to the MACT standards because the 
risk-based emission limit is based on chronic exposure.
    As discussed in Section B.2.e above, although we conclude that the 
chronic exposure Hazard Index would always be higher and thus be the 
basis for the total chlorine emission rate limit, we still must be 
concerned about acute exposure attributable to short-term emission 
rates higher than the maximum average emission rate limit. For example, 
the annual average limit on chlorine (i.e., total chlorine and 
chloride) feedrate would allow a source to feed very high levels of 
chlorine for short periods of time, potentially resulting in 
exceedances of the acute exposure Hazard Index based the AEGL-1 values 
for hydrogen chloride and chlorine gas. We specifically request comment 
on how a short-term limit on chlorine feedrate could be established for 
each hazardous waste combustor to ensure that the acute exposure Hazard 
Index is less than or equal to 1.0. One approach would be for you to 
extrapolate from the chlorine feedrate during the comprehensive 
performance test to the feedrate projected to achieve emission rates of 
hydrogen chloride and chlorine gas that result in an acute exposure 
Hazard Index of 1.0.\181\ This feedrate would be a 1-hour average 
feedrate limit. This approach uses the reasonable assumption that there 
is a proportional relationship between chlorine feedrate and the 
emission rate of hydrogen chloride and chlorine gas. To extrapolate 
feedrates, you would consider the system removal efficiency achieved 
during the performance test for sources equipped with wet or dry acid 
gas scrubbers and for cement kilns.\182\ Other sources would assume a 
zero system removal efficiency because any removal efficiency that may 
be measured would be incidental and not reproducible.
---------------------------------------------------------------------------

    \181\ We also request comment on whether extrapolation of the 
chlorine feedrate should be allowed to 100% of the Hazard Index 
limit of 1.0, or whether a more conservative approach of limited 
extrapolation to a fraction of the Hazard Index (e.g., 0.8) would be 
warranted, given the uncertainties inherent in projecting emissions 
from extrapolated feedrates.
    \182\ We request comment on whether the system removal 
efficiency a cement kiln demonstrates during a performance test 
because of the alkalinity of the raw material is reasonably 
indicative of the system removal efficiency it routinely achieves 
(i.e., is the system removal efficiency reasonably reproducible).
---------------------------------------------------------------------------

    The approach discussed above would be applicable if you use the 
site-specific compliance eligibility demonstration. If you use the 
look-up table for your eligibility demonstration, an alternative 
approach would be needed to establish a short-term chlorine feedrate 
limit. One approach would be to establish a look-up table for maximum 
1-hour average HCl-equivalents based on acute exposure. Acute exposure 
HCl-equivalents would be calculated using the AEGL-1 values for 
hydrogen chloride and chlorine gas, and the look-up table of acute 
exposure maximum emission rate limits would be based on normalized air 
concentrations for maximum 1-hour average ground level 
concentrations.\183\ You would extrapolate the chlorine feedrate from 
the level achieved during the comprehensive performance test to a level 
that would not exceed the acute exposure HCl-equivalent emission rate 
limit for each combustor provided in the look-up table. This feedrate 
would be a 1-hour average feedrate limit.
---------------------------------------------------------------------------

    \183\ We would use the normalized maximum 1-hour average 
concentrations in U.S. EPA, ``A Tiered Modeling Approach for 
Assessing the Risk Due to Sources of Hazardous Air Pollutants,'' 
March 1992, Table 2.
---------------------------------------------------------------------------

    We specifically request comment on these approaches to establish a 
short-term limit on the feedrate of total chlorine and chloride to 
ensure that the acute exposure Hazard Index for hydrogen chloride and 
chlorine gas is less than or equal to 1.0.
2. What Test Methods Would You Use?
    Although you would comply with the MACT standard for total chlorine 
using stack Method 26/26A, certain sources would not be allowed to use 
that method to demonstrate compliance with the risk-based HCl-
equivalent emission rate limit.\184\ Cement kilns and sources equipped 
with a dry acid gas scrubber should use EPA Method 320/321 or ASTM D 
6735-01 to measure hydrogen chloride, and the back-half (caustic 
impingers) of Method 26/26A to measure chlorine gas. Incinerators, 
boilers, and lightweight aggregate kilns should use EPA Method 320/321 
or ASTM D 6735-01 to measure hydrogen chloride, and Method 26/26A to 
measure total chlorine, and calculate chlorine gas by difference if: 
(1) the bromine/chlorine ratio in feedstreams is greater than 5 
percent; or (2) the sulfur/chlorine ratio in feedstreams is greater 
than 50 percent.
---------------------------------------------------------------------------

    \184\ Even though Method 26/26A may bias total chlorine emission 
measurements low for cement kilns for reasons discussed in the text, 
it is appropriate to allow compliance with the technology-based MACT 
emission standards for total chlorine using that method. Because the 
MACT standards are developed using data obtained using Method 26/
26A, allowing that method for compliance will achieve reductions in 
total chlorine emissions. For the same reason, it would be 
inappropriate to require compliance with unbiased methods because 
the average of the best performing sources might not be able to 
achieve the standard.
---------------------------------------------------------------------------

    a. Method 26/26A Has a Low Bias for Hydrogen Chloride in Certain 
Situations. Method 26/26A has a low bias for hydrogen chloride for 
sources that emit particulate matter than can adsorb hydrogen chloride: 
cement kilns and sources equipped with a dry acid gas scrubber. 
Particulate matter caught by the Method 26/26A filter scrubs hydrogen 
chloride from the sample gas, and can result in measurements that are 
biased low by 2 to 30 times.\185\ Chlorine gas is not adsorbed so that 
chlorine gas emissions are not biased by this mechanism.
---------------------------------------------------------------------------

    \185\ USEPA, ``Draft Technical Support Document for HWC MACT 
Replacement Standards, Volume III: Selection of MACT Standards and 
Technologies,'' March 2004.
---------------------------------------------------------------------------

    b. Method 26/26A Can Have a Low Bias for Chlorine Gas and a High 
Bias for Hydrogen Chloride, but Has No Bias for Total Chlorine. Method 
26/26A also has a low bias for chlorine and a high bias for hydrogen 
chloride when bromine is present at significant levels. Bromine has a 
strong effect on the bias. Although the various interhalogen reactions 
are extremely complex and may depend on a variety of system parameters, 
it appears that each bromine molecule can react with a chlorine 
molecule in the acidic impingers of Method 26/26A where hydrogen 
chloride is captured, converting the chlorine to chloride ions which 
are

[[Page 21305]]

reported as hydrogen chloride. Total chlorine measurements (i.e., 
hydrogen chloride and chlorine gas, combined, reported as Cl-
equivalents), however, are not affected. To minimize this bias, we 
propose to require sources that have a bromine/chlorine feedrate 
exceeding 5 percent to use alternative methods discussed below. Given 
the strong bias that bromine can have on M26/26A measurements, we 
believe a 5 percent limit on the ratio is within the range of 
reasonable values that we could select. We specifically request comment 
on this or other approaches to minimize the bromine bias.
    Method 26/26A also has a low bias for chlorine and a high bias for 
hydrogen chloride when sulfur is present at substantial levels relative 
to the levels of chlorine. The capture of chlorine in the acidic 
impingers that collect hydrogen chloride has been shown to rapidly 
increase when the ratio of SO2/HCl (both expressed in ppmv) exceeds 
0.5. Again, total chlorine measurements are not biased. To minimize 
this bias, we believe that a 50 percent limit on the ratio of the 
sulfur/chlorine feedrate is within the range of reasonable values that 
we could select. We specifically request comment on this or other 
approaches to minimize the sulfur dioxide bias.
    c. Unbiased Methods Are Available. The Agency recently developed 
three methods for hydrogen chloride in the context of the Portland 
Cement MACT rule for purposes of area source determinations: Methods 
320, 321, and 322. Although M322 (GFCIR, Gas Filter Correlation Infra-
Red) is easier to use and less expensive than M320/M321 (FTIR, Fourier 
Transform Infra-Red), the Agency did not promulgated M322 in the final 
Portland Cement MACT rule because of accuracy concerns resulting from 
emissions sampling of lime manufacturing kilns in the context of 
developing the Lime Manufacturing MACT rule.
    The Agency has also adopted an American Society of Testing and 
Materials (ASTM) standard for measuring hydrogen chloride emissions: 
ASTM D 6735-01. This method (and M321) is allowed for area source 
determinations under the Lime Manufacturing MACT rule. 69 FR 394 (Jan. 
5, 2004). The method is an impinger method, like M26/26A, but with 
several improvements. For example, the method uses a rejection probe 
(i.e., the probe is directed counter to the gas flow), the filter is 
heated to minimize adsorption of hydrogen chloride on particulate 
matter that may catch on the filter, glassware must be conditioned, and 
improved quality assurance/quality control procedures are prescribed.

H. How Would You Ensure That Your Facility Remains Eligible for the 
Risk-Based Emission Limit?

1. Changes Over Which You Have Control
    Changes in design, operation, or maintenance of a hazardous waste 
combustor that may affect the rate of emissions of HCl-equivalents from 
the combustor are subject to the requirements of Sec.  63.1206(b)(5).
    If you change the information documented in the demonstration of 
eligibility for the HCl-equivalent emission rate limit which is used to 
establish the HCl-equivalent emission rate limit, you would be subject 
to the following procedures.
    a. Changes that Would Decrease the Allowable HCl-Equivalent 
Emission Rate Limit. If you plan to make a change that would decrease 
the allowable HCl-equivalent emission rate limit documented in your 
eligibility demonstration, you would comply with Sec.  
63.1206(b)(5)(i)(A-C) regarding notifying the permitting authority of 
the change, submitting a comprehensive performance test schedule and 
test plan, comprehensive performance testing, and restriction on 
burning hazardous waste prior to submitting a revised Notification of 
Compliance. An example of a change that would decrease the allowable 
HCl-equivalent emission rate limit is location of the property boundary 
closer to the nearest hazardous waste combustor stack when using the 
look-up table to make the eligibility demonstration.
    b. Changes that Would Not Decrease the Allowable HCl-Equivalent 
Emission Rate Limit. If you determine that a change would not decrease 
the allowable HCl-equivalent emission rate limit documented in your 
eligibility demonstration, you would document the change in the 
operating record upon making such change. If the change would increase 
your allowable HCl-equivalent emission rate limit and you elect to 
establish a higher HCl-equivalent limit, you must submit a revised 
eligibility demonstration for review and approval. Upon approval of the 
revised eligibility demonstration, you must comply with Sec.  
63.1206(b)(5)(i)(A)(2), (B), and (C) regarding submitting a 
comprehensive performance test schedule and test plan, comprehensive 
performance testing, and restriction on burning hazardous waste prior 
to submitting a revised Notification of Compliance.
2. Changes Over Which You Do Not Have Control
    Over time, factors and information over which you do not have 
control and which you use to make your eligibility demonstration may 
change. For example, if you use a site-specific compliance 
demonstration, individuals may locate within the area impacted by 
emissions such that the most exposed individual may be exposed to 
higher ground level concentrations than previously estimated. This 
could lower your allowable HCl-equivalent emission rate limit. 
Consequently, you would be required to review the documentation you use 
in your eligibility demonstration every five years on the anniversary 
of the comprehensive performance test and submit for review with the 
test plan either a certification that the information used in your 
eligibility demonstration has not changed in a manner that would 
decrease the allowable HCl-equivalent emission rate limit, or a revised 
eligibility demonstration for a revised HCl-equivalent emission rate 
limit.
    If you determine that you cannot demonstrate compliance with a 
lower allowable HCl-equivalent emission rate limit during the 
(subsequent) comprehensive performance test because you cannot complete 
changes to the design or operation of the source prior to the test, you 
may request that the permitting authority grant you additional time as 
necessary to make those changes, not to exceed three years.

I. Request for Comment on an Alternative Approach: Risk-Based National 
Emission Standards

    As noted earlier, another approach to implement section 112(d)(4)--
and one EPA has used in past MACT rules--would be to establish national 
emission standards for each source category to ensure that the 
emissions from each source within the category are protective of public 
health with an ample margin of safety (and do not pose adverse 
environmental impacts). Under this approach, dispersion modeling of 
representative worst-case sources (or all sources) within a category 
would be used to identify an emission level that meets the section 
112(d)(4) criteria for all sources within the category. Thus, the same 
risk-based national emission standard would be established for each 
source in each source category under this approach, rather than the 
approach we discuss above of establishing a national exposure standard 
based on a uniform level of protection that you would use to establish 
a site-specific emission limit.

[[Page 21306]]

    The approach of establishing a risk-based national emission 
standard for a source category has the advantage of being less 
burdensome to implement both for the regulated community and regulatory 
authorities. It has the disadvantage, however, of requiring 
documentation ``up front'' to support the proposed emission standards. 
EPA does not have the time, data, or resources to conduct the analyses 
required to support this approach.
    The Cement Kiln Recycling Coalition (CKRC), however, has submitted 
documentation supporting a national risk-based emission standard for 
total chlorine for cement kilns.\186\ CKRC uses normalized air 
concentrations from ISC-PRIME and ISCST3 to estimate maximum annual 
average and maximum 1-hour average off-site ground level concentrations 
of hydrogen chloride and chlorine gas for each source. CKRC assumes 
that each kiln emits total chlorine at 130 ppmv, the current Interim 
Standard, and that emissions of hydrogen chloride and chlorine gas 
partition at the same ratio as measured during the most recent 
compliance test. The analysis indicates that the facility Hazard Index 
for 1-hour exposures was below 0.2 for the kilns at all facilities, and 
the facility Hazard Index for long-term exposures was below 0.2 for the 
kilns at 8 of 14 facilities. Emissions from kilns at the remaining 6 
facilities can potentially result in facility Hazard Index values up to 
0.7.
---------------------------------------------------------------------------

    \186\ Trinity Consultants, ``Analysis of HCl/Cl2 Emissions from 
Cement Kilns for 112(d)(4) Consideration in the HWC MACT Replacement 
Standards,'' September 17, 2003.
---------------------------------------------------------------------------

    Notwithstanding that CKRC followed the guidance we suggested to 
identify a section 112(d)(4) risk-based emission standard for a source 
category, we conclude that establishing a stack gas concentration-based 
total chlorine standard of 130 ppmv may not be protective with an ample 
margin of safety. Even though the highest Hazard Index for any facility 
in the category is below the maximum HI of less than 1.0, the Hazard 
Index value for a facility could increase even though sources do not 
exceed an emission standard of 130 ppmv. This is because the Hazard 
Index is affected by the mass emission rate (e.g., lb/hr) of hydrogen 
chloride and chlorine gas individually. Thus the Hazard Index could 
increase from the values CKRC has calculated even though each source 
complies with a 130 ppmv total chlorine emission standard given that: 
(1) The RfC for chlorine gas is 100 times lower than the RfC for 
hydrogen chloride; (2) the partitioning of total chlorine between 
hydrogen chloride and chlorine gas could change so that a greater 
portion is emitted as chlorine; and (3) the mass emission rate of 
hydrogen chloride and chlorine gas would increase if the stack gas 
flowrate increases.
    Because of these concerns, the more appropriate metric for a risk-
based standard for total chlorine would be the toxicity-weighted HCl-
equivalent emission rate discussed above in Section C.1.
    To achieve our dual objective of establishing a protective risk-
based emission standard expressed as a toxicity-weighted HCl-equivalent 
emission rate (lb/hr) and ensuring that the standard does not allow 
total chlorine emission concentrations (ppmv) higher than the current 
interim standard of 130 ppmv, we propose that an HCl-equivalent 
emission rate limit be established that is achievable by all cement 
facilities. This would be an HCl-equivalent emission rate for which on-
site cement kiln emissions of hydrogen chloride and chlorine gas do not 
exceed a Hazard Index of 1.0. To make this determination, facilities 
would assume that emissions of hydrogen chloride and chlorine gas 
partition at the same ratio as measured during the most recent 
compliance test. Finally, the HCl-equivalent emission rate limit would 
be capped, if necessary, at a limit that ensures that total chlorine 
concentrations for each kiln do not exceed 130 ppmv.
    If this information and supporting documentation is provided to us, 
we would promulgate a toxicity-weighted HCl-equivalent emission rate 
that would be applicable to cement kilns.
    On a related matter, we evaluated whether using hydrogen chloride 
and chlorine gas emissions data obtained with stack sampling Method 26/
26A to project hydrogen chloride and chlorine gas emissions in CKRC's 
analysis compromised the results. Method 26/26A is known to 
underestimate hydrogen chloride emissions from cement kilns.\187\ We 
discuss above in Section F.2 concerns about Method 26/26A and the 
rationale for proposing to require sources to use methods other than 
Method 26/26A to measure emissions of hydrogen chloride and chlorine 
gas for compliance with risk-based standards. Briefly, Method 26/26A 
results for hydrogen chloride are biased low for cement kilns, although 
results for chlorine gas are unaffected. Even though CKRC used Method 
26A results to apportion the 130 ppmv total chlorine assumed emissions 
between hydrogen chloride and chlorine gas for each source, the 
calculated Hazard Index values are not compromised. Given that the 
hydrogen chloride emission levels are biased low, the chlorine gas/
hydrogen chloride ratio that CKRC used to apportion the 130 ppmv total 
chlorine emissions between chlorine gas and hydrogen chloride emissions 
for each source is biased high. Thus, CKRC projected chlorine gas 
emissions that are biased high and hydrogen chloride emissions that are 
biased low. These biases result in calculating conservative (i.e., 
higher than actual) Hazard Index values because the health threshold 
values are lower for chlorine gas than for hydrogen chloride.\188\ 
Thus, actual Hazard Index values at an emission level of 130 ppmv total 
chlorine would be lower than those that CKRC calculated.
---------------------------------------------------------------------------

    \187\ See 63 FR at 14196 (March 24, 1998).
    \188\ For the same reasons, HCl-equivalent emission rates that 
CKRC may use in an eligibility demonstration for the source category 
would be biased conservatively high.
---------------------------------------------------------------------------

XIV. How Did EPA Determine Testing and Monitoring Requirements for the 
Proposed Rule?

    The CAA requires us to develop regulations that include monitoring 
and testing requirements. CAA section 114 (a) (3). The purpose of these 
requirements is to allow us to determine whether an affected source is 
operating in compliance with the rule.
    We propose testing and monitoring requirements for solid fuel-fired 
boilers, liquid fuel-fired boilers and hydrochloric acid production 
furnaces that are identical to those applicable to incinerators, cement 
kilns, and lightweight aggregate kilns under Sec. Sec.  63.1207, 
63.1208, and 63.1209.\189\ Please note, however, that we discuss below 
a proposed requirement for boilers that would not be subject to a 
numerical dioxin/furan emission standard to conduct a one-time test for 
dioxin/furan emissions. In addition, in Part Three of today's preamble, 
we request comment on, or propose revisions to, several compliance 
requirements. Any amendments to the compliance requirements that we 
promulgate would be applicable to all hazardous waste combustors. In 
addition, we discuss below in this

[[Page 21307]]

section proposed compliance procedures for emission standards that 
would be based on normal rather than compliance test data and that 
would be applicable to all hazardous waste combustors subject to such a 
standard. Finally, we discuss below in this section proposed compliance 
procedures for emission standards based on hazardous waste thermal 
emissions that would be applicable to all hazardous waste combustors.
---------------------------------------------------------------------------

    \189\ Please note that we also propose to revise the existing 
schedule for the initial comprehensive performance test for 
incinerators, cement kilns, and lightweight aggregate kilns. Under 
the proposed revised schedule, owners and operators of incinerators, 
cement kilns, and lightweight aggregate kilns would be required to 
conduct the initial comprehensive performance test to document 
compliance with the replacement standards proposed today (Sec. Sec.  
63.1219, 63.1220, and 63.1221) within 12 months of the compliance 
date. See discussion in Part Three, Section I.F.
---------------------------------------------------------------------------

    The rationale for the testing and monitoring requirements, and 
implementation of the requirements, is the same as discussed in the 
rulemakings promulgating those requirements for hazardous waste-burning 
incinerators, cement kilns, and lightweight aggregate kilns, and as 
discussed in Part Three of today's preamble. See 61 FR 43501 (August 
23, 1996), 62 FR 24212 (May 2, 1997), 67 FR 6791 (February 13, 2002), 
and 67 FR 6967 (February 14, 2002). For this reason, we only summarize 
those identical requirements and our rationale for them in today's 
notice.\190\
---------------------------------------------------------------------------

    \190\ For this reason, in the technical support documents for 
today's proposed rule we also refer extensively to the technical 
support documents for the Phase I rule.
---------------------------------------------------------------------------

A. What Is the Rationale for the Proposed Testing Requirements?

    The proposed rule requires solid fuel-fired boilers and liquid 
fuel-fired boilers to perform an initial comprehensive performance test 
for dioxin/furan,\191\ mercury, particulate matter, semivolatile 
metals, low volatile metals, and total chloride to demonstrate 
compliance with emission standards. Hydrochloric acid production 
furnaces would be required to perform an initial comprehensive 
performance test for dioxin/furan and total chloride to demonstrate 
compliance with emission standards. All three source categories are 
also subject to the destruction and removal efficiency standard. 
Compliance with the destruction and removal efficiency standard, 
however, is based on a one-time emissions test, and previous 
destruction and removal efficiency testing under RCRA requirements may 
be used for that demonstration if design, operation, or maintenance of 
the source has not changed in a manner that could adversely affect 
combustion efficiency and, thus, destruction and removal efficiency. 
Finally, all three source categories would be required to demonstrate 
compliance with the carbon monoxide/hydrocarbon emission standard 
during the comprehensive performance test (and at all other times).
---------------------------------------------------------------------------

    \191\ Those boilers that would be subject to a numerical dioxin/
furan standard (i.e., liquid fuel-fired boilers equipped with an 
electrostatic precipitator or fabric filter) would be required to 
conduct periodic comprehensive and confirmatory testing. Other 
boilers would be required to conduct a one-time test for dioxin/
furan emissions under the conditions discussed below in the text.
---------------------------------------------------------------------------

    The comprehensive performance test would be conducted every five 
years to ensure that the performance of the air pollution control 
device has not deteriorated and that other factors that may affect 
emissions have not caused an increase in emissions above the standards.
    The proposed rule also requires confirmatory testing to ensure 
compliance with the dioxin/furan emission standards, the test to be 
conducted mid-way between comprehensive performance tests when 
operating under typical conditions rather than at performance test 
conditions. More frequent confirmatory testing for dioxin/furan is 
needed because dioxin/furan emissions can be affected by various and 
interrelated factors, some of which are not fully understood, and 
because of the particular health hazard posed by emissions of dioxin/
furan.
    To ensure continuous compliance with the emissions standards, you 
would be required to establish limits on key operating parameters 
susceptible to continuous monitoring. The limits would be based on 
operating values achieved during the comprehensive performance test 
when the source successfully demonstrates compliance.\192\ Because 
operating limits are calibrated based on operations during the 
comprehensive performance test, sources generally operate at the upper 
end of the range of normal operations during these tests. These 
proposed requirements are discussed below in Section XII.C.
---------------------------------------------------------------------------

    \192\ Because the dioxin/furan confirmatory test is conducted 
under operating conditions that are within the range of normal 
operations rather than at the upper end of the range of normal 
operations as during a comprehensive performance test, you would not 
reestablish operating conditions for dioxin/furan based on the 
confirmatory performance test.
---------------------------------------------------------------------------

B. What Are the Dioxin/Furan Testing Requirements for Boilers That 
Would Not Be Subject to a Numerical Dioxin/Furan Emission Standard?

    As explained earlier, we are not proposing numerical dioxin/furan 
emission standards for solid fuel-fired boilers and for those liquid 
fuel-fired boilers that are equipped with wet scrubbers or no 
particulate control device. Rather, those boilers would be subject to 
the carbon monoxide/hydrocarbon emission standard and the destruction 
and removal efficiency standard to help minimize dioxin/furan 
emissions. See discussion in Part Two, Sections X.A and XI.A.
    We propose that solid fuel-fired boilers and those liquid fuel-
fired boilers that would not be subject to a numerical dioxin/furan 
emission standard conduct a one-time dioxin/furan emission test to 
quantify the effectiveness of today's proposed surrogate dioxin/furan 
emission controls. This test would be performed no later than the 
initial comprehensive performance test required under the proposed 
standards. The results of this one-time test would be reported with the 
test results for the first comprehensive performance test. See proposed 
Sec.  3.1207(b)(3).
1. What Is the Rationale for Requiring the Test?
    We are adopting this provision pursuant to our authority in CAA 
section 114 (a)(1)(D), which allows EPA to require ``any person * * * 
who is subject to any requirement of this chapter'' (which includes 
section 112) on a one-time, periodic or continuous basis, to ``sample 
such emissions (in accordance with such procedures or methods, at such 
locations, at such intervals, during such periods and in such manner as 
the Administrator shall prescribe)''. The purpose of such monitoring is 
``developing or assisting in the development of'' standards under 
various provisions of the Act, including section 112. In this case, 
monitoring will assist in making determinations under both section 
112(d)(6) and section 112(f), which could lead to development of 
standards under either or both of these provisions.
    Section 112(d)(6) of the Act requires us to ``review, and revise as 
necessary emission standards promulgated under this section no less 
than every eight years.'' We believe testing that results from 
compliance with today's proposed standards will, in nearly all cases, 
establish an adequate database for us to perform this review. However, 
we would not have sufficient dioxin/furan emissions data for those 
boilers that are subject to the carbon monoxide/hydrocarbon standard 
and destruction and removal efficiency standard in lieu of a numerical 
dioxin/furan standard. We have data from approximately one-third of the 
boilers that are not subject to a numerical dioxin/furan standard. 
Although those data indicate that these sources emit low concentrations 
of dioxin/furan despite the absence of any dioxin/furan control 
equipment, we are concerned about extrapolating this performance to the 
entire universe of

[[Page 21308]]

the subject boilers because our data set may not be statistically 
random and the potential hazard posed by dioxin/furan is high. In fact, 
the design of these sources would seem to have the potential for 
formation of significant dioxin/furan concentrations.\193\ We think 
this proposed testing would add a one-time cost of approximately 
$10,000 for each source for which dioxin/furan test data are not 
already available, and the cost appears reasonable to enable us to meet 
our section 112(d)(6) and 112(f) mandates. Section 112(d)(6) requires 
EPA, at specified times, to determine if further technology-based 
emission reductions are warranted. Quantified dioxin/furan emission 
information from these sources will assist in this determination. 
Section 112(f) requires EPA (among other things) to determine if 
emissions from all sources subject to section 112(d) standards must be 
further reduced in order to assure an ample margin of safety to protect 
public health. Having actual emission data from these sources obviously 
will assist in making the required section 112(f) determinations for 
these sources.
---------------------------------------------------------------------------

    \193\ Incinerators equipped with waste heat recovery boilers are 
known to emit high levels of dioxin/furan, and hydrochloric acid 
production furnaces with waste heat recovery boilers can also emit 
high levels of dioxin/furan. Because the mechanisms that affect 
formation and control of dioxin/furan are complex and not fully 
understood, we are concerned that some of the factors that cause 
high dioxin/furan emissions from incinerators and hydrochloric acid 
production furnaces equipped with waste heat recovery boilers may 
also affect dioxin/furan emissions from boilers.
---------------------------------------------------------------------------

2. What Are the Operating Requirements for the Test?
    You must perform the dioxin/furan test under feed and operating 
conditions that are most likely to maximize dioxin/furan emissions, 
similar to a dioxin/furan comprehensive performance test. Based on 
currently available research, the following factors should be 
considered for the testing: (1) Dioxin/furan testing should be 
conducted at the point in the maintenance cycle for the boiler when the 
boiler tubes are more fouled and soot-laden, and not after maintenance 
involving soot or ash removal from the tubes; (2) dioxin/furan testing 
should be performed following (or during) a period of feeding normal or 
greater quantities of metals; (3) dioxin/furan testing should be 
performed while feeding normal or greater quantities of chlorine; (4) 
the flue gas temperature in some portion of the heat recovery section 
of the boiler should be within the dioxin formation temperature window 
of 750 to 400[deg]F during the testing; (5) the testing should not be 
conducted under optimal combustion conditions; (6) for units equipped 
with wet air pollution control systems, the testing should be conducted 
after a high solids loading has developed in the scrubber system; and 
(7) for solid fuel-fired boilers, the sulfur content of the coal should 
be equivalent to or lower than normal coal sulfur levels, and the gas 
temperature at the inlet to the electrostatic precipitator or fabric 
filter should be close to the operating limit. In addition, unless 
sulfur compounds are routinely fed to the unit, dioxin/furan testing 
should not be performed after a period of firing high sulfur fuel or 
injection of sulfur additives.
    The majority of these recommendations are based on research 
demonstrating that soot deposits can enhance dioxin/furan formation in 
the presence of chlorine and catalytic metal contaminants, with 
formation continuing even after cessation of those contaminant feeds to 
the system.194, 195 The boiler tube deposits serve as a sink 
and source for dioxin/furan reactants (catalytic metals and chlorine), 
and combined soot-copper deposits have been shown to cause more dioxin/
furan formation than a deposit of soot or copper alone. From analysis 
of soot deposits taken from different sections of a firetube boiler, 
the highest measured dioxin/furan concentrations were found in those 
deposits containing the highest concentrations of copper and chloride. 
Those same deposits were removed from the boiler passages where flue 
gas temperatures ranged from 600-300[deg]C, which is within the often-
cited optimal temperature region for dioxin/furan formation. Tube 
deposits have also been shown to have a negative effect on dioxin 
emissions when those deposits have been affected by sulfur dioxide, 
which is why dioxin/furan testing is not recommended following a period 
of feeding higher-than-normal levels of sulfur to the boiler.
---------------------------------------------------------------------------

    \194\ Lee, C.W.; Kilgroe, J.D.; Raghunathan, K. Environ. Eng. 
Sci. 1998, 15(1), 71-84.
    \195\ Gullett, B.K.; Touati, A.; Lee, C.W. Environ. Sci. 
Technol. 2000, 34, 2069-2074.
---------------------------------------------------------------------------

    The recommendation not to test under optimal combustion conditions 
has been explained previously in the September 1999 Final Rule preamble 
discussion. See 64 FR at 52937. Good combustion practices minimize 
dioxin/furan emissions by: (1) Destroying trace dioxins/furans that may 
be present in feed streams; (2) minimizing gas-phase formation of 
dioxins/furans; and (3) minimizing dioxin/furan precursors that may 
enhance post-combustion formation.
    For units equipped with wet air pollution control systems, it is 
also recommended that testing be conducted after a high solids loading 
has developed in the scrubber system. Research conducted to explore the 
phenomenon of increased dioxin/furan flue gas concentrations across 
some wet scrubber systems has shown differing flue gas outlet dioxin/
furan homologue profiles than flue gas inlet profiles to the scrubber, 
but similar flue gas outlet homologue profiles to scrubber suspended 
solids and sludge profiles.\196\ This result suggests that some type of 
memory effect may be associated with suspended solids in a scrubber 
system which can cause higher dioxin/furans emissions.
---------------------------------------------------------------------------

    \196\ Takaoka, M.; Liao, P.; Takeda, N.; Fujiwara, T.; Oshita, 
K. Chemosphere 2003, 53, 153-161.
---------------------------------------------------------------------------

    You may use data-in-lieu of testing to document dioxin/furan 
emissions for similar on-site boilers. In addition, dioxin/furan 
emission data from previous testing would be acceptable, provided the 
test was performed in a manner likely to maximize dioxin/furan 
emissions.

C. What Are the Proposed Test Methods?

    The proposed emission standards are method-based standards, meaning 
that the stack test methods used for compliance must be the same as 
those used to generate the emissions data we used to calculate the 
standards. Because alternative stack methods may report lower 
emissions, it is appropriate to require use of the same methods for 
compliance as sources used to generate the emissions data in our data 
base.
    For this reason, you would be required to use the following stack 
test methods for compliance: (1) Method 29 for mercury, semivolatile 
metals, and low volatile metals; and (2) Method 26/26A for total 
chlorine.\197\ For dioxin/furan, the rule would require use of Method 
0023A unless you receive approval to use Method 23. We discuss the 
rationale for allowing site-specific approvals to use Method 23 in Part 
Three, Section II.D of today's preamble. In addition, for particulate 
matter, you would be required to use either Method 5, the method used 
to generate the data in our data base or Method 5i. We allow use of 
Method 5i because it is more

[[Page 21309]]

precise than Method 5 at lower particulate matter loadings.
---------------------------------------------------------------------------

    \197\ Please note that we discuss in Section XIII of the 
preamble above concerns with the accuracy of M26/26A for measuring 
emissions of total chlorine for cement kilns. As we explain there, 
although M26/26A is appropriate for demonstrating compliance with 
the MACT standards for cement kilns, it is not acceptable for 
demonstrating compliance with risk-based standards developed under 
authority of section 112(d)(4) of the Act.
---------------------------------------------------------------------------

    These test methods are codified in 40 CFR part 60, appendix A.\198\
---------------------------------------------------------------------------

    \198\ Method 0023A, however, is included in ``Test Methods for 
Evaluating Solid Waste, Physical/Chemical Methods,'' EPA Publication 
SW-846 Third Edition (November 1986), as amended.
---------------------------------------------------------------------------

D. What Is the Rationale for the Proposed Continuous Monitoring 
Requirements?

    The most direct means of ensuring compliance with emissions limits 
is the use of continuous emission monitoring systems (CEMS). We 
consider other options when CEMS are not available or when we consider 
the impacts of including such requirements unreasonable. When 
monitoring options other than CEMS are considered, it is often 
necessary for us to balance more reasonable costs against the quality 
or accuracy of the emissions monitoring data. Although monitoring 
operating parameters cannot provide a direct measurement of emissions, 
it is often a suitable substitute for CEMS. The information provided 
can be used to ensure that air pollution control equipment is operating 
properly. Because most parameter requirements are calibrated during 
comprehensive performance testing,\199\ they provide a reasonable 
surrogate for direct monitoring of emissions. This information 
reasonably assures the public that the reductions envisioned by the 
proposed rule are being achieved.
---------------------------------------------------------------------------

    \199\ Except that some parameters are limited based on the 
recommendations/specifications of the manufacturer of the control 
device.
---------------------------------------------------------------------------

1. What CEMS Requirements Did EPA Consider?
    To comply with the carbon monoxide or hydrocarbon emission limits, 
you would be required to use a carbon monoxide or hydrocarbon CEMS as 
well as an oxygen CEMS to correct the carbon monoxide or hydrocarbon 
values to 7% oxygen. See Sec.  63.1209(a). Because boilers and 
hydrochloric acid production furnaces are currently required to use 
these CEMS to comply with existing RCRA emission standards for carbon 
monoxide or hydrocarbons, there would be a minimal incremental 
compliance cost.\200\
---------------------------------------------------------------------------

    \200\ If you elect to comply with the carbon monoxide standard 
rather than the hydrocarbon standard, you would be required to 
document that hydrocarbon emissions during the comprehensive 
performance test meet the standard.
---------------------------------------------------------------------------

    We also evaluated the cost of applying hydrogen chloride CEMS to 
boilers and hydrochloric acid production furnaces. We estimate the 
capital costs for hydrogen chloride CEMS to be $88,000 per unit and 
annualized costs to be $33,000 per unit. We determined these costs 
would be unreasonably high considering: (1) The CEMS detects hydrogen 
chloride but not chlorine gas, so that compliance with the total 
chlorine emission standard could not be monitored; (2) the 
effectiveness of operating parameter limits to ensure compliance with 
the emission standard for total chlorine; and (3) the relatively low 
level of hazard posed by emissions of total chlorine.
    Finally, we conclude that the use of CEMS to document compliance 
with particulate matter or metal HAP emission standards has not been 
demonstrated on hazardous waste combustors in the United States.
2. What Operating Parameter Limits Would Be Required?
    To ensure continuous compliance with the proposed emission limits, 
you would be required to establish limits on key operating parameters 
and continuously monitor the parameters including: feedrate of metals, 
chlorine, and, for some source categories, ash; key combustor operating 
parameters; and key operating parameters of the control device. See 
Sec.  63.1209(j-o). You would also be required to document monitoring 
by recordkeeping and reporting. We selected the following requirements 
based on reasonable cost, ease of execution, and usefulness of the 
resulting data to both owners and operators and EPA for ensuring 
continuous compliance with the emission limits.
    To ensure continuous compliance with the dioxin/furan emission 
limit, you would be required to establish: (1) A limit on maximum gas 
temperature at the inlet to a dry particulate matter control device; 
(2) a limit on minimum combustion chamber temperature; (3) a limit on 
maximum flue gas flowrate or production rate; (4) a limit on maximum 
waste feedrate; (5) if your combustor is equipped with an activated 
carbon injection system: limits on the particulate matter control 
device, as discussed below; a limit on minimum carbon injection rate; a 
limit on minimum carrier fluid flowrate or pressure drop; and you must 
specify and use the brand (i.e., manufacturer) and type of carbon used 
during the comprehensive performance test, unless you document key 
parameters that affect adsorption and establish limits on those 
parameters based on the carbon used in the comprehensive performance 
test; (6) if your combustor is equipped with a carbon bed: you must 
monitor the bed life to ensure that it has not reached the end of its 
useful life to minimize dioxin/furan (and mercury) emissions at least 
to the levels required by the emission standards; you must replace the 
bed or bed segment before it has reached the end of its useful life; 
you must specify and use the brand (i.e., manufacturer) and type of 
carbon used during the comprehensive performance test, unless you 
document key parameters that affect adsorption and establish limits on 
those parameters based on the carbon used in the comprehensive 
performance test; and you must establish a limit on maximum gas 
temperature either at the bed inlet or outlet; (7) if your combustor is 
equipped with a catalytic oxidizer: limits on minimum and maximum gas 
temperature at the inlet to the catalyst; you must replace the oxidizer 
when it has reached the maximum service time specified by the 
manufacturer; and when replacing the catalyst, the new catalyst must be 
equivalent to or better than the one used during the previous 
comprehensive performance test as measured by catalytic metal loading 
for each metal, space time, and substrate construction; (8) if you feed 
a dioxin/furan inhibitor into the combustion system: a limit on minimum 
inhibitor feedrate; and you must specify and use the brand (i.e., 
manufacturer) and type of inhibitor used during the comprehensive 
performance test, unless you document key parameters that affect the 
effectiveness of the inhibitor and establish limits on those parameters 
based on the inhibitor used in the comprehensive performance test. See 
Sec.  63.1209(k).
    To ensure continuous compliance with the mercury emission limit, 
owners and operators of boilers would be required to establish: (1) A 
limit on the total feedrate of mercury in all feedstreams for solid 
fuel-fired boilers, and a limit on mercury in hazardous waste 
feedstreams per million Btu of hazardous waste fired for liquid-fuel-
fired boilers; \201,\ \202\ (2) if your boiler is equipped with a wet 
scrubber, limits prescribed for control of total chlorine with a wet 
scrubber, except for a limit on minimum pH of the scrubber water; (3) 
if your boiler is equipped with an activated carbon injection system, 
limits on the particulate matter control device as discussed below, and 
limits on the activated carbon injection system as

[[Page 21310]]

discussed above for dioxin/furan; and (4) if your boiler is equipped 
with an activated carbon bed, limits on the carbon bed as discussed 
above for dioxin/furan.
---------------------------------------------------------------------------

    \201\ This is because the mercury emission standard for liquid 
fuel-fired boilers is a hazardous waste thermal emission 
concentration. Liquid fuel-fired boilers would also be required to 
monitor the heating value of hazardous waste feeds to ensure 
compliance with the hazardous waste thermal emission concentration.
    \202\ The mercury feedrate limit would be based on levels fed 
during the comprehensive performance test unless the regulatory 
authority approves a request for you to extrapolate to a higher 
allowable feedrate (and emission rate) limit.
---------------------------------------------------------------------------

    You may comply with mercury feedrate limits only, however, if you 
elect to assume that all mercury in the feed is emitted. For solid 
fuel-fired boilers, you would assume that all mercury in all 
feedstreams is emitted under this alternative approach. You would also 
establish a limit on minimum flue gas flowrate to ensure compliance 
with the mercury emission standard. For liquid fuel-fired boilers where 
the mercury emission standard is expressed as hazardous waste thermal 
emissions, you would assume that all mercury in all hazardous waste 
feedstreams is emitted. You would have to comply with a hazardous waste 
thermal feed concentration that would be expressed as the mass of 
mercury in the hazardous waste per million Btu heat input contributed 
by the hazardous waste. Also, please note that these compliance 
requirements would not apply to hydrochloric acid production furnaces 
because (as explained earlier) we propose to use the total chlorine 
standard as a surrogate for the mercury, particulate matter, 
semivolatile metal, and low volatile metal standards for these sources. 
See Sec.  63.1209(l).
    To ensure continuous compliance with the particulate matter 
emission limit, you would be required to establish: (1) Limits on the 
control device operating parameters; (2) a limit on maximum flue gas 
flowrate or production rate; and a limit on maximum ash feedrate. If 
your boiler is equipped with a wet scrubber, you would establish limits 
on: (1) For high energy scrubbers only, minimum pressure drop across 
the scrubber and either minimum liquid to gas ratio or minimum scrubber 
water flowrate and maximum flue gas flowrate; and (2) for all 
scrubbers, the solids content of the scrubber liquid or a minimum 
blowdown rate. If your boiler is equipped with an electrostatic 
precipitator, ionizing wet scrubber, or fabric filter, please note that 
we discuss in Part Three, Section II.I. below proposed compliance 
parameters for these control devices. Briefly, if your boiler is 
equipped with a fabric filter, you must comply with bag leak detection 
system requirements. If your boiler is equipped with an electrostatic 
precipitator or ionizing wet scrubber, you must either: (1) Install and 
operate a particulate matter loading detector as a process monitor to 
indicate when you must take corrective measures; or (2) establish 
limits on key operating parameters, on a site-specific basis, that are 
representative and reliable indicators that the control device is 
operating within the same range of conditions as during the 
comprehensive performance test, and link those operating limits to the 
automatic waste feed cutoff system. Please note that the particulate 
matter compliance requirements would not apply to hydrochloric acid 
production furnaces, as discussed above. See Sec.  63.1209(m).
    To ensure continuous compliance with the semivolatile and low 
volatile metal emission limits, you would be required to establish: (1) 
A limit on the maximum inlet temperature to the primary dry particulate 
matter control device; (2) a limit on maximum feedrate of semivolatile 
and low volatile metals from all feedstreams for solid fuel-fired 
boilers, and a limit on semivolatile metals and low volatile metals in 
hazardous waste feedstreams per million Btu of hazardous waste fired 
for liquid-fuel-fired boilers; 203, 204 (3) limits (or 
process monitors) on the particulate matter control device as discussed 
above; (4) a limit on maximum feedrate of total chlorine or chloride in 
all feedstreams; and (5) a limit on maximum flue gas flowrate or 
production rate. You may comply with semivolatile and low volatile 
metal feedrate limits only, however, if you elect to assume that all 
semivolatile and low volatile metals in the feed is emitted. For solid 
fuel-fired boilers, you would assume that all semivolatile and low 
volatile metals in all feedstreams are emitted under this alternative 
approach. You would also establish a limit on minimum flue gas flowrate 
to ensure compliance with the semi- and low volatile metals emission 
standard. For liquid fuel-fired boilers where the semivolatile and low 
volatile metals emission standards are expressed as hazardous waste 
thermal emissions, you would assume that all semivolatile and low 
volatile metals in all hazardous waste feedstreams are emitted. You 
would have to comply with a hazardous waste thermal feed concentration 
that would be expressed as the mass of semivolatile (or low volatile) 
metals in the hazardous waste per million Btu heat input contributed by 
the hazardous waste. Also, please note that the semivolatile metal and 
low volatile metal compliance requirements would not apply to 
hydrochloric acid production furnaces, as discussed above. See Sec.  
63.1209(n).
---------------------------------------------------------------------------

    \203\ This is because the semivolatile metal and low volatile 
metal emission standards for liquid fuel-fired boilers are hazardous 
waste thermal emission concentrations. You would also be required to 
monitor the heating value of hazardous waste feedstreams to ensure 
compliance with the hazardous waste thermal emission concentration.
    \204\ The semivolatile and low volatile metal feedrate limits 
would be based on levels fed during the comprehensive performance 
test unless the regulatory authority approves a request for you to 
extrapolate to higher allowable feedrate (and emission rate) limits. 
Please note that the semivolatile and low volatile metal feed limits 
for liquid fuel-fired boilers are hazardous waste thermal 
concentration limits (pounds of metal per million Btu), not mass 
feedrate limits, given that the emission standards are expressed as 
hazardous waste thermal emissions.
---------------------------------------------------------------------------

    To ensure continuous compliance with the total chlorine emission 
limit, you would be required to establish: (1) A limit on maximum 
feedrate of total chlorine and chloride from all feedstreams for solid 
fuel-fired boilers, and a limit on total chlorine and chloride in 
hazardous waste feedstreams per million Btu of hazardous waste fired 
for liquid-fuel-fired boilers;\205\ (2) a limit on maximum flue gas 
flowrate or production rate; (3) if your combustor is equipped with a 
high or low energy wet scrubber: a limit on minimum pH of the scrubber 
water; a limit on either the minimum liquid to gas ratio or the minimum 
scrubber water flowrate and maximum flue gas flowrate; (4) if your 
combustor is equipped with a high energy wet scrubber, a limit on 
minimum pressure drop across the scrubber; (5) if your combustor is 
equipped with a low energy wet scrubber: a limit on minimum pressure 
drop across the scrubber; and a limit on minimum liquid feed pressure 
to the scrubber; and (6) if your combustor is equipped with a dry 
scrubber: a limit on minimum sorbent feedrate; a limit on minimum 
carrier fluid flowrate or nozzle pressure drop; and you must specify 
and use the brand (i.e., manufacturer) and type of sorbent used during 
the comprehensive performance test, unless you document key parameters 
that affect the effectiveness of the sorbent and establish limits on 
those parameters based on the sorbent used in the comprehensive 
performance test. If your combustor is equipped with an ionizing wet 
scrubber, please note that we discuss in Part Three, Section II.I. 
below proposed compliance parameters for this control device. Briefly, 
if your combustor is equipped with an ionizing wet scrubber, you must 
either: (1) Install and operate a particulate matter loading detector 
as a process monitor to indicate when you must take corrective 
measures; or (2)

[[Page 21311]]

establish limits on key operating parameters, on a site-specific basis, 
that are representative and reliable indicators that the control device 
is operating within the same range of conditions as during the 
comprehensive performance test, and link those operating limits to the 
automatic waste feed cutoff system.
---------------------------------------------------------------------------

    \205\ This is because the total chlorine emission standard for 
liquid fuel-fired boilers is a hazardous waste thermal emission 
concentration. You would also be required to monitor the heating 
value of hazardous waste feedstreams to ensure compliance with the 
hazardous waste thermal emission standard.
---------------------------------------------------------------------------

    You may comply with a total chlorine and chloride feedrate limit 
only, however, if you elect to assume that all chlorine in the feed is 
emitted. For solid fuel-fired boilers, you would assume that all 
chlorine in all feedstreams is emitted under this alternative approach. 
You would also establish a limit on minimum flue gas flowrate to ensure 
compliance with the total chlorine standard. For liquid fuel-fired 
boilers where the total chlorine emission standard is expressed as 
hazardous waste thermal emissions, you would assume that all chlorine 
in all hazardous waste feedstreams is emitted. You would have to comply 
with a hazardous waste thermal feed concentration that would be 
expressed as the mass of chlorine in the hazardous waste per million 
Btu heat input contributed by the hazardous waste. See Sec.  
63.1209(o).
    To ensure continuous compliance with the destruction and removal 
efficiency standard, you would be required to: (1) Establish a limit on 
minimum combustion chamber temperature; (2) establish a limit on 
maximum flue gas flowrate or production rate; (3) establish a limit on 
maximum hazardous waste feedrate; and (4) specify operating parameters 
and limits to ensure that good operation of each hazardous waste firing 
system is maintained. See Sec.  63.1209(j).

E. What Are the Averaging Periods for the Operating Parameter Limits, 
and How Are Performance Test Data Averaged To Calculate the Limits?

    Except as discussed in Section XIV.F below, we propose that owners 
and operators of solid fuel-fired boilers, liquid fuel-fired boilers, 
and hydrochloric acid production furnaces establish averaging periods 
for the operating parameter limits and calculate the limits from 
comprehensive performance test data under the same approaches required 
currently for incinerators, cement kilns, and lightweight aggregate 
kilns. A detailed discussion of how those approaches work, and the 
rationale for them, are provided at 64 FR at 52919-22 (September 30, 
1999). That discussion is summarized below.
    We propose the following averaging periods: (1) No averaging period 
(i.e., instantaneous monitoring) for maximum combustion chamber 
pressure to control combustion system leaks; \206\ (2) 12-hour rolling 
averages for maximum feedrate of mercury, semivolatile metals, low 
volatile metals, total chlorine and chloride, and ash; and (3) one-hour 
rolling averages for all other operating parameters. We propose a 12-
hour rolling average for metal, total chlorine and chloride, and ash 
feedrate limits to correspond to the potential duration of three runs 
of a comprehensive performance test, considering that feedrate and 
emissions, are, for the most part, linearly related. We propose an 
hourly rolling average limit for all parameters that are based on 
operating data from the comprehensive performance test, except 
combustion chamber pressure and metal, chlorine, and ash feedrate 
limits. Hourly rolling averages are appropriate for these parameters 
rather than averaging periods based on the duration of the performance 
test because we are concerned that there may be a nonlinear 
relationship between operating parameter levels and emission levels of 
HAP or HAP surrogates.
---------------------------------------------------------------------------

    \206\ Please note, however, that we request comment on the 
appropriateness of these combustion system leak requirements in Part 
Three of today's preamble.
---------------------------------------------------------------------------

    We propose two approaches to calculate limits for operating 
parameters: (1) Calculate the limit as the average of the maximum (or 
minimum, as specified) rolling averages for each run of the test; or 
(2) calculate the limit as the average of the test run averages for 
each run of the test. Hourly rolling averages for two parameters--
combustion gas flowrate or production rate and hazardous waste 
feedrate--would be based on the average of the maximum hourly rolling 
averages for each run. Hourly rolling average and 12-hour rolling 
average limits for all other parameters, however, would be based on the 
average level occurring during the comprehensive performance test. We 
conclude that this more conservative approach is appropriate for these 
parameters because they can have a greater effect on emissions, and 
because it is consistent with how manual emissions results are 
determined.\207\ We also conclude that limits based on the average 
level occurring during the comprehensive performance are readily 
achievable. This is because sources generally conduct performance 
testing at the extreme upper end of the range of normal operations to 
provide the operating flexibility needed after establishing operating 
parameter limits. Because sources can readily control (during the 
performance test and thereafter) the parameters for which limits are 
established, the operating limits based on the average of the 
performance test runs should be readily achievable under routine 
operations.
---------------------------------------------------------------------------

    \207\ Manual method emission test results for each run represent 
average emissions over the entire run.
---------------------------------------------------------------------------

F. How Would Sources Comply With Emissions Standards Based on Normal 
Emissions?

    Several proposed emission standards would be based on emissions 
that are within the normal range of operations for the source rather 
than on compliance test emissions that represent the extreme upper end 
of the range of normal emissions: \208\ mercury standards for cement 
kilns, lightweight aggregate kilns, and liquid fuel-fired boilers, and 
semivolatile metal emissions for liquid fuel-fired boilers. To ensure 
compliance with emission standards based on normal emissions data, you 
would document during the comprehensive performance test a system 
removal efficiency for the metals and back-calculate from the emission 
standard a maximum metal feedrate limit that must not be exceeded on an 
annual rolling average. If your source is not equipped with an emission 
control system (such as activated carbon to control mercury) for the 
metals in question, however, you must assume zero system removal 
efficiency. This is because a source that is not equipped with an 
emission control system may be able to document a positive system 
removal efficiency, but it is not likely to be reproducible. It is 
likely to be an artifact of the calculation of emissions and feeds 
rather than a removal efficiency that is reliable and reproducible.
---------------------------------------------------------------------------

    \208\ Compliance test emissions represent the upper range of 
emissions from a source because operating parameter limits for the 
HAP or HAP surrogate are established based on this compliance test.
---------------------------------------------------------------------------

    To ensure that you can calculate a valid, reproducible system 
removal efficiency for sources equipped with a control system that 
effectively controls the metal in question, you may need to spike 
metals in the feed during the comprehensive performance test at levels 
that may result in emissions that are higher than the standard. This 
would be acceptable because compliance with an emission standard 
derived from normal emissions data is based on compliance with an 
annual average feedrate limit calculated as prescribed here, rather 
than compliance with the emission standard during the comprehensive 
performance test.
    We propose a one-year averaging period for the metal feedrate limit

[[Page 21312]]

because the emission standard represents normal, average emissions. 
Although the averaging period could be substantially shorter or longer, 
a one-year averaging period is within the range of reasonable averaging 
periods and would be readily achievable for a standard based on normal 
emissions. The annual rolling average metal feedrate would be updated 
each hour based on the average of the 60 previous 1-minute averages.
    We propose to retain the hourly rolling average requirement for the 
other operating parameter limits, however, for the reasons discussed 
above (i.e., to be conservative given the nonlinear relationship 
between the operating parameter and emissions, and because the limits 
would be readily achievable).

G. How Would Sources Comply With Emission Standards Expressed as 
Hazardous Waste Thermal Emissions?

    Several proposed emission standards would be expressed as hazardous 
waste thermal emissions: mass of pollutant emissions attributable to 
the hazardous waste feed per million Btu of hazardous waste fed to the 
combustor.
    To demonstrate compliance with a hazardous waste thermal emissions-
based standard during a comprehensive performance test, you would 
calculate the hazardous waste thermal emissions by apportioning mass 
emissions of mercury, semivolatile metals, low volatile metals, or 
total chlorine according to the ratio of the mass feedrate of mercury, 
semivolatile metals, low volatile metals, or total chlorine and 
chloride from hazardous waste feedstreams to the feedrate for all 
feedstreams and dividing by the heat input rate (i.e., million Btu/hr) 
attributable to the hazardous waste.
    To ensure continuous compliance with the hazardous waste thermal 
emissions-based standard, you would calculate an operating limit based 
on the hazardous waste thermal feed concentration during the 
performance test.\209\ The hazardous waste thermal feed concentration 
limit would be calculated as the mass feedrate (lb/hr) of mercury, 
semivolatile metals, low volatile metals, or total chlorine and 
chloride from hazardous waste feedstreams divided by the heat input 
rate (million Btu/hr) from hazardous waste feedstreams. For compliance, 
you would continuously monitor the feedrate of hazardous waste on a 12-
hour rolling average updated each minute or, for standards based on 
normal emissions, on an annual rolling average updated each hour. You 
must know the concentration of mercury, semivolatile metals, low 
volatile metals, or total chlorine and chloride in the hazardous waste 
at all times, and the heating value of the hazardous waste at all 
times. Using this information, you would calculate and record the 
hazardous waste thermal feed concentration on a 12-hour rolling 
average, or for standards based on normal emissions, on an annual 
rolling average updated each hour.
---------------------------------------------------------------------------

    \209\ If the hazardous waste thermal emission standard is 
derived from normal rather than compliance test emissions data, 
however, the hazardous waste thermal feed concentration would be 
calculated as discussed above in Section F of the preamble.
---------------------------------------------------------------------------

H. What Happens if My Thermal Emissions Standard Limits Emissions to 
Below the Detection Limit of the Stack Test Methods?

    Under today's proposed thermal emissions standards, the standard 
may limit emissions to levels that are below the analytical detection 
limit of the stack test method. For example, this may occur with the 
semi-volatile metals standard for liquid fuel boilers when allowable 
emission levels are below the analytical detection capabilities of 
Method 29 when the hazardous waste firing rate or heating value is low. 
To address this issue, we are requesting comment on an approach that 
would allow you to be in compliance with today's proposed thermal 
emission standards if certain sampling and analytical criteria are met.
    The first criterion would ensure that the test crew accumulates 
enough of the analyte (e.g., metal HAP) in the sample train to ensure 
that it is measurable by the laboratory. For example, the amount of HAP 
accumulated in a one hour sample may not be sufficient for the 
laboratory to quantify. On the other hand, a three hour test would be 
more likely to accumulate enough sample, since three times the amount 
of that HAP would be collected. Most Method 29 results that comprise 
our emissions database are from two to three hour samples. The first 
criterion would be met if the facility samples the flue gas for at 
least three hours for each run.
    The second criterion would ensure that the laboratory uses adequate 
quality assurance procedures to measure the HAP in the sample. Section 
13.2 of Method 29 provides the analytical detection limits for the 
various laboratory methods used to determine the amount of HAP 
accumulated in the sample. The second criterion would be met if the 
laboratory reports analytical detection limits that are less than or 
equal to those reported in section 13.2.
    The final criterion is that no HAP represented by the standard can 
be present above the analytical detection limit. For the semi-volatile 
metals standard, this means that neither lead nor cadmium could be 
present above the analytical detection limits for any run of the test. 
You would assume that the HAP is present at the full detection limit, 
if lead or cadmium are present above the analytical detection limit 
during any run of the test.
    If you wish to use this provision to demonstrate compliance with 
the standard, you would be required to show that all three criteria 
have been met in the Notification of Compliance sent to the appropriate 
permitting agency. You would not be required to provide advance notice 
or obtain prior approval from the permitting authority.

I. Are We Concerned About Possible Negative Biases Associated With 
Making Hydrogen Chloride Measurements in High Moisture Conditions?

    Several industry stakeholders have brought several scientific 
papers to our attention that indicate that Method 26A, used for 
compliance with the hydrogen chloride and chlorine gas standards, may 
have a significant low bias at wet stacks with low hydrogen chloride 
concentrations. These stakeholders have asked us not to establish 
standards for hydrogen chloride and chlorine standard below 20 ppmv to 
address this substantial negative bias.
    We agree that there was a concern early in the development and 
deployment of Method 26A that water droplets would not evaporate in the 
sampling train and would therefore dissolve hydrogen chloride in the 
sample train, before the hydrogen chloride can be caught by the 
impingers. EPA determined that this potential problem can be precluded 
by providing enough heat to the sample train to evaporate all water 
droplets that might collect in the sample probe or filter. Once the 
water is evaporated, the hydrogen chloride reenters the sample gas 
stream and is collected by the impingers.
    EPA's Office of Research and Development (ORD) performed laboratory 
studies to document and fully understand this problem. We also 
monitored the application of Method 26A and it's SW-846 equivalent to 
determine how these concerns may impact hydrogen chloride measurements 
made on wet stacks. Our conclusion is that the situations encountered 
in ORD's laboratory studies are not encountered when making stack test 
measurements.
    The Coalition for Responsible Waste Incineration, CRWI, provided a 
paper authored by Joette Steger, et al., which

[[Page 21313]]

illustrates this point. (See memorandum to docket for today's proposed 
rule from H. Scott Rauenzahn, U.S. EPA, entitled ``Method 26A and 
CRWI's Concerns,'' dated March 25, 2004.) Steger found that Method 26A 
has a significant negative bias when 40 to 50 percent of the water in 
the sample is in the form of water droplets. Under similar sample 
conditions, with 60 percent of the water in the form of droplets, 
Steger found that providing more heat to the sample train corrected the 
negative bias concern.
    We also checked our hydrogen chloride emissions data for hazardous 
waste combustors to see if water droplets could be present in the 
sample line. We found that water droplets could be present in three of 
our incinerator test conditions: 327C10 at 5 percent water droplets; 
808C1 at 12.5 percent water droplets; and 3024C1 at 8 percent water 
droplets. None of these stack conditions approach the 40 to 50 percent 
water droplets observed to be a problem by Steger. These stack gas 
conditions most closely resemble Steger's run B-5, with 10% water 
droplets. No negative bias was observed for Steger's run B-5. We 
conclude that this negative bias, while conceptually possible, is not 
encountered at hazardous waste combustors with wet stacks.
    We request comments on our analysis of these trade association's 
concerns, and request more data regarding this issue.

J. What Are the Other Proposed Compliance Requirements?

    We propose other compliance requirements for solid fuel-fired 
boilers, liquid fuel-fired boilers, and hydrochloric acid production 
furnaces that are the same as those currently in place at Sec.  63.1206 
for incinerators, cement kilns, and lightweight aggregate kilns. The 
rationale for the requirements is the same as discussed in previous 
rulemakings for incinerators, cement kilns, and lightweight aggregate 
kilns, and compliance procedures would be the same as currently 
required for those sources.
    The other compliance requirements include provisions for: startup, 
shutdown, and malfunction plans; operation and maintenance plans 
including a requirement for bag leak detector systems for fabric 
filters; automatic hazardous waste feed cutoff systems, including a 
requirement for exceedance reporting; combustion system leak 
requirements; changes in design, operation, or maintenance that could 
adversely affect compliance with emission standards; operator training 
and certification requirements; and requirements for sources that elect 
to comply with the carbon monoxide standard to document one-time that 
hydrocarbons also meet the hydrocarbon standard; and provisions 
allowing a one-time demonstration of compliance with the destruction 
and removal efficiency standard.
    Please note that we propose revisions to, or request comment on, 
some of these compliance requirements in Part Three of the preamble. 
Any revisions to these requirements that we might make in the final 
rule would be applicable to all hazardous waste combustors.

XV. How Did EPA Determine Compliance Times for this Proposed Rule?

    Section 112 of the CAA specifies the dates by which affected 
sources must comply with the emission standards. New or reconstructed 
units must be in compliance with the proposed rule immediately upon 
startup or [DATE THE FINAL RULE IS PUBLISHED IN THE Federal Register], 
whichever is later. A new or reconstructed unit for purposes of 
complying with this proposed rule is one that begins construction after 
April 20, 2004.\210\
---------------------------------------------------------------------------

    \210\ Please note that a new or reconstructed unit for purposes 
of complying with the Interim Standards applicable to incinerators, 
cement kilns, and lightweight aggregate kilns is a unit that began 
operation after September 30, 1999.
---------------------------------------------------------------------------

    Existing sources are allowed up to three years to comply with the 
final rule. See proposed Sec.  63.1206(a)(1)(ii) and (a)(2). This is 
the maximum period allowed by the CAA. We believe that three years for 
compliance is necessary to allow adequate time to design, install, and 
test control systems that will be retrofitted onto existing units.

XVI. How Did EPA Determine the Required Records and Reports for the 
Proposed Rule?

    We propose notification, reporting, and recordkeeping requirements 
for solid fuel-fired boilers, liquid fuel-fired boilers and 
hydrochloric acid production furnaces that are identical to those 
already in place at Sec. Sec.  63.1210 and 63.1211 and applicable to 
incinerators, cement kilns, and lightweight aggregate kilns. Please 
note, however, that we are proposing a new requirement applicable to 
all hazardous waste combustors that would require you to submit a 
Notification of Intent to Comply and a Compliance Progress Report.

A. Summary of Requirements Currently Applicable to Incinerators, Cement 
Kilns, and Lightweight Aggregate Kilns and That Would Be Applicable to 
Boilers and Hydrochloric Acid Production Furnaces

    Owners and operators of solid fuel-fired boilers, liquid fuel-fired 
boilers, and hydrochloric acid production furnaces would be required to 
submit the following notifications to the Administrator in addition to 
those required by the NESHAP General Provisions, subpart A of 40 CFR 
part 63: (1) Notification of changes in design, operation, or 
maintenance (Sec.  63.1206(b)(5)(i)); (2) notification of performance 
test and continuous monitoring system evaluation, including the 
performance test plan and continuous monitoring system performance 
evaluation plan (Sec. Sec.  63.1207(e)); and (3) notification of 
compliance, including results of performance tests and continuous 
monitoring system evaluations (Sec. Sec.  63.1210(b), 63.1207(j); 
63.1207(k), and 63.1207(l)). You would also be required to submit 
notifications to the Administrator if you request or elect to comply 
with various alternative requirements. Those notifications are listed 
at Sec.  63.1210(a)(2).
    Owners and operators of solid fuel-fired boilers, liquid fuel-fired 
boilers, and hydrochloric acid production furnaces would be required to 
submit the following reports to the Administrator in addition to those 
required by the NESHAP General Provisions, subpart A of 40 CFR part 63: 
(1) Startup, shutdown, and malfunction plan (if electing to comply with 
Sec.  63.1206(c)(2)(ii)(B)); (2) excessive exceedances report (Sec.  
63.1206(c)(3)(vi)); and (3) emergency safety vent opening reports 
(Sec.  63.1206(c)(4)(iv)).
    Owners and operators of solid fuel-fired boilers, liquid fuel-fired 
boilers, and hydrochloric acid production furnaces would be required to 
keep records documenting compliance with the requirements of Subpart 
EEE. Recordkeeping requirements are prescribed in Sec.  63.1211(b), and 
include requirements under the NESHAP General Provisions, subpart A of 
40 CFR part 63.

B. Why Is EPA Proposing Notification of Intent to Comply and Compliance 
Progress Report Requirements?

1. What Is the Notification of Intent to Comply?
    In the June 1998 ``fast track'' rule (63 FR 33782), we required 
that sources subject to the Phase I subpart EEE standards complete a 
Notification of

[[Page 21314]]

Intent to Comply (NIC) no later than October 2, 2000 and conduct a NIC 
public meeting no later than July 31, 2000. The NIC and its associated 
public meeting served four primary purposes during the early 
implementation and compliance phases of the Phase I subpart EEE 
requirements which we believe were of benefit to regulators, sources 
and the public alike.
    First, the NIC served as a compliance planning tool for Phase I 
sources because it required you to develop an outline of the key 
activities that needed to be completed in order to meet the subpart EEE 
standards by the compliance date. It also required that you include the 
estimated dates for each of those key activities. Because the NIC was 
required to be completed within the first year of implementing the 
Phase I requirements, it also may have had the added and important 
benefit of encouraging sources to reduce their HAP emissions early. By 
focusing a source's attention on the means by which it would achieve 
compliance well before the actual compliance date, the NIC may have 
prompted some sources to upgrade their combustion design and operations 
earlier, thereby yielding an early reduction in HAP emissions. The NIC 
also may have prompted earlier waste minimization efforts for the same 
reason.
    Second, the NIC also served as a planning tool for regulatory 
authorities. Based on the information provided in the NIC, regulators 
could determine what activities were likely to occur and when over the 
course of the three-year compliance period. For example, they could 
estimate how many sources needed to modify their combustion units and 
existing RCRA permits prior to performance testing, how many sources 
intended to stop burning hazardous waste, and how many sources intended 
to apply for the comparable fuels exclusion. Using this information, 
regulators could plan how to most efficiently allocate their resources 
in response to the forthcoming compliance activities of the sources.
    Third, the NIC promoted early public involvement by fostering an 
open dialogue between sources and the public regarding compliance 
strategies for meeting the Phase I subpart EEE standards. Experience 
has shown that members of the public are interested in being kept 
adequately informed of and having input into the compliance and 
permitting activities of hazardous waste combustion facilities. The NIC 
and its associated public meeting provided an opportunity for the 
public to share their views, thereby allowing the source to develop a 
final compliance strategy that met the goals of both the source and the 
surrounding community.
    Fourth, the public involvement aspect of the NIC also offset any 
public participation opportunities that may have been ``lost'' if 
sources chose to take advantage of the RCRA streamlined permit 
modification process. Many Phase I sources had to modify their 
combustion systems' design and/or operations in order to comply with 
the MACT standards. Sources that were already operating under RCRA 
combustion permits needed to first modify those permits before 
initiating any MACT compliance related changes. Normally, a Class 2 or 
3 modification would be necessary to incorporate into a RCRA permit the 
types of changes we expected would be necessary for sources complying 
with Phase I standards. Given that Class 2 and 3 modifications could 
have consumed a year or more of a source's three-year subpart EEE 
compliance period, we developed a streamlined permit modification 
process solely for the purpose of implementing subpart EEE upgrades. 
Under the streamlined process, you could request a Class 1 modification 
with prior Agency approval to address and incorporate any necessary 
MACT upgrades into your RCRA permit. To be eligible to use the 
streamlined permit modification, however, you first must have complied 
with the NIC requirements, including those related to public 
involvement.
2. What Happened to the NIC Provisions?
    We promulgated the NIC on June 19, 1998 (63 FR 33782) along with 
several other requirements related to the Phase I NESHAP. On May 14, 
2001, we removed the NIC and two other provisions from the federal 
regulations in response to a court mandate to vacate. See 66 FR 24270. 
In Chemical Manufacturers Ass'n v EPA, 217 F. 3d 861 (D.C. Cir. 2000), 
the court vacated three provisions of the Phase I rule: the Early 
Cessation requirement, the NIC and the Compliance Progress Report.\211\ 
While the panel majority held that we possessed the legal authority to 
impose an Early Cessation requirement, the panel also held that we had 
claimed the authority to do so without making a showing of a health and 
environmental benefit (such as reduced HAP emissions or less hazardous 
waste generated) and that this was an impermissible statutory 
interpretation. See 217 F. 3d at 865-67. The panel majority further 
held that because it could not determine whether we would have 
promulgated the NIC and Progress Report requirements absent the Early 
Cessation provision, both the NIC and Progress Report requirements 
should be vacated as well. However, the panel did agree to issue a stay 
of its mandate for a long enough period of time to allow sources to 
submit their NICs so that they would be eligible for the RCRA 
streamlined permit modification.
---------------------------------------------------------------------------

    \211\ Under the Early Cessation provision, we required sources 
that did not intend to comply with the Phase I standards to stop 
burning hazardous waste within two years of the effective date of 
the Phase I rule. Under the Compliance Progress Report provision, we 
required sources to report to their regulatory agencies the status 
of their progress toward compliance with the standards.
---------------------------------------------------------------------------

    As discussed above, the NIC was intended to serve as a compliance 
planning and communication tool. We did not intend the NIC to serve as 
the basis for requiring a source to cease burning hazardous waste. 
However, as a planning and communication tool we expected sources that 
did not intend to comply with the standards to state this in their NIC 
and include a schedule of activities that the source would need to 
complete in order to stop burning hazardous waste within the two-year 
Early Cessation time frame. We believe that the court recognized this 
interpretation as our original intent in their agreement to stay their 
issuance of the mandate until after sources had submitted their final 
NICs on October 1, 2000. By allowing the Phase I sources to complete 
the NIC process, the court provided sources with the opportunity to 
effectively plan their compliance strategies and take advantage of the 
RCRA streamlined permit modification. It also provided the public with 
the opportunity for a level of participation that they may not have had 
otherwise.
3. Why Is EPA Proposing To Re-Institute the NIC for Phase I Sources?
    As stated above, we believe that the NIC was a valuable planning 
and communication tool for sources, regulators, and the public during 
the early implementation and compliance stages of the 1999 Phase I 
subpart EEE requirements. The NIC also provided an additional benefit 
to sources upgrading their combustion systems by compensating for any 
``lost'' public participation opportunities when using the RCRA 
streamlined permit modification process. As discussed in Part One, I. B 
and D, we are proposing in today's notice to supplant the existing 
Phase I standards with final Replacement standards. We anticipate that 
a significant number of Phase I sources may need to conduct additional 
upgrades, or in some cases upgrade for the first time, to comply with 
the Replacements standards. See

[[Page 21315]]

Sec. Sec.  63.1219, 63.1220, and 63.1221. Re-instituting the NIC for 
these sources could provide the same planning and communication 
benefits during the initial Replacement standards compliance period 
that it did for the original Phase I standards.
    Specifically, we expect that by focusing attention early on the 
necessary tasks and strategies for achieving compliance, Phase I 
sources will be in a better position to meet the Replacement standards 
by the compliance date. Regulators will gain insight from the 
information provided in the NIC to effectively allocate their resources 
to accommodate future regulatory activities. And, the NIC will provide 
the public with the opportunity and mechanism to keep abreast of any 
significant changes an existing source might need to make as a result 
of the Replacement standards. We do not believe that the same planning 
and communication opportunities gained from completing the NIC process 
are available from other portions of the air regulatory program. For 
example, although the public will be notified of a source's obligation 
to comply with the Replacement standards during the reopening or 
renewal of the source's title V, this notification, in most cases, will 
not occur as early in the three-year subpart EEE compliance period, nor 
is it likely to include the specific information regarding the source's 
compliance strategy.\212\
---------------------------------------------------------------------------

    \212\ If a major title V source has a remaining permit term of 
three or more years on the date the Replacement standards are 
promulgated, the title V permitting authority must complete a 
reopening of the source's title V permit to incorporate the 
requirements of these standards not later than 18 months after 
promulgation. Major sources having remaining permit terms of less 
than three years on the date the Replacement standards are 
promulgated may wait until permit renewal to incorporate the new 
standards. Area sources with title V permits likewise may wait until 
permit renewal. Permitting authorities must follow the same public 
notice procedures for title V permit reopenings and renewals as is 
required for initial permit issuance under title V, including 
providing public notice of the action, providing a public comment 
period of at least 30 days, and providing an opportunity for a 
public hearing. See 40 CFR 70.7 and 71.7.
---------------------------------------------------------------------------

    In addition, while we believe that there will be fewer Phase I 
sources in the position of having RCRA combustion permit conditions 
after demonstrating compliance with the Interim standards, for those 
that do and wish to use the streamlined permit modification process to 
allow any necessary Replacement standards upgrades, a second NIC would 
provide the same public participation benefits as did the first 
NIC.\213\ 40 CFR 270.42(j) currently allows a source to use the RCRA 
streamlined modification process provided that the source first 
complied with the NIC requirements that were in place prior to October 
11, 2000. Since many sources complied with those NIC requirements in 
1999 and 2000, the existing regulatory language would allow those same 
sources to further modify their RCRA permits for Replacement standards 
upgrades. The regulatory language does not make any distinction 
regarding when the upgrades are to take place in relation to when the 
NIC requirements were to have been fulfilled. We do not believe that it 
is appropriate for a source to rely on previous informational and 
public participation activities carried out to comply with the earlier 
NIC requirements and emission standards to address upgrades occurring 
years later in response to a different set of standards any more than 
it would be appropriate to allow the public participation activities of 
a previous RCRA modification to suffice for a later modification. By 
requiring sources that choose to use the RCRA streamlined permit 
modification process for Replacement standards upgrades to first 
complete a NIC, including its associated public meeting, that 
specifically addresses those Replacement standards upgrades, the 
community will be kept better informed of additional changes to the 
combustion system and the impact on the RCRA permit.
---------------------------------------------------------------------------

    \213\ Once a source conducts its CPT and submits an Notification 
of compliance documenting compliance with the Subpart EEE standards, 
the source may request that its RCRA permit be modified to remove 
any duplicative limits or conditions. Only those risk-based 
provisions that are more stringent than the MACT requirements as 
specified in the Notification of compliance or that address other 
emission hazards will remain in the RCRA permit. We expect that many 
sources will document compliance with the Phase I Interim standards 
between 2003 and 2004 and will request the removal of any 
duplicative, less stringent provisions from their RCRA permits 
shortly thereafter.
---------------------------------------------------------------------------

4. Why Is EPA Proposing To Require the NIC for Phase II Sources?
    We believe that the NIC would provide the same benefits with 
respect to communication and compliance strategy planning for the Phase 
II sources that it has for Phase I sources. In addition, without 
completing the NIC process, Phase II sources will not be eligible to 
take advantage of the RCRA streamlined permit modification when 
upgrading their combustion systems. We are proposing that Phase II 
sources comply with the same NIC requirements as their Phase I 
counterparts.
5. How Will the NIC Process Work?
    We are proposing to apply a similar NIC process to that which we 
promulgated in the June 19, 1998 ``fast track'' rule (63 FR 33782). The 
following is a general description of that process. Within nine months 
of the promulgation of the final Phase I Replacement standards and 
Phase II standards, you would develop and make publicly available a 
draft NIC. The draft NIC would contain general information such as 
whether you are a major or an area source and what waste minimization, 
emission control techniques, and emission monitoring techniques you 
might be considering. At the same time, you would also provide a notice 
to the public of at least one informal NIC public meeting. Within ten 
months, you would hold this public meeting to discuss the activities 
you described in the draft NIC for achieving compliance with the 
subpart EEE standards. The meeting provides an opportunity for a mutual 
understanding between you and the public regarding compliance options, 
including consideration of both technical (e.g., equipment changes to 
upgrade air pollution control devices) and operational (e.g., process 
changes to minimize waste generation) alternatives. We expect the 
exchange between you and the community at the meeting to be similar to 
that which would occur at RCRA pre-application meetings. That is, we 
intend for the meeting to provide an open, flexible and informal 
occasion for you and the public to discuss various aspects of your 
compliance strategy, provide an opportunity for sharing ideas and 
provide an opportunity for building a framework for a solid and 
positive working relationship. Lastly, you would submit a final NIC to 
your regulatory authority that would include the information provided 
in the draft NIC (revised as necessary after the public meeting) as 
well as a summary of the public meeting. This final NIC would be 
submitted to your regulatory authority within one year of the 
promulgation of the final Phase I Replacement standards and Phase II 
standards.
    In summary, we believe that the NIC would provide important 
planning and communication opportunities for both Phase I and Phase II 
sources. It also would allow all Phase I, as needed, and Phase II 
sources to take advantage of the RCRA streamlined permit modification 
procedure. Thus, we are proposing NIC requirements for both Phase I and 
Phase II sources.

[[Page 21316]]

6. What Is the Compliance Progress Report?
    In addition to the NIC, we also promulgated Compliance Progress 
Report requirements in the 1998 ``fast track'' rule. See 63 FR 33782. 
The purpose of the Progress Report was to help regulatory agencies 
determine if sources were making reasonable headway in their efforts to 
come into compliance. The Progress Report was required to be submitted 
at the midpoint of the three-year compliance period and contain 
information that essentially built on the information you previously 
provided in the NIC. For example, if you indicated in the NIC that you 
needed to make specific physical modifications to your combustion 
system in order to comply with the standards, you would be expected to 
describe your progress in making those modifications in your Compliance 
Progress Report. Although the Progress Report was primarily intended as 
a tool for the regulatory agencies, we believe it also may have been 
beneficial to sources as well. For example, the Progress Report could 
have been used by sources as a mechanism to review and make any 
necessary changes to their original strategy for achieving compliance.
    As discussed in the previous section, the Court vacated the early 
cessation, NIC and Compliance Progress Report provisions of the Phase I 
rule in Chemical Manufacturers Ass'n v EPA, 217 F. 3d 861 (D.C. Cir. 
2000). Although the Court's primary focus was the early cessation 
provision, it also vacated the Progress Report requirements because it 
could not determine whether we would have promulgated those 
requirements absent the early cessation provision.
7. Why Is EPA Requesting Comment on Requiring the Compliance Progress 
Report for Phase I and Phase II Sources?
    We believe that the Progress Report would be a useful tool for both 
regulators and sources in measuring progress toward achieving 
compliance with the Subpart EEE standards and determining if any 
revisions to a source's compliance strategy are necessary. Unlike the 
NIC, however, we do not have practical experience with the application 
of the Compliance Progress Report, because the Court vacated its 
requirements prior to their implementation. As a result, we are 
requesting comment on whether or not the Compliance Progress Report 
should be required for Phase I or Phase II sources.
8. How Would the Compliance Progress Report Requirement Work?
    The Compliance Progress Report requirements would be similar to 
those promulgated for Phase I sources in the June 19, 1998 ``fast 
track'' rule (63 FR 33782). Within two years of the promulgation of the 
final standards, you would develop and submit to your regulatory 
authority a Compliance Progress Report. The Report would include 
information which demonstrates your progress toward compliance. This 
could include, for example, completed engineering designs for any 
physical modifications to the combustion unit that are needed to comply 
with the standards; copies of construction applications; and binding 
contractual commitments to purchase, fabricate, and install any 
necessary equipment, devices, and ancillary structures. In addition, 
you would be expected to include a detailed schedule that lists the 
dates for all remaining key activities and projects that will bring you 
into compliance with the standards. For example, you would include bid 
and award dates for construction contracts, milestones for 
groundbreaking, and dates for the approval of permits and licenses. We 
would also expect you to include in your report any updates or changes 
to the information you previously provided in your NIC, including if 
you have changed your compliance plan based on engineering studies or 
evaluations that you have conducted since your NIC submittal.\214\ 
Sources that intend to cease burning hazardous waste prior to or on the 
compliance date would still be expected to submit a report describing 
key activities and projected dates for initiating RCRA closure and 
discontinuing hazardous waste activities at the combustion unit.
---------------------------------------------------------------------------

    \214\ For example, if you reported in your NIC that you intended 
to upgrade your existing unit, but later determined that it was more 
appropriate to replace the unit with a new unit, we would expect you 
to inform your regulatory agency of this change in your compliance 
plan in your Compliance Progress Report.
---------------------------------------------------------------------------

XVII. What Are the Title V and RCRA Permitting Requirements for Phase I 
and Phase II Sources?

    In today's notice of proposed rulemaking, we are maintaining the 
same general approach we took in the 1999 rule with respect to title V 
and RCRA permitting requirements and the Phase I sources. We feel that 
this approach, to place the MACT air emissions and related operating 
requirements in the title V permit and to continue to require RCRA 
permits for all other aspects of the combustion unit and the facility 
that are governed by RCRA, is still the most appropriate method to meet 
our obligations under both statutes. In 1999, our goal in developing a 
permitting scheme to accommodate both statutes with respect to air 
emission limitations and standards, was to avoid duplication to the 
extent practicable and to streamline requirements. We remain committed 
to that goal, as we revise and refine the permitting approach we 
finalized in 1999.

A. What Is the General Approach To Permitting Hazardous Waste 
Combustion Sources?

    In the September 1999 rule, we finalized a permitting approach that 
places the MACT air emissions and related operating requirements in the 
title V permit and retains all other RCRA related requirements (e.g., 
corrective action, general facility standards, other combustor specific 
concerns such as material handling, risk-based emission limits and 
operating requirements, and other hazardous waste management units) in 
the RCRA permit. See 64 FR 52828, 52833-52834 (September 30, 2000). 
Under this approach, sources comply with their RCRA emission limits and 
operating requirements until they demonstrate compliance with the MACT 
standards by conducting a comprehensive performance test and submitting 
a Notification of Compliance (NOC) to the Administrator (or authorized 
State) that documents compliance.\215\ Upon documenting compliance 
through the NOC, sources may begin the transition from RCRA permitting 
to title V permitting.
---------------------------------------------------------------------------

    \215\ There is no change to our decision to subject Phase I area 
sources to the same MACT standards and title V permitting 
requirements as the major sources. For Phase II sources, area 
sources are required to meet the same MACT standards as major 
sources, but only for: dioxin/furan, mercury, carbon monoxide/
hydrocarbons, and destruction and removal efficiency. See Part Two, 
Section I.A. for more information on regulation of area sources. 
Therefore, Phase II area sources will be required to obtain a title 
V permit only for those MACT standards as discussed later in 
Paragraph C.4. of this section.
---------------------------------------------------------------------------

    We believe that this approach still makes the most sense in terms 
of providing flexibility and minimizing duplication between the two 
permitting programs, while ensuring that there is no break in 
regulatory coverage. It is also appropriate given where sources will be 
in the transition process of complying with the MACT Interim Standards 
upon promulgation of the Phase I Replacement standards and the Phase II 
standards. The majority of Phase I sources will have initiated a

[[Page 21317]]

significant modification of their title V permits to include the 
operating requirements of their NOC and a modification of their RCRA 
permits to remove duplicative conditions. By this time, permitting 
authorities and sources are familiar with the current permitting 
approach and have worked through many issues to make compliance with 
the Interim Standards and the ensuing transition successful. We feel 
that permitting authorities and sources would prefer to draw upon their 
experiences and utilize the expertise they have developed, rather than 
exploring ways to implement a new permitting scheme. Therefore, we are 
retaining the same general approach to permitting for Phase I sources 
and are proposing to apply this same general approach to Phase II 
sources in today's Notice of proposed rulemaking: to place the MACT 
emission standards only in the CAA regulation at 40 CFR part 63 subpart 
EEE, and rely on implementation through the air program and operating 
permit programs developed under title V.
1. What Is the Authority for the Proposals Discussed in This Section?
    EPA is issuing these proposals to modify RCRA permits under the 
authority of sections 1006(b), 2002, 3004, 3005 and 7004(b) of RCRA. 
With regard to the regulatory framework that would result from today's 
proposal, we are proposing to eliminate the existing RCRA stack 
emissions national standards for hazardous air pollutants for hazardous 
waste combustors. That is, after submittal of the NOC established by 
today's rule and, where applicable, RCRA permit modifications at 
individual facilities, RCRA national stack emission standards will no 
longer apply to these hazardous waste combustors. We originally issued 
emission standards under the authority of section 3004(a) and (q) of 
RCRA, which calls for EPA to promulgate standards ``as may be necessary 
to protect human health and the environment.'' We believe that the 
proposed MACT standards are generally protective of human health and 
the environment, and that separate RCRA emission standards are not 
needed to protect human health and the environment. Refer to Part Four, 
Section IX. How Does the Proposed Rule Meet the RCRA Protectiveness 
Mandate? for a discussion on this topic.
    In addition, RCRA section 1006(b) directs EPA to integrate the 
provisions of RCRA for purposes of administration and enforcement and 
to avoid duplication, to the maximum extent practicable, with the 
appropriate provisions of the Clean Air Act (and other federal 
statutes). This integration must be done in a way that is consistent 
with the goals and policies of these statutes. Therefore, section 
1006(b) provides further authority for EPA to eliminate the existing 
RCRA stack emissions standards to avoid duplication with the new MACT 
standards.
    We are not proposing, however, that RCRA permit conditions to 
control emissions from these sources will never be necessary, only that 
the national RCRA standards appear to be unnecessary. Under the 
authority of RCRA's ``omnibus'' clause section 3005(c)(3); see 40 CFR 
270.32(b)(2)), RCRA permit writers may impose additional terms and 
conditions on a site-specific basis as may be necessary to protect 
human health and the environment. Thus, if MACT standards are not 
protective of human health and the environment in an individual 
instance, RCRA permit writers will establish permit limits that are 
protective.
    In RCRA, Congress gave EPA broad authority to provide for public 
participation in the RCRA permitting process. Section 7004(b) of RCRA 
requires EPA to provide for, encourage, and assist public participation 
in the development, revision, implementation, and enforcement of any 
regulation, guideline, information, or program under the Act.
2. Is EPA Proposing a Different Permitting Approach for New Sources?
    As discussed above, we are maintaining the same general permitting 
approach as before. However, we are proposing to eliminate the 
unintended result of the previous regulatory construct, which caused 
new sources to initially be subject to the RCRA air emission and 
operating requirements. In particular, we want to specify that any 
hazardous waste burning incinerators, cement kilns, lightweight 
aggregate kilns, boilers, and hydrochloric acid production furnaces 
newly entering the RCRA permitting process (e.g., sources that are 
seeking an initial RCRA permit or permit modification to include a new 
hazardous waste combustion unit) after promulgation of the Phase I 
Replacement standards and Phase II standards are not subject to certain 
specified RCRA permit requirements or performance standards. The 
approach we are proposing today is similar to the one we proposed in 
the July 3, 2001, proposed amendment rule (see 66 FR 35146), but was 
not finalized. The amendment was not finalized due to several 
unresolved issues and thus, it was agreed (during litigation settlement 
discussions), that we would revisit and address the issues in the Phase 
I Replacement standards and Phase II standards rulemaking.
    a. Why Is EPA Proposing a Different Permitting Approach for New 
Sources? In the September 1999 rule, we had amended language in 40 CFR 
264.340, 265.340, 266.100, 270.19, 270.22, 270.62, and 270.66 to 
accommodate the permit transition from RCRA to the CAA. To summarize, 
the amended language in these sections says that once a source 
demonstrates compliance with the standards in 40 CFR part 63 subpart 
EEE, the requirements in specified part 264, 265, 266, and part 270 
sections would no longer apply. However, the amended language neglected 
to specifically address if, how, or when new sources would make the 
transition from RCRA permitting requirements to CAA MACT requirements.
    As we discussed in the preamble to the July 3, 2001, proposed 
amendments, under RCRA, new sources must obtain a permit or a permit 
modification before they may start construction of a new source/unit. 
The way the current part 270 language reads, new sources subject to the 
1999 rule and the Interim Standards rule are not able to demonstrate 
compliance with the part 63 standards until after a RCRA permit is 
issued, the source is built, and they conduct performance testing. This 
means they would have to submit a trial burn plan with their RCRA 
permit application and also submit suggested conditions for the various 
phases of operation--start-up/shake-down, trial burn, and post-trial 
burn. Likewise, RCRA permitted facilities that are adding a new 
combustion source would have to provide the same information with their 
permit modification request. Whether the source is new or adding a new 
combustion source, the permit writer would have to review this 
information and write conditions into the RCRA permit governing all 
phases of combustor operations. This expenditure of resources, on the 
part of the source and the permitting agency, is unnecessary given that 
the conditions will become inactive or be removed from the RCRA permit 
upon compliance with the MACT standards. For new sources, compliance 
with the MACT standards is upon start-up. Therefore, today we are 
proposing that new sources (whether a new source or a new source at an 
existing permitted source) who will be subject to the Phase I 
Replacement standards and Phase II standards upon start-up, not follow 
the RCRA permitting process for establishing combustor emissions and

[[Page 21318]]

operating requirements (i.e., submission of a trial burn plan with the 
RCRA permit application, submission of suggested conditions for the 
various phases of operation--start-up/shake-down, trial burn, and post-
trial burn, and ultimately obtaining a permit with operating and 
emission standards).
    b. How Is EPA Proposing to Change the Current Requirements for New 
Sources? In the July 3, 2001 proposal, we developed regulatory language 
to clarify our intent not to require new sources to obtain a RCRA 
permit with respect to combustor operations and emissions. In response 
to that proposal, we received comments from the Sierra Club expressing 
concerns that the increased opportunities for public participation 
established in the RCRA Expanded Public Participation Rule (60 FR 
63417, December 11, 1995) would be lost. This rule involves communities 
earlier in the permitting process, provides more opportunities for 
participation, expands public access to information, and offers 
guidance on how facilities can improve public participation. In a 
follow-up discussion with the Sierra Club, they specifically expressed 
interest in being able to influence decisions on the construction of 
hazardous waste combustors. Upon consideration, we agree with the 
Sierra Club that in our previous effort to streamline the RCRA 
permitting process for new sources, we did not fully consider that 
important opportunities for public participation may be lost. Although 
we still believe that new sources, whether a new source or an existing 
source adding a new source, should not be required to follow the RCRA 
permitting process, we also believe that the Sierra Club's concerns 
have merit. It makes sense to afford the public the same (or as close 
as possible) public participation opportunities for new units under the 
HWC MACT/CAA framework that they had under the RCRA regulations. 
Therefore we are modifying our earlier proposal as discussed in the 
paragraphs below, to consider several options that will attempt to 
address these concerns, as well as provide a means to improve the 
existing regulatory requirements for new sources.
    The RCRA Expanded Public Participation Rule implemented four new 
requirements for facilities and permitting agencies that enable 
communities to become more active participants throughout the 
permitting process. They are: (1) Permit applicants must hold an 
informal public meeting before applying for a permit; (2) permitting 
agencies must announce the submission of a permit application which 
will tell community members where they can view the application while 
the agency reviews it; (3) permitting agencies may require a facility 
to set up an information repository at any point during the permitting 
process if warranted; and (4) permitting agencies must notify the 
public prior to a trial (or test) burn. Consequently, we will focus on 
each of these and propose mechanisms that mirror or fulfill the RCRA 
public participation requirements.
    We stated earlier in this section that under RCRA, new sources must 
obtain a permit (or a permit modification at an existing source) before 
they may start construction of a new source. This holds true regardless 
of whether we finalize an approach that does not require new sources to 
obtain a RCRA permit that contains the combustor operating and 
emissions standards (i.e., a RCRA permit will still be required to 
address all other activities at the facility including corrective 
action, general facility standards, other combustor specific concerns 
such as material handling, risk-based emission limits and operating 
requirements, and other hazardous waste management units). So, in 
applying for a RCRA permit, new hazardous waste facilities/sources will 
still be required to meet the public participation requirements. 
However, the problem arises if new sources are not required to provide 
information relative to the combustor (i.e., sources were formerly, at 
this point in the process, required to submit a trial burn plan), but 
only for the other proposed hazardous waste management activities at 
the source. Thus, the source would not be required to discuss the 
proposed combustor-specific operations and emissions at the informal 
public meeting, nor would the permit application that is made available 
to the public to review, contain information regarding the combustor 
operations or emissions.
    In an effort to provide an opportunity for public participation 
equivalent to RCRA, we believe that the Notification of Intent to 
Comply (NIC) requirements, as proposed in Part Two, Section XVI.B., 
serve in place of the first two RCRA public participation requirements. 
The primary functions of the NIC are to serve as a compliance planning 
tool and to promote early public involvement in the permitting process. 
In terms of compliance planning, the draft NIC must contain general 
information including the waste minimization, emission control, and 
emission monitoring techniques that are being considered and how the 
source intends to comply with the emission standards. With regard to 
early public involvement, a draft of the NIC must be made available to 
the public for review within 9 months of the effective date of the 
final Replacement Standards and Phase II Standards rule. One month 
later, the source must hold an informal public meeting to discuss the 
activities described in the NIC. The NIC requirements apply to new 
sources as well (see Sec.  63.1212(b)(1) in today's Notice), but the 
timing will vary according to the date a new source begins burning 
hazardous waste. For example, if a new source begins burning 3 months 
after the rule's effective date, then it will have only 6 months before 
it must prepare and make a draft NIC available for public review.\216\ 
More significantly, according to 40 CFR 63.1212(b)(2), as proposed in 
today's Notice, new sources that are to begin burning more than 9 
months after the effective date of the final rule will be required to 
meet all of the NIC and Compliance progress report requirements in 
Sec. Sec.  63.1210(b) and (c), 63.1211(c), and 63.1212(a) prior to 
burning hazardous waste.
---------------------------------------------------------------------------

    \216\ Note that new sources must have prepared and included 
their documentation of compliance in the operating record upon 
start-up. New sources then have 6 months from the date of start-up 
to begin their comprehensive performance test.
---------------------------------------------------------------------------

    We feel that the NIC requirements are commensurate with the public 
participation requirements to hold an informal public meeting to inform 
the community of the proposed combustor operations and to make the 
compliance information available for public review and comment. On the 
other hand, we also recognize that there are a few gaps. For instance, 
the NIC requirements are not associated with a permit action and the 
regulatory agency is not required to be present at the NIC public 
meeting. We would, however, expect the source to consider any comments 
raised during the NIC process as it develops its final compliance 
strategy and final NIC.\217\ Also, if a new source begins burning after 
the effective date of today's rule, but prior to 9 months after the 
effective date, the NIC is not required to be made available for public 
review before a new source begins burning. In other words, the public 
is not provided information relative to the combustor's operations, 
emissions, and compliance schedule prior to it beginning operations. 
Given these gaps, we are proposing a scenario in which the NIC 
requirements for new sources under MACT, could be crafted

[[Page 21319]]

to achieve a comparable level of public participation as under RCRA.
---------------------------------------------------------------------------

    \217\ If necessary, concerns raised regarding the regulation of 
the combustor can be addressed through application of RCRA's omnibus 
provision (RCRA section 3005(c)(3)).
---------------------------------------------------------------------------

    We are proposing to require that all new sources prepare a draft 
NIC and make it available to the public at the same time as their RCRA 
pre-application meeting notice. We also propose that new sources submit 
their comprehensive performance test plan at this time. By submitting 
the NIC and CPT plan together, the public would be provided with 
compliance-related information relevant to the combustor as well as the 
proposed combustor operations and emissions (i.e., the public is 
provided testing information through the CPT that they would have 
received via the trial burn plan). Lastly, as part of this option we 
propose that the NIC public meeting coincide with the informal public 
meeting for the RCRA permit. By holding a simultaneous meeting, the 
public is given the opportunity to inquire and comment on both the 
source's proposed activities and the combustor's proposed operations 
with regulatory officials from both the Air and RCRA programs present. 
We request comment on this discussion.\218\
    With respect to the information repository regulations at 40 CFR 
124.33, the purpose of the information repository is to make 
information (i.e., documents, reports, data, and information deemed 
necessary) available to the public during the permit issuance process 
and during the life of a permit. While the Title V permit procedures 
specify that information relevant to the permitting decision be made 
available to the public,\219\ this information would not be accessible 
prior to construction or operation of the combustor. Under RCRA, the 
information repository would be established some time after submission 
of the permit application, but before construction and operation of the 
combustor. Even though an information repository is not a required 
component of the RCRA permit process, the regulations provide a 
permitting agency with the discretion to evaluate the need for and 
require a source to establish and maintain one. Therefore, so that the 
public is afforded the same opportunities to view and copy information 
such as the NIC, test plans, draft Title V permit and application, 
reports and so forth under MACT, we are considering two options. We 
could include a provision similar to Sec.  124.33 in the NIC 
regulations for new sources. It would allow a regulatory agency, on a 
case-by-case basis, to require a source to establish an information 
repository specific to the combustor. We believe the NIC regulations 
are a suitable location to place such a provision, since the NIC is the 
first opportunity for the public to discuss the combustor operations 
and emissions. Alternatively, rather than incorporate provisions for an 
information repository in the NIC regulations, the applicability 
language in Sec.  124.33 could be amended to include new combustion 
sources that will comply with Part 63, subpart EEE upon start-up. We 
request comment on this discussion.
---------------------------------------------------------------------------

    \218\ Since the public participation requirements of 40 CFR 
124.31 and 124.32 only apply to initial RCRA permits and renewals 
with significant changes, a corresponding regulatory amendment would 
need to be made to the applicability paragraphs to include 
modifications to RCRA permits only for new combustion sources that 
will comply with Part 63, subpart EEE upon start-up. Also, 
63.1212(b) would need to be amended to reference Sec. Sec.  124.31 
and 124.32.
    \219\ 40 CFR Sec.  70.7(h)(2) requires that information 
including the draft Title V permit, the application, all relevant 
supporting materials, and other materials available to the 
permitting authority that are relevant to the permit decision, be 
made available to interested persons.
---------------------------------------------------------------------------

    The last RCRA public participation requirement requires the 
permitting agency to notify the public prior to a trial burn or test 
burn at a combustion facility. If new sources are not required to 
follow the RCRA permitting process with respect to combustor emissions 
and operations, they also would not be required to submit a trial burn 
plan with their permit application or conduct a trial burn. However, 
under MACT, new (and existing) combustion sources are required to 
submit performance test and continuous monitoring system (CMS) 
performance evaluation test plans for approval. The MACT performance 
test serves the same purpose as the RCRA trial burn test: To 
demonstrate compliance with the relevant emission standards and to 
collect data to determine at what levels the corresponding operating 
conditions should be set. Similar, but not identical to the RCRA 
requirements at 40 CFR 270.62 and 270.66 requiring the permitting 
agency to notify the public prior to a trial/test burn, the MACT 
performance test regulations (see Sec.  63.1207(e)(2)), specify that a 
source must issue a public notice announcing the approval of the test 
plans and provide a location where the public may view them. Although 
the timing of the public notices are slightly different, the 
regulations both provide notice to the public about testing. Under 
RCRA, notice is given to the public prior (usually 30 days) to 
commencement of the trial burn, whereas under MACT, notice is given 
when the test plans are approved. The newly amended regulations of 
Sec.  63.1207(e)(2) proposed in this Notice, specify that sources must 
make the test plans available for review at least 60 days prior to 
commencement of the test and must provide the expected time period for 
commencing (and completing) the test. Thus, the public is informed of 
the test and provided estimates of test dates through public notice of 
the approved test plan.
    Thus far, the approach we have proposed is intended to ensure that 
the public will have the same opportunities for participation and 
access to information as they would if new sources continued to be 
subject to the RCRA permit process to include the combustor emission 
and operating requirements. By proposing that new sources not be 
required to obtain a RCRA permit with combustor emission and operating 
requirements, it provides for the smoothest and most practical 
transition from RCRA requirements to MACT requirements.\220\
---------------------------------------------------------------------------

    \220\ This approach does not eliminate the possibility that some 
combustor-specific requirements may be retained in the RCRA permit 
such as: Risk-based conditions, compliance with an alternative MACT 
standard, compliance with startup, shutdown and malfunction events 
under RCRA rather than the CAA, etc. See section XVII, D.2. for a 
more complete discussion. Consequently, sources would be expected to 
include the applicable RCRA conditions in their RCRA permit 
application.
---------------------------------------------------------------------------

    Aside from the approach we have focused on, there are others that 
may be worthy of consideration. We can also look at the option of a 
transition point for new sources that would specify how far a new 
source would proceed down the RCRA permit path before it could 
``transition'' over to compliance with the MACT standards and CAA 
permitting. There are three additional options we can consider relative 
to a transition point: (1) After the RCRA Part B application is 
submitted; (2) after the RCRA permit is issued; and (3) after the 
source places its Documentation of Compliance (DOC) in the operating 
record.
    Beginning with the first option, each successive one moves in the 
direction toward the way new sources currently make the transition from 
RCRA to MACT and includes modifications to the RCRA information 
requirements. We envision each of these options to be a variation of 
the current RCRA permit process. Under the first option, the transition 
point would occur after the source submits its RCRA Part B application. 
The key to this option is that the source would be subject to the 
public participation requirements of 40 CFR 124.31 and 124.32, to hold 
an informal public meeting and to have the submission of the permit 
application noticed. However, new sources would

[[Page 21320]]

not be required to include the combustor's operation and emission 
information in the Part B application. Rather, the source would only be 
required to discuss the compliance-related activities related to the 
combustor as part of the informal public meeting. For the second 
option, the transition point would be after the permitting agency 
issues the RCRA permit. The source would not only discuss the 
combustor's compliance-related activities as part of the RCRA informal 
public meeting as in the first option, but it would also address the 
operations and emissions through development of a trial burn plan, or a 
CPT plan in lieu of the trial burn plan, or even a coordinated CPT/RCRA 
trial burn plan, if it is likely that the source will require some RCRA 
permit conditions (i.e., risk-based conditions). With this option, even 
though all activities pre-permit issuance must address the source and 
the combustor's operations and emissions, the approved permit would not 
contain the operating and emission requirements (with the exception of 
risk-based or alternative standards). For the third option, the 
transition point would be after the source places its DOC in the 
operating record, which indicates the source's compliance with the MACT 
standards. Basically, the source would proceed down the RCRA permit 
path as in option two by complying with the public participation 
requirements, submitting a trial burn plan/CPT plan/coordinated plan, 
suggesting conditions for the various phases of operation, and 
receiving a RCRA permit. However, in this option, the permit would need 
to address combustor operations and emissions to the extent that it 
would cover the construction and start-up/shakedown periods.
    With respect to the public participation requirements, all three 
options automatically factor in the first two RCRA public participation 
requirements (by virtue of where the transition would be made). 
However, we did not discuss how we would account for the remaining two 
public participation requirements. We believe that the information 
repository and the notification of a trial burn requirements can be 
addressed in the same manner as we discussed in our proposed approach. 
So, for these options, we would incorporate an appropriate requirement, 
either through the NIC regulations or the public participation 
regulations, that would allow for an information repository to be 
established. Regarding the notice of a trial burn, we believe that the 
notice of the performance test is equivalent.
    In summary, our proposed approach involves modifying the NIC 
provisions to include RCRA public participation requirements. The 
second group of options consider a range of transition points that are 
also worthy of consideration. We invite comment on this discussion.
3. What Are the Proposed Changes to the RCRA Permitting Requirements 
That Will Facilitate the Transition to MACT?
    To alleviate potential conflicts between the RCRA permit 
requirements and MACT, we are proposing an additional streamlined 
permit modification provision, requiring prior Agency approval, which 
would allow an existing RCRA permit to be better aligned with specific 
provisions contained in the Subpart EEE requirements. The intent of 
this provision is to reduce potential burdens associated with 
compliance with overlapping RCRA and MACT requirements, while still 
maintaining the overall integrity of the RCRA permit.
    a. How Will the Overlap During Performance Testing Be Addressed? 
When we finalized the performance test requirements and the changes to 
the RCRA permitting requirements in the September 30, 1999, rule, we 
did not consider how sources would conduct their performance tests 
while at the same time, maintain compliance with their RCRA permit 
requirements. For instance, during the performance test, a source will 
likely want to conduct testing at the edge of the operating envelope or 
the worst case for certain parameters to ensure operating flexibility. 
This could conflict with established operating and emissions limits 
required in the source's RCRA permit and consequently, prevent the 
source from optimizing its testing range.
    Currently, sources have three options that would allow them to 
resolve any potential conflicts between their performance test and 
their RCRA permit requirements. One option would be for a source to 
submit a RCRA Class 2 or 3 permit modification request to temporarily 
change or waive specific RCRA permit requirements during the MACT 
performance test (see Sec.  270.42, appendix I, L.5). Another option 
would be for a source to request approval for such changes through its 
RCRA trial burn plan or coordinated MACT / RCRA test plan (see Sec.  
270.42, appendix I, L.7.a. or d.). In this case, a source could include 
proposed test conditions in its plan to temporarily waive specific RCRA 
permit requirements during the test. The last option would be for a 
source to request a temporary authorization that would allow specific 
RCRA permit requirements to be waived for a period of 180 days (see 
Sec.  270.42(e)).
    We do not believe that any of the options discussed above provide 
an optimal solution to resolving conflicts between a source's 
performance test protocol and its RCRA permit operating and emissions 
limits. A Class 2 or 3 RCRA permit modification may not be an option 
for many sources due to the time typically involved in processing these 
requests. Sources that choose to modify their permits would need to do 
so well in advance of conducting their performance test to ensure that 
the modification would be processed in time to conduct the test on 
schedule. This may result in sources submitting modification requests 
prior to approval of their performance test plans. We believe that RCRA 
permit writers are unlikely to approve any modifications to RCRA permit 
requirements without the assurance that the source will be operating 
under an approved test plan. Resolving conflicts using a trial burn or 
coordinated test plan is not a viable option for a source that has 
already completed its trial burn/risk burn testing. Lastly, while a 
temporary authorization is relatively streamlined, it is meant to be 
used in unique cases affecting an individual facility. We believe that 
it is most logical and easily implemented to propose a modification 
that can be used consistently to remedy a common problem affecting an 
entire group of facilities with similar operations (e.g., hazardous 
waste burning combustors facing barriers to testing due to RCRA permit 
requirements). Therefore, in today's Notice, we are proposing to allow 
sources to waive specific RCRA permit operating and emissions limits 
during pretesting, initial, and subsequent performance testing through 
a new streamlined permit modification procedure.\221\
---------------------------------------------------------------------------

    \221\ For subsequent performance tests, we anticipate that this 
modificaiton would be useful for sources that may have risk-based or 
alternative requirements in their RCRA permits.
---------------------------------------------------------------------------

    We believe that a process for waiving specific RCRA permit 
requirements during performance testing is consistent with our 
objectives to streamline requirements and minimize conflicts between 
the RCRA and CAA programs without sacrificing the protections afforded 
by RCRA. Moreover, we view this new permit modification to be 
complementary to the provisions of Sec.  63.1207(h) for waiving 
operating parameter limits (OPLs) during

[[Page 21321]]

performance testing. In the February 14, 2002 final amendments rule, we 
reiterated that OPLs in the Documentation of Compliance (DOC) may be 
revised at any time to reflect testing parameters for the initial 
performance test prior to submission of the NOC and so, in effect, are 
automatically waived. Also, we revised the language in Sec.  
63.1207(h)(1) and (2) to not require that subsequent performance test 
plans be approved in order to waive OPLs, but rather that sources only 
record the emission test results of the pretesting.
    b. Are There Other Instances Where the New Streamlined Permit 
Modification Can Be Used? In addition to our efforts today to minimize 
overlapping permit requirements during performance testing, we are also 
proposing to allow the new streamlined permit modification to address 
other potential conflicts. In implementing the 1999 rule, it has become 
clear that there are several other instances when conflicts may arise 
where RCRA permit requirements overlap with MACT requirements. For 
example, the required averaging period for an operating parameter might 
be slightly different between MACT and the RCRA permit, requiring two 
different data acquisition schemes during the interim period between 
submittal of the Documentation of Compliance (DOC) and the final 
modification of the RCRA permit after receipt of the NOC. Or, if a RCRA 
permit requires periodic emissions testing, the specified test schedule 
in the permit might not be aligned with the required test schedule for 
MACT, causing a facility to perform duplicate testing instead of 
allowing a single coordinated RCRA/MACT test event. Conflicts in 
operating limitations, monitoring and recordkeeping requirements, and 
scheduling provisions can be especially prevalent during this interim 
period. Consequently, we believe the new streamlined permit 
modification procedure would be appropriate to address these probable 
overlaps.
    c. Why Is a New Streamlined Permit Modification Procedure Being 
Proposed? This new streamlined modification differs from the one we 
finalized in the June 1998 ``fast track'' rule (63 FR 33782). In 1998, 
we provided for a streamlined RCRA permit modification process whereby 
you could request a Class 1 modification with prior Agency approval to 
address and incorporate any necessary MACT upgrades into your RCRA 
permit (see 40 CFR 270.42, appendix I, L(9)). The streamlined permit 
modification provision, which was intended solely for the purpose of 
implementing physical or operating upgrades, allowed sources that were 
already operating under RCRA combustion permits to modify their 
combustion systems' design and/or operations in order to comply with 
the MACT standards without having to obtain a Class 2 or 3 RCRA permit 
modification. Thus, L(9) was not intended to account for overlapping 
requirements. Further, to be eligible to use L(9), you first must have 
complied with the NIC requirements, including those related to public 
involvement. Refer to Part Two, Section XVI for a discussion of the 
NIC.
    However, similar to the streamlined modification we finalized as 
L(9), we feel that this new streamlined modification warrants a Class 1 
modification with prior Agency approval. We feel that a Class 1 is 
appropriate considering that: we do not expect that there would be 
significant changes when requesting certain RCRA permit requirements to 
be waived; it would be applicable for a relatively short period of 
time; regulatory oversight is incorporated via approval of the 
modification request and; the intended goal of the modification is to 
achieve environmental improvement ultimately through implementation of 
more protective standards.
    d. How Will the New Streamlined Permit Modification Work? Our 
proposed approach allows for a waiver of specific RCRA permit 
requirements provided that you: (1) Submit a Class 1 permit 
modification request specifying the requested changes to the RCRA 
permit, with an accompanying explanation of why the changes are 
necessary and how the revised provisions will be sufficiently 
protective, and (2) obtain Agency approval prior to implementing the 
changes.\222\ When utilized to waive permit requirements during the 
performance test, you also must have an approved performance test plan 
prior to submitting your modification request. (We believe that the 
Class 1 modification with prior Agency approval will ensure that your 
proposed test conditions are reasonable with respect to your existing 
permit limits (i.e. that they are sufficiently protective); and that an 
approved performance test plan confirms that you have met the 
regulatory requirements for performance test plans.)
---------------------------------------------------------------------------

    \222\ Refer to the new section in the RCRA permit modification 
table in 40 CFR 270.42, appendix I, L(10) and new regulatory 
language in 270.42(k), that must be used to waive specified permit 
requirements.
---------------------------------------------------------------------------

    We propose that you submit your streamlined modification request in 
sufficient time to allow the Director a minimum of 30 days (with the 
option to extend the deadline for another 30 days) to review and 
approve your request. For purposes of performance testing, we propose 
that you submit your request at the time you receive approval of your 
performance test plan, which is 90 days in advance of the test and 
coincides with the time limitations imposed on the Director for 
approval. Additionally, we are requiring that the waiver of permit 
limits only be relevant during the actual testing events and during 
pretesting for an aggregate period of up to 720 hours of operation. In 
other words, it would not apply for the duration of time allotted to 
begin and complete the test (i.e., the entire 60 days).
    As a side note, we realize that some sources may not have an 
approved performance test plan by the date their test is scheduled to 
begin because the Administrator failed to approve (or deny) it within 
the specified time period, which could render this new streamlined 
modification impractical. However, we expect that sources would 
petition the Administrator to waive their performance test date for up 
to 6 months, with an additional 6 months possible, rather than to 
proceed with the performance test without the surety of an approved 
test plan.\223\
---------------------------------------------------------------------------

    \223\ See 40 CFR 63.1207(e)(3) for performance test time 
extension requirements.
---------------------------------------------------------------------------

B. How Will the Replacement Standards Affect Permitting for Phase I 
Sources?

1. Where Will Phase I Sources Be in Their Transition to MACT With 
Respect to Their RCRA Permits?
    We discussed earlier that by the time the Phase I Replacement 
standards and Phase II standards are finalized, most Phase I sources 
will have completed their initial comprehensive performance test and 
submitted their NOC documenting compliance with the MACT Interim 
Standards.\224\ This marks the point at which sources will begin to 
transition from RCRA permitting requirements to CAA requirements and 
title V permitting. For sources with RCRA permits, they must continue 
to comply with the operating standards and emission limits in their 
permits until any duplicative requirements are either removed through a 
permit modification, expire, or are automatically inactivated via a 
sunset clause contained in the permit. For sources operating under 
interim status,

[[Page 21322]]

they must comply with the RCRA interim status requirements until they 
demonstrate and document compliance with the MACT Interim Standards. We 
anticipate that sources who are in the process of renewing their RCRA 
permits would work with their permit writers to include sunset clauses 
to inactivate duplicative requirements upon compliance with the MACT 
Interim Standards. Given the permit actions taken during the transition 
period leading up to compliance with the Interim Standards, we believe 
that many sources will have had duplicative requirements removed from 
their permits by the time the Replacement Standards are promulgated. 
For sources that have not had their RCRA permits modified, we expect 
that they will proceed with a modification to remove duplicative 
requirements.\225\
---------------------------------------------------------------------------

    \224\ Some sources will receive extensions of up to one year to 
conduct their initial comprehensive performance test (see 40 CFR 
63.1207(e)(3)). Therefore, their transition point will occur at a 
later time designated by the extension.
    \225\ A streamlined permit modification was developed in the 
1999 rule to allow the removal of duplicative conditions from RCRA 
permits (see Sec.  270.42, appendix I, section A.8).
---------------------------------------------------------------------------

2. Where Will Phase I Sources Be in Their Transition to MACT With 
Respect to Their Title V Permits?
    With regard to title V permits, Phase I major and area sources were 
required to submit a title V permit application 12 months after the 
effective date of the 1999 rule--or were required to reopen existing 
title V permits with 3 or more years remaining in the permit term, 18 
months after the effective date--to include the MACT standards. Sources 
with less than 3 years remaining could wait until renewal to 
incorporate the 1999 standards.\226\ Upon promulgation of the Interim 
Standards on February 13, 2002, major sources were required to reopen 
their permits or could wait until renewal to include the revised 
standards according to the same time frames mentioned above. Therefore, 
we expect that all Phase I sources would have title V permits 
containing the MACT Interim Standards and potentially, operating 
standards in accordance with their DOC, at the time the Replacement 
Standards rule is promulgated. Furthermore, most sources will have 
initiated a significant modification to their permits to include the 
revised operating requirements of their NOC. Regardless of these 
required compliance activities leading up to the promulgation date of 
the Replacement Standards rule, Phase I sources will again need to 
reopen within 18 months or wait until renewal to incorporate the MACT 
Replacement standards.
---------------------------------------------------------------------------

    \226\ Only major sources are required to reopen their title V 
permits when 3 or more years remain in the permit term. Even though 
area sources were subject to the same standards and title V permit 
requirements, they can wait until renewal regardless of the time 
remaining to incorporate new or revised standards. The reopening 
provisions of 40 CFR 70.7(f) and 71.7(f) only apply to major 
sources.
---------------------------------------------------------------------------

3. What Is Different With Respect To Permitting in Today's Notice of 
Proposed Rulemaking?
    Based upon our decision to utilize the same general permitting 
approach as in the 1999 and Interim Standards rules, we expect sources 
to follow the same transition scheme as it relates to RCRA permit 
requirements and the CAA requirements and title V permitting for the 
Replacement Standards rule. One aspect, however, that was not addressed 
in those rules was how the permitting of new sources would be affected. 
Hence, we discuss approaches in this Notice of Proposed Rulemaking (see 
Section A.1. above) that would require them to obtain RCRA permits only 
for corrective action, general facility standards, other combustor 
specific concerns such as material handling, risk-based emission limits 
and operating requirements, and other hazardous waste management units 
at the source. Should the approach we are proposing be finalized, there 
may not be any operating requirements and emission standards to remove 
from their RCRA permits.
    We also discussed a new streamline permit modification procedure in 
section A.2. ``What Are the Proposed Changes to the RCRA Permitting 
Requirements that Will Facilitate the Transition to MACT?''. This new 
procedure allows sources to waive specific RCRA permit operating and 
emission limits during pretesting, performance testing, and other 
instances where there may be conflicts during the interim period 
between submission of the Documentation of Compliance and final RCRA 
permit modification.
    Another important difference is our proposal to codify the 
authority for permit writers to evaluate the need for and, where 
appropriate, require Site-Specific Risk Assessments (SSRA). We are also 
proposing to codify the authority for permit writers to add conditions 
to RCRA permits that they determine, based on the results of an SSRA, 
are necessary to protect human health and the environment. In doing so, 
our intent is to change the regulatory mechanism that is the basis for 
SSRAs, while retaining the same SSRA policy from a substantive 
standpoint. Under this approach, permitting authorities continue to 
have the responsibility to ensure the protectiveness of RCRA permits.
    Next, we have proposed to re-institute the NIC (see Part Two, 
Section XVI for a discussion of the NIC) for Phase I sources and to 
require the NIC for Phase II sources. While the NIC serves as a 
compliance planning tool and to promote early public involvement, it is 
also a requirement before the streamlined permit modification procedure 
in 40 CFR 270.42(j) and 270.42, appendix I, section L.9, can be 
utilized to make changes to either the combustor design or operations, 
in order to comply with the final Replacement Standards. Thus, sources 
who have not yet made the transition from their RCRA permits to title V 
permits must comply with the NIC requirements to take advantage of the 
streamlined permit modification.
    Last, a subtle difference pertaining to the transition scheme stems 
from the time span between compliance with the Interim Standards and 
the effective date of the Replacement Standards relative to RCRA 
permits. Sources who received extensions to the date for commencing 
their initial comprehensive performance test, whether a 6 month or 12 
month extension, will not be required to submit an NOC until either a 
few months before or just after the effective date of the final 
Replacement Standards rule. Therefore, these sources would be modifying 
their RCRA permits just before or after the effective date of the final 
rule. Nevertheless, we anticipate that sources will proceed with 
modification of their RCRA permits to remove duplicative requirements.

C. What Permitting Requirements Is EPA Proposing for Phase II Sources?

    Phase II sources are presently subject to the RCRA permitting 
requirements for hazardous waste combustors provided in 40 CFR 270.22 
and 270.66. We are proposing in today's notice to apply the same 
approach to permitting Phase II sources that we did for Phase I sources 
in the September 1999 rule. Specifically, we propose to:
    (1) Place the new Phase II emission standards only in the CAA 
regulations at 40 CFR part 63, subpart EEE, and rely on their 
implementation through the air program,
    (2) Specify that, with few exceptions, the analogous standards in 
the RCRA regulations no longer apply once a facility demonstrates 
compliance with the MACT standards in subpart EEE, and
    (3) Require that the new standards be incorporated into operating 
permits issued under title V of the CAA rather than be incorporated 
into RCRA permits.
    Our goal with regard to permitting Phase II sources remains the 
same as the goal that we had for Phase I sources--to accommodate the 
requirements of

[[Page 21323]]

both the RCRA and CAA statutes, while at the same time avoiding 
duplication between the two programs to the extent practicable. The 
permitting approach we developed for Phase I sources in the September 
1999 rule enables us to achieve this goal. In that rule, we amended the 
applicability of 40 CFR 270.19, 270.22, 270.62, and 270.66 so that once 
a source demonstrates compliance with the MACT standards, it is no 
longer subject to the full array of RCRA combustion permitting 
activities, unless the Director of the permitting agency decides to 
apply specific RCRA regulatory provisions, on a case-by-case basis, for 
purposes of information collection in accordance with Sec. Sec.  
270.10(k) and 270.32(b)(2). We are proposing to make a similar change 
to 40 CFR 270.22 and 270.66 for Phase II sources. In addition, we are 
proposing for Phase II sources, as we are for Phase I sources, that new 
sources not follow the RCRA permitting process for establishing 
combustor emissions and operating requirements. Of course, as for Phase 
I sources, Phase II sources would remain subject to the RCRA permitting 
requirements for all other aspects of their combustion unit and 
facility operations, including general facility standards, corrective 
action, other combustor-specific concerns such as materials handling, 
risk-based emission limits and operating requirements, as appropriate, 
and other hazardous waste management units at the site.\227\ Also, some 
sources will retain specific RCRA permitting requirements if they 
choose to comply with an alternative MACT standard; address startup, 
shutdown and malfunction events under RCRA rather than the CAA; or, if 
an area source, comply with the RCRA metals, particulate matter, or 
chlorine standards and associated requirements. It is also important to 
note that if you later decide to add a new combustion unit to your 
facility, you must first modify your RCRA permit to include the new 
unit. This is because your RCRA permit must reflect all hazardous waste 
management units at the facility. Although the emissions from the new 
unit will be regulated under the CAA MACT standards, as noted above, 
your RCRA permit must address any other related requirements for the 
new unit.
---------------------------------------------------------------------------

    \227\ Even though the RCRA air emission standards for combustors 
will no longer apply once compliance is demonstrated with MACT 
(except in certain cases), other RCRA air emission standards will 
continue to apply to other hazardous waste management units at the 
facility. For example, part 264, subpart CC, still applies to air 
emissions from tanks, surface impoundments, and containers.
---------------------------------------------------------------------------

1. What Other Permitting Requirements Are We Proposing To Apply To 
Phase II Sources?
    As part of the Phase I rule, we promulgated additional specific 
changes to the RCRA permitting requirements in 40 CFR part 270 to 
facilitate implementation of the new standards and permit transition 
from RCRA to the CAA. First, we added a streamlined RCRA permit 
modification process to allow sources to make changes to either their 
combustor design or operations, as necessary, in order to comply with 
the Phase I standards. This modification process, a Class 1 with prior 
Agency approval, was promulgated in the June 19, 1998 ``Fast Track'' 
rule and is provided in 40 CFR 270.42(j) and 270.42, appendix I, 
section L.9. See 63 FR 33785. Second, we further amended the Sec.  
270.42, appendix I permit modification table to add a new line item 
that streamlines modification procedures for removing conditions from a 
permit that are no longer applicable (e.g., because the standards upon 
which they are based are no longer applicable to the source). This new 
line item is a Class 1 modification requiring prior Agency approval and 
is provided in section A.8 of appendix I.\228\ Third, we added a new 
section, 40 CFR 270.235, to the RCRA permitting requirements that 
address startup, shutdown, and malfunction events and the integration 
of those requirements between the RCRA program and the CAA program. 
Fourth, we amended the requirements in 40 CFR 270.72 governing changes 
that facilities can make while they are operating under interim 
status.\229\ We believe that each of the above changes that we made to 
the RCRA permitting regulations for Phase I sources are also 
appropriate for Phase II sources and thus, are proposing that these 
same features apply to Phase II sources. They will serve to ease 
implementation of the new standards and transition combustion sources 
from RCRA to the CAA.
---------------------------------------------------------------------------

    \228\ It is important to note that you only may request the 
removal of duplicative combustion limits and conditions from your 
RCRA permit. Any risk-based conditions that are more stringent than 
the MACT requirements would be retained.
    \229\ Section 270.72(b) imposes a limit on the extent of the 
changes, stating that they cannot amount to ``reconstruction'' 
(defined in the regulation as ``when the capital investment in the 
changes to the facility exceeds 50 percent of the capital cost of a 
comparable entirely new hazardous waste management facility''). 
Although we did not expect the individual costs to perform changes 
required to comply with the MACT standards to exceed this 50 percent 
limit, the limit is cumulative for all changes at an interim status 
facility. Thus, conceivably there could be situations where MACT-
related changes would cause a source to exceed the limit. To ensure 
that the limit would not be a hindrance to MACT compliance, we added 
an exemption to paragraph (b) of that section for changes necessary 
to comply with standards under 40 CFR part 63, subpart EEE.
---------------------------------------------------------------------------

    We did not amend any title V regulations in 40 CFR parts 70 or 71 
for Phase I sources. It was our intent during the Phase I rulemaking, 
and continues to be our intent for Phase II, to rely on the existing 
air program to implement the new MACT requirements, including their 
incorporation into a title V operating permit. Thus, we are proposing 
that all current CAA title V requirements governing permit 
applications, permit content, permit issuance, renewal, reopenings and 
revisions will apply to air emissions from Phase II sources. In 
addition, the requirements of other CAA permitting programs, such as 
air construction permits, likewise will continue to apply, as 
appropriate. We also included provisions in the subpart EEE 
requirements that address the relationship between the standards and 
title V permits. Specifically, we stated in 40 CFR 63.1206(c)(1)(iv) 
and (v) that the operating requirements in the Notification of 
Compliance are applicable requirements for purposes of parts 70 and 71, 
and that these operating requirements will be incorporated into title V 
permits. We are proposing the same approach for the interface between 
the Phase II standards and title V permits.
2. What Other Permitting Requirements Are We Proposing in Today's 
Notice That Would Also Be Applicable to Phase II Sources?
    In today's notice, we are proposing three changes to the general 
permitting approach for all sources subject to part 63, subpart EEE, 
including Phase II sources. First, we are proposing to allow sources to 
waive specific RCRA permit operating and emission limits using a 
streamlined permit modification procedure. This would apply for 
pretesting, performance testing, and other instances where there may be 
conflicts during the interim period between submittal of the DOC and 
final RCRA permit modification. Second, we are proposing that new units 
not be required to obtain a RCRA permit that includes emission limits 
or conditions, with certain exceptions (e.g., more stringent risk-based 
limits). Third, we are proposing to codify the authority for permit 
writers to evaluate the need for and, where appropriate, require SSRAs. 
We are also proposing to codify the authority for permit writers to add 
conditions to RCRA permits that they determine, based on the results of 
an SSRA, are necessary to protect human health and the environment. We 
believe

[[Page 21324]]

that each of the above proposals are appropriate for Phase II as well 
as Phase I sources and, therefore, are applying them to all hazardous 
waste combustors subject to part 63, subpart EEE. See the discussions 
provided in A.1 and A.2 of this section.
3. How Will the Permitting Approach Work for Phase II Sources?
    In the preamble to the September 1999 rule, we discussed at length 
how to implement the new permitting approach, including aspects such as 
when and how to transition sources from RCRA permitting to title V. See 
64 FR 52981. We have also provided a fact sheet on permit transition in 
our Hazardous Waste Combustion NESHAP Toolkit, which is available at 
the following Internet address: http://www.epa.gov/epaoswer/
hazwaste/combust/toolkit/index.htm. The information provided in the above-
mentioned preamble and the fact sheet is appropriate for Phase II as 
well as Phase I sources. Below is a summary of this information for 
sources that already have RCRA permits and for sources that are 
currently operating under RCRA Interim Status. The permitting approach 
for new sources is discussed earlier in A.1 of this section.
    a. Implementing the New Permitting Approach for Phase II Sources 
that Already Have RCRA Permits. If you already have a RCRA permit, you 
must continue to comply with the conditions in your permit until either 
they expire or your permitting authority modifies your permit to remove 
them. You can request a permit modification, using line item A.8 
provide in appendix I of Sec.  270.42, to request that your permitting 
authority remove any duplicative conditions once you have conducted 
your comprehensive performance test and submitted a Notification of 
Compliance documenting compliance to your CAA regulatory agency. The 
appropriate CAA regulatory agency in most cases will be the state 
environmental agency.
    When you submit your RCRA permit modification request you should 
identify the conditions in your RCRA permit that you believe should be 
removed. We recommend that you also attach a copy of your Notification 
of Compliance. This information will help the RCRA permit writer 
determine whether there are any risk-based conditions that need to 
remain in your RCRA permit. For example, any conditions imposed under 
RCRA omnibus authority, or similar state authority, based on the 
results of a site-specific risk assessment that are more stringent than 
the corresponding MACT standard or limitation documented in the 
Notification of Compliance would have to remain in the RCRA permit. You 
should also inform your RCRA permit writer if you intend to comply with 
any specific RCRA requirements in lieu of those provided in part 63, 
subpart EEE, such as the RCRA startup, shutdown, and malfunction 
requirements. Providing this information to the RCRA permit writer 
likely will expedite review of your permit modification request.
    We expect that in some situations RCRA permit writers may not 
approve a request to remove conditions until they know that their 
counterparts in the Air program have reviewed the Notification of 
Compliance and verified that the facility has successfully demonstrated 
compliance with the MACT standards. This may happen, for example, with 
facilities that have historically generated a lot of interest or 
concern from the community or that have had previous problems in 
maintaining compliance with performance standards. If you have received 
confirmation that the regulatory agency has made a Finding of 
Compliance based on your Notification of Compliance, we recommend you 
include that with your RCRA permit modification request as well. Once 
people in the Air program responsible for reviewing the Notification of 
Compliance have completed their evaluation of the documentation and 
test results, we encourage them to inform their RCRA counterparts. This 
courtesy will help RCRA permit writers complete their review of the 
RCRA permit modification requests, thereby facilitating the permit 
transition.
    b. Implementing the New Permitting Approach for Sources that Are 
Operating under RCRA Interim Status. If you are currently operating 
under RCRA interim status, you must continue to meet RCRA performance 
standards governing emissions of hazardous air pollutants in 40 CFR 
part 266 until you conduct your comprehensive performance test and 
submit your Notification of Compliance documenting compliance with the 
MACT standards to the regulatory agency. The RCRA combustion permitting 
procedures in 40 CFR part 270 also continue to apply until you 
demonstrate compliance.
    There is not a ``one size fits all'' answer to how facilities 
operating under RCRA interim status should make the transition. RCRA 
permit writers, in coordination with facility owners or operators, 
should map out the most appropriate route to follow in each case. In 
mapping out site-specific approaches to transition, both the regulators 
and the facility owners or operators should keep in mind the goal we 
mentioned earlier of minimizing the amount of time a facility might be 
subject to duplicative requirements under the two programs. Factors 
they should take into consideration include, but are not limited to the 
following. (1) The status of the facility in the RCRA permitting 
process at the time the final MACT rule is promulgated. For example--If 
a facility is on the verge of conducting a RCRA trial burn, it should 
proceed with the trial burn and continue through the RCRA permitting 
process. (2) The facility's anticipated schedule for demonstrating 
compliance with the MACT standards. For example--If the facility plans 
to come into compliance with the standards early, it may make sense to 
transition before completing the RCRA permitting process. (3) The 
priorities and schedule of the regulatory agency. For example--A state 
agency may have made certain commitments (e.g., to the public or to its 
state legislature) regarding their RCRA or CAA programs that might 
impact its decisions regarding the transition. (4) The level of 
environmental concern at a given site. For example--To make sure that 
the facility is being operated in a manner protective of human health 
and the environment, the regulatory agency may decide to proceed with 
RCRA permitting, including the site-specific risk assessment, rather 
than delay the RCRA process to coordinate with testing under MACT.
    If after evaluating all the relevant factors a decision is made to 
proceed with a RCRA permit in advance of a source's MACT compliance 
demonstration, we suggest including language to facilitate the eventual 
transition. Regulators can attach ``sunset'' provisions to those 
conditions that will no longer apply once a source demonstrates 
compliance with the part 63 subpart EEE standards.
    In making the transition from one program to the other, testing 
under one program should not be unnecessarily delayed in order to 
coordinate with testing required under the other. As proposed for Phase 
II, sources would be conducting periodic performance testing (every 
five years) anyway, just as the Phase I sources are required to do. In 
both our Hazardous Waste Minimization and Combustion Strategy and in 
the September 1999 Phase I rule, we emphasized the importance of 
bringing hazardous waste combustion units under enforceable controls 
that have been demonstrated to achieve compliance with performance 
standards. Stack testing is essentially

[[Page 21325]]

the way to make this demonstration, whether it is performed under the 
RCRA or CAA regulatory schemes, and so should be performed as 
expeditiously as possible.
4. How Do We Propose Regulating Phase II Area Sources?
    In today's Notice, we are not making a positive area source finding 
as we have with the Phase I area sources. However, we are using the 
``specific pollutants'' authority in section 112(c)(6) of the CAA to 
propose that area sources be subject to MACT standards only for certain 
hazardous air pollutants. Thus, area sources will be subject to title V 
permitting requirements for those pollutants specified per CAA section 
112(c)(6).
    Under 40 CFR 63.1(c)(2), area sources subject to MACT standards are 
also subject to title V permitting, unless the standards for the source 
category specifies that: (1) states will have the option to exclude 
area sources from title V permit requirements; or (2) states will have 
the option to defer permitting of area sources. We did not allow the 
states these options in the September 1999 rule for Phase I sources, 
and we are not proposing to offer them for Phase II sources either. 
Since the RCRA program does not make a distinction between regulating 
major and area sources and would no longer be able to address the 
pollutants covered by MACT (because the underlying RCRA standards in 40 
CFR parts 264, 265, and 266 would no longer be applicable once the 
source demonstrates compliance with subpart EEE), we believe that area 
sources should not be exempt from the title V permitting requirements. 
It is important that there not be a gap in permitting coverage as we 
implement the deferral from regulation under RCRA to regulation under 
the CAA. In addition, section 502(a) of the CAA requires that any area 
source exemptions from the title V permitting requirements be 
predicated on a finding that compliance with the requirements is 
impracticable, infeasible, or unnecessarily burdensome. We do not 
believe that the title V permitting requirements will be impracticable, 
infeasible, or unnecessarily burdensome for Phase II area sources, 
because these sources are already complying with RCRA permitting 
requirements.
    As explained above, we are using the ``specific pollutants'' 
authority to propose that area sources be subject to MACT standards 
only for certain hazardous air pollutants: dioxin/furans, mercury, DRE 
and carbon monoxide/hydrocarbons. (See Part Two, Section II.C.) For 
particulate matter, chlorine and HAP metals other than mercury, we are 
proposing that area sources have the option of complying with the MACT 
standards for Phase II major sources or continuing to comply with the 
RCRA emission standards and requirements. Those Phase II area sources 
that choose to comply with the RCRA standards and requirements will be 
subject to title V permits for some of their emissions and RCRA permits 
for others. In summary, regardless of whether an area source elects to 
comply with all or only the pollutants pursuant to CAA section 
112(c)(6), a title V permit will be required.

D. How Would this Proposal Affect the RCRA Site-Specific Risk 
Assessment Policy?

1. What Is the Site-Specific Risk Assessment Policy?
    In the September 30, 1999 Phase I rule, we articulated a revised 
Site-Specific Risk Assessment (SSRA) policy recommendation for 
hazardous waste burning incinerators, cement kilns and light-weight 
aggregate kilns. Specifically, we recommended that for hazardous waste 
combustors subject to the Phase I MACT standards, permitting 
authorities should evaluate the need for an SSRA on a case-by-case 
basis. We further stated that while SSRAs are not anticipated to be 
necessary for every facility, they should be conducted where there is 
some reason to believe that operation in accordance with the MACT 
standards alone may not be protective of human health and the 
environment. If the permitting authority concludes that a risk 
assessment is necessary for a particular combustor, the permitting 
authority must provide the factual and technical basis for its decision 
in the facility's administrative record. Should the SSRA demonstrate 
that supplemental requirements are needed to protect human health and 
the environment, additional conditions and limitations should be 
included in the facility's RCRA permit pursuant to the omnibus 
authority. The basis and supporting information for those supplemental 
requirements also must be documented in the facility's administrative 
record. For hazardous waste combustors not subject to the Phase I 
standards, we continued to recommend that SSRAs be conducted as part of 
the RCRA permitting process. See 64 FR 52841.\230\
---------------------------------------------------------------------------

    \230\ We provided further clarification of the appropriate use 
of the SSRA policy and technical guidance in an April 10, 2003 
memorandum from Marianne Lamont Horinko, Assistant Administrator for 
OSWER, to the EPA Regional Administrators titled Use of the Site-
Specific Risk Assessment Policy and Guidance for Hazardous Waste 
Combustion Facilities. This document is available in the docket 
(Docket # RCRA-2003-0016) established for today's proposed 
action.
---------------------------------------------------------------------------

2. Are SSRAs Likely To Be Necessary After Sources Comply With the Phase 
I Replacement Standards and Phase II Standards?
    As explained earlier, all Phase I replacement standards must be 
equivalent to or more stringent than the negotiated interim standards. 
Many of the replacement standards proposed in today's notice would be 
more stringent than the interim standards (e.g., 64 [mu]g/dscm as 
opposed to 120 [mu]g/dscm for the existing source cement kiln mercury 
standard). And, with the exception of the mercury standard for both new 
and existing LWAKs and the total chlorine standard for new LWAKs, they 
are also equivalent to or more stringent than the 1999-promulgated 
standards, which EPA determined to be generally protective in a 
national risk assessment conducted for that 
rulemaking.231, 232 For today's proposed action, we 
conducted a comparative risk analysis of the Phase I replacement 
standards to the 1999-promulgated Phase I standards. Specifically, we 
compared certain characteristics of the Phase I source universe as it 
exists today to the 1999 Phase I source universe to determine if there 
were any significant differences that might influence or impact the 
potential risk. We focused on the following four key characteristics: 
emission rates, stack gas characteristics, meteorological conditions, 
and exposed populations. Based on the results of our comparative 
analysis, we believe that the risk to human health and the environment 
from Phase I sources complying with the proposed replacement standards 
will be, for the most part, the same or less than the estimated risk 
from sources complying with the 1999-promulgated standards. See Part 
Four, Section IX, How Does the Proposed Rule Meet the RCRA 
Protectiveness Mandate?.
---------------------------------------------------------------------------

    \231\ The 1999-promulgated total chlorine standard for new LWAKs 
was 41 ppmv. The proposed replacement standard is 150 ppmv. We do 
not view the total chlorine replacement standard as a concern 
because the 1999-promulgated total chlorine standard for existing 
sources was higher (230 ppmv) and found to be generally protective 
in the national risk assessment conducted for that rulemaking. With 
respect to risk from mercury for LWAKs, see ``Inferential Risk 
Analysis in Support of Standards for Emissions of Hazardous Air 
Pollutants from Hazardous Waste Combustors,'' prepared under 
contract to EPA by Research Triangle Institute, Research Triangle 
Park, NC.
    \232\ See Human Health and Ecological Risk Assessment Support to 
the Development of Technical Standards for Emissions from Combustion 
Units Burning Hazardous Wastes: Background Document, July 1999.

---------------------------------------------------------------------------

[[Page 21326]]

    Although the replacement standards are generally equivalent to or 
more stringent than both the interim and 1999-promulgated standards, we 
cannot assess to what extent this may change the frequency with which 
SSRAs are determined to be necessary. In the end, the MACT standards 
are technology-based and so, risk analysis notwithstanding, cannot 
assure that emissions from each affected source will be protective of 
human health and the environment. For example, a particular source 
could emit types and concentrations of non-dioxin PICs different from 
those we modeled, and so could continue to pose risk not accounted for 
in our analysis. Sources' emissions of criteria pollutants, which are 
non-HAPs and so are beyond the direct scope of MACT, also could 
possibly pose risk which could necessitate site specific risk 
assessment.\233\ Another potential example involves emissions of 
nonmercury metal HAP by cement kilns and lightweight aggregate kilns. 
The semivolatile and low volatile metal thermal emission standards 
directly address emissions attributable to the hazardous waste, as 
opposed to a source's total HAP metal emissions. Thus, although these 
proposed limits reflect MACT, by normalizing the standards to thermal 
firing rate (for the appropriate reasons explained earlier), they do 
not create a HAP metal ``emissions cap.'' HAP metal emission 
contributions from nonhazardous waste fuels and raw materials are not 
directly regulated by this type of emission standard, but are rather 
controlled appropriately with the particulate matter standard.\234\
---------------------------------------------------------------------------

    \233\ See 56 FR at 7145 (Feb. 21, 1991) explaining why there can 
be circumstances where a risk-based standard for particulate matter 
(a criteria pollutant) for hazardous waste combustion sources may be 
needed, and how such a standard could be integrated into the 
National Ambient Air Quality Standard implementation process.
    \234\ Particulate matter is an appropriate surrogate to control 
metal emissions in nonhazardous waste fuels and raw material in lieu 
of a numerical metal emission limit because a numerical metal 
emission standard may inappropriately control feedrate of HAP metals 
in the raw materials and fossil fuels (since such control would be 
neither replicable nor duplicable, and is not justified as a beyond-
the-floor standard).
---------------------------------------------------------------------------

    In contrast, RCRA permits can address the total emissions from the 
combustion unit, assuming an appropriate nexus with hazardous waste 
combustion. Thus, for those combustors that must comply with a thermal 
emission standard and that feed materials other than hazardous waste, 
the permitting authority may decide that an SSRA is appropriate to 
determine if additional limits (i.e., a total emissions cap) are 
necessary to ensure that all metal HAP emissions from the combustion 
unit remain at a level that is protective of human health and the 
environment.
    With respect to Phase II sources, the standards we are proposing in 
today's notice are significantly more stringent than the existing 
technical standards required under RCRA (40 CFR part 266, subpart H). 
To evaluate the protectiveness of the proposed Phase II standards, we 
conducted the same comparative risk analysis for Phase II sources that 
we conducted for Phase I sources. Specifically, we evaluated the 
differences between the 1999 Phase I source universe and the existing 
Phase II source universe with respect to the four key source 
characteristics mentioned above to determine if there were any 
significant differences that might influence or impact the potential 
risk. As discussed in the background document, (``Draft Technical 
Support Document for HWC MACT Replacement Standards, Volume V: 
Emissions Estimates and Engineering Costs'') we estimated emissions for 
each facility based on site-specific stack gas concentrations and flow 
rates measured during trial burn or compliance tests. We then assumed 
that sources would design their systems to meet an emission level below 
the proposed standard. For today's proposed standards, the design level 
is generally the lower of: (1) 70% of the standard; or (2) the 
arithmetic average of the emissions data of the best performing 
sources.\235\ We believe the comparative analysis lends support to our 
view that the standards for Phase II sources are generally protective. 
For a detailed discussion of the comparative risk analysis methodology 
and results, see the background document entitled ``Inferential Risk 
Analysis in Support of Standards for Emissions of Hazardous Air 
Pollutants from Hazardous Waste Combustors,'' prepared under contract 
to EPA by Research Triangle Institute, Research Triangle Park, NC.
---------------------------------------------------------------------------

    \235\ If available test data in our data base indicate that the 
source was emitting below the design level, we assumed that the 
source would continue to emit at the levels measured in test.
---------------------------------------------------------------------------

    As with the Phase I sources, we cannot reliably predict to what 
extent SSRAs will continue to be necessary for Phase II sources once 
they have complied with the MACT standards. In view of the standards 
alone there are at least three possible scenarios for which SSRAs may 
continue to be needed. First, we are proposing thermal emission 
standards for liquid fuel-fired boilers. Thus, similar to cement kilns 
and LWAKs, permitting authorities may determine that an SSRA is 
necessary to ensure that all emissions from liquid fuel-fired boilers 
are protective of human health and the environment. Second, we are 
proposing that liquid fuel-fired boilers with wet APCD or no APCD and 
solid fuel-fired boilers comply with a CO or total hydrocarbon limit as 
a surrogate for the dioxin/furan emission standard. Permitting 
authorities may determine that an SSRA is necessary for these sources 
if there is some concern that the CO or total hydrocarbon limit alone 
may not be adequately protective. Third, we are not proposing standards 
for all HAPs emitted by Phase II area sources. Instead, consistent with 
CAA section 112(c)(6), we are proposing MACT standards only for dioxin/
furans, mercury, carbon monoxide and hydrocarbons, and DRE. For the 
remaining metals, particulate matter and TCl, we are providing area 
sources with the option of complying with the MACT standards for major 
sources or continuing to comply with the existing RCRA technical 
standards. Sources that choose to comply with the RCRA standards may 
need to consider an SSRA, because the RCRA standards alone may not be 
sufficiently protective (i.e., since they do not address the potential 
risk from indirect exposures to long-term deposition of metals onto 
soils and surface waters). To date, we have identified only three area 
sources in the Phase II universe. Thus, the number of sources that 
could decide to continue complying with the above-mentioned RCRA 
standards is expected to be very limited.
    It is useful to note that there are other site-specific factors or 
circumstances beyond the standards themselves that can be important to 
the SSRA decision making process for an individual combustor. For 
example, a source's proximity to a water body or an endangered species 
habitat, repeated occurrences of contaminant advisories for nearby 
water bodies, the number of hazardous air pollutant emission sources 
within a facility and the surrounding community, whether or not the 
waste feed to the combustor is comprised of persistent, bioaccumulative 
or toxic contaminants, and sensitive receptors with potentially 
significantly different exposure pathways, such as Native Americans, 
will likely influence a permitting authority's decision of whether or 
not an SSRA is necessary. In addition, uncertainties inherent in our 
comparative risk analysis and the national risk assessment conducted in 
support of the 1999-promulgated

[[Page 21327]]

standards also may influence a permitting authority's decision. For 
example, the 1999 national risk assessment contained some uncertainties 
regarding the fate and transport of mercury in the environment and the 
biological significance of mercury exposures in fish. Another example 
relates to nondioxin products of incomplete combustion. Due to 
insufficient emissions data and parameter values, the 1999 national 
risk assessment did not include an evaluation of risk posed by 
nondioxin products of incomplete combustion. See 64 FR 52840 and 52841 
for additional discussion of uncertainties regarding the national risk 
assessment. Also, the comparative risk analysis conducted in support of 
today's action did not account for cumulative emissions at a source or 
background exposures from other sources.
3. What Changes Are EPA Proposing With Respect To the Site-Specific 
Risk Assessment Policy?
    As stated earlier in this section, we recommended in the preamble 
to the 1999 rulemaking that permitting authorities evaluate the need 
for an SSRA on a case-by-case basis for hazardous waste combustors 
subject to the Phase I MACT standards. For hazardous waste combustors 
not subject to the Phase I standards, we continued to recommend that 
SSRAs be conducted as part of the RCRA permitting process if necessary 
to protect human health and the environment. We indicated that the RCRA 
omnibus provision authorized permit writers to require applicants to 
submit SSRA results where an SSRA was determined to be necessary. 
Today, we are proposing to codify the authority for permit writers to 
evaluate the need for and, where appropriate, require SSRAs. We are 
also proposing to codify the authority for permit writers to add 
conditions to RCRA permits that they determine, based on the results of 
an SSRA, are necessary to protect human health and the environment. In 
doing so, our intent is to change the regulatory mechanism that is the 
basis for SSRAs, while retaining the same SSRA policy from a 
substantive standpoint. Under this approach, permitting authorities 
continue to have the responsibility to ensure the protectiveness of 
RCRA permits. We are requesting comment on this proposal.
    RCRA sections 3004(a) and (q) require that we promulgate standards 
for hazardous waste treatment, storage and disposal facilities and 
hazardous waste energy recovery facilities as may be necessary to 
protect human health and the environment. RCRA section 1006(b) directs 
us to integrate the provisions of RCRA with the appropriate provisions 
of the CAA and other federal statutes to the maximum extent 
practicable. Thus, to the extent that the RCRA emission standards and 
associated requirements promulgated under section 3004(a) or (q) are 
duplicative of the CAA MACT standards, section 1006(b) provides us with 
the authority to eliminate duplicative RCRA standards and associated 
requirements. For this reason, we have provided that most RCRA emission 
standards and associated requirements no longer apply to incinerators, 
cement kilns, and lightweight aggregate kilns once these sources 
demonstrate compliance with MACT requirements. As explained earlier, we 
are proposing to do the same in today's notice for solid fuel-fired 
boilers, liquid fuel-fired boilers and HCl production furnaces.
    Although the Phase I replacement and Phase II standards provide a 
high level of protection to human health and the environment, thereby 
allowing us to nationally defer the RCRA emission requirements to MACT, 
additional controls may be necessary on an individual source basis to 
ensure that adequate protection is achieved in accordance with RCRA. We 
believe that this will continue to be the case even after the Phase I 
replacement and Phase II standards are promulgated as discussed earlier 
in this section. Up to this point in time, we have relied exclusively 
on RCRA section 3005(c)(3) and its associated regulations (e.g., 40 CFR 
270.10(k)) when conducting or requiring a risk assessment on a site-
specific basis. Because risk assessments are likely to continue to be 
necessary at some facilities, we are proposing to explicitly codify the 
authority to require them on a case-by-case basis and add conditions to 
RCRA permits based on SSRA results under the authority of sections 
3004(a) and (q) and 3005 of RCRA. We continue to believe that section 
3005(c)(3) and its associated regulations provide the authority to 
require and perform SSRAs and to write permit conditions based on SSRA 
results. Indeed, as explained below, EPA will likely continue to 
include permit conditions based on the omnibus authority in some 
circumstances when conducting these activities, and state agencies in 
states with authorized programs will continue to rely on their own 
authorized equivalents, at least for some period of time. However, 
since we foresee that SSRAs will likely continue to be necessary at 
some hazardous waste combustion facilities, we are proposing to 
expressly codify these authorities for the convenience of both 
regulators and the regulated community.
    We are not proposing that SSRAs automatically be conducted for 
hazardous waste combustion units, because we continue to believe that 
the decision of whether or not a risk assessment is necessary must be 
made based upon relevant site-specific factors associated with an 
individual combustion unit and that there are combustion units for 
which an SSRA will not be necessary. We further believe that it is the 
permitting authority, with information provided by hazardous waste 
combustion facilities, that is best equipped to make this decision.
4. How Would the New SSRA Regulatory Provisions Work?
    The SSRA regulatory provisions are proposed under both base program 
authority (sections 3004(a) and 3005(b)) and HSWA authority (section 
3004(q)). Thus, where EPA or a state regulator has determined that a 
risk assessment is necessary, the applicability of the new provisions 
will vary according to the nature of the combustion unit in question 
(whether it is regulated under 3004(q), or only 3004(a) and 3005(b)), 
and the authorization status of the state. Depending on the facts, the 
new authority would be applicable, or the omnibus provision would 
remain the principal authority for requiring site-specific risk 
assessments and imposing risk-based conditions where appropriate.
    As explained in the state authorization section of this preamble 
(see Part Two, Section XIX.C), EPA does not consider these provisions 
to be either more or less stringent than the pre-existing federal 
program, since they simply make explicit an authority that has been and 
remains available under the omnibus authority and its implementing 
regulations. Thus, states with authorized equivalents to the federal 
omnibus authority will not be required to adopt these provisions, so 
long as they interpret their omnibus authority broadly enough to 
require risk assessments where necessary. Nonetheless, we encourage 
states to adopt these provisions to promote regulatory transparency.
    We are proposing to add a paragraph to the general permit 
application requirements of 40 CFR 270.10 to specifically allow a 
permit writer to require that a permittee or an applicant submit an 
SSRA or the information necessary for the regulatory agency to conduct 
an SSRA, if one is determined to be necessary. The permit writer may 
decide that an SSRA is needed if there

[[Page 21328]]

is some reason to believe that additional controls beyond those 
required pursuant to 40 CFR parts 63, 264 or 266 may be needed to 
ensure protection of human health and the environment under RCRA. We 
are also proposing to allow the permit writer to require that the 
applicant provide information, if needed, to make the decision of 
whether a risk assessment should be required. In addition, we are 
proposing to amend the applicability language of 40 CFR 270.19, 270.22, 
270.62, and 270.66 to allow a permit writer that has determined that an 
SSRA is necessary for a specific combustion unit to continue to apply 
the relevant requirements of these sections on a case-by-case basis and 
as they relate to the performance of the SSRA after the source has 
demonstrated compliance with the MACT standards.
    The basis for the decision to conduct the risk assessment must be 
included in the administrative record for the facility and made 
available to the public during the comment period for the draft permit. 
If the facility, or any other party, files comments on a draft permit 
decision objecting to the permitting authority's conclusions regarding 
the need for a risk assessment, the authority must respond fully to the 
comments. In addition, the risk assessment itself also must be included 
in the administrative record and made available to the public during 
the comment period for the permit. Any resulting permit conditions from 
the SSRA also must be documented and supported in the administrative 
record. We are proposing to add a paragraph to 40 CFR 270.32 to address 
the inclusion of conditions and limitations in RCRA permits as a result 
of the findings of an SSRA.
5. Why Is EPA Not Providing National Criteria for Determining When an 
SSRA Is or Is Not Necessary?
    We are not proposing national criteria for determining when an SSRA 
is necessary. In the preamble to the April 1996 Phase I NPRM, we 
provided a list of guiding factors which we later updated and modified 
in the preamble to the September 1999 final rulemaking. See 61 FR 17372 
and 64 FR 52842. We view these guiding factors as items that, because 
they may be relevant to the potential risk from a hazardous waste 
combustion unit, could be considered by a permitting authority when 
deciding if an SSRA is necessary. We did not, and do not, intend for 
them to be definitive criteria from which permitting authorities would 
make their decision. As we stated in 1999, we believed that the 
complexity of multi-pathway risk assessments precluded the conversion 
of these qualitative guiding factors into more definitive criteria. 
Since that time, we have reaffirmed our belief that the decision 
process regarding SSRAs does not lend itself to the application of 
required national criteria. Most combustors may be characterized using 
one or more of the qualitative guiding factors we provided in 1999, but 
not all. These factors were not intended to be an exclusive list of 
considerations, nor do we believe that this decision is necessarily 
susceptible to an exclusive list of factors. The decision whether to 
require a risk assessment is inherently site specific, and permitting 
authorities need to have the flexibility to evaluate a range of factors 
that can vary from facility to facility. In addition, it is useful to 
recognize that as risk assessment science continues to mature, the 
factors may change in terms of relative importance and it may not be 
prudent to obligate permitting authorities to an exclusive list that 
could not be easily adjusted to keep pace with scientific advancements.
    In a study conducted by U.S. EPA Region 4, the guiding factors were 
used to rank 13 hazardous waste combustion facilities into high, medium 
and low risk potential groupings to ascertain if the factors could be 
used as a prioritization tool for determining whether or not an SSRA 
was necessary. The region found that all facilities evaluated exhibited 
a ``high'' level of concern with respect to at least one or more site-
specific characteristics relating to the guiding factors and that 
further analysis was required before the region could be assured that 
the source would operate in a manner that is adequately protective 
under RCRA. As a result, the region concluded that the guiding factors 
alone could not be used to make a protectiveness finding. The region's 
study, which is entitled Technical Support Assistance of MACT 
Implementation Qualitative Risk Check is available in the docket 
(Docket #RCRA-2003-0016) established for today's notice.
    Moreover, simply determining whether a combustor fits a particular 
guiding factor does not address the complex interplay that may exist 
between the guiding factors. Nor, does it measure the level of relative 
importance of one factor over another. For example, is the proximity of 
potentially sensitive receptors more important than multiple on-site 
emission points? For all of these reasons, we believe that codification 
of a list of factors would not be appropriate here.
6. What Is the Cement Kiln Recycling Coalition's SSRA Rulemaking 
Petition?
    On February 28, 2002, the Cement Kiln Recycling Coalition (CKRC) 
submitted a petition for rulemaking ``Petition Under RCRA Sec.  7004(a) 
For (1) Repeal of Regulations Issued Without Proper Legal Process and 
(2) Promulgation of Regulations If Necessary With Proper Legal 
Process'' to the Administrator containing two independent requests with 
respect to SSRAs. First, CKRC requested that we repeal the existing 
SSRA policy and technical guidance because it believes that the policy 
and guidance ``are regulations issued without appropriate notice and 
comment rulemaking procedures.'' Second, CKRC requested that after we 
repeal the policy and guidance, ``should EPA believe it can establish 
the need to require SSRAs in certain situations, CKRC urges EPA to 
undertake an appropriate notice and comment rulemaking process seeking 
to promulgate regulations establishing such requirements.''
    As stated in the petition, ``CKRC does not believe that these SSRA 
requirements are in any event necessary or appropriate.'' In addition, 
CKRC disagrees with our use of the RCRA omnibus provision as the 
authority to conduct SSRAs or to collect the information and data 
necessary to conduct SSRAs and further contends that the regulations 
associated with the omnibus provision are insufficient in detail. CKRC 
asserts that we have chosen to establish SSRA requirements through 
guidance documents. CKRC also raised the following three general 
concerns: (1) Whether an SSRA is needed for hazardous waste combustors 
that will be receiving a RCRA permit when the combustor is in full 
compliance with the RCRA boiler and industrial furnace regulations and/
or with the MACT regulations; (2) How an SSRA should be conducted; and 
(3) What is the threshold level for a ``yes'' or ``no'' decision that 
additional risk-based permit conditions are necessary. In support of 
its petition, CKRC refers to Appalachian Power Co. v. EPA, 208 F.3d 
1015 (D.C. Cir. 2000), GE v. EPA, 290 F.3d 377 (D.C. Cir. 2002), and 
Ethyl Corporation v. EPA, 306 F.3d 1144 (D.C. Cir. 2002). The petition 
is available in the docket established for today's proposed action.
    CKRC filed the petition filed under RCRA section 7004(a), which 
provides that: ``Any person may petition the Administrator for the 
promulgation, amendment, or repeal of any regulation under this Act. 
Within a reasonable time following receipt of such a petition, the

[[Page 21329]]

Administrator shall take action with respect to the petition and shall 
publish notice of such action in the Federal Register, together with 
the reasons therefor.''
    Shortly after receiving the petition, we conducted a preliminary 
evaluation of CKRC's concerns as stated in the petition.\236\ We 
determined that any decision regarding the petition should be made in 
coordination with our development of the proposed Replacement MACT 
standards for Phase I sources and the proposed new MACT standards for 
Phase II sources. Thus, we decided that today's notice was the most 
appropriate vehicle to announce and request comment on our tentative 
decision concerning the petition.
---------------------------------------------------------------------------

    \236\ EPA does not consider the request to repeal EPA's guidance 
documents to be a valid petition under this section, since the 
documents are guidance documents, not regulations. Nonetheless, 
because CKRC has also petitioned the Agency to issue regulations, 
and to be responsive to issues raised by the regulated community, 
EPA has decided to use the procedure established in 40 CFR 260.20 
for section 7004 petitions to respond to both of CKRC's requests. 
EPA does not concede by relying on the section 7004(a) procedure 
that its guidance documents are regulations.
---------------------------------------------------------------------------

    In the meantime, we believed that it was important to take certain 
measures to ensure that the SSRA policy and guidance were being used in 
the manner that we had intended. In an April 10, 2003 memorandum from 
Marianne Lamont Horinko, Assistant Administrator of the Office of Solid 
Waste and Emergency Response, to the U.S. EPA Regional Administrators, 
we took two of these measures. First, we requested that the regions 
review certain documents (e.g., regional memoranda, policy and guidance 
documents, Memoranda of Agreement of Grant Workplans with the states) 
to determine if any contained misleading or incorrect information 
concerning the SSRA policy and technical guidance. If any were found to 
contain misleading or incorrect information, we requested that the 
region take immediate measures to clarify or correct the information. 
Second, we reiterated, in detail, the appropriate use of the SSRA 
policy and guidance for hazardous waste combustors, as well as the 
appropriate use of the RCRA omnibus authority as it relates to SSRAs. 
In a May 15, 2002, memoranda from Robert Springer, Director of the 
Office of Solid Waste, to the RCRA Senior Policy Advisors, we took the 
third measure to ensure proper application of the SSRA policy by our 
regional permit writers. In this memorandum, we instituted an EPA 
headquarters review process of future regional decisions concerning the 
need for an SSRA for hazardous waste combustion units seeking a RCRA 
permit determination. Specifically, we requested that the regions 
provide us with a written summary of the basis for any future decisions 
to conduct or not conduct an SSRA. It is our intention that the review 
process focus on whether or not permit writers have adequately 
supported their decisions. It is important to point out that because 
many of the decisions regarding SSRAs are now being made at the state 
level, we do not yet know how many regional SSRA decision summaries 
will be submitted for our review. Both the April 10, 2003, and May 15, 
2003, memoranda are provided in the docket established for today's 
proposed action.
    EPA is in the process of an additional effort to ensure proper use 
of the guidance: we are reviewing the guidance documents themselves, 
and, to the extent we find language that could be construed as limiting 
discretion, we intend to revise the documents to make clear that they 
are non-binding. CKRC indicated in its petition that, in its view, the 
documents contain language that could be construed as mandatory. While 
EPA does not necessarily agree, and believes that, in context, it is 
clear that the guidance in the documents is discretionary, EPA is 
nonetheless reviewing the documents to ensure that they are carefully 
drafted.
    After consideration of the petition, we have made a tentative 
decision to partially grant and partially deny CKRC's requests. 
Specifically, we are proposing to deny CKRC's request that we repeal 
the SSRA policy and guidance and we are proposing to grant CKRC's 
request in part by promulgating an explicit authority to require SSRAs 
on a site-specific basis using notice and comment rulemaking 
procedures. We are requesting comment on our tentative decision.
    With respect to CKRC's first request that we repeal the SSRA policy 
and guidance, and in response to their specific concern of whether an 
SSRA is necessary for combustors that are in full compliance with the 
RCRA and/or MACT regulations, we believe that SSRAs do serve a useful 
purpose and can be necessary even if a facility is in full compliance 
with the existing RCRA and/or MACT technical standards. RCRA requires 
that all hazardous waste permits be protective of human health and the 
environment. As discussed in the preamble to the 1999 Phase I 
rulemaking, the existing RCRA incinerator and Boiler and Industrial 
Furnace (BIF) regulations do not address the potential risk that may be 
posed from indirect exposures to combustor emissions. See 64 FR 52828, 
52839-52842 (September 30, 1999). Further, the technical requirements 
associated with the RCRA standards have not been updated to reflect 
changes in technology or science for a decade or more and, thus, may 
not be sufficiently protective with respect to the potential risk from 
direct exposures either. For example, our knowledge regarding the 
formation, control and toxicity of dioxin/furans has vastly improved 
since the promulgation of the RCRA standards. Therefore, until such 
time that hazardous waste combustors comply with the MACT standards, 
SSRAs can serve a useful function in ensuring that RCRA combustor 
permits will be protective of human health and the environment.
    Moreover, even once the MACT standards are fully implemented for 
incinerators and BIFs, we believe that there may continue to be 
instances in which the permitting authority determines that additional 
protections are necessary (e.g., where site-specific conditions 
indicate that there may be a potential risk to a sensitive ecosystem or 
population), as was explained above in Section 2, Are SSRAs Likely to 
be Necessary After Sources Comply with the Phase I Replacement 
Standards and Phase II Standards? See also, the explanations at 64 FR 
52840-52841. Because there may continue to be a need for SSRAs at some 
level, we agree with CKRC that it would be appropriate to explicitly 
codify the authority to require SSRAs and SSRA-based permit conditions, 
for the sake of regulatory clarity and transparency (although we 
continue to believe that the RCRA omnibus provision provides sufficient 
authority to conduct SSRAs). EPA requests comment on the variety of 
site-specific circumstances that might give rise to the need for an 
SSRA, and whether other mechanisms might exist to address those 
circumstances.
    As stated earlier, CKRC raised three general concerns, the first of 
which we discussed in the preceding paragraphs. The second concern 
relates to the technical recommendations that EPA has offered for 
conducting an SSRA. CKRC disagrees with our use of guidance, instead 
arguing that EPA's recommendations should have been issued through the 
notice and comment rulemaking process.
    We disagree that the Agency's technical recommendations either must 
or should be issued as a regulation. Risk assessment--especially multi-
pathway, indirect exposure assessment--is a highly technical and 
evolving field. Any regulatory approach EPA might codify in this area 
is likely to become outdated, or at least artificially constraining, 
shortly after promulgation in ways that

[[Page 21330]]

EPA cannot anticipate now. In EPA's view, this is an area that is 
uniquely fitted for a guidance approach, rather than regulation. In 
fact, across Agency programs, EPA has generally adopted a guidance 
approach to risk assessment for exactly this reason. See, e.g., 
Guidelines for Reproductive Toxicity Risk Assessment, 61 FR 56274 
(October 31, 1996). EPA's Superfund program has not promulgated 
regulations specifying risk assessment methods. Instead, the program 
uses site-specific approaches for determining risk, employing methods 
offered in EPA guidance as appropriate. The same is true for the RCRA 
corrective action program. Although we have attempted to provide our 
guidance recommendations in a form that responds to or encompasses many 
of the issues that can arise when conducting an SSRA, we recognize that 
the flexibility to apply other methodologies, assumptions, or 
recommendations has been important to both regulators and the regulated 
community in terms of developing an appropriate site-specific 
protocol.\237\ For example, some of EPA's technical recommendations may 
not be appropriate for the combustion device in question, and risk 
assessors must have the flexibility to make adjustments for the 
specific conditions present at the source, and the state of risk 
assessment science at the time that the SSRA is being performed. As an 
obvious example, sources that are located in a dry, desert climate with 
no nearby permanent or temporary water bodies of concern should not be 
required to include a fisher exposure scenario in an SSRA. In addition, 
risk assessors should be free to use the most recent air modeling tools 
and toxicity values available rather than be limited to those that may 
be out-of-date because a regulation has not been revised following the 
development of the new tools or values. Guidance allows for this 
flexibility.
---------------------------------------------------------------------------

    \237\ Permitting authorities, in some cases, have developed 
their own guidance methodologies responsive to the specific needs 
associated with their facilities. For example, North Carolina, 
Texas, and New York have each developed their own risk assessment 
methodologies. We think this flexibility employed in the field 
supports our judgment that risk assessment methodologies should not 
be codified.
---------------------------------------------------------------------------

    CKRC points out the EPA codified certain parameters for BIF risk 
assessments, to show that it is possible to do so. While EPA agrees it 
is possible, the codification in the BIF area is the exception, not the 
rule. It has been our experience in implementing the BIF regulations 
that codification of certain risk parameters has proven to be overly 
constraining because risk science is a continually changing field. For 
example, by codifying the toxicity values, risk managers were not able 
to utilize more recent values available through EPA's Integrated Risk 
Information System (IRIS) \238\ and other resources. Also, shortly 
after we codified the air modeling guidelines in support of the risk 
parameters and procedures, the Air program revised their air modeling 
guidelines, rendering some of the BIF air modeling guidelines 
inconsistent and so, they were removed. Further, it is important to 
note that at the time of codification, BIF risk assessments were not 
intended to address indirect routes of exposure, thus making the 
parameters easier to implement. Today, however, risk assessments are 
more complex due to the necessary inclusion of multi-pathway and 
indirect exposure routes. Given the complexity of multi-pathway and 
indirect exposure assessments and the fact that risk science is 
continuously evolving, it would be difficult and again, overly 
constraining, to codify risk parameters today.
---------------------------------------------------------------------------

    \238\ IRIS is a collection of continuously updated chemical 
files which contain descriptive and quantitative information with 
respect to: oral reference doses and inhalation reference 
concentrations (RfDs and RfCs, respectively) for chronic 
noncarcinogenic health effects; and hazard identification, oral 
slope factors, and oral and inhalation unit risks for carcinogenic 
effects. For more information, see http://www.epa.gov/iris/index.html
.

---------------------------------------------------------------------------

    We also believe that a guidance approach is consistent with the 
fact that permit writers must make site-specific decisions whether to 
do risk assessments at all. We expect that permit writers will reach 
their decisions based on different factors and concerns--in some cases, 
factors and concerns that we may not have identified at this time. We 
think that it makes little sense to allow this kind of flexibility 
regarding whether to do a risk assessment and for what purposes, while 
prescribing how one must be conducted if one is required.
    CKRC further contends that the guidance is overly conservative and 
constitutes ``a confusing pattern of drafts over a number of years in a 
seemingly endless fashion'' that has resulted in their members 
incurring significant costs. Because of the variability in the many 
factors that influence the risk from hazardous waste combustors, the 
guidance contains some conservative recommendations and assumptions in 
order to address this wide range. However, based on input from users of 
the guidance, we have attempted to correct the recommendations and 
assumptions that we consider to be overly conservative and, as stated 
previously, because they are guidance recommendations and not 
requirements, the risk assessor may choose not to follow them. More 
recently, we have solicited public and peer review comments on the 1998 
guidance,\239\ and are in the process of revising it based on the 
comments received. This includes comments CKRC submitted related to the 
components of the guidance they contended were overly 
conservative.\240\
---------------------------------------------------------------------------

    \239\ USEPA. ``Human Health Risk Assessment Protocol for 
Hazardous Waste Combustion Facilities'' EPA-520-D-98-001A, B&C. 
External Peer Review Draft, 1998. 
(http://www.epa.gov/epaoswer/hazwaste/combust/risk.htm)
    \240\ We are not responding to the specific comments here, but 
will respond to them as part of the public process for developing 
the final guidance documents.
---------------------------------------------------------------------------

    With respect to CKRC's assertion that the guidance is ``a confusing 
pattern of drafts over a number of years'', we acknowledge that we have 
issued a number of guidance documents since 1990. However, we disagree 
that this has resulted in a confusing pattern of drafts. The 
development and release of the guidance documents correspond to three 
specific regulatory time periods in the area of hazardous waste 
combustion. In addition, the issuance of subsequent versions relates to 
the fact that the Agency has repeatedly solicited public and peer 
review comments on its technical guidance, and has built upon the 
experience of regulators and facilities in using earlier guidance.
    In 1990, EPA developed its initial guidance document during the 
same time period as the RCRA BIF emission standards. In 1993, we 
released an addendum to the 1990 guidance in response to the draft 
Hazardous Waste Minimization and Combustion Strategy and our increasing 
concerns about the potential impacts from indirect routes of exposure, 
and solicited comments from the public and the Science Advisory Board. 
A revised document taking into account these comments was issued one 
year later.\241\
---------------------------------------------------------------------------

    \241\ USEPA. ``Guidance for Performing Screening Level Risk 
Analyses at Combustion Facilities Burning Hazardous Wastes'' Draft, 
April 1994. USEPA. ``Implementation of Exposure Assessment Guidance 
for RCRA Hazardous Waste Combustion Facilities'' Draft, 1994. (These 
documents are available as part of the ``Exposure Assessment 
Guidance for RCRA Hazardous Waste Combustion Facilities'' EPA530-R-
R-94-021. Copies may be ordered through the National Service Center 
for Environmental Publications' Web site at 
http://www.epa.gov/ncepihom/)

---------------------------------------------------------------------------

    At the time that we were developing the Phase 1 MACT standards, we 
again updated our combustion risk assessment guidance by releasing a 
document specifically addressing human health risk in 1998 and one 
addressing ecological risk in 1999, again soliciting public input and 
peer review on these

[[Page 21331]]

documents.\242\ For purposes of clarity, both of these documents refer 
to all earlier guidance where appropriate and discuss briefly the 
progression of the guidance. Although the 1998 human health guidance 
and the 1999 ecological guidance provide our current thinking regarding 
SSRA methodology for hazardous waste combustors, we noted to our permit 
writers that we recommended that they should continue to use the 1994 
guidance for those SSRAs that were in progress.
---------------------------------------------------------------------------

    \242\ We noted earlier that the 1998 guidance is currently being 
revised in consideration of public and peer review comments 
received. With respect to the 1999 guidance (USEPA. ``Screening 
Level Ecological Risk Assessment Protocol for Hazardous Waste 
Combustion Facilities'' EPA-530-D-99-001A, B&C. Peer Review Draft, 
1999), we solicited public comment and plan to conduct a peer 
review. (http://www.epa.gov/epaoswer/hazwaste/combust/ecorisk.htm)

---------------------------------------------------------------------------

    Although CKRC claims to find these guidance documents confusing, 
EPA's judgment is that most interested parties--both regulators and the 
regulated community--have found the guidance to be useful, and that the 
documents have substantially reduced the uncertainty and confusion that 
surrounded multi-pathway risk assessments a decade ago. As stated 
above, no one is obligated to follow this guidance, and regulators 
often depart from it; but EPA believes it has been extremely helpful on 
the whole, rather than confusing.
    CKRC has alleged that SSRA's typically cost between $200,000 and 
$1,000,000 for an individual facility. We are aware that prior to the 
release of the 1998 guidance, combustion risk assessments were more 
costly than we understand them to be today. For an individual facility, 
we do not know to what extent these costs are attributed to the act of 
conducting a risk assessment, to recommendations provided in our 
guidance, to changes that the facility chose to make during the risk 
assessment, or the facility's desire to develop its own site-specific 
protocol. Not including the collection and analysis of emission risk 
data, we have been advised that the cost of an average SSRA today is 
approximately $84,000. (See document entitled Hazardous Waste 
Combustion MACT--Replacement Standards: Proposed Rule. Preliminary Cost 
Assessment for Site Specific Risk Assessment, November, 2003, as 
provided in the docket for today's action.) The emission risk data is 
projected to add on average between $57,000 (if the facility collects 
its emission risk data at the same time as its emission standards 
performance data) and $285,000 (if the facility must conduct a separate 
emission test solely for the purpose of collecting data for the SSRA). 
Therefore, including emission data collection, the average cost of an 
SSRA is between $141,000 and $370,000. This is considerably less than 
the cost range provided by CKRC of $200,000 to $1,000,000. 
Additionally, EPA's upper bound cost of $370,000 is significantly less 
than the upper bound cost of $1,300,000, as reported by CKRC in their 
petition (and the attached affidavit).\243\ We believe that the cost of 
SSRAs has decreased over time, particularly since the release of the 
1998 guidance. This may be in large part because the 1998 guidance is 
much more comprehensive than previous guidance documents and because 
private software companies have developed computer programs based on 
the guidance, which can further decrease costs associated with the risk 
calculations for each exposure scenario.
---------------------------------------------------------------------------

    \243\ The cost ranges for CKRC include both the cost of risk 
assessments and emission data collection. In its petition, CKRC 
provided a range of costs ($100,000 to $500,000 for risk assessments 
and $100,000 to $500,000 for emission data collection), but also 
provided an upper bound cost ($728,297 for a risk assessment and 
$588,790 for emission data collection, plus additional permit costs 
to equate to $1.3M).
---------------------------------------------------------------------------

    CKRC also expressed specific concern that it and its members have 
been denied an opportunity to comment on the combustion risk assessment 
guidance documents. We strongly disagree with this assertion. We have 
repeatedly sought public comment on the guidance documents. For the 
1998 human health guidance we not only requested public comment, but 
also submitted the document for an external peer review and held a peer 
review meeting which was open to the public. Since the peer review 
meeting, we have been incorporating both the public and peer review 
comments into the human health guidance. While we have not yet 
completed this task and released a final document, any member of the 
public may at any time discuss any concerns that they have with our 
recommendations. In addition, regardless of whether a risk assessor 
uses the recommendations provided in our guidance or not, we have 
encouraged the permit writer and facility representatives to meet prior 
to any analysis to discuss the appropriate risk methodology and data 
input needs for an SSRA. Such a meeting allows both the permitting 
authority and the facility the opportunity to raise questions and 
objections concerning the appropriateness of different methodologies, 
assumptions, or default values and their application to the hazardous 
waste combustor. Facility representatives and any member of the public 
also may comment on the risk assessment methodology as part of the 
public comment process associated with the RCRA permit.
    The third general concern raised by CKRC in its petition was that 
we had not provided a threshold level for a ``yes'' or ``no'' decision 
to trigger the need for additional risk-based permit conditions. EPA 
agrees that its guidance does not establish a bright-line threshold 
level for determining whether to impose additional permit conditions; 
such a binding requirement would only be appropriately established 
through rulemaking. However, EPA has provided recommendations about the 
overall targets for acceptable risk levels. See USEPA. Implementation 
of Exposure Assessment Guidance for RCRA Hazardous Waste Combustion 
Facilities, Draft, 1994. Moreover, we do not intend to codify our 
recommended target levels for some of the same reasons that we are not 
proposing to codify the risk assessment technical guidance. Our 
recommended target levels provide risk managers with a starting point 
from which to determine if a combustor's potential risk may or may not 
be acceptable. However, we believe that it is important, and indeed 
essential, that risk managers be afforded sufficient flexibility to 
apply different target levels as dictated by the circumstances 
surrounding the combustor. For example, a risk manager may wish to 
apply a more stringent carcinogenic target level for a combustor that 
is located in a densely populated area with a high concentration of 
industrial emission sources.
    In summary, we have made a tentative decision to deny CKRC's 
request that we repeal the SSRA policy and guidance and to grant CKRC's 
request in part by proposing to codify the authority to require SSRAs. 
We are not proposing to codify the SSRA guidance or our recommended 
risk methodology for hazardous waste combustors. We are requesting 
comment on our tentative decision.

XVIII. What Alternatives to the Particulate Matter Standard Is EPA 
Proposing or Requesting Comment On?

    As discussed in Part Two, Section IV.C, we are proposing 
particulate matter standards as surrogates to control metal HAP.\244\ 
We are not proposing numerical metal HAP emission standards that would 
have accounted for all metal HAP because we generally do not have as 
much compliance test

[[Page 21332]]

emissions information in our database for the nonenumerated metal HAP 
compared to the enumerated metal HAP,\245\ and because we believe that 
a particulate matter standard, in lieu of emission standards that 
directly regulate all the metals in all feedstreams, simplifies 
compliance activities.
---------------------------------------------------------------------------

    \244\ Particulate matter is not a listed HAP pursuant to CAA 
112(b).
    \245\ ``Enumerated'' metals are those HAP metals that are 
directly controlled with an emission limit, i.e., lead, cadmium, 
arsenic, beryllium, and chromium. The remaining nonmercury metal HAP 
are controlled using particulate matter as a surrogate.
---------------------------------------------------------------------------

    Nonetheless, we are today proposing an alternative to the 
particulate matter standard for incinerators, liquid fuel-fired 
boilers, and solid fuel-fired boilers that is conceptually similar to 
the alternative metal emission control requirements that were 
previously promulgated for incinerators. We are also requesting comment 
on another alternative to the particulate matter standard that would 
apply to all source categories that would be subject to particulate 
matter standards (i.e., all source categories except hydrochloric acid 
production furnaces).
    We discuss these two different alternatives below.

A. What Alternative to the Particulate Matter Standard Is EPA Proposing 
For Incinerators, Liquid Fuel-Fired Boilers, and Solid Fuel-Fired 
Boilers?

    We promulgated an alternative to the particulate matter standard 
for incinerators feeding low levels of metals in the July 3, 2001, 
direct final rule. See 66 FR at 35093. Today we propose a simplified 
alternative to the particulate matter standard for incinerators, and we 
propose to expand the provision to also apply to liquid and solid fuel-
fired boilers. Below, we first describe the alternative that was 
originally promulgated for incinerators, after which we describe the 
simplified approach and our rationale for proposing it.
    The July 3, 2001, final rule allows incinerators to operate under 
alternative metal emission control requirements reflecting MACT in lieu 
of complying with the 0.015 gr/dscf particulate emission standard. 
Under the alternative, no particulate matter emission standard applies 
to incinerators under subpart EEE; however, the incinerator remains 
subject to the RCRA particulate matter standard of 0.08 gr/dscf 
pursuant to Sec.  264.343(c). This is because Clean Air Act standards 
can supplant RCRA standards only when the CAA standard is sufficiently 
protective of human health and the environment to make the RCRA 
standard duplicative (within the meaning of RCRA section 1006 (b) 
(3)).\246\ See Part Two, Section XVII.D.
---------------------------------------------------------------------------

    \246\ Sources electing to comply with these alternative 
requirements thus remain subject to the RCRA PM standard in their 
RCRA permit. The RCRA permit must include applicable operating 
limits that ensure compliance with the RCRA PM limit.
---------------------------------------------------------------------------

    This previously promulgated alternative to the particulate matter 
standard has three components. The first component is simply to meet 
metal emission limitations for semivolatile and low volatile metals. 
The emission limitations apply to both enumerated and non-enumerated 
metal HAP, excluding mercury. Enumerated semivolatile metals are those 
metals that are directly controlled with the numerical semivolatile 
emission standard, i.e., cadmium and lead. Enumerated low volatile 
metals are those metals that are directly controlled with the numerical 
low volatile metals emission standard, i.e., arsenic, beryllium and 
chromium. Non-enumerated metals are those remaining metal HAP: 
antimony, cobalt, manganese, nickel, and selenium that are not 
controlled directly with an emission standard, but are rather 
controlled through the surrogate particulate matter standard.\247\ For 
purposes of these alternative requirements, the non-enumerated metals 
are classified as either a semivolatile or a low volatile metal, and 
included in the calculation of compliance with the corresponding 
emissions limit. The level of the standard is the same as that which 
applies to other incinerators, but the standard would apply to all 
metal HAP, not just those enumerated in the present low volatile metal 
and semivolatile metal standards.
---------------------------------------------------------------------------

    \247\ Please note that the particulate matter standard is not 
redundant to the semivolatile and low volatile metal standards. 
Although controlling particulate matter also controls semivolatile 
and low volatile metals in combustion gas, these metals can also be 
controlled by feedrate control. Thus, sources can achieve the 
emission standard for semivolatile and low volatile metals primarily 
by feedrate control. In such cases, the particulate matter standard 
would be controlling nonenumerated metals primarily.
---------------------------------------------------------------------------

    The second component is a requirement for the incinerator to 
demonstrate that it is using reasonable hazardous waste metal feed 
control, i.e., a defined metal feedrate that is better than the MACT-
defining metal feed floor control level. The third component is a 
requirement for the incinerator to demonstrate that its air pollution 
control system achieves, at a minimum, a 90 percent system removal 
efficiency for semivolatile metals.
    Today we propose a simplified version of the above described 
alternative in that we propose to require you to comply only with the 
first component described above, which is to achieve metal emission 
standards for semivolatile and low volatile metals. As discussed above, 
the level of the proposed standard is the same as that which applies to 
other sources, but the standard would apply to all metal HAP, not just 
those enumerated in the present semivolatile and low volatile metal 
standards. As with the previously promulgated alternative, no 
particulate matter emission standard would apply to these sources under 
subpart EEE; however, sources would remain subject to the RCRA 
particulate matter standard of 0.08 gr/dscf pursuant to Sec. Sec.  
264.343(c) or 266.105.
    We propose to eliminate the requirements for you to demonstrate 
that: (1) You are using reasonable hazardous waste metal feed control, 
i.e., a defined metal feed control that is better than the MACT-
defining feed control level; and (2) your source is equipped with an 
air pollution control system that achieves at least a 90 percent system 
removal efficiency for semivolatile metals. We believe these two 
requirements are not necessary to ensure you are in fact controlling 
metals below MACT levels given that all sources electing to comply with 
this alternative must limit both the enumerated metals and non-
enumerated metals to levels below the proposed levels that apply only 
to enumerated metals. Today's proposed approach, in effect, lowers the 
existing semivolatile and low volatile metal emissions limits because 
the contribution of nonenumerated metals must be accounted for when 
achieving the same numerical semivolatile and low volatile emission 
limits. We believe this is appropriate because this effectively lower 
emissions limit for enumerated metals compensates for the lower 
emission levels that would have been achieved if the source used a 
particulate matter control device capable of achieving the particulate 
matter standard. Put another way, we regard this emission limitation as 
an equivalent means of meeting the standard for HAP metals (except 
mercury) already established in the rule.
    As discussed above, the approach we promulgated on July 3, 2001 
required you, in practice, to feed low levels of metals on a continuous 
basis in order to qualify for the alternative. The rule required that 
the source's feed control level must be equivalent to or lower than 25% 
of the MACT-defining hazardous waste feed control level. We considered 
whether it would be appropriate to also apply such a

[[Page 21333]]

qualification requirement to today's proposed alternative. 
Unfortunately, the methodology used to calculate today's proposed 
emission standards does not base the standards on a specific MACT-
defining feed control level. Thus, we do not have a MACT feed control 
level that we can readily use to define an appropriate low feed control 
level. We request comment on whether it is appropriate and/or necessary 
to establish a minimum feed control level, and if so, how it could be 
determined.
1. What Emission Limitation Must Incinerators Comply With Under This 
Alternative?
    For existing incinerators, the emissions limits under this 
alternative would be: (1) A semivolatile metal emission limit of 59 
[mu]g/dscm for the combined emissions of lead, cadmium, and selenium; 
and (2) a low volatile metal emission limit of 84 [mu]g/dscm for 
combined emissions of arsenic, beryllium, chromium, antimony, cobalt, 
manganese, and nickel (all emissions corrected to 7% oxygen).
    For new sources, the emissions limits would be: (1) a semivolatile 
emission limit of 7 [mu]g/dscm for combined emissions of lead, cadmium, 
and selenium; and (2) a low volatile emission limit of 9 [mu]g/dscm for 
emissions of arsenic, beryllium, chromium, antimony, cobalt, manganese, 
and nickel (all emissions corrected to 7% oxygen).
2. What Emission Limitation Must Liquid Fuel-Fired Boilers Comply With 
Under This Alternative?
    For existing liquid fuel-fired boilers, the emissions limits under 
this alternative would be: (1) A semivolatile metal emission limit of 
1.1E-5 lb/MM BTU for the combined emissions of lead, cadmium, and 
selenium; and (2) a low volatile metal emission limit of 7.7E-5 lb/MM 
BTU for combined emissions of arsenic, beryllium, chromium, antimony, 
cobalt, manganese, and nickel (all emissions corrected to 7% oxygen).
    For new sources, the emissions limits would be: (1) A semivolatile 
metal emission limit of 4.3E-6 lb/MM BTU for combined emissions of 
lead, cadmium, and selenium; and (2) a low volatile metal emission 
limit of 3.6E-5 lb/MM BTU for emissions of arsenic, beryllium, 
chromium, antimony, cobalt, manganese, and nickel (all emissions 
corrected to 7% oxygen).
3. What Emission Limitation Must Solid Fuel-Fired Boilers Comply With 
Under This Alternative?
    For existing solid fuel-fired boilers, the emissions limits under 
this alternative would be: (1) A semivolatile metal emission limit of 
170 [mu]g/dscm for the combined emissions of lead, cadmium, and 
selenium; and (2) a low volatile metal emission limit of 210 [mu]g/dscm 
for combined emissions of arsenic, beryllium, chromium, antimony, 
cobalt, manganese, and nickel (all emissions corrected to 7% oxygen).
    For new sources, the emissions limits would be: (1) A semivolatile 
metal emission limit of 170 [mu]g/dscm for combined emissions of lead, 
cadmium, and selenium; and (2) a low volatile metal emission limit of 
190 [mu]g/dscm for emissions of arsenic, beryllium, chromium, antimony, 
cobalt, manganese, and nickel (all emissions corrected to 7% oxygen).
4. Why Don't We Offer This Alternative to Lightweight Aggregate Kilns 
and Cement Kilns?
    This alternative is intended to apply to sources that feed de 
minimis levels of metal HAP. We do not believe hazardous waste burning 
lightweight aggregate kilns and cement kilns feed these metals at de 
minimis levels primarily because raw materials and coal that is co-
fired may contain these metal HAP, and because hazardous waste that is 
combusted by sources that receive off-site hazardous waste shipments 
(i.e., commercial hazardous waste combustors) typically contain these 
metal HAP. Thus, we think that allowing this alternative would not be 
of practical significance because we do not believe these sources could 
meet the standard. As a result, we are not proposing this alternative 
for these source categories.

B. What Alternative to the Particulate Matter Standard Is EPA 
Requesting Comment On?

    As previously discussed, we do not have sufficient metal HAP 
compliance data to calculate MACT floors that would account for all the 
nonmercury metal HAP in all feedstreams. We discuss below, however, an 
alternative approach to the particulate matter standard that could be 
implemented if sources monitor and collect nonmercury metal HAP feed 
concentration data prior to the compliance date. Such an approach, if 
promulgated, would result in site-specific metal HAP emission limits 
that would be dependent, in part, on each source's average feed 
concentration levels of metal HAP in their hazardous and nonhazardous 
waste feedstreams, and, for energy recovery units, each source's 
hazardous waste firing rate. We discuss this alternative below, and we 
request comment as to whether this approach is appropriate given the 
complexities associated with its implementation. Also see USEPA, 
``Draft Technical Support Document for HWC MACT Replacement Standards, 
Volume IV: Compliance With MACT Standards,'' March 2004, Chapter 23.9, 
for more discussion.
1. What Are the Components of the Total Metal Emissions Limitations?
    This total metal emission limitation would regulate all nonmercury 
metal HAP with separate semivolatile HAP metal and low volatile HAP 
metal emission limits. Each semivolatile and low volatile metal limit 
would have separate MACT components that would control and limit 
enumerated and nonenumerated metal HAP emissions that are attributable 
to: (1) Hazardous waste feedstreams; (2) nonhazardous waste, non-fuel 
feedstreams (e.g., cement kiln raw material); and (3) nonhazardous 
waste fuels (e.g., coal). Some of these components may or may not apply 
depending on the source category. Each semivolatile and low volatile 
metal component is converted to a mass emission limitation, and each 
source's resultant total metal emissions would be limited to the 
summation of each of the applicable components. We describe these MACT 
components below.
    a. Energy Recovery Units: Allowable Enumerated Semivolatile and Low 
Volatile Metal Emissions Attributable to the Hazardous Waste. This 
first component limits enumerated metal emissions attributable to 
hazardous waste feedstreams from energy recovery units, i.e., liquid 
boilers, cement kilns, and lightweight aggregate kilns, and is 
equivalent to the enumerated semivolatile and low volatile metal mass 
emission rate that would be allowed by today's proposed standards. Each 
source's allowable mass emission rate limit for this component would be 
equivalent to its associated hazardous waste thermal feed rate 
(expressed as million Btu hazardous waste per hour) multiplied by the 
proposed semivolatile and low volatile metal thermal emission standard.
    b. Solid Fuel-Fired Boilers and Incinerators: Allowable Enumerated 
Semivolatile and Low Volatile Metal Emissions Attributable to All 
Feedstreams. This second component applies only to solid fuel-fired 
boilers and incinerators, and limits enumerated

[[Page 21334]]

metal mass emissions attributable to all feedstreams, i.e., hazardous 
waste, nonhazardous waste, and nonhazardous waste fuels. This component 
limit is equivalent to the enumerated semivolatile and low volatile 
metal mass emission rate that would be allowed by today's proposed 
standards. Today's proposed standards for incinerators and solid-fuel-
fired boilers limits total emissions from all feedstreams, and are 
expressed as stack gas concentration limits. Each source's allowable 
mass emission rate limit for this component would be equivalent to its 
gas flowrate multiplied by the proposed standard.
    c. All Source Categories: Allowable Nonenumerated Semivolatile and 
Low Volatile Metal Emissions Attributable to the Hazardous Waste. This 
third component limits nonenumerated semivolatile and low volatile 
metal emissions attributable to hazardous waste feedstreams, and is 
applicable to all source categories. We currently do not have 
sufficient data to calculate a MACT emission limitation for 
nonenumerated metals in the hazardous waste. As a result, sources 
complying with this alternative would be required to collect three 
years of nonenumerated semivolatile and low volatile metal hazardous 
waste feed control concentrations.\248\ Incinerators and solid fuel-
fired boilers would be required to collect hazardous waste maximum 
theoretical emissions concentrations, and energy recovery units would 
be required to collect three years of hazardous waste thermal feed 
concentration data for these metal groups.\249\ Each incinerator and 
solid fuel-fired boiler's allowable semivolatile and low volatile metal 
mass emission rate for this component would be equivalent to its 
associated three year average hazardous waste maximum theoretical 
emissions concentrations for each metal group multiplied by: (1) One 
minus the MACT system removal efficiency; and (2) its associated 
volumetric gas flow rate. Each energy recovery unit's allowable mass 
emission rate for this component would be equivalent to its associated 
three year average hazardous waste thermal feed concentration for each 
metal group multiplied by: (1) One minus the MACT system removal 
efficiency; and (2) its associated hazardous waste thermal feedrate 
(expressed as million Btu hazardous waste per hour). The MACT system 
removal efficiency that would be applied separately for semivolatile 
metals and low volatile metals would be determined as described in Part 
Two, Section VI.G.5 for each source category.
---------------------------------------------------------------------------

    \248\ We request comment on how such an approach would work for 
new sources, given that new sources may not have historical feed 
concentration data at the time they begin operations.
    \249\ Each source would be required to calculate its associated 
three year average nonenumerated metal hazardous waste 
concentrations for both semivolatile metals (selenium) and low 
volatile metals (antimony, cobalt, manganese, and nickel) expressed 
in either hazardous waste thermal concentrations, i.e., pounds per 
million Btus (for energy recovery units) or maximum theoretical 
emissions concentrations, i.e., pounds per dry standard cubic feet 
(for incinerators and solid fuel-fired boilers).
---------------------------------------------------------------------------

    d. Energy Recovery Units: Enumerated and Nonenumerated Metal HAP 
Emissions Attributable to Nonhazardous Waste Fuels. The fourth 
component limits enumerated and nonenumerated semivolatile and low 
volatile metal mass emissions attributable to nonhazardous waste fuels 
(e.g., coal) and is applicable to energy recovery units, i.e., cement 
kilns, lightweight aggregate kilns, and liquid fuel-fired boilers. 
Energy recovery units complying with this alternative would be required 
to collect three years of enumerated and nonenumerated semivolatile and 
low volatile metal nonhazardous waste fuel thermal feed concentration 
levels.\250\ Each source's allowable mass emission rate for this 
component would be equivalent to its associated three year average 
metal nonhazardous waste fuel thermal feed concentration for each metal 
group \251\ multiplied by: (1) One minus the MACT system removal 
efficiency for the specified metal group; and (2) its associated 
nonhazardous waste thermal feedrate.\252\ As discussed above, the MACT 
system removal efficiency that would be applied separately for 
semivolatile metals and low volatile metals would be determined as 
described in Part Two, Section VI.G.5 for each source category.
---------------------------------------------------------------------------

    \250\ Sources would not be required to collect three years of 
data if the nonhazardous waste fuels such as natural gas do not 
contain metal HAP.
    \251\ Each source would be required to calculate its associated 
three year average metal concentrations in their coal for both 
semivolatile metals (lead, cadmium, and selenium) and low volatile 
metals (arsenic, beryllium, chromium, antimony, cobalt, manganese, 
and nickel) expressed in pounds per million Btu of coal.
    \252\ This would be equivalent to a kiln's coal feedrate 
expressed in million Btus per hour.
---------------------------------------------------------------------------

    e. Incinerators and Solid Fuel-Fired Boilers: Nonenumerated Metal 
HAP Emissions Attributable to Nonhazardous Waste Fuels. The fifth 
component limits nonenumerated semivolatile and low volatile metal mass 
emissions attributable to nonhazardous waste fuels (e.g., coal, fuel 
oil) and is applicable to incinerators and solid fuel-fired boilers. 
Sources complying with this alternative would be required to collect 
three years of nonenumerated semivolatile and low volatile metal 
nonhazardous waste fuel thermal feed concentrations. Each source's 
allowable mass emission rate for this component would be equivalent to 
its associated three year average metal nonhazardous waste fuel thermal 
feed concentration for each metal group \253\ multiplied by: (1) One 
minus the MACT system removal efficiency for the specified metal group; 
and (2) its associated nonhazardous waste fuel thermal feedrate 
(expressed as million btu per hour). As discussed above, the MACT 
system removal efficiency that would be applied separately for 
semivolatile metals and low volatile metals would be determined as 
described in Part Two, Section VI.G.5 for each source category.
---------------------------------------------------------------------------

    \253\ Each source would be required to calculate its associated 
three year average nonenumerated metal concentrations in their 
nonhazardous waste fuel for both semivolatile metals (selenium) and 
low volatile metals (antimony, cobalt, manganese, and nickel) 
expressed in pounds per million Btu.
---------------------------------------------------------------------------

    f. Incinerators and Solid Fuel-Fired Boilers: Nonenumerated Metal 
HAP Emissions Attributable to Nonfuel Nonhazardous Waste. The sixth 
component limits nonenumerated metal HAP emissions attributable to 
nonfuel nonhazardous waste feedstreams from incinerators and solid 
fuel-fired boilers. Sources complying with this alternative would be 
required to collect three years of nonenumerated semivolatile and low 
volatile metal nonfuel nonhazardous waste feedstream concentration 
data, expressed as mass of metal fed in its nonfuel nonhazardous waste 
feedstream per total thermal input into the combustor. Each source's 
allowable mass emission rate for this component would be equivalent to 
its associated three year average metal nonfuel nonhazardous waste 
thermal feed concentration for each metal group \254\ multiplied by: 
(1) One minus the MACT system removal efficiency for the specified 
metal group; and (2) its associated total thermal feedrate (expressed 
as million Btus per hour). As discussed above, the MACT system removal 
efficiency that would be applied separately for semivolatile metals and 
low volatile metals would be determined as described in Part Two, 
Section VI.G.5 for each source category.
---------------------------------------------------------------------------

    \254\ Each source would be required to calculate its associated 
three year average nonenumerated metal thermal feed concentrations 
in their nonfuel nonhazardous waste feedstreams for both 
semivolatile metals (selenium) and low volatile metals (antimony, 
cobalt, manganese, and nickel) expressed in pounds per million Btu.
---------------------------------------------------------------------------

    g. Cement Kilns and Lightweight Aggregate Kilns: Enumerated and 
Nonenumerated Metal HAP Emissions Attributable to Raw Materials. The

[[Page 21335]]

seventh component limits enumerated and nonenumerated metal HAP 
emissions attributable to raw material from cement kilns and 
lightweight aggregate kilns. Cement kilns and lightweight aggregate 
kilns complying with this alternative would be required to collect 
three years of enumerated and nonenumerated semivolatile and low 
volatile metal raw material feed concentration data, expressed as mass 
of metal fed in raw material per total thermal input into the 
kiln.\255\ Each cement kiln and lightweight aggregate kiln's allowable 
mass emission rate for this component would be equivalent to its 
associated three year average metal raw material thermal feed 
concentration for each metal group \256\ multiplied by: (1) one minus 
the MACT system removal efficiency for the specified metal group; and 
(2) its associated total thermal feedrate. As discussed above, the MACT 
system removal efficiency that would be applied separately for 
semivolatile metals and low volatile metals would be determined as 
described in Part Two, Section VI.G.5 for each source category.
---------------------------------------------------------------------------

    \255\ Total thermal input to kiln would include both hazardous 
and nonhazardous fuel thermal input.
    \256\ Each source would be required to calculate its associated 
three year average metal thermal feed concentrations in their raw 
material for both semivolatile metals (lead, cadmium, and selenium) 
and low volatile metals (arsenic, beryllium, chromium, antimony, 
cobalt, manganese, and nickel) expressed in pounds per million Btus.
---------------------------------------------------------------------------

2. Would Sources Still Be Required To Comply With a Particulate Matter 
Standard if They Comply With This Alternative?
    As previously discussed in Part Two, Section VI.F, we conclude that 
today's proposed floor levels can be no higher than the interim 
standards because all sources, not just the best performing sources, 
are achieving the interim standards. It is not clear whether this 
alternative total metal emission limitation is less stringent than the 
current interim particulate matter standard for incinerators, cement 
kilns, and lightweight aggregate kilns.\257\ As a result, incinerators, 
cement kilns, and lightweight aggregate kilns complying with this 
alternative would also be required to comply with the interim standard 
for particulate matter. Liquid and solid fuel-fired boilers complying 
with this alternative would remain subject to the RCRA particulate 
matter standard of 0.08 gr/dscf pursuant to Sec.  264.343(c).\258\
---------------------------------------------------------------------------

    \257\ There is not a direct correlation between particulate 
matter emissions and metal emissions given that metal emission 
levels are both a function of feed control and particulate matter 
control.
    \258\ As previously discussed, this is because Clean Air Act 
standards can supplant RCRA standards only when the CAA standard is 
sufficiently protective of human health and the environment to make 
the RCRA standard duplicative (within the meaning of RCRA section 
1006 (b) (3)).
---------------------------------------------------------------------------

3. How Would Sources Demonstrate Compliance With This Alternative?
    Sources complying with this alternative would be required to 
calculate its site-specific semivolatile and low volatile metal mass 
emission rate limitation as described above. Each source's emission 
limitation would not only be a function of its average three years of 
metal concentration data collected, but also would be a function of 
either its gas flowrate (for incinerators and solid fuel fired 
boilers), hazardous waste thermal firing rate (for cement kilns, 
lightweight aggregate kilns, and liquid fuel-fired boilers), and total 
thermal input rate (for all sources). As a result each source's mass 
emission limitation would vary over time as the dependent variables 
change (e.g., a cement kiln's allowable mass emission limitation would 
increase if its hazardous waste thermal firing rate increases).
    Sources would demonstrate compliance with these site-specific metal 
emission rate limitations during its comprehensive performance test and 
would establish operating parameter limits on its air pollution control 
device to ensure that the source achieves the metal system removal 
efficiency that was demonstrated during the test during normal day-to-
day operations. Sources would then establish total metal feedrate 
limits that would assure compliance with this site-specific metal 
emission limitation. Given that these metal emission limitations may 
vary over time, we request comment as to whether these emission 
limitations (and associated feedrate operating limits) should be 
instantaneous limits based on each source's current operating levels 
(e.g., hazardous waste thermal input rate for energy recovery units, or 
gas flowrate for incinerators), or rather 12 hour rolling average 
limits that would be updated each minute.

XIX. What Are the Proposed RCRA State Authorization and CAA Delegation 
Requirements?

A. What Is the Authority for This Rule?

    Today's rule amends the promulgated standards located at 40 CFR 
part 63, subpart EEE. It amends the standards for the Phase I source 
categories--incinerators, cement kilns, and lightweight aggregate kilns 
that burn hazardous waste, and it also amends subpart EEE to establish 
MACT standards for the Phase II source categories--boilers and 
hydrochloric acid production furnaces that burn hazardous waste. 
Additionally, this rule amends several RCRA regulations located in 40 
CFR part 270 to reflect changes in applicability, addition of a new 
permit modification procedure and additions related site-specific risk 
assessments and permitting.
1. How Is This Rule Delegated Under the CAA?
    Consistent with the September 1999 rule, we recommend that state, 
local, and tribal (S/L/T) air pollution control agencies apply for 
delegation of this subpart (and all NESHAP) under section 112(l) of the 
CAA, if they have not done so already, so that they can exercise 
delegable authorities for the final Phase I Replacement standards and 
Phase II standards. Delegable authorities are the discretionary 
activities, such as approving changes to the reporting schedule, that 
are part of each NESHAP. EPA retains some of those authorities, but 
allows most to be implemented by those S/L/T agencies who accept 
straight delegation of the NESHAP; in this case, subpart EEE. The 
delegable authorities, those that can and cannot be delegated, are 
described in section 63.1214 of this subpart. (For more information on 
delegation of part 63 provisions, see 65 FR 55810-55846.) All major 
sources of air pollutants, such as all sources subject to this subpart, 
must have a title V operating permit which would contain all applicable 
requirements, including those for this subpart. (For more information, 
please see 40 CFR part 70.) While S/L/T agencies can implement and 
enforce MACT standards through their approved title V programs, 
approval of title V programs alone do not allow S/L/T authorities to be 
the primary enforcement authority and they cannot exercise delegable 
provisions' authorities. An approved title V program means that S/L/T 
agencies commit to incorporating all MACT standards into title V 
permits as permit conditions and to enforcing all the terms and 
conditions of the permit.\259\ Having an approved title V program, for

[[Page 21336]]

instance, does not automatically allow S/L/T agencies to approve test 
plans, requests for (minor and intermediate) changes to monitoring, 
performance test waivers, document notifications, or other Category I 
Authorities (see 40 CFR 63.91(g)(1)(i)). For those S/L/T agencies who 
have been previously delegated authority for the MACT standards under 
40 CFR part 63 subpart EEE, we encourage you to request approval of the 
revisions to emission standards and various other compliance 
requirements of today's proposal when promulgated.
---------------------------------------------------------------------------

    \259\ Accordingly, S/L/T agencies are required to reopen 
existing title V permits that have 3 or more years remaining in the 
permit term to include the promulgated standards. If there are less 
than 3 years remaining, S/L/T agencies may wait until renewal to 
incorporate the standards. Provided that a source is not required to 
reopen its title V permit, it must still fully comply with the 
promulgated standards (40 CFR 70.7(f)(1)(i)).
---------------------------------------------------------------------------

B. Are There Any Changes to the CAA Delegation Requirements for Phase I 
Sources?

    With regard to CAA delegation requirements for Phase I sources, we 
intend to clarify which provisions in 40 CFR part 63 subpart EEE are 
delegable and those that are not in today's Notice of proposed 
rulemaking. We recently published a final rule, Clarifications to 
Existing National Emissions Standards for Hazardous Air Pollutants 
Delegations' Provisions on June 23, 2003 (see 68 FR 37334), that 
clarifies and streamlines delegable provisions for each existing 
NESHAP. Prior to finalization of this rule, many permitting authorities 
and sources alike were left to interpret which Category I authorities 
were delegable according to provisions specific to one NESHAP versus 
another. In light of this final rule, which outlines the non-delegable 
provisions for subpart EEE, some confusion remains today as to which 
actions can be taken by a delegated S/L/T agency. Therefore, we intend 
to clarify specific actions in subpart EEE that can or cannot be taken 
by permitting agencies who have received delegation under 112(l) of the 
CAA for subpart EEE.
    Sections 63.91(g)(1)(i) and (g)(2)(i) list authorities that are 
generally delegable to S/L/T agencies and those that are not, 
respectively. These apply to all NESHAP. Similar information contained 
in Sec.  63.1214 explains that some of the discretionary authorities, 
such as approval of alternative reporting schedules, under subpart EEE, 
can be implemented and enforced by a delegated authority. It also lists 
the authorities that are retained by EPA and are not delegable to S/L/T 
agencies even if they have received delegation for subpart EEE. These 
non-delegable authorities are: (1) Approval of alternatives to 
requirements in Sec. Sec.  63.1200, 63.1203 through 63.1205, and 
63.1206(a); (2) approval of major alternatives to test methods under 
Sec.  63.7(e)(2)(ii) and (f); (3) approval of major alternatives to 
monitoring under Sec.  63.8(f) and; (4) approval of major alternatives 
to recordkeeping and reporting under Sec.  63.10(f). It is important to 
note that if the alternatives mentioned in items (2) through (4) are 
determined to be minor or intermediate according to the definitions in 
Sec.  63.90(a), then they are considered delegable and can be approved 
by a S/L/T agency who has been granted authority for subpart EEE.\260\ 
To aid in the determination of whether a request is major, 
intermediate, or minor, we recommend that you consult the September 14, 
2000 final rule, Hazardous Air Pollutants: Amendments to the Approval 
of State Programs and Delegation of Federal Authorities (65 FR 55810). 
The preamble to this rule provides examples, as well as the regulatory 
definitions as they exist today in 40 CFR 63.90(a). Additionally, you 
may consult a guidance document entitled, How to Review and Issue Clean 
Air Act Applicability Determinations and Alternative Monitoring (EPA 
305-B-99-004, February 1999).
---------------------------------------------------------------------------

    \260\ EPA Regions may choose whether they will or will not 
delegate authority to S/L/T agencies to approve minor and 
intermediate changes.
---------------------------------------------------------------------------

    While Sec.  63.1214(c) and Sec.  63.90(a) provide which authorities 
are not delegable for subpart EEE sources and define degrees of 
changes, they may not be clear in certain applications. We will address 
specific sections in subpart EEE, through the following preamble 
discussion and through regulatory amendments, where we believe there is 
a need for clarity based upon our experiences with the implementation 
of the Phase I standards thus far. Also, there are some alternatives in 
subpart EEE that were inadvertently left out of Sec.  63.1214(c) which 
we are adding through this Notice of proposed rulemaking.
    Beginning with test methods, major alternatives are not delegable. 
(See 40 CFR 63.90(a) for definitions of major, intermediate, and minor 
changes to test methods.) We noted in Sec.  63.1214(c)(2) that major 
alternatives to the test methods as addressed in the general provisions 
at Sec.  63.7(e)(2)(ii) and (f) were not delegable, however, we did not 
specifically include test methods relevant to subpart EEE. Section 
63.1208(b) specifies the test methods sources must use to determine 
compliance with emission standards in subpart EEE. This section is 
delegable in its entirety to S/L/T agencies who have been delegated 
authority for subpart EEE, as long as the request is not a major 
change. Additionally, the CEMS required in Sec.  63.1209(a)(1), 
although a monitoring requirement, is considered to be a test method 
since it serves as the benchmark measurement method for demonstrating 
compliance with emission standards. The authority to approve changes to 
the CEMS-related requirements is also delegable to S/L/T agencies as 
long as the request is not a major change. To summarize, if a source 
proposes a major change to a test method specified in Sec. Sec.  
63.1208(b) and 63.1209(a)(1), it must send the request to the 
appropriate EPA Region and EPA's Office of Air Quality Planning and 
Standards,\261\ since major changes to test methods are not delegable. 
We are adding Sec. Sec.  63.1208(b) and 63.1209(a)(1), to the 
authorities in Sec.  63.1214(c)(2) that are not delegable for major 
changes.
---------------------------------------------------------------------------

    \261\ Send requests to: Conniesue B. Oldham, Ph.D., Group 
Leader, Source Measurement Technology Group (D205-02), Office of Air 
Quality Planning and Standards, U.S. Environmental Protection 
Agency, Research Triangle Park, NC 27711.
---------------------------------------------------------------------------

    Consistent with the major alternatives to test methods, major 
alternatives to monitoring are not delegable. (See 40 CFR 63.90(a) for 
definitions of major, intermediate, and minor changes to test methods.) 
We noted in Sec.  63.1214(c)(2) that major alternatives to monitoring 
as addressed in the general provisions in Sec.  63.8(f) were not 
delegable, but we did not specifically address the relevant monitoring 
requirements in subpart EEE. Section 63.1209 specifies the monitoring 
requirements sources must use to determine compliance with emission 
standards in EEE. Depending upon the pollutant to be monitored, either 
a CEMS or COMS is required.
    Before discussing whether changes to monitoring in subpart EEE are 
delegable, it is important first to review how requests for changes to 
monitoring are handled under the general provisions of Sec.  63.8(f). 
In general, requests for alternative monitoring follow the same 
approach, with respect to delegation authority, as requests for 
alternative test methods discussed above; requests that are defined as 
major should be sent to the appropriate EPA Region and requests that 
are intermediate or minor should be sent to the delegated S/L/T agency. 
A request to use other monitoring in lieu of a CEMS is always 
considered a major change. However, if a source proposes to use a CEMS 
in lieu of an operating parameter, the request may be considered an 
intermediate change, so long as the CEMS to be used is regarded as a 
``proven technology'' and could be submitted to a S/L/T agency for 
approval. The rationale for this is that the use of a CEMS, rather than 
monitoring via an operating parameter, provides a better measure of 
compliance

[[Page 21337]]

and thus, we want to encourage the use of CEMS when possible. While we 
want to encourage the use of CEMS, we recognize that S/L/T agencies may 
not always have the technical resources to review these applications, 
particularly when there are no federally promulgated performance 
specifications for the CEMS. In such cases, we expect that the S/L/T 
agency will rely on EPA Regions for approval.
    In subpart EEE, Sec.  63.1209, there are two alternative approaches 
to monitoring that sources may use. One is located at Sec.  
63.1209(a)(5), Petitions to use CEMS for other standards, and the other 
is at Sec.  63.1209(g)(1), Alternative monitoring requirements other 
than continuous emissions monitoring systems. Section 63.1209(a)(5) 
allows sources to request to use CEMS to monitor particulate matter, 
mercury, semivolatile metals, low volatile metals, and/or hydrochloric 
acid/chlorine gas in lieu of compliance with operating parameter 
limits. In these cases, a source would be monitoring the pollutant of 
concern and comparing the emissions measurements directly against an 
emission limitation rather than comparing the measurements to an 
operating parameter. We consider a request under Sec.  63.1209(a)(5) to 
be a major change to monitoring and consequently, it is not delegable. 
We classify Sec.  63.1209(a)(5) to be a major change (rather than an 
intermediate change which can be delegable) mainly because we have not 
yet promulgated Performance Specifications for the CEMS that may be 
used. In other words, it could be argued that these CEMS do not yet 
qualify as fully ``proven technology''. We understand that it could be 
argued either way, but for the reasons discussed in the previous 
paragraph and as an added measure of consistency, requests to use CEMS 
in lieu of operating parameters should be submitted to the EPA Region 
for approval. Therefore, we are adding Sec.  63.1209(a)(5) to the 
authorities in Sec.  63.1214(c)(2) that are not delegable for major 
changes.
    The other alternative monitoring provision, Sec.  63.1209(g)(1), 
allows sources to use alternative monitoring methods, with the 
exception of the standards that must be monitored with a CEMS, and to 
request a waiver of an operating parameter limit. Section 63.1209(g)(1) 
applies to requests for alternative parameter monitoring that involve 
the use of a different detector (i.e., thermocouple, pressure 
transducer, or flow meter), a different monitoring location, a 
different method as recommended by the manufacturer, or a different 
averaging period that is more stringent than the applicable standard. 
For example, sources equipped with wet scrubbers are required to 
establish a minimum pressure drop limit to assure adequate contact 
between the gas and liquid. A source may petition to have this 
monitoring requirement waived if the manufacturer does not recommend 
pressure drop as a critical control parameter that affects the unit's 
operating efficiency. Depending upon the type of wet scrubber, an 
appropriate minimum limit may be specified for steam injection rate, 
disk spin rate, or a maximum temperature limit on liquid and flue gas, 
rather than pressure drop. Also, sources could request more stringent 
averaging periods in order to ``mirror'' the averaging periods required 
under RCRA. This may facilitate an easier transition from RCRA to MACT 
during the time period sources may need to comply with both sets of 
requirements. Since we do not consider these changes to be major, 
requests under Sec.  63.1209(g)(1) should be sent to the delegated S/L/
T agency for approval. Accordingly, we are amending the language in 
Sec.  63.1209(g)(1) to specify that a source may submit an application 
to the Administrator or a State with an approved Title V program. Also, 
we are revising the title under Sec.  63.1209(g)(1) so that it is more 
specific regarding its intended use.
    Lastly, major alternatives to recordkeeping and reporting also are 
not delegable. (See 40 CFR 63.90(a) for definitions of major, 
intermediate, and minor changes to test methods.) We noted in Sec.  
63.1214(c)(2) that major alternatives to the general provisions of 
Sec.  63.10(f) were not delegable, but we did not specifically address 
any relevant recordkeeping and reporting requirements in subpart EEE. 
Section 63.1211 specifies the recordkeeping and reporting requirements 
sources must comply with in subpart EEE. This section is delegable in 
its entirety to S/L/T agencies who have been delegated authority to 
implement and enforce subpart EEE, as long as the request is not a 
major change. It is worthwhile to note that paragraph (e), Data 
compression, may be incorrectly interpreted as a major change itself to 
the recordkeeping and reporting requirements, because it appears as 
though there are no criteria to define fluctuation or data compression 
limits. However, this is not the case. In the preamble to the September 
1999 final rule (see 64 FR 52961 and 52962), we provided guidance for 
preparing a request to use data compression techniques and recommended 
fluctuation and data compression limits. This guidance was not affected 
by the court's vacatur of portions of this rule, so it remains in 
effect. Consequently, this allows permitting authorities to be 
consistent in their evaluation of requests. We view paragraph (e) to be 
a minor change itself and so a written request to use data compression 
techniques can be submitted to a delegated S/L/T agency. We are adding 
Sec.  63.1211(a)--(d) to the authorities in Sec.  63.1214(c)(2) that 
are not delegable for major changes.
    In addition to the clarifications and amendments addressed above, 
there are two important delegation issues we would like to emphasize. 
The first is simply to remind sources and permitting authorities alike 
that, if a provision in this subpart specifies that you may petition or 
request that the ``Administrator or State with an approved Title V 
program * * *,'' then a state that has not been delegated for that 
requirement, but has an approved Title V program, does have the 
authority to approve or disapprove the request. For instance, Sec.  
63.6(i)(1) and Sec.  63.1213(a) both specify that the ``Administrator 
(or a State with an approved permit program)'' can grant a compliance 
extension request. The second is that EPA Regions can decide whether or 
not to delegate the authority to approve intermediate changes to state 
and local agencies. In some cases, a state may have received delegation 
to approve only minor changes. Where there is uncertainty, we recommend 
that sources try to determine if a request is major, intermediate, or 
minor based on the definitions in 40 CFR 63.90(a), and then consult 
with their S/L/T agency and/or EPA Region to determine where to submit 
the request. Or, sources may submit requests to the S/L/T agency or EPA 
Region who will then determine where it should go for approval.

C. What Are the Proposed CAA Delegation Requirements for Phase II 
Sources?

    With respect to CAA delegation requirements for Phase II sources, 
they are the same as those for Phase I sources. Since both Phase I and 
Phase II MACT standards are located in the same subpart, EEE, the same 
delegation provisions apply to both. Generally speaking, authority to 
approve alternatives to standards or major changes to test methods, 
monitoring, and recordkeeping and reporting are not delegated to S/L/T 
agencies. Authority to approve intermediate and minor changes to test 
methods, monitoring, and recordkeeping and reporting are delegated to 
S/L/T agencies who have been delegated authority to implement

[[Page 21338]]

subpart EEE. All other subpart EEE implementation requirements may be 
handled by the delegated S/L/T agency. For specific information, please 
refer to the previous section, A.1. What are the clarifications and 
changes to CAA delegable authorities for this rule?
    How Would States Become Authorized under RCRA for this Rule? Under 
section 3006 of RCRA, EPA may authorize qualified states to administer 
their own hazardous waste programs in lieu of the federal program 
within the state. Following authorization, EPA retains enforcement 
authority under sections 3008, 3013, and 7003 of RCRA, although 
authorized states have primary enforcement responsibility. The 
standards and requirements for state authorization are found at 40 CFR 
part 271.
    Prior to enactment of the Hazardous and Solid Waste Amendments of 
1984 (HSWA), a State with final RCRA authorization administered its 
hazardous waste program entirely in lieu of EPA administering the 
federal program in that state. The federal requirements no longer 
applied in the authorized state, and EPA could not issue permits for 
any facilities in that state, since only the state was authorized to 
issue RCRA permits. When new, more stringent federal requirements were 
promulgated, the state was obligated to enact equivalent authorities 
within specified time frames. However, the new federal requirements did 
not take effect in an authorized state until the state adopted the 
federal requirements as state law.
    In contrast, under RCRA section 3006(g) (42 U.S.C. 6926(g)), which 
was added by HSWA, new requirements and prohibitions imposed under HSWA 
authority take effect in authorized states at the same time that they 
take effect in unauthorized states. EPA is directed by the statute to 
implement these requirements and prohibitions in authorized states, 
including the issuance of permits, until the state is granted 
authorization to do so. While states must still adopt HSWA related 
provisions as state law to retain final authorization, EPA implements 
the HSWA provisions in authorized states until the states do so.
    Authorized states are required to modify their programs only when 
EPA enacts federal requirements that are more stringent or broader in 
scope than existing federal requirements. RCRA section 3009 allows the 
states to impose standards more stringent than those in the federal 
program (see also 40 CFR 271.1). Therefore, authorized states may, but 
are not required to, adopt federal regulations, both HSWA and non-HSWA, 
that are considered less stringent than previous federal regulations.
    The amendments to the RCRA regulations proposed today in sections 
40 CFR 270.10, 270.22, 270.32, 270.42, 270.66, and 270.235 are 
considered to be either less stringent or equivalent to the existing 
Federal program. Thus, states are not required to modify their programs 
to adopt and seek authorization for these provisions, although we 
strongly encourage them to do so to facilitate the transition from the 
RCRA program to the CAA program and to promote national consistency. 
Additionally, EPA will not implement those provisions promulgated under 
HSWA authority that are not more stringent than the previous federal 
regulations in States that have been authorized for those previous 
federal provisions.
    The amendments in sections 40 CFR 270.22 and 270.66 in today's 
notice are proposed under the HSWA amendments to RCRA. Further, today's 
proposed amendment in 40 CFR 270.235 to apply this provision to solid 
and liquid fuel-fired boilers and HCL production furnaces, is proposed 
under HSWA statutory authority. The amendments to the RCRA regulations 
proposed today in sections 40 CFR 270.10 and 270.32 are proposed under 
both non-HSWA and HSWA authority, depending on the type of unit to 
which these amendments are applied (under HSWA authority if applied to 
BIFs or non-HSWA authority if applied to incinerators). Refer to Part 
Two, Section XVII.D.4 for a more detailed discussion of the 
implementing authorities for proposed regulations in 40 CFR 270.10 and 
270.32. The following RCRA sections, enacted as part of HSWA, apply to 
today's rule: 3004(o), 3004(q), and 3005(c)(3). As a part of HSWA, 
these RCRA provisions are federally enforceable in an authorized State 
until the necessary changes to a State's authorization are approved by 
us. See RCRA section 3006, 42 U.S.C. 6926. The Agency is adding these 
requirements to Table 1 in 271.1(j), which identifies rulemakings that 
are promulgated pursuant to HSWA.

Part Three: Proposed Revisions to Compliance Requirements

    In this section, we discuss proposed revisions to compliance 
requirements that may affect all hazardous waste combustors. We also 
request comment on whether we should make revisions to other compliance 
requirements, and explain why we conclude not to make revisions to 
other compliance requirements that we proposed (or requested comment 
on) previously.

I. Why Is EPA Proposing To Allow Phase I Sources To Conduct the Initial 
Performance Test To Comply With the Replacement Rules 12 Months After 
the Compliance Date?

    We propose to allow owners and operators of incinerators, cement 
kilns, and lightweight aggregate kilns to commence the initial 
comprehensive performance test to comply with the replacement standards 
proposed at Sec. Sec.  63.1219, 63.1220, and 63.1221 within 12 months 
of the compliance date rather than within six months of the compliance 
date. See proposed Sec.  63.1207(c)(3). Owners and operators of solid 
fuel-fired boilers, liquid fuel-fired boilers, and hydrochloric acid 
production furnaces, however, must commence the initial comprehensive 
performance test within six months of the compliance date.
    During development of the joint motion by petitioners to the United 
States Court of Appeals for the District of Columbia Circuit that 
resulted in the Agency promulgating the Interim Standards Rule on 
February 13, 2002,\262\ stakeholders representing owners and operators 
of incinerators, cement kilns, and lightweight aggregate kilns 
requested that we propose to allow them 12 months after the compliance 
date to commence the initial comprehensive performance test. These 
stakeholders request a 12 month window rather than the six month window 
currently required under Sec.  63.1207(c) to give them longer to 
amortize the cost of the comprehensive performance test demonstrating 
compliance with the Interim Standards before having to retest to 
demonstrate compliance with the replacement standards proposed 
today.\263\ We believe this request has merit and so are proposing to 
allow them to commence the initial comprehensive performance test 
within 12 months after the compliance date.\264\
---------------------------------------------------------------------------

    \262\ See discussion in Part One, Section I.B.1.
    \263\ These stakeholders assumed, correctly, that today's 
proposed replacement emission standards would be substantially more 
stringent than the current (September 1999 Final Rule) standards.
    \264\ Please note that this does not affect the compliance date. 
You must be in compliance with the replacement standards on the 
compliance date, and certify in the Documentation of Compliance that 
you have established operating parameter limits that you believe 
will ensure compliance with the standards. You must record the 
Documentation of Compliance in the operating record by the 
compliance date.

---------------------------------------------------------------------------

[[Page 21339]]

II. Why Is EPA Requesting Comment on Requirements Promulgated as 
Interim Standards or as Final Amendments?

    As discussed in Part One, Section I.B., EPA promulgated interim 
standards (called the Interim Standards Rule) on February 13, 2002 that 
amended compliance and implementation provisions of the September 1999 
Final Rule. The amended provisions were specified in a joint motion by 
petitioners to the United States Court of Appeals for the District of 
Columbia Circuit (the Court). Although petitioners agreed that the 
amendments should be promulgated (see 67 FR at 6794), petitioners 
requested that EPA reopen certain amended provisions for public 
comment.
    Also as discussed in Part One, Section I.B, EPA promulgated 
amendments (called Final Amendments) to the September 1999 Final Rule 
on February 14, 2002 that revised certain implementation and compliance 
requirements. These amendments were also specified in the joint motion 
to the Court, and petitioners requested that EPA reopen specific 
amended provisions for public comment.
    We discuss these provisions in this section, and reopen them for 
public comment. (We note, however, that we are not reopening for 
comment any RCRA rules, and are not soliciting comment on any aspect of 
those rules, or otherwise reconsidering or reexaming any such rules. 
Any references to RCRA rules in the discussion which follows is solely 
as an aid to readers.) Although we are not proposing additional 
revisions to these provisions, we may determine after review of public 
comments on the issues we raise that revisions are appropriate. If so, 
we would promulgate those amendments in the Replacement Rule.
    Although these provisions currently apply only to incinerators, 
cement kilns, and lightweight aggregate kilns, we are proposing today 
to apply them to boilers and hydrochloric acid production furnaces as 
well. (See Part Two, Sections XIII-XV.) Accordingly, any amendments to 
these requirements that we may promulgate would also apply to boilers 
and hydrochloric acid production furnaces.

A. Interim Standards Amendments to the Startup, Shutdown, and 
Malfunction Plan Requirements

    The September 1999 Final Rule required compliance with the emission 
standards and operating requirements at all times that hazardous waste 
is in the combustion system, including during startup, shutdown, and 
malfunctions. Industry stakeholders noted that requiring compliance 
with emission standards and operating requirements during startup, 
shutdown, and malfunctions is inconsistent with the General Provisions 
of subpart A, part 63, that apply to MACT sources (unless alternative 
requirements are prescribed for a source category). Stakeholders stated 
that it is inappropriate to penalize a source for exceeding emission 
standards and operating requirements during malfunctions because some 
exceedances are unavoidable and sources are already required to take 
corrective measures prescribed in the startup, shutdown, and 
malfunction plan (SSMP) to minimize emissions.
    In response to industry stakeholder concerns, the Interim Standards 
Rule amended the SSMP requirements to: (1) Exempt sources from the 
Subpart EEE emission standards and operating requirements during 
startup, shutdown, and malfunctions; (2) continue to subject sources to 
RCRA requirements during malfunctions, unless they comply with 
alternative MACT requirements including expanding the SSMP to minimize 
the frequency and severity of malfunctions, and submit the plan to the 
delegated CAA authority for review and approval \265\; (3) continue to 
subject sources that burn hazardous waste during startup and shutdown 
to RCRA requirements for startup and shutdown, unless they comply with 
alternative MACT requirements, and require sources to include waste 
feed restrictions and operating conditions and limits in the startup, 
shutdown, and malfunction plan; (4) require sources to include in the 
SSMP a requirement to comply with the automatic hazardous waste feed 
cutoff system during startup, shutdown, and malfunctions; and (5) make 
conforming revisions to the emergency safety vent opening requirements. 
See 67 FR at 6798-6802.
---------------------------------------------------------------------------

    \265\ These requirements are needed to minimize emissions of HAP 
during startup, shutdown, and malfunctions and, thus, help meet our 
RCRA mandate to ensure that emissions from hazardous waste 
combustors do not pose a hazard to human health and the environment. 
Sources may elect either to remain under RCRA control during these 
events or to comply under MACT with requirements to develop and 
implement a comprehensive and proactive startup, shutdown, and 
malfunction plan that is reviewed and approved by the delegated 
regulatory authority.
---------------------------------------------------------------------------

    In response to Sierra Club's request during development of the 
joint motion to the Court, we specifically request comment on the 
following issues. Notwithstanding the rationale for revising the 
September 1999 Final Rule to exempt sources from the subpart EEE 
emission standards and operating requirements during malfunctions, 
would it be appropriate to require compliance with those standards and 
operating requirements during malfunctions to ensure that owners and 
operators have an incentive to minimize the frequency and duration of 
malfunctions that result in exceedances of the standards or operating 
requirements. Given that most excess emissions would occur during 
startup, shutdown, and malfunctions, should the SSMP be submitted for 
review by the delegated regulatory authority and made available for 
public review under all options for controlling emissions during 
startup, shutdown, and malfunctions? Providing a mechanism for public 
review may help ensure that the SSMP is complete, proactive, and 
provides appropriate corrective measures.\266\ And finally, should the 
final rule clarify the definitions of startup, shutdown, and 
malfunctions to preclude, for example, an owner or operator incorrectly 
classifying an exceedance of an operating limit while hazardous waste 
remains in the combustion chamber as a malfunction when, in fact, the 
exceedance occurred because of a not infrequent event that could have 
been prevented by proper operation and maintenance of equipment?
---------------------------------------------------------------------------

    \266\ We also request comment on whether the startup, shutdown, 
and malfunction plan should be expanded beyond the scope required 
under Sec.  63.6(e)(3) (requiring appropriate corrective measures in 
reaction to a malfunction) to address specific, proactive measures 
that the owner and operator have considered and are taking to 
minimize the frequency and severity of malfunctions.
---------------------------------------------------------------------------

B. Interim Standards Amendments to the Compliance Requirements for 
Ionizing Wet Scrubbers

    The September 1999 Final Rule required sources to establish a limit 
on minimum total power to an ionizing wet scrubber. The Interim 
Standards Rule deleted that requirement to conform with the 
requirements for electrostatic precipitators given that an ionizing wet 
scrubber is essentially an ESP integrated with a packed bed scrubber. 
See 67 FR at 6802-03.\267\ In lieu of establishing a limit on the 
minimum total power requirement to an ionizing wet scrubber, sources 
and delegated CAA authorities will use the alternative monitoring 
provisions of Sec.  63.1209(g) to identify appropriate controls for an 
ionizing wet scrubber on a site-specific basis. This is

[[Page 21340]]

the same approach that is used for electrostatic precipitators.
---------------------------------------------------------------------------

    \267\ EPA voluntarily vacated operating parameter limits for 
electrostatic precipitators (and fabric filters) on May 14, 2001. 
See 66 FR at 24272. Until new operating parameter limits are 
promulgated, sources and delegated CAA authorities will use Sec.  
63.1209(g) to establish operating parameter limits for electrostatic 
precipitators (and fabric filters) on a site-specific basis.
---------------------------------------------------------------------------

    Please note that we are requesting comment today on compliance 
requirements for electrostatic precipitators and fabric filters. In 
that discussion (see Section III.I below), we explain that we are 
proposing to apply the same compliance requirements to both 
electrostatic precipitators and ionizing wet scrubbers.

C. Why Is EPA Requesting Comment on the Fugitive Emission Requirements?

    The September 1999 Final Rule required sources to control 
combustion system leaks by either: (1) Keeping the combustion zone 
sealed; (2) maintaining the maximum combustion zone pressure lower than 
ambient pressure using an instantaneous monitor; or (3) using an 
alternative means to provide control of system leaks equivalent to 
maintaining the maximum combustion zone pressure lower than ambient. 
After publication of the September 1999 Final Rule, stakeholders 
expressed concern that the option to maintain combustion zone pressure 
lower than ambient pressure (option 2 above) could result in overly 
prescriptive requirements. Stakeholders believed that this regulatory 
language could be interpreted to require sources to monitor and record 
combustion zone pressure at a frequency of every 50 milliseconds. 
Stakeholders also requested that we clarify that combustion system 
leaks refers to fugitive emissions resulting from the combustion of 
hazardous waste, and not fugitive emissions that originate from 
nonhazardous process streams.
    In response to these concerns, we proposed amendments to the 
combustion system leak provisions on July 3, 2001. See 66 FR at 35132. 
We promulgated several revisions in the Final Amendments Rule after 
considering stakeholder comments. See 67 FR at 6973.
    The amended provisions that we are reopening for public comment 
today are discussed below. First, we amended the definition of an 
instantaneous pressure monitor to better clarify that the intent of the 
combustion system leak requirements is to prevent fugitive emissions 
from the combustion of hazardous waste rather than from nonhazardous 
feedstreams. The revised definition also clarifies that instantaneous 
pressure monitors must detect and record pressure at a frequency 
adequate to detect combustion system leak events, as determined on a 
site-specific basis. See Sec.  63.1201(a) and Sec.  63.1209(p). Second, 
we added a provision that requires sources to specify the method used 
to control combustion system leaks in the performance test workplan and 
Notification of Compliance. See Sec.  63.1206(c)(5)(ii). Finally, in 
response to numerous comments, we added a provision that will allow 
sources, upon prior written approval of the Administrator, to use other 
techniques that can be demonstrated to prevent fugitive emissions 
without the use of instantaneous pressure limits. See Sec.  
63.1206(c)(5)(i)(D).
    The provision allowing sources, upon prior written approval, to use 
other techniques that are demonstrated to prevent fugitive emissions 
without the use of instantaneous pressure limits was the most 
controversial. Specifically, some stakeholders believe this revised 
regulatory language is inappropriate because it suggests sources can 
sustain a positive pressure event and still prevent fugitive emissions. 
We believe that all positive pressure events do not necessarily result 
in fugitive emissions. As discussed in detail in the Final Amendments 
Rule, there are state-of-the-art rotary kiln seal designs (such as 
shrouded and pressurized seals) which are capable of handling positive 
pressures without fugitive releases. However, we believe these kilns 
are highly unusual, and that other conventional rotary kilns used in 
the hazardous waste combustion industry may not have seals which are 
designed for such positive pressure operation. In fact, we believe 
that, for most rotary kilns in use today, positive pressure events can 
result in fugitive releases. The level of such fugitive releases will 
be dependent on factors including the magnitude and duration of the 
pressure excursion and the design and operation of the kiln.
    Furthermore, one commenter recommends that sources should be 
allowed to petition the regulatory official to use an alternative 
approach, i.e., an approach that does not require instantaneous 
pressure limits, only if they meet specific combustor design criteria. 
For example, it may be appropriate to apply this provision only to 
sources that we know are designed in manner that would not necessitate 
use of instantaneous pressure limits to prevent fugitive emissions 
(e.g., kilns with multiple graphite seals with pressurized chambers 
between the seals to prevent out-leakage, or overlapping spring plate 
seals to form an air seal). We request comment on whether this 
specificity is necessary, or whether it is more appropriate to 
determine this on a site-specific basis (as is currently required). We 
also request comment on whether all the previously discussed combustion 
system leak regulatory revisions are appropriate.

D. Why Is EPA Requesting Comment on Bag Leak Detector Sensitivity?

    The September 1999 Final Rule required sources equipped with fabric 
filters to install a bag leak detection system where the detector has 
the capability to detect PM emissions at concentrations of 1.0 
milligrams per actual cubic meter, or less. In response to industry 
stakeholder concerns that a detector need not be able to detect levels 
as low as 1.0 mg/acfm to detect subtle changes in baseline, normal 
emissions of PM, we proposed in the July 3, 2001, proposed rule (66 FR 
at 35134-35) to allow sources to use detectors with less sensitivity 
provided that the detector could detect subtle increases in normal 
emissions (e.g., caused by pinhole leaks in the bags). The stakeholders 
noted that sources equipped with well designed and operated fabric 
filters can have normal, baseline emissions well above 1.0 mg/acfm and 
be in compliance with the particulate matter emission standards. 
Stakeholders recommended that we revise the bag leak detection 
requirements to explicitly allow detectors with lower sensitivity in 
lieu of source's having to petition the delegated regulatory authority 
under the alternative monitoring provisions of Sec.  63.1209(g)(1) to 
receive case-by-case approval. All commenters on the proposed amendment 
supported the revision, and we finalized the amendment in the February 
14, 2002, Final Amendments. See 67 FR at 6981.
    In response to a petitioner's request during development of the 
joint motion to the Court, however, we specifically request additional 
comment on whether allowing detectors that have a level of detection 
that is higher than 1.0 mg/acfm will enable the detector to detect 
subtle increases in normal emissions. The petitioner is concerned that 
a detector with a level of detection higher than 1.0 mg/acfm may not 
have the same sensitivity as a detector that can detect PM at 1.0 mg/
acfm. Thus, petitioner is concerned that the less sensitive detector 
may not be able to detect subtle increases in PM emissions due to bag 
degredation as readily as a detector that can detect at 1.0 mg/acfm. We 
specifically request comment on this issue.
    We reopen this issue for comment without prejudice to the existing 
regulations which allow for less sensitive bag leak detectors. You may 
use less sensitive bag leak detectors until the compliance date for any 
change we may make in the final rule.

[[Page 21341]]

E. Final Amendments Waiving Operating Parameter Limits During Testing 
Without an Approved Test Plan

    The September 1999 Final Rule waived operating parameter limits 
during subsequent performance testing under an approved performance 
test plan. In response to stakeholder concerns, we addressed two issues 
in the Final Amendments: (1) Applicability of operating parameter 
limits, established in the Documentation of Compliance, during an 
initial performance test conducted without an approved test plan; and 
(2) applicability of operating parameter limits, established in the 
Notification of Compliance, during subsequent performance tests 
conducted without an approved test plan. See 67 FR at 6978.
    Regarding the initial performance test, we explained that a source 
can revise the operating parameter limits specified in the 
Documentation of Compliance at any time based on supporting 
information. This information would also be included in the performance 
test plan to support deviating from the operating limits established in 
the previous Documentation of Compliance. Given that sources operate 
after the compliance date until the Notification of Compliance is 
submitted under operating limits established in the Documentation of 
Compliance, and that the technical support for the operating limits 
established in the Documentation of Compliance is the same as would be 
included in the test plan, it is appropriate to allow initial 
performance testing and associated pretesting without an approved test 
plan.
    Regarding subsequent performance testing, we amended the rule to 
waive the operating parameter limits during performance testing and 
associated pretesting even when testing without an approved test plan. 
We reasoned that stack emissions data obtained during the testing would 
document whether the source maintained compliance with the emission 
standards. (Please note that during testing, including pretesting, 
stack emissions must be documented for any emissions standard for which 
the source waives an operating parameter limit.) Absent approval of the 
test plan, documentation of potential violation of an emission standard 
is nonetheless an ample incentive to operate within the emission 
standards.
    In response to a petitioner's request during development of the 
joint motion to the Court, however, we request comment on whether 
documentation of stack emissions during subsequent performance testing 
and associated pretesting is adequate to ensure compliance with the 
emission standards absent an approved test plan.

III. Why Is EPA Requesting Comment on Issues and Amendments That Were 
Previously Proposed?

    In a July 3, 2001, proposed rule, EPA proposed several revisions to 
implementation and compliance requirements, and discussed other 
implementation and compliance issues. See 66 FR 35126. We promulgated 
several of those amendments in the February 14, 2002, Final Amendments 
Rule, and we stated in that rule that we would address the remaining 
proposed amendments and other issues in a future rulemaking. See 67 FR 
at 6970-71. We discuss below those remaining proposed amendments and 
issues.
    Although these issues and proposed amendments originally pertained 
only to incinerators, cement kilns, and lightweight aggregate kilns, 
any amendments that we may promulgate subsequent to this notice would 
also apply to boilers and hydrochloric acid production furnaces.

A. Definition of Research, Development, and Demonstration Source.

    In response to industry stakeholder concerns, EPA requested comment 
in the July 3, 2001, proposed rule on approaches to preclude 
inappropriate use of the exemption for research, development, and 
demonstration sources. See 66 FR at 35128. We indicated we were 
considering two approaches: (1) Clearly distinguishing between research 
and development sources, and limiting the exemption for demonstration 
sources to one year or less; or (2) requiring documentation of how a 
source's demonstration of an innovative or experimental hazardous waste 
treatment technology or process is different from the waste management 
services provided by a commercial hazardous waste combustor.
    Two stakeholders provided comments, and both recommended that EPA 
not revise the definition of research, development, and demonstration 
source. One commenter suggested that EPA should be able to determine if 
a source is inappropriately claiming the exemption for research, 
development, and demonstration source without amending the regulation. 
The other commenter suggested that, rather than amend the regulation, 
EPA should reiterate that RCRA regulations continue to apply to exempt 
research, development, and demonstration sources.\268\
---------------------------------------------------------------------------

    \268\ Hazardous waste research, development, and demonstration 
sources remain subject to RCRA permit requirements under Sec.  
270.65, which direct the Administrator to establish permit terms and 
conditions that will assure protection of human health and the 
environment.
---------------------------------------------------------------------------

    We concur with the commenters and are not proposing to amend the 
definition of research, development, and demonstration source.

B. Identification of an Organics Residence Time That Is Independent of, 
and Shorter Than, the Hazardous Waste Residence Time

    In response to industry stakeholder recommendations, EPA requested 
comment in the July 3, 2001, proposed rule on whether it is practicable 
to calculate a hazardous waste organics residence time that defines 
when organic constituents in solid materials have been destroyed. See 
66 FR at 35128-30. Under stakeholders' recommendation, after the 
hazardous waste organics residence time expires, sources could comply 
with standards the Agency has promulgated under sections 112 or 129 of 
the Clean Air Act to control organic emissions for source categories 
that do not burn hazardous waste in lieu of the hazardous waste 
combustor standards and associated compliance requirements under 
subpart EEE, part 63, for dioxin/furan, destruction and removal 
efficiency, and carbon monoxide or hydrocarbon emissions.\269\
---------------------------------------------------------------------------

    \269\ Stakeholders also wanted the hazardous waste residence 
time (for organics) to expire as soon as possible to avoid 
violations associated with exceedances of an organics emission 
standard or associated operating requirement during malfunctions 
when hazardous waste remained in the combustion chamber. The rule 
has been amended, however, to state that an exceedance of an 
emission standard or operating requirement during a malfuncation is 
not a violation provided that the source has developed an 
appropriate startup, shutdown, and malfuncation plan, and follows 
the corrective measures provided by the plan. See 67 FR at 6798-
6801.
---------------------------------------------------------------------------

    In the July 3, 2001, proposed rule, we raised several concerns 
regarding the approach recommended by stakeholders to calculate an 
organics residence time, and specifically requested comment on how 
these concerns could be addressed. See 66 FR at 35130. Although several 
stakeholders provided comment on the discussion we presented in the 
July 3, 2001, proposed rule, commenters did not address the concerns we 
raised. Rather, commenters generally note that calculation of an 
organics residence time for solid waste streams would be difficult to 
characterize generically. Accordingly, commenters suggest that the rule 
be amended to specifically allow calculation of an organics residence 
time on a site-specific basis.
    We are reluctant to encourage site-specific petitions to calculate 
an

[[Page 21342]]

organics residence time, however, given that the concerns we raised in 
the July 3, 2001, proposal have not been addressed.\270\ Moreover, we 
believe that stakeholders' primary motive for identifying an organics 
residence time has been eliminated by the February 13, 2002, amendment 
to the rule stating that an exceedance of an emission standard or 
operating requirement during a malfunction when hazardous waste remains 
in the combustion chamber is not a violation provided that the source 
follows the corrective measures provided by an appropriate startup, 
shutdown, and malfunction plan.
---------------------------------------------------------------------------

    \270\ We questioned whether available information on low oxygen 
destruction would adequately model destruction under the pyrolytic 
conditions that occur within solid matrices and whether it is 
practicable to perform valid engineering calculations for multiple 
waste streams that are not homogeneous and that contain multiple 
organic constituents of concern.
---------------------------------------------------------------------------

    For these reasons, we are not proposing an organics residence time 
or explicitly encouraging sources to petition the delegated CAA 
authority on a site-specific basis to identify an organics residence 
time.

C. Why Is EPA Not Proposing To Extend APCD Controls After the Residence 
Time Has Expired When Sources Operate Under Alternative Section 112 or 
129 Standards?

    In the July 3, 2001, proposed rule, we proposed to extend 
applicability of operating requirements for dry particulate matter 
emission control devices before you could switch modes of operation and 
become subject to Section 112 or 129 standards for sources that do not 
burn hazardous waste. See 66 FR at 35130-32. We proposed to require you 
to maintain compliance with applicable emission standards for 
semivolatile metals, low volatile metals, and particulate matter, 
including the operating parameter limits for dry control systems, after 
the hazardous waste residence time has expired until the control device 
undergoes a complete cleaning cycle. We were concerned that dry 
particulate matter control devices such as electrostatic precipitators 
and baghouses retain collected particulate matter contaminated with 
waste-derived metals; and dioxin/furan when activated carbon injection 
is used. In such cases, we were concerned that waste-derived metals and 
dioxin/furan may be emitted at levels exceeding the hazardous waste 
combustor emission standards if you were to switch modes of operation 
and comply with potentially less stringent alternative MACT standards 
for sources that do not burn hazardous waste (e.g., subpart LLL for 
cement kilns, section 129 standards the Agency is developing for 
commercial and industrial solid waste incinerators, and MACT standards 
the Agency is developing for boilers).\271\
---------------------------------------------------------------------------

    \271\ Please note that you are subject to the standards under 
subpart EEE at all times, including after the hazardous waste 
residence time has expired, unless you have established an 
alternative mode of operation under Sec.  63.1209(q)(1).
---------------------------------------------------------------------------

    Commenters raised several concerns about the practicability of 
maintaining compliance with the semivolatile metals, low volatile 
metals, and particulate matter standards after the hazardous waste 
residence time has expired until the particulate matter device 
undergoes a complete cleaning cycle. Commenters explained that it is 
difficult to determine when a cleaning cycle has been completed for 
multi-field electrostatic precipitators and multi-compartment fabric 
filters because fabric filter cleaning is typically a continuous 
process, and electrostatic precipitator plate cleaning frequency varies 
significantly depending on the plate position within the electrostatic 
precipitator. Commenters also stated that the proposed requirement 
would encourage more frequent cleaning of electrostatic precipitators 
and fabric filters than normal, which could increase emissions of HAP 
and adversely affect bag life.
    After review of comments and further consideration, we conclude 
that it is not necessary to revise the standards to extend 
applicability of the operating requirements for dry particulate matter 
control devices before you could switch modes of operation and become 
subject to MACT standards for sources that do not burn hazardous waste. 
We now believe that it is highly unlikely that entrained particulate 
matter contaminated with hazardous waste derived metals would be 
released from the electrostatic precipitator or fabric filter at rates 
higher than when feeding hazardous waste when the source begins 
operating under the alternative MACT (or section 129) standards for 
sources that do not burn hazardous waste. In addition, incinerators, 
cement kilns, and solid-fuel-fired boilers would be subject to 
alternative standards and operating limits for particulate matter. 
Although lightweight aggregate kilns would not be subject to 
alternative standards for particulate matter,\272\ lightweight 
aggregate kilns that burn hazardous waste are equipped with fabric 
filters where their performance is not highly sensitive to operating 
conditions. And, although liquid fuel-fired boilers would not be 
subject to alternative Section 129 standards for particulate 
matter,\273\ over 80% of liquid fuel-fired boilers that burn hazardous 
waste are not equipped with a control device, and only about one third 
of those with a control device are equipped with an electrostatic 
precipitator or fabric filter. Thus, the absence of particulate matter 
controls under the alternative section 129 standards is not a 
significant concern.
---------------------------------------------------------------------------

    \272\ The Agency determined that lightweight aggregate kilns 
that do not burn hazardous waste are not a significant source of HAP 
emissions and, thus, that MACT standards are not necessary for that 
source category.
    \273\ The Agency did not propose PM standards for existing 
liquid fuel-fired industrial, commercial, and institutional boilers 
and process heaters. See 68 FR 1660.
---------------------------------------------------------------------------

    For these reasons, we are not proposing to extend applicability of 
the operating requirements for dry particulate matter control devices 
before you could switch modes of operation and become subject to MACT 
standards for sources that do not burn hazardous waste.

D. Why Is EPA Proposing To Allow Use of Method 23 as an Alternative to 
Method 0023A for Dioxin/Furan?

    The September 1999 Final Rule requires use of Method 0023A for 
stack sampling of dioxin/furan emissions. In response to industry 
stakeholder requests, we proposed in the July 3, 2001, proposed rule to 
allow you to petition the delegated regulatory authority to use Method 
23 found in 40 CFR part 60, appendix A, instead of Method 0023A. See 66 
FR at 35137. We are revising the proposal today to allow you to use 
Method 23 in lieu of Method 0023A after justifying use of Method 23 as 
part of your performance test plan that must be reviewed and approved 
by the delegated regulatory authority. See proposed Sec.  
63.1208(b)(1)(i)(B). This approach would achieve the same objectives as 
a petition, but would be simpler to implement because it would not 
require a separate petition/document.
    In the July 3, 2001, proposed rule, we explain that Method 0023A is 
an improved version of Method 23 in that it can improve the quality 
assurance of the method. By analyzing the sampling train front half 
catch (filter and probe rinse) separately from the back half catch 
(sorbent and rinses), Method 0023A provides quality assurance of 
recovery of dioxin/furan contained in solid phase particulate and 
collected on the filter and probe. Under Method 23, poor recovery of 
dioxin/furan contained in solid phase particulate may go unnoticed 
because the front half catch and back half catch are combined before 
analysis. This may be of particular

[[Page 21343]]

importance for sources that use activated carbon injection or sources 
that have carbonaceous material in particulate matter.
    Although Method 0023A can improve quality assurance, it is slightly 
more expensive than Method 23 and, in many situations, quality 
assurance may not be improved. For example, Method 0023A may not be 
warranted in the future if Method 0023A analyses document that dioxin/
furan are not detected, are detected at low levels in the front half of 
Method 0023A, or are detected at levels well below the emission 
standard, and the design and operation of the combustor has not changed 
in a manner that could increase dioxin/furan emissions.
    Environmental stakeholders comment that use of Method 23 would 
allow sources to emit dioxin/furan in excess of the standards without 
being detected. We disagree. Owners and operators seeking to use Method 
0023A would be required to document using data or information that 
Method 23 would provide front half recoveries comparable to Method 
0023A.
    Industry stakeholders comment that we should simply revise the rule 
to allow use of either method, rather than requiring a petitioning 
process to use Method 23. As discussed above (and in the July 3, 2001, 
proposal), we believe that there are situations where the quality 
assurance and added cost of Method 0023A may be warranted, and, so, are 
not proposing to allow use of Method 23 without justification and prior 
approval. We agree, however, that the formal petitioning process that 
we proposed is not necessary. Rather, we propose today to require you 
to justify use of Method 23 as part of the performance test plan that 
you submit to the delegated regulatory authority for review and 
approval. See proposed Sec.  63.1207(f)(1)(xxv).
    In the interim, you may request to use Method 23 in lieu of Method 
0023A under Sec.  63.7(e)(2)(i) which allows use of a test method with 
minor changes in methodology. You should submit your request and the 
supporting justification to the delegated regulatory authority.

E. Why Is EPA Not Proposing the ``Matching the Profile'' Alternative 
Approach To Establish Operating Parameter Limits?

    In response to stakeholder concerns about the stringency of 
calculating most operating parameter limits as the average of the test 
run averages of the comprehensive performance test, EPA requested 
comment in the July 3, 2001, proposed rule on an alternative approach 
to establish operating parameter limits. See 66 FR at 35138-39.
    The alternative approach, called ``matching the profile'', was 
intended to allow sources to identify limits for operating parameters 
that would allow the operating parameters to have the same average 
variability as experienced during the comprehensive performance test. 
The parameter could exceed the average achieved during the performance 
test for a period of time, provided that it was equivalently lower than 
the average for the same duration of time.
    Commenters generally note that the matching the profile approach 
has a significant disadvantage in that multiple limits would be 
established for each parameter. Accordingly, commenters recommend that 
we not include this approach in the regulation, but rather continue to 
offer it as guidance. Moreover, commenters note that sources can 
request approval of alternative monitoring approaches under Sec.  
63.1209(g)(1), and they are concerned that codification of only one 
approach, and particularly an approach with potentially limited 
utility, could lead the delegated CAA authority to conclude incorrectly 
that other approaches may not be appropriate.
    We believe that this matter is best dealt with on a site-specific 
basis, but note that by specifying one approach in the rule, we do not 
mean to preclude use of a different approach pursuant to Sec.  
63.1209(g)(1). Sources thus may request approval of the profiling 
approach, or another approach, to establish operating limits on a site-
specific basis under Sec.  63.1209(g)(1).

F. Why Is EPA Not Proposing To Allow Extrapolation of OPLs?

    In response to industry stakeholder concerns, we requested comment 
in the July 3, 2001, proposed rule on whether the rule should allow 
extrapolation of an operating parameter limit to a higher limit using a 
site-specific, empirically-derived relationship between the parameter 
and emissions of the pollutant in question.\274\ See 66 FR at 35139-40. 
We also requested comment on whether the rule should allow use of 
established engineering principles that define the relationship between 
operating parameter and emissions to extrapolate operating limits and 
emissions in lieu of a site-specific, empirically-derived relationship.
---------------------------------------------------------------------------

    \274\ Please note that the rule already allows extrapolation of 
mercury feedrates (Sec.  63.1209(l)(1)(i)) and semivolatile and low 
volatile metal feedrates (Sec.  63.1209(n)(2)(ii)).
---------------------------------------------------------------------------

    Industry stakeholders are concerned that the rule inappropriately 
penalizes sources that achieve comprehensive performance test emission 
levels well below the standard by requiring them to establish operating 
limits based on performance test operations at those low emission 
levels. They note that operating under conditions to artificially 
increase emissions during testing (e.g., by detuning emission control 
equipment) may not be feasible or desirable from a worker/public health 
and cost perspective.
    Although stakeholders acknowledge that they may request such 
extrapolation as an alternative monitoring approach under Sec.  
63.1209(g)(1), they note that explicitly defining an extrapolation 
approach in the rule may better facilitate their efforts to obtain 
approval from the delegated regulatory authority.
    Several industry stakeholders agreed with the principle of 
extrapolation as we discussed it in the July 3, 2001, notice, but 
disagreed with the requirements for, and limits on, extrapolation that 
we recommended. Several other stakeholders oppose the use of 
extrapolation generally because of concern that it is difficult to 
define completely and accurately the relationship between an operating 
parameter and emissions.
    Given the extent of the issues associated with explicitly providing 
for extrapolation of operating parameter limits, particularly on a 
categorical rather than a site-specific level, and given that you 
already have the ability to request approval of extrapolation 
procedures under Sec.  63.1209(g)(1), we are not proposing to revise 
the rule to explicitly allow extrapolation. We believe that 
extrapolation must be justified by a site-specific analysis.

G. Why Is EPA Proposing To Delete the Limit on Minimum Combustion 
Chamber Temperature for Dioxin/Furan for Cement Kilns?

    In response to stakeholder concerns that it is technically 
impracticable for cement kilns to establish a minimum combustion 
chamber temperature based on the average of the test run averages for 
each run of the comprehensive performance test, EPA requested comment 
in the July 3, 2001, proposed rule on whether the rule should continue 
to require cement kilns to establish and comply with a minimum 
combustion chamber temperature limit. See 66 FR at 35140.
    We received a total of five comments to the July 3, 2001, proposed 
rule. Three commenters opposed deleting the requirement for cement 
kilns to establish and comply with a minimum combustion chamber 
temperature.

[[Page 21344]]

Currently, cement kilns are required to establish a minimum combustion 
chamber temperature as an operating parameter limit to ensure 
compliance with the destruction and removal efficiency and dioxin/furan 
standards. See Sec. Sec.  63.1209(j)(1) and (k)(2). These commenters 
generally cited the need for monitoring combustion chamber temperature 
by noting that combustion chamber temperature is a principal factor in 
ensuring combustion efficiency and destruction of toxic organic 
compounds.
    Two commenters support deleting the minimum combustion chamber 
temperature requirements. Commenters state that a cement kiln 
inherently controls the kiln temperature to produce clinker because the 
required material temperatures must exceed approximately 2,500[deg]F 
with combustion gas temperatures higher still. These commenters note 
that a cement kiln operates well above minimum temperatures required to 
destroy the organic compounds in the hazardous waste, and, therefore, a 
minimum combustion chamber temperature limit is not necessary to 
control organic hazardous air pollutant emissions.
    Commenters also state that combustion chamber temperatures cannot 
be maintained at low enough levels for the duration of the 
comprehensive performance test to establish workable operating limits 
that would allow them to burn hazardous waste fuels economically 
without frequent waste feed cutoffs because of potential exceedances of 
the limit. Commenters indicate that combustion chamber temperature 
levels are fairly constant within a narrow range and note that there is 
a very narrow range of temperatures and feed composition in which a 
cement kiln must operate in order to produce quality clinker and a 
marketable product. Moreover, commenters state that cement kiln 
operators must take extreme actions, including potentially equipment-
damaging steps, to lower kiln temperatures to establish an economically 
viable minimum combustion chamber limit. Finally, commenters indicate 
that these problems are compounded by the requirement in the MACT rule 
to establish the hourly rolling limit based on the average of the test 
run averages (Sec. Sec.  63.1209(j)(1)(ii) and (k)(2)(ii)).
    We are not proposing to delete the requirement for cement kilns to 
establish and comply with a minimum combustion chamber temperature to 
help ensure compliance with the destruction and removal efficiency 
standard. Even though we remain reluctant to delete this requirement, 
commenters may, if they choose, provide additional comments on whether 
the rule should continue to require cement kilns to establish a minimum 
combustion chamber temperature limit as specified in Sec.  
63.1209(j)(1).
    We are, however, proposing to delete the requirement to establish a 
minimum combustion chamber temperature limit for dioxin/furan under 
Sec.  63.1209(k)(2). As mentioned above, sources are currently required 
to establish a minimum combustion chamber temperature as an operating 
parameter limit for both the destruction and removal efficiency and 
dioxin/furan standards. This proposed amendment would not affect the 
requirement for cement kilns to establish a minimum combustion chamber 
temperature under Sec.  63.1209(j)(1) during the destruction and 
removal efficiency demonstration. Currently, the destruction and 
removal efficiency demonstration need be made only once during the 
operational life of a source provided that the design, operation, and 
maintenance features do not change in a manner that could reasonably be 
expected to affect the ability to meet the destruction and removal 
efficiency standard. See Sec.  63.1206(b)(7). If a facility wishes to 
operate under new operating parameter limits that could be expected to 
affect the ability to meet the destruction and removal efficiency 
standard, then the source will need to conduct another destruction and 
removal efficiency test. In addition, if a source feeds hazardous waste 
at locations other than the flame zone, the destruction and removal 
efficiency demonstration must be verified during each comprehensive 
performance test and new operating parameter limits must be 
established.
    Sources that fire hazardous waste only at the flame zone (i.e., the 
kiln end where clinker product is normally discharged) are required to 
make only one destruction and removal efficiency demonstration test 
during the operational life of the kiln. During this destruction and 
removal efficiency demonstration test, the source would set a minimum 
combustion chamber temperature limit under Sec.  63.1209(j)(1) that 
would be the limit for the operational life of the kiln. However, as 
the rule is currently written, such sources would need to establish a 
minimum combustion chamber temperature limit during subsequent 
comprehensive performance tests for the dioxin/furan test under Sec.  
63.1209(k)(2). The source would be required to comply with the more 
stringent (higher) of two minimum combustion chamber temperature 
limits, which could lead to a situation where the controlling minimum 
combustion chamber temperature limit is based on the dioxin/furan test 
rather than the destruction and removal efficiency demonstration.
    We believe that this may be an inappropriate outcome given that the 
operating limit for minimum combustion chamber temperature is a more 
important parameter to ensure compliance with the destruction and 
removal efficiency standard than to ensure compliance with the dioxin/
furan standard. Our data indicate that limiting the gas temperature at 
the inlet to the particulate matter control device, an operating 
parameter limit established during each comprehensive performance test 
(Sec.  63.1209(k)(1)), is a critical dioxin/furan control parameter. We 
are, therefore, inviting comment on deleting the requirement to 
establish a minimum combustion chamber temperature limit when complying 
with the dioxin/furans standard. This proposed amendment does not 
affect the other operating parameter limits under Sec.  63.1209(k) that 
must be established for dioxin/furan such as establishing a limit on 
the gas temperature at the inlet to the particulate matter control 
device.
    For cement kilns that fire hazardous wastes at locations other than 
the flame zone, the current requirements would effectively remain the 
same. Given that a source conducts the destruction and removal 
efficiency demonstration and dioxin/furan test simultaneously and that 
a source is also required to establish a minimum combustion chamber 
temperature limit when demonstrating compliance with and establishing 
operating parameter limits for the destruction and removal efficiency 
standard, the minimum combustion chamber temperature limits is 
effectively retained.

H. Why Is EPA Requesting Additional Comment on Whether To Add a Maximum 
pH Limit for Wet Scrubbers To Control Mercury Emissions?

    We requested comment in the July 3, 2001, proposed rule as to 
whether it is appropriate to establish a limit on maximum pH to control 
mercury. See 66 FR at 35142-43. We are requesting additional comment 
today on this issue given the results of a recent study indicating that 
increasing the pH of scrubber liquid can increase mercury emissions.

[[Page 21345]]

1. What Were the Major Comments on the Discussion in the July 3, 2001, 
Proposed Rule?
    One commenter supports placing limits on the maximum pH of wet 
scrubber liquids for mercury control, but did not provide any 
additional rationale on the technical validity of the limit. Other 
commenters oppose the imposition of a maximum pH limit. One commenter 
wants to see stronger evidence that pH has an impact, and suggests a 
reproposal is needed. Another suggests that EPA conduct source testing 
to confirm that pH has an impact. Others suggest that if EPA continues 
to believe that wet scrubber operating parameter limits are important 
for mercury control, then the wet scrubber mercury operating parameter 
limits should be determined on a case-by-case basis because the 
relationship between mercury control and wet scrubber pH is not well 
established and there are numerous other factors that affect mercury 
control in wet scrubbers, especially for facilities that burn waste 
with various chemical compositions.
2. What Is the Rationale for Considering a Maximum pH Limit To Control 
Mercury?
    The use of a low pH liquid scrubber solution has been suggested to 
be beneficial for mercury control because it helps prevent the re-
release of captured mercury. Ionic mercury (Hg+\2\) is 
highly soluble in wet scrubber liquid; as opposed to Hg\o\, which has a 
very low solubility in a typical water/alkali scrubber solution. Once 
absorbed, Hg+\2\ can be reduced to Hg\o\ by compounds in the 
liquid scrubber solution such as SO₂ and HSO₃. 
Hg\o\ may then be revolatilized back into the stack gas. This is 
supported by numerous observations of Hg\o\ at the wet scrubber outlet 
which are higher than Hg\o\ at the scrubber inlet 
275, 276, 277. These studies suggest that the low scrubber 
liquid pH prevents captured mercury from revolatilizing from the 
scrubber liquid by: (1) limiting the capture of reducing agents; and 
(2) favoring the formation of stable mercury-chlorine compounds such as 
HgCl₂ due to available Cl-. In contrast, other 
studies postulate that a high scrubber liquid pH might actually be 
beneficial for the control of mercury, particularly elemental Hg \278\. 
Basic, high pH solutions have the increased ability to absorb chlorine 
gas. Dissolved chlorine gas is suggested to enhance the scrubber's 
ability to oxidize and capture Hg\o\ (specifically, dissolved chlorine 
gas dissociates in basic solutions to produce OCl- ions 
which oxidize Hg\o\ to soluble Hg+\2\). In contrast, the 
presence of hydrogen chloride or sulfur as SO₂ or 
H2SO₃ in the scrubber solution reduces the liquid scrubber 
pH, reduces OCl-, and reduces the Hg\o\ oxidative potential 
of the scrubber liquid.
---------------------------------------------------------------------------

    \275\ B. Siret and S. Eagleson, ``A New Wet Scrubbing Technology 
for Control of Elemental (Metallic) and Ionic Mercury Emissions,'' 
Proceedings of 1997 Conference on Incineration and Thermal Treatment 
Technology, pp. 821-824, 1997.
    \276\ G. T. Amrhein, G. Kudlac, D. Madden, ``Full-Scale Testing 
of Mercury Control for Wet FGD Systems,'' Presented at the 27th 
International Technical Conference on Coal Utilization and Fuel 
Systems, Clearwater, Fl, March 4-7, 2002.
    \277\ C.S. Krivanek, ``Mercury Control Technologies for MWCs: 
The Unanswered Questions,'' 1993 Air and Waste Management Sponsored 
Municipal Solid Waste Combustor Specialty Conference, 1993.
    \278\ W. Linak, J. Ryan, B. Ghorishi, and J. Wendt, ``Issues 
Related to Solution Chemistry in Mercury Sampling Impingers,'' 
Journal or Air and Waste Management Association, Vol. 51, pp. 688-
698, May 2001.
---------------------------------------------------------------------------

    Although limited test data from full-scale coal fired boiler 
evaluations indicate an inconsistent impact of scrubber liquid pH on 
mercury control,\279\ a recent study \280\ confirms that ionic mercury 
(e.g., HgCl₂) that is initially captured in the scrubber can 
be reduced in the liquid to elemental Hg (i.e., H\o\) and then 
revolatilized to the stack gas. The study concludes that the reduction 
of ionic mercury in the liquid is likely due to dissolved sulfur 
compounds and that decreasing the pH of the liquid will decrease the 
reduction process and subsequently decrease mercury emissions. This new 
work is additional evidence that a maximum pH limit might be 
appropriate, especially if sulfur is present in feeds.
---------------------------------------------------------------------------

    \279\ For example, McDermott Technology (McDermott Technology, 
Internet Web page at http://www.mcdermott.com  on ``Mercury 
Emission Results,'' date unknown) report no impact, while DeVito and 
Rosenhoover (M. DeVito and W. Rosenhoover, CONSOL Coal Inc., ``Flue 
Gas Hg Measurements from Coal-fired Boilers Equipped with Wet 
Scrubbers,'' date unknown) observe that mercury control efficiency 
appears to increase with increasing pH.
    \280\ J. Chang and S. Ghorishi, ``Simulation and Evaluation of 
Elemental Mercury Concentration Increase in Flue Gas Across a Wet 
Scrubber,'' Environmental Science and Technology, Vol 37, No. 24, 
2003, pp. 5763-5766.
---------------------------------------------------------------------------

    Other recent work indicates that there are numerous factors that 
influence the control of mercury in wet scrubbers. Mercury speciation 
in the flue gas is vitally important to the ability to control mercury 
in wet scrubbers. In hazardous waste combustor flue gases, mercury 
tends to be predominately in two forms: (1) elemental (Hg\o\); and (2) 
ionic (Hg+\2\, typically as HgCl₂). Speciation 
depends on numerous factors including the presence of chlorine or 
sulfur, both of which are reactive with mercury. For example, increased 
levels of chlorine may increase the amount of HgCl₂ and 
reduce the amount of Hg\o\. This might suggest that a minimum chlorine 
feedrate limit is needed to ensure Hg scrubber efficiency is 
maintained, which is counter to the maximum chlorine feedrate limit 
used to control emissions of total chlorine and semivolatile and low 
volatile metals. Speciation is also affected by the flue gas 
temperature cooling profile, which can impact mercury reaction 
kinetics. For example, rapid cooling may limit the equilibrium 
formation of HgCl₂ (i.e., super equilibrium levels of Hg\o\ 
can survive from rapid cooling). This might suggest that a maximum flue 
gas cooling limit is needed, which is counter to that for controlling 
dioxin/furan.
    Control of mercury in wet scrubbers is also affected by the 
scrubber liquid chemical composition. As discussed above, scrubber 
liquid composition has a dramatic impact on the control of mercury. 
Specifically, the presence of reducing compounds such as SO₂ 
and HSO₃ can lead to increased mercury emission by reducing 
soluble HgCl₂ to insoluble Hg\o\ which can be desorbed while 
oxidative compounds such as chlorine gas and special oxidation 
additives such as NaClO₂, acidified KMnO₃, 
Na₂S, and TMT (tri-mercapto-triazine) would generally help 
control mercury emissions by inhibiting reduction of HgCl₂ 
to Hg\o\ and/or enhancing the capture of Hg\o\.
    Finally, control of mercury in wet scrubbers is affected by the 
scrubber liquid to gas ratio.
    Given the recent study discussed above indicating that increasing 
the pH of scrubber liquid can increase mercury emissions, we request 
additional comment on whether it would be appropriate to establish a 
limit on the maximum pH of scrubber liquid to ensure compliance with 
the mercury emission standard. We also request comment on issues 
relative to establishing and complying with both a maximum limit on pH 
to control mercury emissions and a minimum limit on pH to control total 
chlorine. For example, you would establish the maximum and minimum pH 
limits under separate performance tests. You would establish the 
minimum pH limit during a performance test to demonstrate compliance 
with the total chlorine standard while you would establish the maximum 
pH limit during a performance test to demonstrate compliance with the 
mercury standard. In addition, we request comment on the anticipated 
range of pH levels between the maximum and minimum limits and whether 
the range could potentially be small enough to inhibit operations 
substantially. For example, if the pH

[[Page 21346]]

required to achieve your desired scrubber control efficiency for total 
chlorine (i.e., the minimum pH limit) is just below the pH level 
required to achieve your desired control efficiency for mercury (i.e., 
the maximum pH limit), you may have limited operating flexibility.
    Finally, we note that, in the interim until we determine whether to 
promulgate a maximum pH limit to control mercury emissions, site-
specific or other information may lead the delegated regulatory 
authority to conclude under Sec.  63.1209(g)(2) that a limit on the 
maximum pH of wet scrubber liquid may be warranted to ensure compliance 
with the mercury emission standard.

I. How Is EPA Proposing to Ensure Performance of Electrostatic 
Precipitators, Ionizing Wet Scrubbers, and Fabric Filters?

    If your combustor is equipped with a fabric filter, you would be 
required to use the bag leak detection system under Sec.  
63.1206(c)(7)(ii) to ensure performance of the fabric filter is 
maintained in lieu of operating parameter limits.\281\ In addition, we 
propose to revise the bag leak requirements under Sec.  
63.1206(c)(7)(ii) to require you to operate and maintain the fabric 
filter such that the bag leak detection system alarm does not sound 
more than 5 percent of the operating time during a 6-month period.
---------------------------------------------------------------------------

    \281\ As discussed below in the text, we propose to revise the 
current rules to delete the exemption for cement kilns from the bag 
leak detection system requirements.
---------------------------------------------------------------------------

    If your combustor is equipped with an electrostatic precipitator or 
ionizing wet scrubber, we propose to give you the option of: (1) Using 
a particulate matter continuous emissions detector for process 
monitoring to signal when you must take corrective measures to address 
maintenance or other factors causing relative or absolute mass 
particulate matter loadings to be higher than the levels achieved 
during the performance test; or (2) establishing site-specific 
operating parameter limits. If you choose to use a continuous emissions 
detector, you must not exceed the alarm set-point you establish based 
on the performance test more than 5 percent of the operating time 
during a 6-month period. If you choose to establish site-specific 
operating parameter limits, you must link each limit to the automatic 
waste feed cutoff system.
1. What Is the Background of this Issue?
    The current regulations require you to establish site-specific 
operating parameter limits to ensure performance of electrostatic 
precipitators, ionizing wet scrubbers, and fabric filters. See Sec.  
63.1209(m)(1)(iv).\282\ Regulatory officials review and approve those 
operating parameter limits and may require additional or alternative 
limits under Sec.  63.1209(g)(2).
---------------------------------------------------------------------------

    \282\ Please note that Sec.  63.1209(m)(1)(iv) inadvertently 
indicates that the requirement to establish site-specific operating 
limits applies to control devices other than ionizing wet scrubbers, 
baghouses, and electrostatic precipitators. We should have revised 
that paragraph to require site-specific operating parameter limits 
for those control devices when we revised paragraph (m)(1) to delete 
the operating parameter limits for those devices. The delegated 
regulatory authority can use Sec.  63.1209(g)(2) to require you to 
establish site-specific operating parameter limits for those control 
devices prior to the effective date of the final rule based on 
today's proposed rule.
---------------------------------------------------------------------------

    In the July 3, 2001 proposed rule, we requested comment on how to 
establish prescriptive requirements to ensure performance of these 
control devices. See 66 FR at 35143-45. We requested comment on four 
approaches to ensure performance of electrostatic precipitators: (1) 
Requiring an increasing kVA pattern across the electrostatic 
precipitator; (2) limiting kVA on only the back \1/3\ of fields; (3) 
use of a CMS that measures relative particulate matter loadings; and 
(4) use of predictive emission monitoring systems. These approaches 
would also be applicable to ionizing wet scrubbers. We also requested 
comment on whether and how cell pressure drop should be used to ensure 
performance of fabric filters.
    We received comments in favor of and opposing most of these 
approaches.\283\ Some stakeholders also recommend other approaches. One 
commenter favors use of specific power as an operating parameter for 
electrostatic precipitator performance. Specific power is the secondary 
power/gas flow rate. Another commenter suggests continuing with 
establishing site-specific operating parameter limits.
---------------------------------------------------------------------------

    \283\ USEPA, ``Response to Comments on July 2001 Proposed 
Rule,'' March 2004.
---------------------------------------------------------------------------

 2. What Is the Rationale for Proposing to Revise the Compliance 
Requirements for Fabric Filters?
     After reviewing comments and further investigation, we conclude 
that controls in addition to a bag leak detection system are not needed 
to ensure performance of fabric filters. Use of pressure drop to ensure 
performance is problematic for reasons we discussed in the July 3, 2001 
proposed rule. Moreover, the bag leak detection system provides a 
direct measure of small (and greater) increases in particulate matter 
loading that enable you to take immediate corrective measures.
    We conclude, however, that the bag leak detection system 
requirements under Sec.  63.1206(c)(7)(ii) are not prescriptive enough 
to ensure proper operation and maintenance of the fabric filter. 
Current provisions require you to take immediate corrective measures 
when the bag leak detection system alarm sounds, indicating that 
particulate loadings exceed the set-point. There is no limit on the 
duration of time, however, that the bag house may be operating under 
these conditions. To ensure that you take both corrective and proactive 
measures to minimize the frequency and duration of bag leak detection 
system alarms, you must operate and maintain the fabric filter to 
ensure that the bag leak detection system alarm does not sound more 
than 5 percent of the operating time during a 6-month period.\284\ We 
note that the Agency also proposed this requirement for boilers and 
process heaters that do not burn hazardous waste. See 68 FR at 1708 
(January 13, 2003). If you exceed the alarm set-point more than 5 
percent of the time during a 6-month period, you would be required to 
notify the delegated regulatory authority within 5 days. In the 
notification, you must describe the causes of the excessive exceedances 
and the revisions to the design, operation, or maintenance of the 
combustor or baghouse you are taking to minimize exceedances. This 
notification would alert the regulatory authority of the excessive 
exceedances so that they may review and confirm the corrective measures 
you are undertaking. See proposed Sec.  63.1206(c)(7)(ii)(C).
---------------------------------------------------------------------------

    \284\ Periods of time when the combustor is operating but the 
bag leak detection system is malfunctioning must be considered 
exceedances of the set-point.
---------------------------------------------------------------------------

    We also conclude that the current exemption from the bag leak 
detection system requirements for cement kilns should be eliminated. We 
did not require bag leak detection systems for cement kilns in the 
September 1999 Final Rule because cement kilns are subject to an 
opacity standard and must monitor opacity with a continuous monitor. As 
a practical matter, however, the opacity levels achieved during the 
comprehensive performance test will be lower, often substantially 
lower, than the opacity standard. Thus, absent effective operating 
parameter limits on the fabric filter based on performance test 
operations, we cannot ensure that performance is maintained at the 
level achieved during the performance test (and that you remain in 
compliance with the particulate matter and other

[[Continued on page 21347]]
Notices For
2009
2008
2007
2006
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994
National Emission Standards for Hazardous Air Pollutants: Proposed Standards for Hazardous Air Pollutants for Hazardous Waste Combustors (Phase I Final Replacement Standards and Phase II) [[pp. 21297-21346]]

Local Navigation
