Proposed Test Rule for Hazardous Air Pollutants

SUMMARY: EPA is proposing a test rule under section 4(a) of the Toxic

Substances Control Act (TSCA) to require manufacturers and processors

of 21 hazardous air pollutants (HAPs) (biphenyl, carbonyl sulfide,

chlorine, chlorobenzene, chloroprene, cresols [3 isomers],

diethanolamine, ethylbenzene, ethylene dichloride, ethylene glycol,

hydrochloric acid, hydrogen fluoride, maleic anhydride, methyl isobutyl

ketone, methyl methacrylate, naphthalene, phenol, phthalic anhydride,

1,2,4-trichlorobenzene, 1,1,2-trichloroethane, and vinylidene chloride)

to test these substances for certain health effects. EPA is also

soliciting proposals for enforceable consent agreements (ECAs)

regarding the performance of pharmacokinetics studies which would

permit extrapolation from oral data to predict risk from inhalation

exposure. EPA is also withdrawing the oncogenicity testing proposed for

vinylidene chloride on August 12, 1986 (51 FR 28840).

DATES: Written comments on this proposed HAPs test rule must be

received by EPA on or before December 23, 1996. Proposals for

pharmacokinetics studies must be received by EPA on or before October

24, 1996. EPA will hold a public meeting in Washington, DC prior to the

close of the comment period. If any person requests an additional

public meeting by November 25, 1996, EPA will hold a second public

meeting in Washington, DC.

ADDRESSES: Submit three copies of written comments on this proposed

HAPs test rule, identified by document control number (OPPTS-42187A;

FRL-4869-1) and three copies of proposals for pharmacokinetics studies,

identified by document control number (OPPTS-42187B; FRL-4869-1) to:

U.S. Environmental Protection Agency, Office of Pollution Prevention

and Toxics (OPPT), Document Control Office (7407), Rm. G-099, 401 M

St., SW., Washington, DC, 20460.

A public version of the rulemaking record supporting this action,

excluding confidential business information (CBI), is available for

inspection at the TSCA Nonconfidential Information Center, Rm. NE-B607,

401 M St., SW., Washington, DC 20460, from 12 noon to 4 p.m., Monday

through Friday, except on legal holidays.

All comments which contain information claimed as CBI must be

clearly marked as such. Three additional sanitized copies of any

comments containing information claimed as CBI must also be submitted.

Nonconfidential versions of comments on this proposed rule will be

placed in the rulemaking record and will be available for public

inspection at the TSCA Nonconfidential Information Center. Unit IX of

this preamble contains additional information on submitting comments

containing information claimed as CBI.

Comments and data may also be submitted in electronic form by

sending electronic mail (e-mail) to: [email protected]. Such

comments and data must be submitted in an ASCII file avoiding the use

of special characters and any form of encryption. Comments and data

will also be accepted on disks in WordPerfect in 5.1 file format or

ASCII file format. All comments and data in electronic form must be

identified by (OPPTS-42187A) (FRL-4869-1). No information claimed as

CBI should be submitted through e-mail. Comments in electronic form may

be filed online at many federal depository libraries. Additional

information on electronic submissions can be found under Unit X of this

preamble.

FOR FURTHER INFORMATION CONTACT: Susan B. Hazen, Director,

Environmental Assistance Division (7408), Office of Pollution

Prevention and Toxics, U.S. Environmental Protection Agency, Room E-

543B, 401 M St., SW., Washington, DC 20460; telephone: (202) 554-1404;

TDD: (202) 554-0551; e-mail: TSCA-H[email protected]. For specific

information regarding this action or related activities, contact Gary

E. Timm, Chemical Control Division, OPPT; telephone: (202) 260-1859; e-

mail: [email protected].

SUPPLEMENTARY INFORMATION:

Regulated persons. Potentially regulated persons are manufacturers

(including importers) and processors of the chemical substances

included in this proposed test rule. Processors, small-quantity

manufacturers, and manufacturers of small quantities of these

substances solely for research and development purposes, while legally

subject to the rule, would be required to comply with the rule only if

directed to do so in a subsequent notice.

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

Examples of regulated

Category persons

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

Manufacturers Persons who manufacture or

import 500 kg (1,100 lbs)

or more of a subject

chemical per year.

Persons who produce a

subject chemical as a

byproduct.

Processors Persons who process one or

more subject chemicals.

Small-quantity manufacturers Persons who manufacture or

import less than 500

kg(1,100 lbs) per year of a

subject chemical.

Manufacturers of small quantities of these Persons who manufacture

substances solely for research and quantities of these

development purposes substances no greater than

those necessary for

purposes of scientific

experimentation or analysis

for research and

development purposes.

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

This table is not intended to be exhaustive, but, rather, provides

a guide for readers regarding entities likely to be regulated by this

action. This table lists the types of persons of which EPA is now aware

that potentially could be regulated by this action. To determine

whether you would be subject to this rule, you should examine Unit

IV.F. of the preamble entitled ``Persons Required to Test'' and consult

40 CFR 790.42.

I. Statutory Authority

This notice proposes a test rule under section 4 of the Toxic

Substances Control Act (TSCA), 15 U.S.C. 2603 et seq., that would

require certain health effects testing for 21 chemical substances

listed as hazardous air pollutants (HAPs) in section 112 of the Clean

Air Act (CAA), 42 U.S.C. 7412.

Section 2(b)(1) of TSCA, 15 U.S.C. 2601(b)(1), states that it is

the policy of the United States that ``adequate data should be

developed with respect to the effect of chemical substances and

mixtures on health and the environment and that the development of such

data should be the responsibility of those who manufacture and those

who process such chemical substances and mixtures[.]'' To implement

this policy, section 4(a) of TSCA authorizes EPA to require by rule

that manufacturers and processors of chemical substances conduct

testing if the Administrator finds that:

[[Page 33179]]

(1)(A)(i) the manufacture, distribution in commerce, processing,

use, or disposal of a chemical substance or mixture, or that any

combination of such activities, may present an unreasonable risk of

injury to health or the environment, [or]

(1)(B)(i) a chemical substance or mixture is or will be produced

in substantial quantities, and (I) it enters or may reasonably be

anticipated to enter the environment in substantial quantities or

(II) there is or may be significant or substantial human exposure to

such substance or mixture, [and]

(1)(A)(ii) and (1)(B)(ii) there are insufficient data and

experience upon which the effects of the manufacture, distribution

in commerce, processing, use, or disposal of such substance or

mixture or of any combination of such activities on health or the

environment can reasonably be determined or predicted, and

(1)(A)(iii) and (1)(B)(iii) testing of such substance or mixture

with respect to such effects is necessary to develop such data[.]

Thus once the Administrator has made a finding under TSCA section

4(a)(1)(A)(i) that a chemical substance may present an unreasonable

risk of injury to health or the environment or a finding under section

4(a)(1)(B)(i) that a chemical substance is or will be produced in

substantial quantities and either it may enter the environment in

substantial quantities or there may be significant substantial human

exposure to the chemical substance, EPA may require any type of health

effects or environmental testing necessary to address unanswered

questions about the effects of the chemical substance. EPA need not

limit the scope of testing required to the factual basis for the

section 4(a)(1)(A)(i) or (B)(i) findings as long as EPA finds that data

relevant to a determination of whether a substance does or does not

present an unreasonable risk of injury to health or the environment are

insufficient and that testing is necessary to develop such data. This

concept is explained in more detail in EPA's statement of policy for

making findings under TSCA section 4(a)(1)(B) (frequently described as

the ``B policy'') in the Federal Register of May 14, 1993 (58 FR

28736). Unit V of this preamble also describes the B policy. Moreover,

EPA need not limit the scope of the requirement only to testing needed

to support regulatory action under TSCA. For further discussion of

findings under TSCA section 4, see Unit V of this preamble and the

document entitled ``TSCA Section 4 Findings for 21 Hazardous Air

Pollutants'' in the record for this rulemaking.

In this proposed rule, EPA intends to use its TSCA section 4

authority to obtain data necessary to implement section 112 of the CAA,

which provides a detailed strategy for the assessment and management of

HAPs. EPA has used this broad TSCA section 4 authority in the past to

support regulatory programs requiring health and environmental effects

testing data. See, e.g., final test rule for the Office of Solid Waste

chemicals (53 FR 22300, June 15, 1988); final test rule for the Office

of Water Chemicals (58 FR 59667, November 10, 1993). Additional users

of information collected under this test rule would include other

federal agencies (e.g. the Agency for Toxic Substances and Disease

Registry (ATSDR), the National Institute for Occupational Safety and

Health (NIOSH), the Occupational Safety and Health Administration

(OSHA), the Consumer Product Safety Commission (CPSC), other program

areas within EPA (such as the hazardous waste program under the

Resource Conservation and Recovery Act (RCRA), the Toxics Release

Inventory (TRI), the Integrated Risk Information System database

(IRIS), and the Office of Pesticide Programs (OPP)), and state and

local environmental authorities.

Supporting statutory authority for this proposed rule is provided

by section 112(b)(4) of the CAA, 42 U.S.C. 7412(b)(4), which

specifically authorizes EPA to use any authority available to EPA to

obtain the information needed to make determinations regarding the

addition or deletion of substances to the statutory list of HAPs in CAA

section 112(b)(1), 42 U.S.C. 7412(b)(1). If the data collected under

this proposed test rule show that a chemical substance is not a concern

to human health, this information would be helpful in making decisions

concerning delisting the substance from the Clean Air Act HAPs list.

This toxicity testing program is also intended to fulfill in part

EPA's statutory obligation under section 103(d) of the CAA, 42 U.S.C.

7403(d), to conduct a research program on the health effects of HAPs.

This preamble, along with the supporting material in the record,

provides information that would be used in the research program under

CAA section 103(d) for the HAPs proposed for testing in this rule.

II. Uses for Data

EPA will primarily use the data proposed to be collected under this

rule to implement several provisions of section 112 of the CAA,

including the determination of residual risk (see below), the

estimation of the risks associated with accidental releases of

chemicals, and determinations whether or not substances should be

removed from the CAA section 112(b)(1) list of hazardous air pollutants

(delisting). The acute toxicity test data and developmental toxicity

test data will be useful in judging risks from accidental release. The

term ``accidential release'' is used broadly in this proposal to

include any short-term, relatively high-level chemical exposure lasting

from several minutes to several hours. Such a release may result from

various causes, including spills, transportation accidents, process-

upset conditions, or short bursts during charging of reaction vessels.

All data are relevant to delisting decisions and all non-acute data

will be used by EPA in meeting its statutory obligation under CAA

section 112(f), 42 U.S.C. 7412(f), to assess the risk remaining (i.e.

residual risk) after the imposition of technology-based emission

standards (maximum achievable control technology or MACT standards)

required by CAA section 112(d), 42 U.S.C. 7412(d). Section 112(e) of

the CAA, 42 U.S.C. 7412(e), directs EPA to promulgate these standards

between 1992 and 2000.

Section 112(f)(1) of the CAA requires EPA to submit, by November

1996, a report to Congress that will describe the methods for assessing

the risk remaining after the application of technology-based standards

under section 112(d) of the CAA. These methods will be used to assess

any residual risk for persons exposed to MACT-regulated emissions. The

assessment will include an analysis of both cancer and noncancer

endpoints. Data generated by the proposed test rule would be used in

the analysis to determine the nature and magnitude of any residual

risk.

Within eight years after the promulgation of technology-based

standards, EPA may need to set additional standards (``post-MACT

standards'') to protect public health with an ample margin of safety.

Section 112(f)(2) of the CAA specifies that if MACT standards have not

reduced lifetime cancer risk to the individual most exposed to known or

suspected carcinogenic emissions from a source to a level of less than

1 in a million (1 x 10-6), health-based emission standards must be

promulgated in order to protect public health with an ample margin of

safety. EPA, therefore, would use data obtained under this proposed

rule to determine whether health-based post-MACT standards are needed

and, if they are needed, to assist in establishing the appropriate

level of these standards.

For noncancer health effects, EPA applies an appropriate

mathematical model to toxicity data in order to determine the benchmark

dose level. The benchmark dose or concentration (BMD/C) is defined as

the statistical

[[Page 33180]]

lower confidence limit on the dose estimated to produce a predetermined

level of change in response (the benchmark response--BMR) relative to

controls. If the data are not amenable to modeling, a no-observed-

adverse-effect level (NOAEL) or a lowest-observed-adverse-effect level

(LOAEL) may be obtained from an evaluation of the toxicity database.

``Uncertainty factors'' are then applied to these levels to account for

uncertainties in deriving a dose-response estimate for human exposure

(reference concentration (RfC)) from experimental data. An RfC is

defined as ``an estimate (with uncertainty perhaps spanning an order of

magnitude) of a continuous inhalation exposure to the human population

(including sensitive subgroups) that is likely to be without

appreciable risk of deleterious noncancer health effects during the

lifetime'' (Ref. 1).

Uncertainties due to the extrapolation of effects data between

species and for individual susceptibility within a species are

accounted for by uncertainty factors. Because RfCs are intended to

characterize risk for lifetime exposures, an uncertainty factor may be

applied if the effects data are extrapolated from a subchronic study.

An additional uncertainty factor is applied if a LOAEL is used to

derive the RfC rather than a NOAEL. To provide an accurate

characterization of lifetime risk, the database for a chemical should

be comprehensive and, in principle, should address all potential

endpoints at critical life stages. Therefore, an uncertainty factor

also may be applied if data for appropriate endpoints are not

available. Thus while each uncertainty factor may range up to 10, the

composite factor used to derive an RfC for a chemical with a limited

database may be on the order of up to 3,000 for inhalation studies and

up to 10,000 for oral studies. Five uncertainty factors are never

applied at the same time (i.e. no composite uncertainty factor can be

greater than 10,000) because such derivations are considered too

inaccurate to be used.

Large composite uncertainty factors result in lower RfCs (higher

risk estimates) than do smaller composite uncertainty factors. If the

RfC is low, EPA may be required to promulgate more stringent emission

standards. Industrial plants subject to these standards would, in turn,

be required to meet such standards, perhaps necessitating the

installation of more costly emission controls. Better and more complete

health effects data, on the other hand, may permit EPA to use smaller

composite uncertainty factors, resulting in higher RfCs and, as a

consequence, less stringent emission standards. Thus the economic cost

of using poor-quality health effects data to make residual risk

determinations under CAA section 112(f) could be considerable.

In addition, secondary--though as important--uses of the data to be

collected under this proposed rule would be:

(1) Helping to better inform communities and citizens of toxic

chemical hazards in their own localities. Understanding health effects

associated with these chemicals is integral to furthering the public's

involvement in environmental decisionmaking, especially at the state

and local level. To be an effective participant in this process, the

public needs information on both the inherent toxicity (i.e., hazard)

of a chemical and the potential sources of exposure to the chemical.

This rule will provide valuable information on health effects related

to the affected chemicals, and under TSCA, such health and safety data

are available to the public. Taken together with such publicly

available information sources such as the Toxics Release Inventory

(TRI), which provides site-specific information on chemical releases

into the environment, the health effects data generated under this rule

will allow all segments of the public to better assess the risks

associated with the releases of these chemicals. Taken as part of a

comprehensive right-to-know program, these data will provide the basis

for individuals, communities, governments, producers, and users to

assess the nature and relative severity of toxicity among different

chemicals, as well as to assess site-specific, individual chemical

risk.

(2) Assisting other agencies (e.g., ATSDR, NIOSH, OSHA, CPSC) in

assessing chemical risks and in taking appropriate action within their

programs. For example, OSHA has expressed a need for the data that will

be acquired under the proposed rule. Fifteen of the 21 HAPS are

candidates for OSHA's Permissible Exposure Limit (PEL) update and an

additional 3 have no corresponding PEL. OSHA does not have authority to

require testing, and must rely on toxicology data collected by other

agencies for their risk assessments (Ref. 2). Establishing an ongoing

mechanism for updating its PELs continues to be a high priority for

OSHA. Five of the HAPs are on ATSDR's list of hazardous substances

found at National Priorities List sites and are the subject of

toxicological profiles. CPSC noted that 11 of the 21 substances are

found in or are emitted by consumer products (Ref. 3).

(3) Assisting EPA in evaluating delisting petitions received under

the CAA and the Emergency Planning and Community Right-to-Know Act

(EPCRA), 42 U.S.C. 1101 et seq., in making better clean-up decisions

under the Comprehensive Environmental Response, Compensation, and

Liability Act (CERCLA), 42 U.S.C. 9601 et seq., in assessing inert

ingredients in pesticide products, in setting more appropriate

standards for hazardous wastes under RCRA, and by providing support for

chemical risk assessment activities under TSCA.

(4) Assisting state and local permitting authorities in setting

standards within their programs.

(5) Supporting assessments of ``burst'' exposures (high-level

releases of short duration), such as in the accidental release

prevention program under section 112(r) of the CAA, 42 U.S.C. 7412(r),

due to the inclusion of an acute testing protocol.

Many HAPs are of broad programmatic interest, and are included in

the Agency's Integrated Risk Assessment System (IRIS) database. Thus, a

secondary benefit of this rule is that the health effects data

generated by the rule may result in improvement to the data and

increased confidence in the RfCs contained in IRIS. Improvements to the

IRIS data can result in considerable benefits to the public since IRIS

is publicly available and is used by a wide variety of governmental and

non-governmental entities to assess the safety of chemicals.

In some cases where EPA has had access to better data, the Agency

has been able to revise standards to make them less stringent, thus

mandating less economically costly levels of control. For example, EPA

has revised the maximum contaminant levels (MCL) of barium (from 1 mg/L

to 2 mg/L) and selenium (from 0.01 mg/L to 0.05 mg/L) in drinking water

and withdrew the MCL for silver (0.05 mg/L) based on new data (Ref. 4).

Data have also been used to remove chemical substances from lists

of regulatory significance. Acrylic acid was removed from a list of

``high-risk'' pollutants developed under section 112(i)(5) of the CAA,

42 U.S.C. 7412(i)(5), for the early reductions program. The high-risk

listing for this chemical was based on its predicted environmental

exposure being at least 10 times larger than its RfC. Designation of

acrylic acid as ``high risk'' had the effect of limiting the use of

offsetting reduction of other pollutants in meeting early reduction

goals. Recent data,

[[Page 33181]]

including those related to reproductive effects, developmental

toxicity, and bioavailability, resulted in a decrease in the

uncertainty factor for database deficiency by a factor of three and an

increase in the RfC by the same amount (i.e. from 0.0003 mg/m3 to

0.001 mg/m3). Consequently, acrylic acid no longer met the

criteria for the high-risk list.

Serious deficiencies exist in the current toxicity database for the

189 HAPs listed in the CAA, in that no toxicity data exist for many

HAPs regarding various endpoints of concern. This problem is expected

to be especially serious because post-MACT residual risks will arise

primarily from exposures to emissions that contain different

combinations of HAPs in varying concentrations. In view of the large

number of HAPs of concern and the much larger number of combinations of

those HAPs found in the mixtures of emissions subject to residual risk

evaluation, the toxicity database should provide consistent

characterization of individual HAPs. Therefore, EPA is proposing to

obtain an even, across-the-board database for the HAPs listed in this

proposed rule.

In its report regarding the risk assessment of HAPs, the National

Academy of Sciences (NAS) states that ``[a]vailability of requisite

data varies widely among the 189 [HAPs] chemicals'' (Ref. 5). According

to this report, ``the toxicity data are incomplete on almost all 189

chemicals'' (Ref. 5). For example, the level of carcinogenic risk for

approximately 40% of the 189 HAPs cannot be classified under EPA's

current cancer risk classification system (Ref. 6). Moreover, while

quantitative estimates exist for 70% of the HAPs that have been

classified for cancer risk, about 70% of these estimates are based only

on oral data and thus may not reliably characterize the potential risk

encountered through inhalation exposure.

In evaluating 124 of the 189 HAPs for noncancer risk, EPA found

that the databases for about 62% of the 124 HAPs were not adequate for

deriving an RfC (Ref. 7). Even for those HAPs that have an RfC, the

level of uncertainty associated with such figures is often high. EPA

would use data from the testing proposed in this rule to identify the

critical health risks posed by many individual HAPs and to characterize

the adverse impact posed by exposure to mixtures of HAPs. EPA

anticipates that the test data produced in response to this rule would

provide the consistent database that the National Academy says is

lacking at this time.

III. Testing Approach and Selection of Chemical Substances for This

Proposed Rule

A. Testing Approach Chosen by EPA

With respect to EPA's responsibilities for meeting the requirements

under section 112 of the CAA, the central question is: How broad and

deep a data set should EPA require on each HAP?

Regarding specific endpoints and routes of exposure, CAA section

112(b)(2), 42 U.S.C. 7412(b)(2), indicates that Congress intended that

adverse effects from any endpoint by any route of exposure be taken

into account in listing substances as HAPs. According to this

subsection, substances added to the Clean Air Act HAPs list shall

include:

* * * pollutants which present, or may present, through

inhalation or other routes of exposure, a threat of adverse human

health effects (including, but not limited to, substances which are

known to be, or may reasonably be anticipated to be, carcinogenic,

mutagenic, teratogenic, neurotoxic, which cause reproductive

dysfunction, or which are acutely or chronically toxic) * * * .

Thus the CAA indicates that Congress was very concerned about the

wide variety of health risks attributable to HAPs and intended that

data necessary for characterizing both cancer and noncancer health

risks from exposure to HAPs be developed.

Faced with a broad range of options and little specific guidance

from Congress, EPA decided that some provision should be made for

evaluating the health effects endpoints listed in the CAA, including

respiratory tract toxicity, systemic effects, reproductive toxicity,

developmental toxicity, genotoxicity, neurotoxicity, and

carcinogenicity. EPA's objective was to select endpoints representing

serious health effects that could occur as a result of exposure to

HAPs. Each endpoint represents a health effect of concern arising from

one or more of the following exposures--local concentrations (e.g.,

hotspots and plumes), area-wide sources, or accidental releases of

HAPs.

EPA believes that it is critical to evaluate the respiratory tract

thoroughly in addition to examining extra-respiratory effects (i.e.,

systemic toxicity) because inhalation is an important exposure route of

concern. Carcinogenicity testing is significant because cancer is a

serious health effect that may be caused by long-term, low-level

exposure to toxic substances. Developmental toxicity addresses the

potential of chemical substances to interfere adversely with human

development (i.e., to cause death, structural abnormalities, growth

alterations, and/or functional deficits in the immature organism that

may be more sensitive than the adult to many chemical substances).

Reproductive testing is designed to assess the effects of an

environmental agent on male and female fertility and general

reproductive function to humans exposed prenatally as well as

postnatally. There is general consensus among toxicologists regarding

the assessment of cancer and reproductive and developmental effects,

and further explanation can be found in EPA's risk assessment

guidelines (Guidelines for Carcinogen Risk Assessment (51 FR 33992,

September 24, 1986); Guidelines for Developmental Toxicity Risk

Assessment (56 FR 63798, December 5, 1991); Guidelines for Reproductive

Toxicity Risk Assessment (Pub. No. EPA/600/AP-94/001, February 1994).

Finally, certain aspects of the neurotoxicity and immunotoxicity

testing required in this proposed rule warrant more explanation, which

is provided below. EPA recently published a proposed revision of the

1986 Guidelines for Carcinogen Risk Assessment (see ``Proposed

Guidelines for Carcinogen Risk Assessment'' at 61 FR 17960, April 23,

1996 (FRL-5460-3)).

Neurotoxicity resulting from chemical exposure can affect an

organism in many ways, causing, for example, functional and structural

deficits as well as behavioral effects. To assess neurotoxicity, EPA is

proposing a screening-level battery, consisting of the functional

observational battery and motoractivity and neuropathology tests. At

this time, EPA is not proposing to require additional, more specialized

testing for cognitive functions such as learning, memory, and

performance.

The interest in the potential toxic effects of chemicals on the

immune system arises from the critical role that the immune system

plays in maintaining health. EPA considers the field of immunotoxicity

a promising, scientifically sound, and important area in public health

protection. From time to time, the Agency has considered information on

the effects of chemicals on the immune system in risk assessments. For

example, in its draft report reassessing the effects of dioxin

compounds on human health (Ref. 8), EPA considered the effects of

dioxin on the immune system to serve as an important health endpoint

that provides useful information in developing a hypothesis about

toxicity. In the draft reassessment report, however, EPA arrived at the

preliminary conclusion that ``the impact of dioxin and related

compounds on the immune system and

[[Page 33182]]

implications for characterizing risks are largely unknown at this

time.''

This rule calls for an immunotoxicity screening test which can be

performed as a satellite test to either a 90-day subchronic test or a

reproductive effects test. This immunotoxicity screen will help

identify chemicals as potential immunotoxicants. EPA is not proposing

more comprehensive immunotoxicity testing at present because the

application of immunotoxicity data in risk assessment has not yet

sufficiently matured. As EPA's science policy develops and the Agency's

use of immunotoxicity data in risk assessment increases, EPA will

reconsider this position. Meanwhile, EPA seeks comments on its proposed

approach of using a minimal screen and its preliminary conclusion that

it is premature at this time to include more comprehensive

immunotoxicity testing in this proposed HAPs test rule.

In developing this proposed rule, EPA considered the following

range of options to select the information needed to characterize

health effects of concern to implement section 112 of the CAA.

Option 1. One-species 90-day inhalation subchronic plus follow-up

for known or suspect toxicities. Under this option, a 90-day inhalation

subchronic test would be required, as well as testing for endpoints

that have already been identified as existing or potential concerns,

including cancer, by previous test results of the HAP at issue or

structurally similar agents. A one-species 90-day inhalation test is

considered the minimum information for the development of an RfC. The

inhalation RfC takes into account toxic effects both for the

respiratory tract (portal-of-entry effects) and peripheral to the

respiratory system (extra-respiratory effects). Well-defined and well-

conducted inhalation subchronic toxicity studies--that provide for

histopathologic evaluation of organ toxicities, including the

respiratory tract--are considered to be reliable predictors of certain

kinds of chronic toxicity. But such studies do not, or do not

adequately, account for neurological, developmental and reproductive

toxicities. An RfC based solely on a 90-day subchronic test is, thus,

usually given a low confidence rating because some potentially

important toxic endpoints are not characterized.

In addition, EPA believes that for 90-day inhalation subchronic

testing to constitute a minimally credible option, such a test should

at least be augmented by testing for adverse health effects that are

suggested or indicated, but not adequately characterized, by existing

information such as short-term test data, mechanistic information or

structure-activity relationships (SAR). Even with this modification,

however, Option 1 still provides no test data on those health endpoints

of concern for which no current information exists. If such testing

were included, these effects might become critical in evaluating dose-

response relationships or in demonstrating that a standard uncertainty

factor is inadequate or inappropriate.

EPA thus believes that Option 1 is insufficient to meet EPA's

mandate under section 112 of the CAA because the endpoints listed in

section 112(b)(2) of the CAA, in particular, reproductive,

developmental, and neurological toxicities, would be considered for

testing only if data already exist that indicate or suggest the

potential for these adverse effects. Moreover, this option also does

not adequately address health risks associated with acute or accidental

releases.

EPA did not select Option 1 for the reasons stated above. EPA

believes that the TSCA section 4 program adopted for testing HAPs must

go further toward ensuring that no serious health threat exists from

both long- and short-term exposure for endpoints of potential concern

for which there are no existing data.

Option 2. Option 1 plus inhalation screening for untested toxicity

endpoints. The second option considered by EPA would require the

incorporation of screening level testing for certain untested

toxicities into Option 1. Like Option 1, Option 2 would include testing

for endpoints (including cancer) that have already been identified as

existing or potential concerns. At a minimum Option 2 would consist

of--a 90-day subchronic inhalation study, a screening test for

reproductive effects (i.e., a one-generation reproductive effects

study), a subchronic inhalation neurotoxicity screening battery

(consisting of the functional observation battery and motor activity

and neuropathology tests), an E. coli reverse mutation assay, gene

mutation in somatic cells in culture detection, an in vivo cytogenetics

test (chromosomal analysis or micronucleus assay), and an

immunotoxicity screening test. Any toxicity suggested but not

characterized by existing studies in the toxicological literature would

still be followed up on through more rigorous protocols.

Although Option 2 would conserve resources while allowing for the

testing of a broader range of endpoints, including cancer, it has

serious shortcomings. First, a one-generation reproductive test does

not adequately address reproductive and developmental risk. Two-

generation tests (in which animals have been exposed prenatally as well

as postnatally, including the prepubertal period) are generally needed

to evaluate the effects on reproduction from most exposures to chemical

substances (Ref. 9). Two-generation tests permit the evaluation of

delayed or latent manifestations of some toxicities, detection of

effects absent in the first generation, and the expression and

detection of some effects that may have a heritable basis. Because the

standard two-generation reproductive test would not detect internal

malformations, however, developmental toxicity testing is also needed

for an adequate assessment of developmental risk. The Agency's policy

is to require developmental testing in two species to adequately

characterize the risk because of species-specific differences.

EPA did not select Option 2 because this level of testing would not

provide an adequate evaluation of developmental or reproductive

toxicity. Additional follow-up testing would be required to confirm

suggestive results obtained in screening studies and provide data

adequate for risk assessment under this option. Such testing would

require an additional rulemaking cycle, costing further resources and

incurring so much delay that data would not be available to meet the

deadlines for setting risk-based standards. Moreover, this option does

not adequately address health risks, such as respiratory tract effects

and neurotoxicity, associated with acute or accidental releases.

Option 3. Option 1 plus less than chronic testing for noncancer

endpoints of concern. In addition to the 90-day inhalation subchronic

testing specified in Option 1, this option would add inhalation testing

to assess reproductive effects (i.e., two-generation reproductive test)

and developmental effects (developmental toxicity tests in two

mammalian species). Option 3 includes an acute toxicity testing

guideline for histopathology of the respiratory tract, kidney, and

liver and a bronchoalveolar lavage after four hours of exposure. EPA

believes that it is necessary to characterize the acute effects

associated with accidental releases of HAPs. In addition, a respiratory

sensory irritation assay is included. Acute and subchronic inhalation

neurotoxicity screening batteries consisting of the functional

observation battery, and motor activity and neuropathology tests would

also be conducted. As in Option 2, first-tier tests would be required

for mutagenicity

[[Page 33183]]

(i.e., an E. coli reverse mutation assay, gene mutation in somatic

cells in culture detection, an in vivo cytogenetics test (chromosomal

analysis or micronucleus assay)), as well as immunotoxicity.

Option 3 would follow Options 1 and 2 in requiring a cancer

bioassay where concern for cancer is indicated by short-term data,

general toxicity data, mechanistic information or structure-activity

relationships (SAR). Where no cancer bioassay data exist, testing two

species in both sexes would be required. If cancer bioassay data exist

but are found to be too uncertain for inhalation dose-response

assessment, a modified test, such as testing of the opposite sex in two

species, may be required (Ref. 10).

The Option 3 level of testing would enable EPA to better

characterize risk associated with both acute and longer-term exposures

by providing data to identify and evaluate all the health effects

listed under section 112 of the CAA and by providing data for dose-

response evaluation within the general time frame for risk-based

standards under CAA section 112(f). Accordingly, EPA has selected the

Option 3 level as its preferred option for testing under this proposed

rule.

Option 4. Option 3 plus chronic testing. Under this option, in the

absence of existing adequate data, EPA would require chronic inhalation

bioassays (for both cancer and noncancer effects) in two different

mammalian species for each chemical substance. The balance of the test

program would be the same as under Option 3 (developmental studies in

two mammalian species, a two-generation reproductive study, acute and

subchronic neurotoxicity screening batteries, first-tier mutagenicity

tests, an immunotoxicity screening test, and acute testing). In

general, cancer bioassay data in two species, a two-generation

reproductive test, and a developmental study in two species are

required to establish a high-confidence RfC. Because the RfC is

intended to serve as a lifetime estimate, lifetime exposure studies to

evaluate potential health endpoints at various critical life stages

should be considered.

To a greater degree than under other options, the broad and deep

database that would be produced by this comprehensive testing scheme

could help defuse complaints that EPA frequently regulates industrial

activities without sufficient data regarding either the need for an

appropriate level of regulation or what such a level should be. EPA has

decided, however, that the disadvantages of choosing this option

outweigh its considerable benefits. The extensive chronic testing

required under Option 4 would impose a significant cost on industry. In

addition, as compared to Option 3, the strain that choosing this option

would place on certain resources--such as inhalation testing facilities

and supplies of laboratory animals--would significantly diminish the

cost-effectiveness of compiling the data. For these reasons, EPA did

not select this option. EPA is soliciting comments on the testing

approach to the HAPs that it has selected in this proposal.

It should be noted that, regardless of the test option chosen, if

adequate toxicity data on a HAP is produced by testing using a route of

exposure other than inhalation, route-to-route extrapolation may be

possible (see Unit IV.D. of this preamble).

B. National Academy of Sciences Approach

In section 112(o) of the CAA, 42 U.S.C. 7412(o), Congress directed

EPA to arrange for the NAS to review EPA's risk assessment methodology

relevant to HAPs subject to section 112. EPA has considered the

recommendations of the NAS regarding the assessment of risks associated

with HAPs. The NAS recommended that EPA ``* * * compile for each of the

189 chemicals an inventory of the existing and relevant chemical,

toxicologic, clinical, and epidemiologic literature'' (Ref. 5). It also

recommended that EPA ``screen the 189 chemicals for priorities for the

assessment of health risks, identify the data gaps, and develop

incentives to expedite generation of the needed data by other public

agencies (such as the National Toxicology Program, the Agency for Toxic

Substances and Disease Registry, and state agencies) and by other

organizations (industry, academia, etc.)'' (Ref. 5). As discussed in

Unit III.C. of this preamble, EPA agrees and has taken this approach.

To identify testing needs and help prioritize HAPs testing, Syracuse

Research Corporation, an EPA contractor, has identified and summarized

the existing health and exposure literature on the HAPs, and has

identified testing programs currently in progress (Refs. 11, 12).

The NAS report also discussed how a gradual, highly iterative

testing approach to the generation of health effects data on HAPs might

work. The report recommended that HAPs could be prioritized on the

basis of their acute toxicity and chemical structure, and testing might

proceed stepwise, on a case-by-case basis, from acute toxicity to

studies of the uptake, distribution, retention, and excretion of the

substance, to subchronic toxicity, and ultimately, if needed, to

endpoint testing in animals. Depending on the animal toxicity data

produced by this iterative testing scheme, according to the NAS, EPA

might decide that further studies of human toxicity or mechanisms of

toxicity are warranted.

Although EPA agrees with the need to prioritize testing, it has

taken a different approach to prioritization that is based on

consideration of exposure potential and the rulemaking schedule of

section 112 of the CAA. The amount and type of existing data vary

greatly among the chemical substances that Congress designated as HAPs.

In practice, therefore, no single uniform iterative approach based on

toxicity factors alone would apply to all chemical substances. For

example, one HAP might have only acute and short-term test data, while

longer-term studies might exist for another HAP. Nevertheless, both the

NAS and EPA approaches recognize that existing data must be considered

if EPA is to avoid requiring duplicative testing that previously

produced adequate data.

An iterative testing approach based on toxicity factors alone would

be time consuming and require multiple rulemakings. This process would

take too long to collect useful data for making decisions needed to

meet upcoming statutory deadlines established in the CAA. Furthermore,

multiple iterative rulemakings to develop needed test data would be

prohibitively costly to EPA and would not recognize limitations on EPA

resources. For Option 3, EPA's preferred testing level, follow-up

testing would rarely be required beyond that level proposed in this

rule. Such testing, if necessary, would be required in a separate

rulemaking.

To make the multichemical decisions required under section 112 of

the CAA regarding, for example, residual risk and delisting HAPs, EPA

believes that it needs a consistent, even database covering HAPs across

the same broad set of endpoints. EPA believes that Option 3 will permit

timely gathering of a consistent database on HAPs more efficiently and

at less cost to industry and EPA than is possible with other

approaches.

C. Review of Data and Selection of HAPs

In choosing candidates for this proposed test rule, EPA considered,

consistent with TSCA section 4 requirements, the potential for a

chemical substance to present an unreasonable risk of injury to health

or the environment, the production

[[Page 33184]]

volume of the substance, the amount of emissions produced by the

chemical substance entering or reasonably anticipated to enter the

environment or become a source of exposure for humans, the sufficiency

of the existing database, and the need for further testing to develop

needed data. Consequently, as indicated in Unit IV of this preamble,

and explained in a separate document in the record entitled ``TSCA

Section 4 Findings for 21 Hazardous Air Pollutants'', each candidate

listed for testing in this proposed rule is:

(1) Considered to have the potential of presenting an unreasonable

risk of injury to health or the environment (except in the case of

ethylene glycol, for which no A finding was made--see the table in Unit

V of this preamble);

(2) Produced in quantities exceeding 1,000,000 pounds per year;

(3) Emitted (i.e., released into the atmosphere) in the amount of

50 tons (100,000 pounds) per year or more according to the 1993 Toxics

Release Inventory (TRI);

(4) Considered to have health effects data needs not addressed in

other testing and research programs;

(5) Considered to have health effects data that are insufficient

under TSCA section 4 for determining effects of the HAP on health; and

(6) Considered to need further testing to develop the needed data.

The determination that data are insufficient to ascertain the

effects of the HAPs on human health is based on several factors. First,

EPA determined the effects of concern (toxicological endpoints) and the

depth and quality of data which the Agency needs in order to make

residual risk determinations. This decision and the range of options

EPA considered are discussed in Unit III.A. of this preamble. Having

made the decision that standard endpoint tests are appropriate, EPA

reviewed existing studies and, for the purposes of this rule, compared

such studies against the testing methodology used in the 1985 version

of the EPA test guidelines for these endpoints. The 1985 test

guidelines were the first test guidelines issued by EPA for its TSCA

chemical testing program and represent widely accepted, peer-reviewed

methods for characterizing chemical toxicity.

The reasons why existing studies were judged to be inadequate are

explained in a separate document in the record entitled ``TSCA Section

4 Findings for 21 Hazardous Air Pollutants'', and summarized in the

table in Unit V of this preamble. The reasons are varied but include

the following examples--not studying the appropriate endpoint; too few

dose levels; inappropriately high- or low-dose levels; and too few

animals to have statistical confidence in the results. Nevertheless, in

some cases, EPA toxicologists determined that data were adequate when

the weight of evidence from several flawed studies, which, when

considered individually were determined to be inadequate, gave an

adequate characterization of the toxicity of the substance. Thus expert

judgment must always play a role in determinations of data adequacy.

Indeed, the determination of adequacy is so intimately connected to the

unique characteristics of study design for each toxicological endpoint

that EPA is unable to articulate a universal test of data adequacy that

might be applied consistently in all situations. EPA is soliciting

comments on its approach to determining data adequacy for the HAPs.

To select HAPs for testing, EPA initially reviewed the production

data and TRI data for all 189 HAPs. EPA realizes that TRI data

represent estimates of environmental emissions of the TRI-listed

chemicals and do not account for all chemical substances in the United

States. Nevertheless, TRI figures offer the most complete, readily

available emissions data, and EPA has determined that this database is

sufficient for the purpose of helping EPA select high-emission HAPs for

consideration as potential test candidates. While publicly available

sources of production data are cited in the analysis supporting this

rule, data from these sources were checked against the TSCA chemical

inventory update production data, most of which are claimed as CBI.

After reviewing TRI data for all HAPs, EPA decided to select a

number of HAPs for initial consideration by focusing its attention on

HAPs with TRI emissions of 50 tons or more per year. The 66 HAPs in

this group constituted a reasonably sized group for further review. The

selection of 50 tons per year or more as a cutoff is appropriate

because this number captures high-emission HAPs and because section 112

(a)(1) of the CAA, 42 U.S.C. 7412(a)(1), defines ``major source'' as

emitting ``* * * 10 tons per year or more of any hazardous air

pollutant or 25 tons per year or more of any combination of hazardous

air pollutants * * *.''

A survey of testing conducted by EPA under TSCA section 4, the

Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA), and other

testing programs supplemented a preliminary review of health findings

from secondary source documents and the IRIS database. This initial

survey revealed that certain HAPs having high emissions (50 tons or

more) already have a large inhalation toxicology database or are

subject to testing or research in existing programs. Therefore, EPA

decided not to pursue additional testing under this rule for benzene,

butadiene, carbon disulfide, chromium, cyanide, ethylene oxide,

formaldehyde, lead, methanol, methyl tertiary butyl ether, methylene

chloride, tetrachloroethylene, toluene, trichloroethylene, vinyl

chloride, and vinyl acetate. Additional testing under TSCA for these

chemicals may be considered at some time in the future.

EPA decided that the remaining group of 50 HAPs could be handled

most efficiently by promulgating more than one rule. Consequently, 21

high-emission HAPs were scrutinized further and were selected as

candidates for this proposed rule, and the remaining 29 HAPs were

deferred for consideration in subsequent HAPs test rulemaking efforts.

In the second HAPs test rule, EPA plans to focus on persistent HAPs

that may bioaccumulate. EPA may, therefore, require ecological and

environmental testing for these HAPs. EPA also may require testing in

the second rule to collect data needed to implement the ``Great

Waters'' program of section 112(m) of the CAA, 42 U.S.C. 7412(m).

During the selection process for this proposed rule, EPA's

contractor undertook a comprehensive search of the toxicological,

health, and exposure literature for the 21 HAPs proposed for testing in

the current rule (Refs. 11, 12). EPA's contractor performed the

literature search in a stepwise manner to save both time and expense.

The first step was to review secondary source health effects documents.

EPA's contractor identified documents published by EPA, the

International Agency for Research on Cancer (IARC), and ATSDR and

extracted relevant data. EPA performed online Environmental Mutagen

Information Center (EMIC) searches for genetic toxicity information

(Ref. 13). In addition, an article entitled ``Genetic Activity Profiles

of 110 Hazardous Air Pollutants Listed Under Title III of the Clean Air

Act'' (Ref. 14) and the International Commission for Protection against

Environmental Mutagens and Carcinogens (ICPEMC) have provided useful

summary information.

EPA realizes that using secondary sources of information is not

ideal. For example, it is possible that a secondary source document

could miss an important study or that the document could fail to

properly interpret a study. Consequently, whenever essential

[[Page 33185]]

information appeared to be missing from the review documents or was not

explained clearly, EPA's contractor consulted original articles.

Through its contractor, EPA next checked several sources for

relevant published and unpublished studies. It obtained unpublished but

publicly available studies submitted to EPA under TSCA, searched the

Toxic Substances Control Act Test Submissions (TSCATS) database by CAS

Registry number, and reviewed the National Toxicology Program (NTP)

Results Report (generated from NTP's CHEMTRACK database) to locate

completed but unpublished NTP studies. With the contractor's

assistance, EPA next undertook an update search of the open literature

to locate any as yet unidentified studies published either shortly

before or after the review documents appeared. For this purpose, EPA's

contractor searched the National Library of Medicine (NLM) TOXLINE

database for studies published during a period of time beginning three

years prior to the date of the review document initially used to obtain

toxicity information and ending on the date of the search.

In addition, EPA's contractor consulted with representatives of

NIOSH, OSHA, the Food and Drug Administration (FDA), the National

Institute of Environmental Health Sciences (NIEHS), and various

chemical companies. The purpose of these inquiries was to determine if

these organizations had any information on completed or ongoing studies

that might not be found in any readily available database. Through its

contractor, EPA also contacted the Chemical Industry Institute of

Toxicology (CIIT), as well as trade associations and allied

organizations to determine whether these organizations were sponsoring

or knew of any relevant studies currently in progress. Finally, EPA's

contractor closely reviewed data sheets compiled by EPA's IRIS RfD/RfC

Working Group to ascertain if the group had identified any additional,

otherwise unlocated information.

Varied levels of scrutiny were applied to different types of

toxicity testing information throughout the literature search. Because

the primary focus of the review of acute, subchronic, and chronic

systemic toxicity literature was inhalation exposure, only inhalation

studies were reviewed for these endpoints. Although oral studies can

provide important information on target organ toxicity and should be

considered in the design of any testing protocol, these studies usually

provide limited information on the effects of a compound on the

respiratory tract. In addition, the systemic dose remote to the

respiratory tract for many compounds is affected by modulation of

uptake at the portal of entry into the body. This modulation is not

only from first-pass effects but from other influences of anatomy and

physiology (Ref. 15). Because EPA's literature search did not encompass

oral acute and subchronic toxicity studies, the preliminary findings of

risk that EPA is making below, under TSCA section 4(a)(1)(A), are not

based on such studies. Thus, oral acute and subchronic studies may

provide additional evidence of potential toxicity.

The contractor reviewed studies of carcinogenicity, neurotoxicity,

and reproductive and developmental toxicity, regardless of the route of

administration.

EPA took a different approach to identify HAP candidates for

immunotoxicity testing. EPA relied on an EPA document ``Hazardous Air

Pollutants: Profiles of Non-Cancer Toxicity from Inhalation Exposures''

(Ref. 16), containing a database that was developed from EPA and ATSDR

documents and data files, and from the Hazardous Substances Data Bank

(HSDB) of NLM. The contractor searched recent literature (i.e., 1989 to

present) for immunotoxicity data on the 21 HAPs in both MEDLINE and

TOXLINE. For chemicals with ATSDR Toxicological Profiles, the profile

was used to identify immunotoxicity data. Much of the identified

immunotoxicity literature used rather insensitive indicators of impact

(e.g., organ weight changes, histopathology, leukocyte counts, and

total serum protein determinations), that were judged to constitute an

inadequate evaluation of suppression of immune system responsiveness

(Ref. 17). Thus, an immunotoxicity screening test is being proposed in

this rule for many of these HAPs.

Although EPA has made intense and thorough attempts to identify all

relevant studies, EPA recognizes the limitations inherent in relying on

secondary sources and realizes that its literature search may have

failed to locate studies recently undertaken or completed. Therefore,

EPA solicits comments bringing to its attention any valid studies not

identified in its search efforts.

D. Previous TSCA Testing Actions Affecting These Chemical Substances

Eight of the substances included in this proposed rule have been

the subject of previous testing under TSCA section 4. Testing by the

inhalation route was not generally required, however, and acute

effects--including respiratory tract effects--were not generally a

target endpoint. This subunit will briefly summarize previous testing

decisions and explain the relationship between those activities and

this proposed rule.

1,1'-Biphenyl was recommended by the Interagency Testing Committee

(ITC) in its 10th report for environmental effects and chemical fate

testing (47 FR 22585, May 25, 1982). Focusing only on environmental

testing, EPA found that 1,1'-biphenyl may present an unreasonable risk

to the environment and issued a test rule requiring environmental

effects and chemical fate testing of the chemical on September 12, 1985

(50 FR 37182). This proposed rule complements the earlier action by

requiring health effects testing of 1,1'-biphenyl, namely, acute

toxicity, respiratory sensory irritation, subchronic toxicity,

developmental toxicity, reproductive toxicity, neurotoxicity and

immunotoxicity.

Chlorobenzene was recommended to EPA for health and environmental

effects testing in the first report of the ITC (42 FR 55026, October

12, 1977). Subsequently, EPA found that the chemical may present an

unreasonable risk to human health (an A finding) and issued a rule

requiring reproductive effects testing (51 FR 24657, July 8, 1986).

Although the preamble of the proposed rule described specific

neurotoxicity concerns, EPA stated that neurotoxicity testing

requirements were not being proposed because it had not issued

neurotoxicity test guidelines at that time. Instead EPA explained its

then-current views on neurotoxicity testing in the preamble and

solicited public comment on those views (45 FR 48524, July 18, 1980).

Because a neurotoxicity screening battery guideline (OPPTS 870.6200)

has since been proposed, this rule proposes the testing of

chlorobenzene for neurotoxicity, acute toxicity, respiratory sensory

irritation, subchronic toxicity, and immunotoxicity.

Cresols are members of a chemical category consisting of three

isomers: ortho-, para-, and meta-cresol. Based on both A and B

findings, a test rule proposed on July 11, 1983 (48 FR 31812) would

have required testing of cresols for subchronic toxicity, mutagenicity,

carcinogenicity, developmental toxicity, reproductive effects,

neurotoxicity, and skin sensitization. The final rule, published on

April 28, 1986 (51 FR 15771), which specified testing for all three

isomers and provided a rationale for this decision, required testing

for mutagenicity, developmental toxicity, and reproductive effects.

Data received

[[Page 33186]]

under this test rule satisfy the HAPs data needs for these endpoints.

Based on the results from this first tier of tests, a conditionally

required cancer bioassay was not triggered. In addition, oral

subchronic toxicity studies and subchronic neurotoxicity studies were

conducted by EPA's Office of Solid Waste. In accordance with the need

for data on respiratory tract effects, today's rule proposes acute and

subchronic inhalation toxicity, respiratory sensory irritation, acute

neurotoxicity, and immunotoxicity tests for all three cresol isomers

(see Unit IV.B. of this preamble). For the purposes of this proposal,

the three cresol isomers are counted as a single chemical.

Methyl isobutyl ketone was the subject of a negotiated testing

agreement between EPA and industry for mutagenicity, developmental

toxicity, and subchronic testing (47 FR 58025, December 29, 1982, and

48 FR 44905, September 30, 1983). Data received under the negotiated

testing agreement satisfy the HAPs data needs for these endpoints.

Methyl isobutyl ketone is also being tested for neurotoxicity under a

TSCA enforceable consent agreement (ECA) with industry (announced at 60

FR 4514, January 23, 1995 (FRL-4924-8)). This rule proposes testing for

reproductive toxicity, acute toxicity, respiratory sensory irritation,

and immunotoxicity to complement ongoing testing and existing data.

Phenol is the subject of a test rule proposed on the basis of A and

B findings on November 22, 1993 (58 FR 61654). That rule proposed

subchronic toxicity, neurotoxicity, and reproductive and developmental

toxicity testing and a study of phenol's pharmacokinetics. EPA has

received a proposal for an ECA for this chemical substance that would

cover the testing proposed in the 1993 rule. The rule proposed herein

would add acute inhalation toxicity, respiratory sensory irritation,

and immunotoxicity to the testing program for phenol. Under the

procedures set forth at 40 CFR 790.22, members of the CMA Phenol Panel

and EPA have negotiated an ECA which provides for the testing proposed

in November 1993 as well as additional testing, including

immunotoxicity. Such testing would meet the HAPs-related data needs for

phenol. If the ECA is successfully concluded, EPA will drop the testing

requirement for phenol from the final HAPs rule.

1,2,4-Trichlorobenzene has been tested for carcinogenicity under a

test rule (51 FR 24657, July 8, 1986) based on an A finding. Data

received under this test rule satisfy the HAPs data needs for this

endpoint. Although the preamble of the proposed rule described specific

neurotoxicity concerns, EPA stated that neurotoxicity testing

requirements were not being proposed because it had not issued

neurotoxicity test guidelines at that time (45 FR 48545, July 18,

1980). Because a neurotoxity screening battery guideline (OPPTS

870.6200) has since been proposed, this rule proposes the testing of

1,2,4-trichlorobenzene for acute toxicity, respiratory sensory

irritation, neurotoxicity, immunotoxicity, and developmental toxicity.

Oncogenicity testing for vinylidene chloride was called for in a

proposed test rule based on an A finding on August 12, 1986 (51 FR

28840). The rule proposed that distribution, metabolism, and excretion

studies and an inhalation oncogenicity study be conducted in mice on

behalf of EPA's Office of Air Quality Planning and Standards (OAQPS).

EPA has not finalized the vinylidene chloride proposal and is hereby

withdrawing it. EPA is not pursuing these studies because the Agency

has concluded that, at this time, an oncogenicity bioassay would do

little to add to EPA's understanding of the oncogenic potential of the

substance. Today's rule proposes testing for acute toxicity,

respiratory sensory irritation, and neurotoxicity.

IV. Proposed Testing

A. Testing and Reporting Requirements

EPA is proposing specific testing and reporting requirements for

each of the 21 HAPs as specified in table 1 in Sec. 799.5053(a)(5) of

this proposed rule. EPA is proposing for the first time in a TSCA

section 4 rule to require an immunotoxicity screen and an acute

inhalation toxicity test that focuses on respiratory damage and

sublethal systemic toxicity. These and other test guidelines are

discussed below in Unit IV.C. of this preamble.

EPA is proposing to require a modified inhalation carcinogenicity

bioassay using only the male rat and female mouse when existing oral

carcinogenicity data and supporting information for a chemical

substance are deemed too uncertain to determine its carcinogenicity via

inhalation (Ref. 10). The reduced protocol is less expensive than a

traditional bioassay. However, test sponsors would also have the

alternative of performing pharmacokinetics studies and using route-to-

route extrapolation from existing adequate oral toxicity data under

enforceable consent agreements (EDAs) in lieu of this and other test

requirements if the Agency decides to use this approach (see Units

IV.D. and IV.E. of this preamble).

A total of 21 months would be given for the submission of final

reports for acute toxicity testing because the acute inhalation

toxicity with histopathology guideline proposes to make certain

histopathology studies contingent upon the results of the 90-day

subchronic studies. The time for the submission of immunotoxicity

studies would vary as a function of the test with which they can be

combined (e.g., subchronic and reproductive effects).

B. Test Substance

EPA is proposing that a substance of at least 97% purity be used as

the test substance. EPA recognizes that exposure to HAPs will occur as

exposure to complex mixtures and that ideally one would like data on

the mixtures themselves. However, it is not practical to test mixtures

due to the huge number of possible combinations. EPA will thus evaluate

the toxicity of HAP mixtures using data on the relatively pure

components in order to avoid the possible confounding effects of

impurities that might be found in technical grade substances. These

impurities, if substantial contributors to air pollution, should also

be captured as separate entries on the CAA list of HAPs. EPA believes

that a purity of 97% is available or readily achievable for all

substances covered by this rule.

For cresols, the subject of the test rule is a mixture (CAS No.

1319-77-3) of three isomers: ortho- (CAS No. 95-48-7), para- (CAS No.

106-44-5) and meta- (CAS No. 108-39-4). The mixture and individual

isomers are contained in the CAA section 112(b)(1) list of hazardous

air pollutants. Most human exposure is to the mixture. However, because

the mixture is of variable composition, EPA believes that it would be

very burdensome to test every possible variation of the mixture, which

would have different proportions of isomers. Therefore EPA is proposing

to follow the approach taken in the final test rule for cresols (51 FR

15771, 15776, April 28, 1986) and test each isomer (see Unit III.D. of

this preamble).

Another critical factor in study design for HAP testing is the low

vapor pressure of several of these substances (diethanolamine, 1,1'-

biphenyl, phthalic anhydride). This raises two questions. To which

forms of the chemical are humans exposed--vapor, aerosol or particle?

How does one design a valid toxicity study that can be used to assess

human risk to such exposures? Given the reported TRI releases to the

atmosphere for these substances, EPA has assumed that exposures are to

[[Page 33187]]

aerosols or particulates from the condensation of high temperature

stack gases. EPA is proposing that diethanolamine, 1,1'-biphenyl and

phthalic anhydride be tested via aerosol exposure. EPA invites

manufacturers and processors to submit information about the forms of

these substances that are encountered in ambient exposures and the

forms that should be tested, and encourages the development of

pharmacokinetics data that would permit testing by the less expensive

oral route for HAPs with low vapor pressure.

C. Test Guidelines

The 11 guidelines being proposed for use in testing HAPs under this

rule are included in the recently harmonized health effects test

guidelines proposed by EPA's Office of Prevention, Pesticides, and

Toxic Substances (OPPTS). When final, these harmonized guidelines will

incorporate an updated version of the test guidelines previously

developed for use under TSCA and FIFRA into a single set. A notice of

availability and request for comments on the proposed guidelines was

published in the Federal Register of June 20, 1996 (61 FR 31522).

The 11 guidelines proposed for use in testing HAPs are included in

the public version of the record for this rulemaking at the address

specified in the ``ADDRESSES'' section of this document. The complete

set of proposed guidelines is available electronically from the EPA

Public Access Gopher (gopher.epa.gov) under the heading:

``Environmental Test Methods and Guidelines''; by internet e-mail:

[email protected]; by mail: Public Docket and Freedom of

Information Section, Field Operations Division (7506C), Office of

Pesticide Programs, U.S. Environmental Protection Agency, Washington,

DC 20460; or in person or for courier pick-up: Room 1132, CM #2, 1921

Jefferson Davis Highway, Arlington, VA; telephone: (703) 305-5805.

The 11 guidelines proposed to be used for testing HAPs are as

follows:

(1) Acute Inhalation Toxicity with Histopathology, OPPTS 870.1350,

EPA Pub. No. 712-C-96-291, June 1996;

(2) Subchronic Inhalation Toxicity, OPPTS 870.3465, EPA Pub. No.

712-C-96-204, June 1996;

(3) Inhalation Developmental Toxicity Study, OPPTS 870-3600, EPA

Pub. No. 712-C-96-206, June 1996;

(4) Reproduction and Fertility Effects, OPPTS 870.3800, EPA Pub.

No. 712-C-96-208, February 1996;

(5) Carcinogenicity, OPPTS 870.4200, EPA Pub. No. 712-C-96-211,

June 1996;

(6) Escherichia coli WP2 and WP2 uvrA Reverse Mutation Assays,

OPPTS 870.5100, EPA Pub. No. 712-C-96-247, June 1996;

(7) Detection of Gene Mutations in Somatic Cells in Culture, OPPTS

870.5300, EPA Pub. No. 712-C-96-221, June 1996;

(8) In Vivo Mammalian Cytogenetics Tests: Bone Marrow Chromosomal

Analysis, OPPTS 870.5385, EPA Pub. No. 712-C-96-225, June 1996;

(9) In Vivo Mammalian Cytogenetics Tests: Erythrocyte Micronucleus

Assay, OPPTS 870.5395, EPA Pub. No. 712-C-96-226, June 1996;

(10) Neurotoxicity Screening Battery, OPPTS 870.6200, EPA Pub. No.

712-C-96-238, June 1996; and

(11) Immunotoxicity, OPPTS 870.7800, EPA Pub. No. 712-C-96-351,

June 1996.

To be considered in this rulemaking, comments on the 11 proposed

test guidelines that are specific to HAPs testing must be submitted to

the OPPT Document Control Office. The comments must be submitted in the

manner specified in the ``DATES'' and ``ADDRESSES'' sections at the

beginning of this document. Comments on the 11 proposed guidelines,

which are not specific to the HAPs test rule must be submitted to the

Office of Pesticide Programs by August 19, 1996, at the address

identified in the Federal Register of June 20, 1996 (61 FR 31522).

The process of developing OPPTS harmonized guidelines described

above is proceeding at the same time as the development of the HAPs

test rule. The OPPTS harmonization process may result in the revision

of the guidelines prior to the end of the ocmment period for this

proposed rule. If so, EPA will announce the availability of those of

the 11 guidelines used in the HAPs rule that have been revised in order

to allow for public comment on the applicability of the revised

guidelines to the HAPs rule. If any of these 11 guidelines has not been

revised by the end of the comment period for this proposed rule, EPA

may issue the corresponding HAPs-specific guideline independent of the

OPPTS harmonization process.

EPA is proposing to modify the subchronic inhalation toxicity test

guideline (OPPTS 870.3465) for the purposes of this rulemaking to

include enhanced histopathology of the respiratory tract and an assay

for cell damage via lung lavage. EPA is requesting comment on adding

these parameters to the subchronic test guideline for testing HAPs.

As part of this rulemaking, EPA proposes to use the acute

inhalation toxicity with histopathology test guideline (OPPTS

870.1350). As indicated in Unit III.A. of this preamble, the study of

sublethal effects, especially effects on the respiratory system,

associated with accidental release and acute exposures is necessary for

the HAPs. The standard acute inhalation toxicity test guideline (OPPTS

870.1300) focuses on gross lesions, body weight changes, and effects on

mortality. The acute inhalation toxicity with histopathology test

guideline assesses two endpoints: (1) histopathology of the respiratory

tract, kidney, liver, and other target organs; and (2) cell damage via

lung lavage. The guideline takes a stepwise approach to the evaluation

of acute toxicity and initially requires a 4-hour exposure at three

concentration levels. If the 4-hour study shows positive results in

histopathology or the bronchoalveolar lavage, a 1-hour study and an 8-

hour study would be required to define better the time and

concentration dependence of acute exposures. Histopathology is being

proposed for the respiratory tract, liver, and kidney. Other target

organs identified by either gross pathology in the 4-hour acute study

or by histopathology in the 90-day study would also have to be examined

by histopathology in the 4-hour acute study. If these results are

positive in the 4-hour study, histopathology in the 1-hour and 8-hour

studies would be required. The 4-hour acute testing may be combined

with acute neurotoxicity testing.

A respiratory sensory irritation test using American Standard Test

Method (ASTM) E 981-84 is also being proposed to provide a quantitative

estimate of the sensory irritant potential of an inhaled chemical.

Irritation is detected by a characteristic change in the breathing

pattern of mice, which results in a reduction in the breathing rate

during exposure to a test atmosphere.

For all testing proposed in this rule, test sponsors would have to

conduct testing and generate data in accordance with the specified test

guideline. Data developed under the final rule must be reported in

accordance with TSCA Good Laboratory Practice (GLP) Standards, 40 CFR

part 792.

EPA is considering three alternative procedures for handling these

test guidelines in the context of the final HAPs test rule. The first

alternative is for the final HAPs rule to incorporate the guidelines by

reference. Under this alternative, the text of the guideline would not

appear in the Code of Federal Regulations. Instead the rule would

include a reference to the guideline which would be available on the

internet and elsewhere, as noted above.

[[Page 33188]]

A copy of the applicable guideline would also be maintained in the

public version of the rulemaking record.

The second alternative would be for the final HAPs rule to refer to

the guideline, the text of which would be available on the internet and

elsewhere, as the pre-approved protocol. However, test sponsors may use

other protocols after such protocols have been approved by EPA

(``previously approved equivalents''). If EPA decides on this course of

action, the Agency may issue a supplemental notice proposing specific

implementation procedures if they are significantly different from the

following procedures. A test sponsor would be required to submit to EPA

for review and approval each test protocol that such sponsor believes

is equivalent to the corresponding OPPTS test guideline. A submission

would have to demonstrate equivalency, include a description of the

differences between the sponsor's protocol and the corresponding OPPTS

guideline, and indicate the rationale for changing the guideline. The

deadline for these submissions would be 90 days after the effective

date of the final HAPs rule. In the case of a study where the design

depends upon the results of an earlier test (such as carcinogenicity

where the dose level is contingent upon the results of a subchronic

study), the deadline is 90 days following the date of submission of the

final report for that study.

The third alternative is for the final HAPs rule to reference the

guidelines currently in part 798 of title 40 of the Code of Federal

Regulations and modify these guidelines to make them as nearly

identical as possible to the harmonized OPPTS guidelines. The

modifications that the Agency currently believes would be appropriate

are set forth in a separate document in the record entitled

``Modifications to Health Effects Test Guidelines Currently in 40 CFR

Part 798 for Use in the HAPs Test Rule''. EPA is soliciting comment on

these three alternative procedures.

D. Route-to-Route Extrapolation

EPA would consider route-to-route extrapolation of toxicity data

from routes other than inhalation when it is scientifically defensible

to empirically derive the inhalation risk. Derivation of the inhalation

risk is generally only reasonable when portal-of-entry effects (toxic

effects on the respiratory tract from inhalation exposure) and/or

first-pass effects can be ruled out or adequately characterized.

``First-pass'' effects refer to the metabolism that can take place

in portal-of-entry tissue, prior to a chemical's entry into the

systemic circulation. For example, after oral administration, many

chemicals are delivered to the liver via the portal vein from the

gastrointestinal (GI) tract before they enter into the systemic

circulation. The respiratory tract can also exhibit a first-pass effect

after inhalation due to its various cell types and metabolic enzyme

systems. The first-pass action can alter the disposition of the parent

and metabolites, thereby modulating the dose to remote or systemic

target tissues in a route-dependent fashion. Therefore, unless this

first-pass effect and dosimetry are adequately understood and taken

into account, substantial error can be introduced in route-to-route

extrapolation.

In the absence of data to determine dosimetry via inhalation,

quantitative route-to-route extrapolation is subject to substantial

error when a chemical is thought to be susceptible to first-pass

effects (e.g., metabolized) or when a potential for portal-of-entry

effects is indicated (e.g., skin irritation after dermal

administration). There are situations where oral data should not be

used for route extrapolation to inhalation. For example, chemicals with

a short active half-life that were administered by gavage may result in

high short-term blood concentrations and consequently much greater

effects than the much lower constant blood levels that occur with

inhalation exposure. Conversely, if a chemical requires metabolic

activation via a rate-limited reaction, bolus dosing via gavage may

underestimate the dose. Consideration of factors such as these is

important in judging whether the oral study of interest qualifies for

route extrapolation.

Regardless of the toxic endpoint considered, EPA's ability to

perform quantitative route-to-route extrapolation is critically

dependent on the amount and type of data available. The minimum

information generally needed includes both the nature of the toxic

effect and a description of the relationship between exposure and the

toxic effect. The actual impact of exposure by different routes can

best be estimated by taking account of factors that influence

absorption at the portal of entry, such as:

(1) Physicochemical characteristics of the chemical (e.g.,

disassociation state, molecular weight, partition coefficient,

reactivity, solubility);

(2) Exposure factors (e.g., concentration, duration, regimen); and

(3) Physiologic parameters (e.g., barrier capacity as related to

variability in species, blood flow, cell types and morphology,

metabolism, pH, specialized absorption sites, storage in cells) and

those parameters that influence dose that are remote to the portal of

entry including metabolism, clearance, tissue binding, tissue blood

flows, tissue:blood partition coefficients, and tissue volumes.

Oral toxicity data are the most commonly available data as

alternatives to inhalation data. Oral data are problematic for route-

to-route extrapolation in the following instances:

(1) When groups of chemicals are expected to have different

toxicities by the two routes, for example, metals, irritants, and

sensitizers.

(2) When a first-pass effect by the liver is expected.

(3) When a respiratory tract effect is established but nodosimetry

comparison can be clearly established between the two routes.

(4) When the respiratory tract was not adequately studied in the

oral studies.

(5) When short-term inhalation studies, dermal irritation, or in

vitro studies indicate potential portal-of-entry effects at the

respiratory tract, but the studies themselves are not adequate for risk

assessment.

Dose-response data from other routes of exposure, such as

intravenous, intraperitoneal, subcutaneous, dermal, and intramuscular

routes also may be available. Intravenous data can provide reliable

information on blood levels, but such information should be

supplemented by knowledge of the quantitative relationship between

exposure concentration and blood levels in order to be useful. The

other routes usually are less useful in route-to-route extrapolation

because the pharmacokinetics are, in general, poorly characterized.

Methods for route-to-route extrapolation vary in accuracy and,

therefore, in inherent uncertainty. The simplest approach is to use

default absorption values for each exposure route dependent on the

chemical class in question. Such values have only been developed for a

few classes of organic chemicals. Because this approach entails

increased uncertainty compared with those that use pharmacokinetics

(PK) data and physiologically based pharmacokinetics (PBPK) modeling,

use of default absorption values is generally considered highly

uncertain for quantitative dose-response assessment.

EPA's optimal but most complex and data intensive method for

performing route-to-route extrapolation involves the development of a

PBPK model that describes the disposition (deposition, absorption,

distribution, metabolism, and elimination) of the chemical for the

routes of interest (Ref. 15). Such models

[[Page 33189]]

account for fundamental physiological and biochemical parameters and

processes such as blood flows, ventilatory parameters, metabolic

capacities, and renal clearance tailored by the physicochemical (e.g.,

blood:air and tissue:blood partitions) and biochemical properties

(e.g., binding, depletion of cofactors) of the chemical in question.

PBPK models should be used in conjunction with toxicity and mechanistic

studies in order to relate the effective dose associated with an

adverse effect for the test species and conditions to other scenarios.

Although the development of a full PBPK model can involve greater

effort than other methods using pharmacokinetics data, the application

of pharmacokinetics modeling to determine health risk provides a

considerable improvement in the reliability of an extrapolation across

routes. The use of an existing model structure, essentially a template,

can greatly reduce the effort required for model development of

analogous chemicals.

More limited pharmacokinetics data such as measurement of

bioavailability and disposition of an internal dose marker (e.g., blood

cholinesterase activity, enzyme elevation, and amount of chemical bound

to protein) may be used for route-to-route extrapolation in conjunction

with a consideration of the uncertainties involved in each case. As

above, if the portal of entry is affected by the agent, then more

elaborate data may be required.

EPA realizes that the use of pharmacokinetics data for route-to-

route extrapolation, as well as for the broader purpose of generally

identifying the mechanisms by which exposure to a specific agent causes

particular health effects, is a fast-developing and often controversial

area of science at this time. However, under certain circumstances, as

explained above, route-to-route extrapolation based on valid

pharmacokinetics data can offer a useful and less expensive alternative

to testing or retesting by another route of exposure.

E. Opportunity To Submit Proposals for Enforceable Consent Agreements

for Pharmacokinetics Studies

Basic pharmacokinetics parameters provide information on a

substance's absorption, distribution, biotransformation, and excretion

which can aid in understanding the potential for accumulation of the

substance in various tissues or organs and the mechanism of toxicity.

Basic PK parameters can be determined through use of the OPPTS

harmonized test guideline for pharmacokinetics studies (870.7485). EPA

considered but rejected the option of requiring the use of this

guideline in this proposed rule because the Agency is interested in a

more sophisticated level of study that could potentially support PBPK

modeling.

EPA believes that enforceable consent agreements (ECAs) and testing

consent orders offer an opportunity to obtain this more in-depth

understanding of the pharmacokinetics of HAPs. The Agency, therefore,

is inviting manufacturers to submit proposals for pharmokinetics

studies for HAPs to be used in the ECA process. Each study proposal

should include the name of the chemical(s), a detailed description of

the proposed pharmacokinetics study, and discussion of the application

of the pharmacokinetics data in performing route-to-route

extrapolations. Study proposals should reflect an understanding of the

scientific reasoning presented in Unit IV.D. of this preamble, the

existing database on the chemical and testing required under this

proposed test rule. EPA expects to use a previously published decision

tree (Ref. 15) as an element in the evaluation of these proposals. As

noted in Unit IV.D., these data may be used for route-to-route

extrapolation with a level of uncertainty in inverse proportion to

their level of complexity and sophistication.

Each study proposal should be labeled: ``Proposal for

Pharmacokinetics Study of (name of chemical),'' identified by document

control number (OPPTS-42187B, FRL-4869-1), and sent to: U.S.

Environmental Protection Agency, Office of Pollution Prevention and

Toxics, Document Control Office (7407), Room G-099, 401 M St., SW.,

Washington, DC 20460. Proposals for pharmacokinetics studies must be

received by EPA no later than October 24, 1996. Enforceable consent

agreements must be negotiated and signed no later than 12 months after

the date of proposal of this rule in order to permit timely development

of the final HAPs rule.

EPA will review the submissions and will select promising

candidates for negotiation under the procedures in 40 CFR 790.22. If

the Agency decides to proceed with the ECA process, it will publish a

notice in the Federal Register soliciting persons interested in

participating in or monitoring negotiations for the development of ECAs

for PK studies to notify the Agency in writing.

EPA noted in Unit IV.D. that the development and use of a PBPK

model represents the optimal approach to route-to-route extrapolation.

The development of such models is often a complex and uncertain task

that in most cases lies beyond the expectations of performance that

could be embodied in an ECA. However, EPA would like to encourage

extension of the data generated under the ECAs described above to the

development of PBPK models. EPA envisions that PBPK models could be

developed through voluntary cooperative arrangements and is interested

in a dialogue with industry and others on ways to encourage and support

PBPK model development.

F. Persons Required To Test

Based on the findings in Unit V of this preamble, EPA is proposing

that persons who manufacture (including import) or process, or who

intend to manufacture or process 1,1'-biphenyl, carbonyl sulfide,

chlorine, chlorobenzene, chloroprene, cresols (all three isomers),

diethanolamine, ethylbenzene, ethylene dichloride, ethylene glycol,

hydrochloric acid, hydrogen fluoride, maleic anhydride, methyl isobutyl

ketone, methyl methacrylate, naphthalene, phenol, phthalic anhydride,

1,2,4-trichlorobenzene, 1,1,2-trichloroethane, and vinylidene chloride,

other than as an impurity, at any time from the effective date of the

final test rule to the end of the reimbursement period, be subject to

the testing requirements in this rule. Manufacturers would be required

to submit letters of intent to conduct testing or exemption

applications (40 CFR 790.45). However, under 40 CFR 790.42, processors,

small-quantity manufacturers, and manufacturers of small quantities of

these substances solely for research and development purposes would not

be required to submit letters of intent or exemption applications

unless directed to do so in a subsequent notice as described in 40 CFR

790.48(b).

EPA is proposing to exempt those manufacturers and processors that

produce the chemical substances listed above only as an impurity, as

defined in 40 CFR 790.3, because it would be difficult and

prohibitively expensive for EPA, manufacturers, and processors to

identify with complete assurance all chemical substances that contain

the 21 substances solely as an impurity. In addition, EPA would find it

difficult to apply both the exemption and reimbursement processes to

those who manufacture and/or process these chemical substances solely

as an impurity. EPA's reimbursement regulations, issued pursuant to

TSCA section 4(c), 15 U.S.C. 2603(c), state that those persons who

manufacture or process chemical substances as impurities are not

subject to test

[[Page 33190]]

requirements unless a particular test rule specifically states

otherwise (40 CFR 791.48(b)). EPA finds no basis to propose such a

requirement in this rule.

Persons who manufacture these substances as byproducts, as defined

in 40 CFR 791.3(c), would be subject to the testing requirements set

forth in this proposed rule. The total amount of imports and domestic

production of these chemical substances, including the amount produced

as a byproduct, would be used in determining reimbursement shares under

the TSCA section 4 data reimbursement regulations in 40 CFR part 791.

In a previous multichemical test rule (undertaken for EPA's Office of

Solid Waste) for which EPA had likewise proposed that byproducts be

subject to the rule, an industry commenter objected to this inclusion

based on historical grounds. The commenter said, ``The historical roots

of section 4 in the Eckart Subcommittee work on TSCA were the sharing

of the costs of test generation in direct proportion to the economic

benefits which producers derived from the chemicals.'' In response to

this comment, EPA explained that,

EPA does not agree that the intention of Congress to have

producers share the cost of testing should be interpreted to exclude

producers of byproducts from TSCA section 4 testing requirements.

While economic benefit is not derived directly from the production

of the subject chemical, the production and disposal of the

byproduct are a result of a production process by which the company

does derive economic benefit (an indirect benefit). (53 FR 22300,

22305, June 15, 1988)

Carbonyl sulfide would be the first chemical substance subject to a

TSCA section 4 test rule that is produced almost exclusively as a

byproduct. Although some carbonyl sulfide is reported to be used in

chemical synthesis, its large production and release, as reported in

the TRI, is due to its creation as a byproduct which is unwanted.

Consistent with EPA's position on byproducts testing, as explained

above, all persons reporting the release of carbonyl sulfide in the TRI

would be considered to be manufacturers of carbonyl sulfide and would

be subject to the provisions of this proposed rule.

V. Findings

As explained in Unit I of this preamble, EPA is proposing findings

under TSCA sections 4(a)(1)(A) and 4(a)(1)(B) for the 21 HAPs subject

to this rule. The findings are summarized in the table below. The

detailed discussion of the findings for each chemical substance

included in this rule is contained in a separate document entitled

``TSCA Section 4 Findings for 21 Hazardous Air Pollutants'' that is

available in the rulemaking record. Requirements for sections

4(a)(1)(A) and 4(a)(1)(B) findings appear in Unit I of this preamble.

In articulating its policy for making findings under TSCA section

4(a)(1)(B) (frequently described as the ``B policy'', see Unit I of

this preamble), EPA has defined ``substantial production'' as aggregate

annual production of 1 million pounds or more and ``substantial

release'' as an annual release, from all sources, into the environment

of 1 million pounds or 10% of production, whichever is lower (58 FR

28736, 28746, May 14, 1993). These definitions apply to the terms

``substantial production'' and ``substantial release'' as used in this

preamble. (As explained in Unit III.C. of this preamble, all the

chemical substances proposed for testing in this proposed rule are

emitted into the atmosphere in the amount of 50 tons per year or more

according to the TRI.) EPA also defined ``substantial human exposure''

as an annual exposure of 100,000 members of the general population,

10,000 consumers, or 1,000 workers. Id.

TSCA Section 4(a) Statutory Findings

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

4(a)(1)(A)(ii)/(iii)

Chemical substance 4(a)(1)(A)(i) Finding 4(a)(1)(B)(i) Findinga and 4(a)(1)(B)(ii)/

is based on: is based on: (iii) Finding are for:

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

1,1'-Biphenyl--------------------------Reproductive toxicity----Substantial production:--Acute toxicity\6\,\9\--

(CAS No. 92-52-4) Respiratory toxicity 53.5 million lbs Subchronic

Substantial human toxicity\5\,\9\

exposure: 20,351 Developmental

workers toxicity\2\

Consumer exposure Reproductive

toxicity\1\

Neurotoxicity\7\

Immunotoxicity\7\

Respiratory sensory

irritation\7\

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

Carbonyl Sulfide Oncogenicity Substantial production: Acute toxicity\6\

(CAS No. 463-58-1) Neurotoxicity production is at least Subchronic toxicity\7\

as much as Developmental

environmental release toxicity\7\

(produced as a Reproductive

byproduct) toxicity\5\,\6\

Substantial Neurotoxicity\6\,\7\

environmental release: Oncogenicity\7\

16.7 million lbs Immunotoxicity\7\

Genetic toxicity\7\

Respiratory sensory

irritation\7\

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

Chlorine Respiratory toxicity Substantial production: Acute toxicity\5\,\8\

(CAS No. 7782-50-5) 22.3 billion lbs

Substantial human

exposure: 170,000

workers

Consumer exposure

Substantial

environmental release:

78,498 million lbs

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

[[Page 33191]]

Chlorobenzene Respiratory toxicity Substantial production: Acute toxicity\6\

(CAS No. 108-90-7) Developmental toxicity 210 million lbs Subchronic

Reproductive toxicity Substantial human toxicity\3\,\4\,\6\,\8

Liver toxicity exposure: 17,056 \

Kidney toxicity workers Neurotoxicity\7\

Neurotoxicity General population Immunotoxicity\6\,\7\

Substantial

environmental release:

2.58 million lbs

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

Chloroprene Respiratory toxicity Substantial production: Acute toxicity\3\,\6\

(CAS No. 126-99-8) Reproductive toxicity 321 million lbs Reproductive

Liver toxicity Substantial human toxicity\7\,\8\

Neurotoxicity exposure: 17,749 Neurotoxicity\6\,\8\

Hematotoxicity workers Immunotoxicity\7\

Developmental toxicity General population Respiratory sensory

Substantial irritation\7\

environmental release:

1.7 million lbs

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

Cresols

(CAS No. 1319-77-3)

mixture of 3 isomers:

ortho-isomer

(CAS No. 95-48-7)

para-isomer

(CAS No. 106-445)

Respiratory toxicity Substantial production: Acute toxicity\8\

meta-isomer Developmental toxicity 84.3 million lbs Subchronic

(CAS No. 108-39-4) Neurotoxicity Substantial human toxicity\5\,\8\

exposure: 132,742 Acute

workers neurotoxicity\5\,\6\

Consumer exposure Immunotoxicity\6\,\7\

General population Respiratory sensory

Substantial irritation\7\

environmental release:

1.5 million lbs

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

Diethanolamine Reproductive toxicity Substantial production: Acute toxicity\8\

(CAS No. 111-42-2) Neurotoxicity 198 million lbs Subchronic toxicity\5\

Substantial human Developmental

exposure: 573,025 toxicity\6\

workers Reproductive

Consumer exposure toxicity\7\

Neurotoxicity\5\,\6\,\7

\

Immunotoxicity\7\

Respiratory sensory

irritation\7\

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

Ethylbenzene Developmental toxicity Substantial production: Acute toxicity\6\

(CAS No. 100-41-4) Kidney toxicity 11.4 billion lbs Developmental

Neurotoxicity Substantial human toxicity\2\

exposure: 80,726 Reproductive

workers toxicity\7\

Consumer exposure Neurotoxicity\6\,\7\

General population Immunotoxicity\6\,\7\

Substantial Respiratory sensory

environmental release: irritation\7\

8.8 million lbs

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

Ethylene dichloride Oncogenicity Substantial production: Acute

(CAS No. 107-06-2) General systemic 14.3 billion lbs toxicity\1\,\4\,\5\,\6

toxicity General population \

Substantial human Subchronic

exposure: 77,111 toxicity\4\,\5\,\8\

workers Developmental

Consumer exposure toxicity\2\

Substantial Reproductive

environmental release: toxicity\9\

4 million lbs Neurotoxicity\6\,\7\

Respiratory sensory

irritation\7\

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

Ethylene glycol Substantial production: Acute toxicity\7\

(CAS No. 107-21-1) 7.2 billion lbs Subchronic

Substantial human toxicity\4\,\6\

exposure: 1,133,792 Neurotoxicity\7\

workers Immunotoxicity\5\

Consumer exposure Respiratory sensory

Substantial irritation\7\

environmental release:

17.5 million lbs

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

Hydrochloric acid Respiratory toxicity Substantial production: Acute toxicity\5\,\6\

(CAS No. 7647-01-0) 5.75 billion lbs

Substantial human

exposure: 1,131,879

workers

Consumer exposure

Substantial

environmental release:

287.7 million lbs

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

[[Page 33192]]

Hydrogen fluoride Respiratory toxicity Substantial production: Acute

(CAS No. 7664-39-3) Liver toxicity 322 million lbs toxicity\3\,\5\,\6\,\8

Eye irritation Substantial human \

exposure: 182,589 Subchronic

workers toxicity\8\,\10\

Substantial Developmental

environmental release: toxicity\7\

9.2 million lbs Reproductive

toxicity\7\

Neurotoxicity\7\

Immunotoxicity\7\

Respiratory sensory

irritation\7\

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

Maleic anhydride Respiratory toxicity Substantial production: Acute toxicity\7\

(CAS No. 108-31-6) Eye irritation 382 million lbs Developmental

Substantial human toxicity\2\

exposure: 37,897 Neurotoxicity\7\

workers Oncogenicity\5\,\6\

Immunotoxicity\7\

Respiratory sensory

irritation\7\

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

Methyl isobutyl ketone Developmental toxicity Substantial production: Acute toxicity\5\,\6\

(CAS No. 108-10-1) Neurotoxicity 175 million lbs Reproductive

Substantial human toxicity\7\

exposure: 467,763 Immunotoxicity\7\

workers Respiratory sensory

Consumer exposure irritation

General population

Substantial

environmental release:

27.7 million lbs

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

Methyl methacrylate Respiratory toxicity Substantial production: Acute

(CAS No. 80-62-6) Liver toxicity 1,200 million lbs toxicity\1\,\3\,\4\,\6

Kidney toxicity Substantial human \

Neurotoxicity exposure: 120,788 Developmental

workers toxicity\2\

Consumer exposure Reproductive

Substantial toxicity\7\

environmental release: Neurotoxicity\3\,\4\,\6

2.8 million lbs \,\8\

Immunotoxicity\6\,\7\

Respiratory sensory

irritation\7\

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

Naphthalene Respiratory toxicity Substantial production: Acute toxicity\5\,\6\

(CAS No. 91-20-3) Neurotoxicity 235 million lbs Reproductive

Substantial human toxicity\5\,\6\,\7\

exposure: 23,092 Immunotoxicity\6\,\7\

workers Respiratory sensory

Consumer exposure irritation\7\

General population

Substantial

environmental release:

2.8 million lbs

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

Phenolb Respiratory toxicity Substantial production: Acute

(CAS No. 108-95-2) 3.9 billion lbs toxicity\4\,\5\,\8\

Substantial human Immunotoxicity\4\,\5\,\

exposure: 192,739 6\

workers Respiratory sensory

Consumer exposure irritation

General population

Substantial

environmental release:

10 million lbs

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

Phthalic anhydride Respiratory Substantial production: Acute toxicity\6\,\7\

(CAS No. 85-44-9) sensitization 874 million lbs Subchronic toxicity\7\

Substantial human Developmental

exposure: 62,644 toxicity\5\,\8\

workers Reproductive

toxicity\7\

Neurotoxicity\7\

Oncogenicity\5\

Immunotoxicity\7\

Respiratory sensory

irritation\7\

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

1,2,4-Trichlorobenzene Oncogenicity Substantial production: Acute

(CAS No. 120-82-1) Developmental toxicity CBI toxicity\1\,\3\,\6\

Substantial human Developmental

exposure: 4,032 toxicity\1\,\5\,\6\

workers Neurotoxicity\6\

General population Immunotoxicity\7\

Respiratory sensory

irritation\7\

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

[[Page 33193]]

1,1,2-Trichloroethane Oncogenicity Substantial production: Acute toxicity\7\

(CAS No. 79-00-5) Liver toxicity estimated - 210 Subchronic toxicity\7\

Kidney toxicity million lbs Developmental

Neurotoxicity Substantial human toxicity\5\,\6\

exposure: 1,036 Reproductive

workers toxicity\7\

General population Neurotoxicity\6\,\7\

Oncogenicity\5\

In vivo cytogenicity\7\

Immunotoxicity\6\

Respiratory sensory

irritation\7\

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

Vinylidene chloride Oncogenicity Substantial production: Acute toxicity\5\,\6\

(CAS No. 75-35-4) Respiratory toxicity 230 million lbs Neurotoxicity\7\

Developmental toxicity Substantial human Respiratory sensory

Liver toxicity exposure: 2,675 irritation\7\

Kidney toxicity workers

Consumer exposure

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

\1\ Too few animals were tested.

\2\ Only one species was adequately tested.

\3\ Only one sex was tested.

\4\ Too few exposure levels were tested.

\5\ Inadequate exposure duration, schedule, or route.

\6\ Only limited endpoints were assessed.

\7\ No study addressing the specific endpoint was found.

\8\ Insufficient reporting of data to make the study useful.

\9\ No no-observed-adverse-effect level (NOAEL) was identified.

\10\ No lowest-observed-adverse-effect level (LOAEL) was identified.

Notes to table:

a ``B'' findings are based on Ref. (10). It should be noted that

all HAPs meet the 50 tons of emissions per year selection criteria

discussed in Unit III.C. of this preamble.

b Findings made in 58 FR 61654, 61659-60, November 22, 1993.

VI. Economic Analysis of the Proposed Rule

EPA has prepared and placed in the record for this proposed rule an

economic analysis that evaluates the potential for significant economic

impacts as a result of the testing proposed in this notice. The total

cost of this proposed rule is estimated to range up to $41.4 million.

The total cost of testing for each chemical substance has been

annualized and compared with annual revenues (defined as the product of

sales price and total supply) as an indication of potential economic

impact. Annualized test costs, calculated over 15 years using a 7%

discount rate, represent the equivalent constant costs that would have

to be recouped each year of the payback period to finance the testing

expenditure in the first year. Annualized test costs are then divided

by the total supply of the chemical substance to derive the annualized

unit test costs. The percent price impact is calculated by dividing the

annualized unit test costs by the sales price and multiplying by 100.

The upper-bound estimated total costs of testing (including both

laboratory costs and administrative costs), annualized tests costs, and

price impact for the chemicals in this proposed rule are as follows:

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

Chemical Substances Total test cost ($) Annualized test cost ($) Price impact (%)

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

1,1'-Biphenyl 2,213,900 243,074 0.64

Carbonyl sulfide 5,509,163 609,876 NA

Chlorine 85,400 9,376 0.0003

Chlorobenzene 972,900 106,819 0.098

Chloroprene 1,603,488 176,054 0.076

Cresols (all 3 isomers) 2,139,600 234,917 0.39

Diethanolamine 2,327,838 255,584 0.23

Ethylbenzene 1,732,050 190,170 0.013

Ethylene dichloride 2,007,325 220,393 0.0071

Ethylene glycol 986,638 108,327 0.0097

Hydrochloric acid 85,400 9,376 0.0040

Hydrogen fluoride 2,135,888 234,509 0.094

Maleic anhydride 4,148,588 454,834 0.25

Methyl isobutyl ketone 1,228,913 134,928 0.16

Methyl methacrylate 1,732,050 190,170 0.023

Naphthalene 1,242,650 136,436 0.16

Phenol 85,400 9,376 0.0010

[[Page 33194]]

Phthalic anhydride 5,650,338 620,377 0.21

1,2,4-Trichlorobenzene 963,163 105,750 CBI

1,1,2-Trichloroethane 3,837,900 421,381 0.41

Vinylidene chloride 708,700 77,811 0.12

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

Note: The table shows the maximum costs and impacts estimated by

EPA. The full range of estimates is given in the economic analysis

document placed in the record for this proposed rule.

EPA believes, on the basis of these calculations, that the proposed

testing of the HAPs presents a low potential for adverse economic

impact. Because these chemical substances have relatively large

production volumes, with the exception of carbonyl sulfide (to which

this methodology does not apply) the annualized costs of testing,

expressed as a percentage of annual revenue, are very small--ranging

from 0.0003% to 0.64%. Costs of testing are therefore found to be

insignificant relative to revenues for these chemical substances.

VII. Availability of Test Facilities and Personnel

Although earlier studies indicated that test facilities and

personnel were available to perform the testing specified in this

proposed rule (Ref. 18), the impact of this rule combined with other

testing requirements may exceed capacity for inhalation testing

facilities in the short term. While EPA believes that over the longer

term, additional inhalation facilities will become available, any

short-term effects can be dealt with by adjusting study due dates in

response to comments on this rule or in response to a request for

modification of reporting deadlines.

VIII. Public Meeting

EPA will hold a public meeting in Washington, DC prior to the close

of the comment period. Announcemment of this meeting will be published

in the Federal Register. If requested, EPA will hold an additional

public meeting in Washington, DC.

IX. Comments Containing Confidential Business Information

All comments will be placed in the public version of the rulemaking

record unless they are clearly labeled as containing information

claimed as CBI when they are submitted. CBI claims will be deemed to

have been waived if they are not made at the time of submission of the

document containing the information claimed as CBI, and such document

may be made public with no further notice to the submitter. While a

part of the rulemaking record, comments claimed as CBI will be treated

in accordance with 40 CFR part 2. A sanitized version of all comments

containing information claimed as CBI must be submitted to EPA for

inclusion in the public version of the rulemaking record.

It is the responsibility of the submitter to comply with 40 CFR

part 2 so that all materials claimed as CBI may be properly protected.

This includes, but is not limited to, clearly indicating on the face of

the revelant section of the comment (as well as on any revelant

associated correspondence) that information claimed as CBI is included

and marking ``CONFIDENTIAL,'' ``TSCA CBI,'' or similar designation on

the face of each section of any document or attachment in the comment

that contains information claimed as CBI. Should putatively private

information be put into the public file because of the submitter's

failure to clearly claim and designate its confidential claim on the

face of the comment, EPA will presume any such information that has

been in the public file for more than 30 days to be in the public

domain.

X. Rulemaking Record

EPA has established a record for this rulemaking (docket number

OPPTS-42187) (including comments and data submitted electronically).

This record contains the basic information considered by EPA in

developing this proposal and appropriate Federal Register notices. EPA

will supplement this record as necessary.

A public version of this record, including printed, paper versions

of electronic comments, which does not include any information claimed

as CBI is available for inspection from 12 noon to 4 p.m., Monday

through Friday, except legal holidays. The public record is located in

the TSCA Nonconfidential Information Center, Rm. NE-B607, 401 M St.,

SW., Washington, DC 20460.

Comments in electronic form may be submitted directly to EPA at:

[email protected]

Comments in electronic form must be submitted in an ASCII file

avoiding the use of special characters and any form of encryption.

The official record for this rulemaking, as well as the public

version, will be kept in paper form. Accordingly, EPA will transfer all

comments received electronically into printed, paper form as they are

received and will place the paper copies in the official rulemaking

record which will also include all comments submitted directly in

writing.

The record includes the following information.

A. Supporting Documentation

(1) Federal Register notices/EPA documents pertaining to this

proposed rule consisting of:

(a) ``TSCA Interagency Testing Committee; Initial Report to the

Administrator, Environmental Protection Agency'' (42 FR 55026, October

12, 1977).

(b) ``Chloromethane and Chlorinated Benzenes Proposed Test Rule;

Amendment to Proposed Health Effects Standards'' (45 FR 48524, July 18,

1980).

(c) ``Dichloromethane, Nitrobenzene and 1,1,1-Trichloroethane;

Proposed Test Rule'' (46 FR 30300, June 5, 1981).

(d) ``Tenth Report of the Interagency Testing Committee to the

Administrator; Receipt of Report and Request for Comments Regarding

Priority List of Chemicals'' (47 FR 22585, May 12, 1982).

(e) ``Methyl Isobutyl Ketone and Methyl Ethyl Ketone; Response to

the Interagency Testing Committee'' (47 FR 58025, December 29, 1982).

(f) ``Toxic Substances Control Act; Data Reimbursement'' (48 FR

31786, July 11, 1983).

(g) ``Cresols; Proposed Test Rule'' (48 FR 31813, July 11, 1983).

(h) ``Methyl Isobutyl Ketone and Methyl Ethyl Ketone Decision to

Adopt Negotiated Testing Program'' (48 FR 44905, September 30, 1983).

(i) ``Toxic Substances; Biphenyl; Test Rule'' (50 FR 37182,

September 12, 1985).

(j) ``Cresols; Testing Requirements'' (51 FR 15771, April 28,

1986).

(k) ``Chlorinated Benzenes; Final Test Rule'' (51 FR 24657, July 8,

1986).

(l) ``Toxic Substances, 1,1,-Dichloroethylene; Proposed Test Rule''

(51 FR 28840, August 12, 1986).

(m) ``Guidelines for Carcinogen Risk Assessment'' (51 FR 33992

(September 24, 1986).

[[Page 33195]]

(n) ``Office of Solid Waste Chemicals; Final Test Rule'' (53 FR

22300, June 15, 1988).

(o) ``Toxic Substances Control Act (TSCA); Good Laboratory Practice

Standards'' (54 FR 34034, August 17, 1989).

(p) ``Metabolism and Pharmacokinetics Test Guideline'' (56 FR

32537, July 17, 1991).

(q) ``Guidelines for Developmental Toxicity Risk Assessment'' (56

FR 63798, December 5, 1991).

(r) ``TSCA section 4(a)(1)(B) Final Statement of Policy; Criteria

for Evaluating Substantial Production, Substantial Release, and

Substantial or Significant Human Exposure'' (58 FR 28736, May 14,

1993).

(s) ``Office of Water Chemicals; Final Test Rule'' (58 FR 59667,

November 10, 1993).

(t) ``Acetophenone, Phenol, N,N-Dimethylaniline, Ethyl Acetate, and

2,6-Dimethylphenol; Proposes Test Rule, Notice of Opportunity to

Initiate Negotiations for TSCA Section 4 Testing Consent Agreements''

(58 FR 61654, November 22, 1993).

(u) ``Testing Consent Orders for Acetone, n-Amyl Acetate, n-Butyl

Acetate, Ethyl Acetate, Isobutyl Alcohol, Methyl Isobutyl Ketone, and

Tetrahyrofuran'' (60 FR 4514, January 23, 1995).

(v) ``Executive Order 12866 of September 30, 1993; Regulatory

Planning and Review'' (58 FR 51735, October 4, 1993).

(w) ``Executive Order 12898 of February 11, 1994; Federal Actions

to Address Environmental Justice in Minority Populations and Low-Income

Populations'' (59 FR 7629, February 16, 1994).

(x) ``Guidelines for Reproductive Toxicity Risk Assessment'' (Pub.

No. EPA/600/AP-94/001, February 1994).

(2) OPPTS test guidelines used in this proposed rule:

(a) Acute Inhalation Toxicity with Histopathology, OPPTS 870.1350,

EPA Pub. No. 712-C-96-291, June 1996;

(b) Subchronic Inhalation Toxicity, OPPTS 870.3465, EPA Pub. No.

712-C-96-204, June 1996;

(c) Inhalation Developmental Toxicity Study, OPPTS 870-3600, EPA

Pub. No. 712-C-96-206, June 1996;

(d) Reproduction and Fertility Effects, OPPTS 870.3800, EPA Pub.

No. 712-C-96-208, February 1996;

(e) Carcinogenicity, OPPTS 870.4200, EPA Pub. No. 712-C-96-211,

June 1996;

(f) Escherichia coli WP2 and WP2 uvrA Reverse Mutation Assays,

OPPTS 870.5100, EPA Pub. No. 712-C-96-247, June 1996;

(g) Detection of Gene Mutations in Somatic Cells in Culture, OPPTS

870.5300, EPA Pub. No. 712-C-96-221, June 1996;

(h) In Vivo Mammalian Cytogenetics Tests: Bone Marrow Chromosomal

Analysis, OPPTS 870.5385, EPA Pub. No. 712-C-96-225, June 1996;

(i) In Vivo Mammalian Cytogenetics Tests: Erythrocyte Micronucleus

Assay, OPPTS 870.5395, EPA Pub. No. 712-C-96-226, June 1996;

(j) Neurotoxicity Screening Battery, OPPTS 870.6200, EPA Pub. No.

712-C-96-238, June 1996; and

(k) Immunotoxicity, OPPTS 870.7800, EPA Pub. No. 712-C-96-351, June

1996.

(3) Technical Support Documents consisting of:

(a) TSCA Section 4 Findings for 21 Hazardous Air Pollutants.

(b) Exposure Profiles for HAPs--Group 1.

(c) Summary Tables on the Health Effects Data for Hazardous Air

Pollutants (HAPs)--Group 1.

(d) Economic Analysis of the Impact of the Test Rule.

(4) Communications consisting of:

(a) Written letters and memoranda.

(b) Contact reports of telephone conversations.

(c) Meeting summaries.

B. References

(1) U.S. Environmental Protection Agency. ``Methods for derivation

of inhalation reference concentrations and application of inhalation

dosimetry.'' p. xxviii. Prepared by the Office of Health and

Environmental Assessment, Washington, DC. EPA/600/8-90/066F (1994).

(2) Letter from John F. Martonik, OSHA, to Susan B. Hazen, EPA, May

31, 1995.

(3) Letter from Val Schaeffer, CPSC, to Joe Carra, EPA, June 2,

1995.

(4) Orme-Zavaleta, J. ``OMB Question-Reply.'' Memorandum to Vicki

Dellarco. May 9, 1996.

(5) NAS. National Academy of Sciences, Washington DC, ``Science and

Judgment in Risk Assessment'' pp. 154, 157, 265 (1994).

(6) Siegel-Scott, C. ``Slope factors for hazardous air

pollutants.'' Memorandum to Vicki Dellarco. September 7, 1994.

(7) Shoaf, C.R. ``Clean Air Act chemicals with adequate databases

for development of RfC's.'' Memorandum to Vicki Dellarco. September 20,

1994.

(8) U.S. Environmental Protection Agency. Health assessment

document for 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and related

compounds, Vol. III of III, pp. 9-48 to 9-50. Prepared by the Office of

Health and Environmental Assessment, Washington, DC. EPA/600/PB-92/001C

(1994).

(9) Francis, E.Z., and Kimmel, G.A. ``Proceedings of the workshop

on one versus two-generation reproductive effects studies.'' Journal of

the American College of Toxicology. 7:911-925 (1988).

(10) Lai, D.Y., Baetcke, K.P., Vu, V.T., Cotruvo, J.A., and Eustis,

S.L. ``Evaluation of reduced protocols for carcinogenicity testing of

chemicals: Report of a joint EPA/NIEHS workshop.'' Regulatory

Toxicology and Pharmacology. 19:183-201 (1994).

(11) Syracuse Research Corporation. ``Summary tables on the health

effects data for hazardous air pollutants (HARs)-Group 1.'' Syracuse,

New York. (1995a).

(12) Syracuse Research Corporation. ``Exposure profiles for HAPs-

Group 1.'' Syracuse, New York. (1995b).

(13) Valcovic, L.R. Memorandum: ``Genetic toxicity evaluation of

HAPs'' to Vicki Dellarco, July 14, 1994.

(14) Waters, M.D., Stack, H.F., and Jackson, M.A. 1990. ``Genetic

Activity Profiles of 110 Hazardous Air Pollutants Listed Under Title

III of the Clean Air Act, as amended.'' U. S. Environmental Protection

Agency. Internal Report. October 30, (1990).

(15) Gerrity, T.R., and Henry, C.J. ed. Principles of Route-to-

Route Extrapolation for Risk Assessment. pp 1-12. Elsevier Science

Publishing Co., Inc. New York, N.Y. (1990).

(16) U.S. Environmental Protection Agency. ``Hazardous Air

Pollutants: Profiles of Non-Cancer Toxicity from Inhalation

Exposures.'' Washington, DC. EPA/600/R-93/142. September 1993.

(17) Luster, M.I., Portier, C., Pait, D.G., White, K.L., Genings,

C., Munson, A.E., and Rosenthal, G.J. ``Risk assessment in

immunotoxicology. I. Sensitivity and predictability of immune tests.''

Fundamental and Applied Toxicology. 18:200-210 (1992).

(18) U.S. Environmental Protection Agency. ``EPA census of the

toxicological testing industry final report.'' (1990).

XI. Regulatory Assessment Requirements

A. Executive Order 12866

Pursuant to Executive Order 12866 (58 FR 51735, October 4, 1993),

it has been determined by OMB that this is a ``significant regulatory

action.'' OMB was concerned that the amount of inhalation testing

required by this rule may exceed the capacity of the testing industry,

at least in the short run. This action was submitted to OMB for review,

and any comments or changes made during that review have been

documented in the public record.

In addition, EPA has prepared an economic analysis of the impact of

this

[[Page 33196]]

action, which is contained in a document entitled ``Test Rule Support

for 21 Hazardous Air Pollutants.'' This document is available as a part

of the public record at the address listed in Unit X of this preamble

and is briefly summarized in Unit VI of this preamble.

B. Regulatory Flexibility Act

Under the Regulatory Flexibility Act (5 U.S.C. 601 et seq.), EPA

has determined that this test rule, if promulgated, would not have a

significant impact on small businesses. EPA has identified the entities

that currently manufacture or import the chemical substances required

to be tested under this proposed rule and examined the practices that

industry uses in carrying out chemical testing in response to EPA test

rules. EPA believes that: (1) small businesses would not be expected to

perform testing themselves, or to participate in the organization of

the testing effort, because health effects testing of chemical

substances is generally carried out by consortia of the large

manufacturers or importers of the chemical substances; (2) small

businesses would experience only very minor costs, if any, in securing

exemption from testing requirements because exemption request

requirements, described at 40 CFR 790.82, are minimal--particularly

when, as in this proposed rule, EPA is not requiring exemption

applicants to submit equivalence data (see Unit IV.F of this

preamble)--and EPA does not charge a fee for filing such requests; and

(3) small businesses are unlikely to be affected by reimbursement

requirements because under the reimbursement rules (at 40 CFR 791.40

through 791.52), manufacturers or importers with a significant share of

production or importation are the entities that must share testing

costs under the reimbursement rules, and small businesses generally do

not manufacture or import a significant portion of high-volume chemical

substances.

C. Paperwork Reduction Act

OMB has approved the information collection requirements contained

in this proposed rule under the provisions of the Paperwork Reduction

Act, 44 U.S.C. 3501 et seq., and has assigned OMB control number 2070-

0033.

The public reporting burden for this collection of information is

estimated to average approximately the following number of hours per

response for the chemicals listed below, including time for reviewing

instructions, searching existing data sources, gathering and

maintaining the data needed, and completing and reviewing the

collection of information.

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

Total

Chemical substance burden

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

1,1'-Biphenyl 20,620

Carbonyl sulfide 47,644

Chlorine 693

Chlorobenzene 7,707

Chloroprene 13,039

Cresols (all 3 isomers) 6,048

Diethanolamine 21,826

Ethylbenzene 14,400

Ethylene dichloride 16,707

Ethylene glycol 7,816

Hydrochloric acid 693

Hydrogen fluoride 18,068

Maleic anhydride 35,849

Methyl isobutyl ketone 10,471

Methyl methacrylate 14,400

Naphthalene 10,580

Phenol 693

Phthalic anhydride 51,032

1,2,4-Trichlorobenzene 8,091

1,1,2-Trichloroethane 33,133

Vinylidene chloride 5,439

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

The total public reporting burden is estimated to be 357,045 hours

for all responses. The overall average per chemical is 15,524 hours.

Burden means the total time, effort, or financial resources

expended by persons to generate, maintain, retain, or disclose or

provide information to or for a Federal agency. This includes the time

needed to review instructions; develop, acquire, install, and utilize

technology and systems for the purposes of collecting, validating, and

verifying information, processing and maintaining information, and

disclosing and providing information; adjust the existing ways to

comply with any previously applicable instructions and requirements;

train personnel to be able to respond to a collection of information;

search data sources; complete and review the collection of information;

and transmit or otherwise disclose the information.

An Agency may not conduct or sponsor, and a person is not required

to respond to a collection of information unless it displays a

currently valid OMB control number. The OMB control numbers for EPA's

regulations are listed in 40 CFR part 9 and 48 CFR chapter 15.

Comments are requested on the Agency's need for this information,

the accuracy of the provided burden estimates, and any suggested

methods for minimizing respondent burden, including through the use of

automated collection techniques. Send comments on the ICR to Sandy

Farmer, OPPE Regulatory Information Division; U.S. Environmental

Protection Agency (2136), 401 M St., SW., Washington, DC 20460, (202)

260-2740, or electronically by sending an e-mail message to:

[email protected]. Send a copy of these comments to the

Office of Information and Regulatory Affairs, Office of Management and

Budget, 725 17th St., NW., Washington, DC 20503, marked ``Attention:

Desk Officer for EPA.'' Please remember to include the ICR number in

any correspondence. The final rule will respond to any comments on the

information collection requirements contained in this proposal.

D. Unfunded Mandates Reform Act and Executive Order 12875

Pursuant to Title II of the Unfunded Mandates Reform Act of 1995

(UMRA) (Pub. L. 104-4), EPA has determined that this action does not

contain a Federal mandate that may result in expenditures of $100

million or more for State, local, and tribal governments, in the

aggregate, or the private sector in any one year. (The analysis of the

costs associated with this action is referenced in Unit XI.A. of this

preamble.) Therefore, this action is not subject to the requirements of

sections 202 and 205 of the UMRA.

E. Executive Order 12898

Pursuant to Executive Order 12898 (59 FR 7629, February 16, 1994),

entitled ``Federal Actions to Address Environmental Justice in Minority

Populations and Low-Income Populations,'' the Agency has considered

environmental justice-related issues with regard to the potential

impacts of this action on the environmental and health conditions in

low-income and minority communities. Because many sources of HAP

emissions are located near populations of lower socioeconomic status

and with a higher proportion of minorities, the improved health

database that will be generated by this action will help to protect

these individuals and communities.

List of Subjects in 40 CFR Part 799

Environmental protection, Chemicals, Hazardous substances, and

Reporting and recordkeeping requirements.

Dated: June 20, 1996.

Lynn R. Goldman,

Assistant Administrator for Prevention, Pesticides and Toxic

Substances.

Therefore, it is proposed that 40 CFR, chapter I, subchapter R, be

amended as follows:

[[Page 33197]]

PART 799--[AMENDED]

3. The authority citation for part 799 would continue to read as

follows:

Authority: 15 U.S.C. 2603, 2611, 2625.

4. By adding Sec. 799.5053 to Subpart D of part 799 to read as

follows:

Sec. 799.5053 Chemical testing requirements for hazardous air

pollutants.

(a) General testing provisions--(1) Identification of test

substances. Table 1 in paragraph (a)(5) of this section identifies

those chemical substances that shall be tested in accordance with this

section. The purity of each test substance shall be 97 percent or

greater unless otherwise specified.

(2) Persons required to submit study plans, conduct tests, and

submit data. All persons who manufacture (including those who import

the substance or manufacture it as a byproduct) or intend to

manufacture one or more of the substances listed in table 1 after the

effective date listed in table 1 until the end of the reimbursement

period shall submit letters of intent to conduct testing, submit study

plans, conduct tests and submit data, or submit exemption applications,

as specified in this section, subpart A of this part and parts 790 and

792 of this chapter. Persons who manufacture or process these

substances only as an impurity are not subject to these requirements.

As explained in part 790 of this chapter, processors, small-quantity

manufacturers, and manufacturers of small quantities of these

substances solely for research and development purposes would become

subject to these requirements only after notification in the Federal

Register that no manufacturer had notified EPA of its intent to conduct

testing.

(3) Applicability of test guidelines. The guidelines and other test

methods cited in table 1 in paragraph (a)(5) of this section are

referenced here as they exist on the effective date listed in table 1

for that specific test. Testing shall be conducted in accordance with

the designated Series 870--Health Effects Test Guidelines and other

test methods. This incorporation by reference was approved by the

Director of the Federal Register in accordance with 5 U.S.C. 552(a) and

1 CFR part 51. Copies may be obtained from the Public Docket and

Freedom of Information Section, Field Operations Division (7506C),

Office of Pesticide Programs, Environmental Protection Agency, 401 M

St., SW., Washington, DC 20360. Copies may be inspected at the above

address or at the Office of the Federal Register, 800 North Capitol

Street, NW., suite 700, Washington, DC.

(4) Interim reporting requirements. All testing requirements in

this section are subject to the submission of interim progress reports

every 6 months beginning 6 months after the effective date for any

specific test listed in table 1 in paragraph (a)(5) of this section.

The date for the submission of final reports is specified as the number

of months after the effective date for the specific test listed in

table 1.

(5) Testing and reporting requirements. The substances identified

by CAS Registry number and chemical name in the following table 1 shall

be tested in accordance with the designated OPPTS Harmonized Guideline

testing requirements and any additional requirements and limitations

specified in the ``Specific requirements under this section'' column of

table 1. The numbers and letters in this column refer to the specific

requirements set forth in paragraph (b) of this section. Final reports

shall be submitted by the deadlines indicated as the number of months

after the effective date shown in table 1.

Table 1

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

Specific

Chemical substance/ OPPTS harmonized requirements Final Effective

CAS No. required testing guidelines under this report date

section

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

75-35-4-------------------------Vinylidene---------- ----------------- ----------------- ---------- --------

chloride:

Acute 870.1350 (b)(2) 21 mo

Neurotoxicity 870.6200 (b)(1)(ii)(A), 21 mo

(b)(1)(iii)(A),

(b)(1)(iii)(B)

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

79-00-5 1,1,2-

Trichloroethane:

Acute 870.1350 (b)(2) 21 mo

Subchronic 870.3465 (b)(3) 18 mo

Developmental 870.3600 12 mo

Reproductive 870.3800 (b)(1)(ii)(A), 29 mo

(b)(5)

Neurotoxicity 870.6200 (b)(1)(ii)(A), 21 mo

(b)(1)(iii)(A),

(b)(1)(iii)(B)

Carcinogenicity 870.4200 (b)(1)(i)(D), 60 mo

(b)(1)(ii)(A)

In vivo 870.5385 or (b)(1)(ii)(A) 14 mo

cytogenetics 870.5395

Immunotoxicity 870.7800 (b)(1)(ii)(A) 18 mo

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

80-62-6 Methyl

methacrylate:

Acute 870.1350 (b)(2) 21 mo

Developmental 870.3600 (b)(1)(i)(A) 12 mo

Reproductive 870.3800 (b)(1)(ii)(A), 29 mo

(b)(5)

Neurotoxicity 870.6200 (b)(1)(ii)(A), 21 mo

(b)(1)(iii)(A),

(b)(1)(iii)(B)

Immunotoxicity 870.7800 (b)(1)(ii)(A) 21 mo

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

85-44-9 Phthalic

anhydride:

Acute 870.1350 (b)(2) 21 mo

Subchronic 870.3465 (b)(1)(ii)(B), 18 mo

(b)(3)

Developmental 870.3600 (b)(1)(ii)(B) 12 mo

Reproductive 870.3800 (b)(1)(ii)(B), 29 mo

(b)(5)

Neurotoxicity 870.6200 (b)(1)(ii)(B), 21 mo

(b)(1)(iii)(A),

(b)(1)(iii)(B)

Carcinogenicity 870.4200 (b)(1)(ii)(B) 60 mo

[[Page 33198]]

Immunotoxicity 870.7800 (b)(1)(ii)(B) 18 mo

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

91-20-3 Naphthalene:

Acute 870.1350 (b)(2) 21 mo

Reproductive 870.3800 (b)(1)(ii)(A), 29 mo

(b)(5)

Immunotoxicity 870.7800 (b)(1)(ii)(A) 21 mo

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

92-52-4 1,1'-Biphenyl:

Acute 870.1350 (b)(2) 21 mo

Subchronic 870.3465 (b)(1)(ii)(B), 18 mo

(b)(3)

Developmental 870.3600 (b)(1)(i)(A), 12 mo

(b)(1)(ii)(B)

Reproductive 870.3800 (b)(1)(ii)(B), 29 mo

(b)(5)

Neurotoxicity 870.6200 (b)(1)(ii)(B), 21 mo

(b)(1)(iii)(A),

(b)(1)(iii)(B)

Immunotoxicity 870.7800 (b)(1)(ii)(B) 18 mo

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

95-48-7, Cresols:

106-44-5, and

108-39-4

Acute 870.1350 (b)(2) 21 mo

Subchronic 870.3465 (b)(3) 18 mo

Neurotoxicity 870.6200 (b)(1)(ii)(A), 21 mo

(b)(1)(iii)(A)

Immunotoxicity 870.7800 (b)(1)(ii)(A) 18 mo

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

100-41-4 Ethylbenzene:

Acute 870.1350 (b)(2) 21 mo

Developmental 870.3600 (b)(1)(i)(A) 12 mo

Reproductive 870.3800 (b)(1)(ii)(A), 29 mo

(b)(5)

Neurotoxicity 870.6200 (b)(1)(ii)(A), 21 mo

(b)(1)(iii)(A),

(b)(1)(iii)(B)

Immunotoxicity 870.7800 (b)(1)(ii)(A) 21 mo

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

107-06-2 Ethylene

dichloride:

Acute 870.1350 (b)(2) 21 mo

Subchronic 870.3465 (b)(3) 18 mo

Developmental 870.3600 (b)(1)(i)(C) 12 mo

Reproductive 870.3800 (b)(1)(ii)(A), 29 mo

(b)(5)

Neurotoxicity 870.6200 (b)(1)(ii)(A), 21 mo

(b)(1)(iii)(A),

(b)(1)(iii)(B)

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

107-21-1 Ethylene glycol:

Acute 870.1350 (b)(2) 21 mo

Subchronic 870.3465 (b)(3) 18 mo

Neurotoxicity 870.6200 (b)(1)(ii)(A), 21 mo

(b)(1)(iii)(A),

(b)(1)(iii)(B)

Immunotoxicity 870.7800 (b)(1)(ii)(A) 18 mo

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

108-10-1 Methyl isobutyl

ketone:

Acute 870.1350 (b)(2) 21 mo

Reproductive 870.3800 (b)(1)(ii)(A), 29 mo

(b)(5)

Immunotoxicity 870.7800 (b)(1)(ii)(A) 29 mo

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

108-31-6 Maleic anhydride:

Acute 870.1350 (b)(2) 21 mo

Developmental 870.3600 (b)(1)(i)(A) 12 mo

Neurotoxicity 870.6200 (b)(1)(ii)(A), 21 mo

(b)(1)(iii)(A),

(b)(1)(iii)(B)

Carcinogenicity 870.4200 (b)(1)(ii)(A) 60 mo

Immunotoxicity 870.7800 (b)(1)(ii)(A) 21 mo

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

108-90-7 Chlorobenzene:

Acute 870.1350 21 mo

Subchronic 870.3465 (b)(3) 1

This text is long and has been trimmed here. Open the source document for the complete record.