1
October
23,
2005
MEMORANDUM
TO:
Docket
No.
OAR­
2003­
0028
FROM:
Mark
Morris,
REAG
SUBJECT:
Summary
of
and
Response
to
Public
Comments
Received
on
the
May
30,
2003
Proposed
Rule
to
Delete
Methyl
Ethyl
Ketone
1.0
Overview
The
U.
S.
Environmental
Protection
Agency
(
EPA)
is
proposing
to
amend
the
list
of
hazardous
air
pollutants
(
HAP)
contained
in
section
112(
b)
of
the
Clean
Air
Act
(
CAA)
by
removing
the
compound
methyl
ethyl
ketone
(
MEK)
(
2­
butanone)
(
CAS
No.
78­
93­
3).
This
action
is
being
taken
in
response
to
a
petition
submitted
by
the
Ketones
Panel
of
the
American
Chemistry
Council
(
formerly
the
Chemical
Manufacturers=
Association)
on
behalf
of
MEK
producers
and
consumers
to
delete
MEK
from
the
HAP
list.
Petitions
to
remove
a
substance
from
the
HAP
list
are
permitted
under
section
112(
b)(
3)
of
the
CAA.

The
proposal
to
delist
MEK
appeared
in
the
Federal
Register
on
May
30,
2003
(
68
FR
32606).
The
notice
summarized
the
American
Chemistry
Council=
s
petition
to
delist
MEK
and
EPA's
analysis
of
that
petition,
and
provided
a
detailed
rationale
for
delisting
MEK.

The
proposal
notice
invited
comment
from
interested
parties
on
the
proposal
to
delist
MEK.
In
addition,
the
notice
specifically
requested
comments
on
the
prospective
reference
concentration
(
RfC)
for
MEK
(
the
interim
health
value
EPA
developed
for
the
proposal).
The
notice
also
solicited
comment
on
the
portion
of
the
human
health
risk
characterization
based
on
this
RfC,
and
whether
it
would
be
appropriate
to
delist
MEK
if
the
RfC
resulting
from
an
updated
Integrated
Risk
Information
System
(
IRIS)
review
were
different
from
the
prospective
RfC.

Comments
were
received
from
a
number
of
industry,
State,
environmental
group,
and
other
commenters.
These
comments
are
summarized
below
in
Sections
2.0
through
5.0.
Section
7.0
presents
a
table
identifying
the
commenters.

2.0
General
Support
for
and
Opposition
to
the
Proposed
Delisting
2.1
General
Support
2
Many
commenters
(
58,
61­
65,
69­
75,
77,
79­
82,
85,
86,
88­
90,
92­
94,
98,
101,
103,
106­
108,
112­
114,
115,
119,
124,
126,
127,
131)
generally
supported
the
proposed
rule
to
delist
MEK
from
the
HAP
list
as
specified
in
section
112(
b)
of
the
CAA.
These
commenters
also
believe
that
the
proposed
delisting
is
supported
by
adequate
data.

Several
commenters
(
80,
92,
124)
noted
that
delisting
MEK
would
allow
switching
to
MEK
from
more
costly
solvents,
and
would
reduce
volatile
organic
compound
(
VOC)
emissions.
One
commenter
(
80)
stated
that
if
ethyl
acetate
is
replaced
with
MEK,
VOC
emissions
are
reduced
by
20
percent;
that
MEK
is
not
as
sensitive
to
temperature
changes
as
replacement
solvents;
and
that
MEK
costs
less
than
other
solvents
currently
in
use
(
e.
g.,
16
percent
less
than
ethyl
acetate).
Additionally,
two
commenters
(
92,
124)
stated
that
should
MEK
be
delisted,
some
sources
would
change
from
major
to
area,
potentially
saving
them
money.
Two
commenters
(
82,
92)
noted
that
delisting
MEK
would
not
change
the
VOC
status
of
MEK,
which
subjects
it
to
regulation
under
New
Source
Review
(
NSR),
National
Ambient
Air
Quality
Standards
(
NAAQS),
State
Implementation
Plans
(
SIP),
and
Reasonably
Available
Control
Technology
(
RACT)
standards.
Another
commenter
(
65)
suggested
that
substitutes
for
MEK
currently
in
use
may
pose
an
increased
risk
to
health
and
environment;
therefore,
by
delisting
MEK,
the
risks
would
be
decreased.

2.2
General
Opposition
Several
commenters
(
57,
59,
60,
76,
83,
87,
91,
97,
99,
102,
104,
120,
128)
were
generally
opposed
to
the
delisting
of
MEK.
Some
of
these
commenters
(
83,
104,
128)
maintained
that
the
proposed
rule
did
not
take
into
account
the
future
increases
in
MEK
use.
According
to
one
commenter
(
91),
MEK
was
not
delisted
in
1998
because
it
contributes
to
ozone
formation.
Another
commenter
(
120)
asserted
that
the
delisting
of
MEK
is
not
supported
by
new
or
compelling
scientific
evidence.

3.0
Comments
Pertaining
to
the
Prospective
RfC
EPA
specifically
requested
comments
regarding
the
prospective
RfC
for
MEK.
The
prospective
RfC
was
9
mg/
m3,
while
the
RfC
at
the
time
the
proposed
rule
was
published
was
1
mg/
m3.
The
following
comments
either
oppose
or
support
the
prospective
RfC.

3.1
Opposition
Comment:
One
commenter
(
83)
asserted
that
the
RfC
of
1
mg/
m3
was
set
to
protect
against
birth
defects
and
should
not
be
changed.
Another
commenter
(
120)
maintained
that
the
2003
RfC
is
based
on
the
same
study
from
1991
that
was
used
to
determine
the
1992
RfC.
In
2002,
EPA
reduced
the
uncertainty
after
correcting
the
no­
observed­
adverse­
effect­
level
(
NOAEL)
to
reflect
the
lifetime
continuous
exposure,
using
the
same
1991
data.
In
2003,
EPA
3
abandoned
the
NOAEL
for
a
benchmark
dose.
The
commenter
expressed
the
opinion
that
the
benchmark
dose
and
the
RfC,
which
were
based
on
the
same
study
from
1991,
do
not
adequately
provide
an
estimate
that
is
likely
to
be
without
appreciable
risk
of
deleterious
effects
during
a
lifetime.

Response:
The
RfC
is
designed
to
consider
all
adverse
noncancer
effects
associated
with
lifetime
exposure
to
a
chemical.
The
2003
RfC
is
based
on
developmental
effects,
and
is
based
on
the
methodologies
that
were
in
place
at
the
time
of
derivation,
including
(
1)
the
methods
for
the
use
of
inhalation
dosimetry
to
extrapolate
from
animal
to
human
exposures
(
U.
S.
EPA,
1994)
and
(
2)
benchmark
dose
methods
(
U.
S.
EPA,
2000,
external
review
draft).
Those
methods
have
been
subject
to
peer
review.

Comment:
One
commenter
(
99)
stated
that
the
toxicological
database
is
not
complete
regarding
developmental
effects,
and
that
there
is
inadequate
evidence
to
assess
the
carcinogenic
potential
of
MEK
(
i.
e.,
there
are
no
2­
year
animal
cancer
bioassays).

Response:
There
are
adequate
data
on
developmental
effects
and
on
cancer
effects
to
support
a
decision
to
delist
MEK.
The
principal
study
(
Schwetz
et
al.,
1991),
a
developmental
toxicity
study
in
the
mouse,
is
well­
designed
and
tests
several
exposure
concentrations
over
a
reasonable
range
that
include
maximum
tolerated
doses
for
dams
and
fetuses.
Also,
animal
studies
in
a
second
species
(
rats)
corroborate
the
effect
level
for
developmental
toxicity
(
Deacon
et
al.,
1981;
Schwetz
et
al.,
1974).

Regarding
carcinogenicity,
the
current
IRIS
file
(
completed
in
September
of
2003)
states
that
the
data
for
MEK
are
characterized
as
"
inadequate
for
an
assessment
of
human
carcinogenic
potential."
The
"
Toxicological
Review
of
Methyl
Ethyl
Ketone"
(
U.
S.
EPA,
2003)(
Toxicological
Review
of
MEK),
upon
which
the
IRIS
file
is
based
states,
"
Under
EPA's
draft
revised
cancer
guidelines
(
U.
S.
EPA,
1999),
data
are
inadequate
for
an
assessment
of
human
carcinogenic
potential
for
MEK
because
studies
of
humans
chronically
exposed
to
MEK
are
inconclusive,
and
MEK
has
not
been
tested
for
carcinogenicity
in
animals
by
the
oral
or
inhalation
routes."
Recent
revision
of
these
guidelines
does
not
materially
affect
this
conclusion.

The
traditional
2­
year
animal
cancer
study
has
not
been
conducted
for
MEK,
nor
is
EPA
aware
of
any
organization
planning
to
conduct
one.
EPA
believes
one
reason
no
cancer
assay
has
been
done
is
that
the
results
from
the
majority
of
the
genotoxicity
tests
(
which
are
often
used
as
an
indicator
of
the
need
to
pursue
a
2­
year
cancer
study)
are
negative,
indicating
that
MEK
is
a
low
priority
for
further
study.
In
1997,
the
Organization
for
Economic
Cooperation
and
Development
(
OECD)
reached
this
conclusion.
OECD's
report
states
that
"
MEK
is
not
genotoxic
and
is
not
likely
to
be
carcinogenic."
(
OECD,
1997).
The
report
also
states
that
MEK
is
".
.
.
currently
of
low
priority
for
further
work."
(
OECD,
1997).
4
The
general
descriptors
recommended
by
EPA's
"
Guidelines
for
Carcinogen
Risk
Assessment"
(
U.
S.
EPA,
1999)
for
characterizing
the
weight
of
evidence
with
regard
to
a
chemical's
potential
for
human
carcinogenicity
did
not
explicitly
recognize
this
situation.
The
descriptor
applied
to
MEK
in
the
2003
IRIS
assessment
(
i.
e.,
"
data
are
inadequate
for
an
assessment
of
human
carcinogenic
potential")
pertains
to
cases
where
".
.
.
there
is
a
lack
of
pertinent
or
useful
data."
(
U.
S.
EPA,
1999).
While
lacking
data
or
studies
that
would
clearly
support
their
placement
in
other
categories
(
e.
g.,
the
traditional
2­
year
rodent
study),
chemicals
included
within
this
broad
category
may,
however,
have
pertinent
or
useful
data
which
do
not
indicate
any
potential
for
carcinogenicity,
consequently
providing
no
support
for
the
performance
of
the
traditional,
resource­
intensive
studies.

Accordingly,
EPA's
Toxicological
Review
of
MEK
also
states,
"
the
majority
of
short­
term
genotoxicity
testing
of
MEK
has
demonstrated
no
activity,
and
the
Structure
Activity
Relationship
(
SAR)
analysis
suggests
that
MEK
is
unlikely
to
be
carcinogenic."
(
U.
S.
EPA,
2003).
One
study
(
Woo
et
al.,
2002)
has
given
MEK
and
other
unsubstituted
mono­
ketones
(
a
compound
class
to
which
MEK
belongs)
a
low
concern
rating
(
unlikely
to
be
of
cancer
concern)
because
these
chemicals
lack
electrophilic
activity
(
i.
e.,
a
structural
alert
of
carcinogenicity)
and
are
generally
not
associated
with
carcinogenicity.

There
is
an
absence
of
positive
results
in
the
majority
of
mutagenicity
and
genotoxicity
tests
which
are
designed
to
indicate
the
potential
for
carcinogenicity.
Methyl
ethyl
ketone
has
been
tested
for
activity
in
an
extensive
spectrum
of
in
vitro
and
in
vivo
genotoxicity
assays
and
has
shown
no
evidence
of
genotoxicity
in
most
conventional
assays
(
National
Toxicology
Program,
no
date;
World
Health
Organization
1992;
Zeiger
et
al.,
1992).
Methyl
ethyl
ketone
tested
negative
in
bacterial
assays
(
both
the
S.
typhimurium
(
Ames)
assay,
with
and
without
metabolic
activation,
and
E.
coli),
the
unscheduled
deoxyribonucleic
acid
(
DNA)
synthesis
assay,
the
assay
for
sister
chromatid
exchange
(
SCE)
in
Chinese
hamster
ovary
(
CHO)
cells,
the
mouse
lymphoma
assay,
the
assay
for
chromosome
aberrations
in
CHO
cells,
and
the
micronucleus
assay
in
the
mouse
and
hamster.
The
only
evidence
of
mutagenicity
was
mitotic
chromosome
loss
at
high
concentrations
in
a
study
of
aneuploidy
in
yeast
S.
cerevisiae
(
Zimmerman
et
al.,
1985),
but
the
relevance
of
this
finding
to
humans
is
questionable.
Overall,
studies
of
MEK
yield
little
or
no
evidence
of
genotoxicity.

However,
the
finding
of
low
potential
for
genotoxicity
alone
is
not
the
sole
criterion
for
an
assessment
of
carcinogenic
potential,
as
non­
genotoxic
mechanisms
can
also
result
in
carcinogenesis.
While
developing
the
final
rule,
EPA
learned
that
preliminary
results
of
a
recent
cancer
bioassay
by
the
National
Toxicology
Program
suggested
that
methyl
isobutyl
ketone
(
MIBK)
appears
to
be
a
weak
or
marginally
active
carcinogen
in
rats
and
mice,
possibly
by
a
nongenotoxic
mode
of
action.
Both
MEK
and
MIBK
are
small
molecular
weight
alkyl
ketones,
and
this
similarity
raised
some
questions
regarding
the
possible
relevance
of
the
preliminary
MIBK
results
to
MEK.
To
investigate
this
further,
EPA
undertook
SAR
analysis
of
MIBK
and
MEK.
These
two
ketones
have
a
key
difference
in
their
chemical
structure:
MIBK
is
branched,
while
5
MEK
is
linear.
EPA's
SAR
analysis
indicates
that
MIBK's
toxicity
and
possible
carcinogenicity
are
likely
due
to
its
branched
alkyl
structure.
Methyl
ethyl
ketone,
like
acetone,
is
linear
and
lacks
this
structure.
Thus,
the
analysis
concluded
that
in
analogy
to
acetone
and
its
metabolite
isopropanol
(
which
has
shown
no
evidence
of
carcinogenicity),
MEK
and
its
metabolite
(
2­
butanol)
are
linear
and,
therefore,
have
low
concern
for
carcinogenicity
potential.
A
short
document
describing
the
analysis,
"
Acetone,
MEK,
and
MIBK
­
SAR
Analysis
on
Carcinogenicity/
Toxicity,"
is
included
in
the
docket.
Subsequently,
EPA
conducted
an
external
peer
review
of
this
document.
All
three
reviewers
found
the
reasoning
to
be
sound
and
supported
the
conclusions
of
the
analysis.
These
reviews
are
also
included
in
the
docket.
Thus,
EPA
concludes
that
the
available
scientific
evidence
shows
a
low
potential
for
carcinogenicity
in
MEK.

Comment:
One
commenter
(
104)
suggested
that
the
uncertainty
factors
(
UFs)
for
the
prospective
RfC
were
not
adequate.
The
commenter
disagreed
with
the
reduction
of
the
interspecies
UF
and
stated
that
it
should
have
remained
at
10
because
there
are
no
developmental
and
reproductive
studies
available
for
humans
and
animals.
Another
commenter
(
120)
suggested
that
the
human
equivalent
concentration
(
HEC)
resulted
in
low
confidence
because
it
was
based
on
the
same
mouse
study
(
1991)
as
the
1992
RfC,
and
the
prospective
RfC
was
not
robust
enough
to
warrant
decreasing
the
interspecies
UF
from
10
to
3.
This
commenter
also
asserted
that
the
chronic
and
reproductive
studies
are
still
missing
and,
therefore,
EPA=
s
proposal
of
reducing
the
database
UF
is
not
valid.
The
commenter
contended
that
the
lack
of
current
information
results
in
continued
low
confidence
in
the
database
because
the
data
used
are
from
the
original
studies
used
to
develop
the
1992
RfC.
The
commenter
believes
that
the
Dick
study
did
not
provide
adequate
statistical
power.
Consequently,
the
commenter
believes
that
the
lack
of
toxicity
was
not
demonstrated,
and
that
the
modifying
factor
should
be
maintained
at
3.
The
commenter
concluded
that
the
Aabsence
of
data
should
not
conclude
an
absence
of
toxicity.@

Response:
An
interspecies
UF
of
3
was
applied
in
deriving
both
the
prospective
RfC
and
the
2003
RfC,
consistent
with
EPA
guidance
for
deriving
RfCs
in
effect
at
the
time
(
U.
S.
EPA,
1994).
The
UF
for
interspecies
extrapolation
is
not
intended
to
address
database
deficiencies.
A
database
UF
of
10
was
used
in
developing
the
2003
RfC
to
account
for
the
lack
of
a
chronic
inhalation
toxicity
study
and
multigeneration
reproductive
toxicity
study.

Modifying
factors
have
been
used
in
the
past
in
RfC
derivations,
where
the
magnitude
of
the
factor
reflected
the
scientific
uncertainties
of
the
study
and
database
that
were
not
explicitly
treated
with
standard
uncertainty
factors.
For
the
2003
RfC,
the
default
modifying
factor
of
one
was
used
because
EPA
concluded
that
the
modifying
factor
was
sufficiently
subsumed
in
the
general
database
UF.

Comment:
One
commenter
(
104)
advocated
maintaining
the
modifying
factor
of
3
because
the
commenter
contends
that
toxicity
data
addresses
the
potentiation
of
the
toxic
effects
from
co­
pollutant
exposures.
6
Response:
The
1992
RfC
of
1
mg/
m3
incorporated
a
modifying
factor
of
3
to
account
for
the
lack
of
unequivocal
data
for
respiratory
tract
(
portal­
of­
entry)
effects
suggested
by
earlier
studies
with
MEK
(
Altenkirch
et
al.,
1978).
More
recent
data
concerning
the
portal­
of­
entry
effects
from
MEK
address
the
applicability
of
a
modifying
factor
to
this
assessment.
Dick
et
al.
(
1984,
1988,
1989,
1992)
found
no
evidence
of
a
statistically
significant
increase
in
respiratory
tract
irritation
among
humans
who
were
exposed
to
MEK
at
590
mg/
m3
for
4
hours.
In
addition,
Oleru
and
Onyekwere
(
1992)
found
no
statistically
significant
pulmonary
effects
among
MEKexposed
leather
workers
(
mean
duration
of
employment
was
approximately
10
years).
While
these
studies
do
not
directly
address
the
potential
for
portal­
of­
entry
effects
for
MEK
in
continuous
lifetime
exposure
scenarios
as
animal
studies
evaluating
histology
would,
they
do
address
the
concerns
raised
in
the
1992
IRIS
assessment.
Further,
as
discussed
above,
EPA
contends
that
the
purpose
of
the
modifying
factor
is
sufficiently
subsumed
in
the
general
database
UF,
and
has
discontinued
the
use
of
modifying
factors.

3.2
Support
Comment:
One
commenter
(
103)
asserted
that
the
prospective
RfC
of
9
mg/
m3
was
adequately
recorded
and
justified
in
the
docket
to
quantify
non­
cancer
health
risks
from
MEK,
and
this
commenter
supported
the
delisting
of
MEK
as
long
as
the
RfC
was
not
reduced
below
1
mg/
m3
in
the
final
rule.
Two
commenters
(
82,
108)
suggested
that
delisting
should
be
based
on
the
proposed
RfC
of
15
mg/
m3
rather
than
the
prospective
RfC
of
9
mg/
m3.

Response:
EPA
believes
that
the
most
appropriate
dose­
response
value
to
use
in
evaluating
whether
to
delist
MEK
is
the
2003
RfC
of
5
mg/
m3.

4.0
Comments
Pertaining
to
the
Human
Health
Risk
Characterization
EPA
specifically
requested
comments
regarding
the
portion
of
the
human
health
risk
characterization
based
on
the
prospective
RfC.
The
following
comments
either
support
or
oppose
the
human
health
risk
characterization.

4.1
Support
Two
commenters
(
73,
82)
agreed
that
available
data
indicated
that
MEK
has
low
acute
and
chronic
toxicity.
These
commenters
also
stated
that:

$
The
data
are
adequate
to
determine
that
emissions,
ambient
concentrations,
bioaccumulation,
or
deposition
of
MEK
may
not
reasonably
be
anticipated
to
cause
adverse
human
or
environmental
effects.

$
In
1997,
the
international
community
concluded
that
MEK
was
not
toxic
at
ambient
concentrations.
7
$
The
Screening
Information
Assessment
Meeting
concluded
that
MEK
had
no
toxicity
following
acute
or
repeated
exposures.

$
MEK
was
also
not
mutagenic,
neurotoxic,
or
a
skin
sensitizer.

One
commenter
(
108)
stated
that
MEK
had
been
investigated
through
the
Screening
Information
Data
Set
(
SIDS)
process,
an
international
program,
and
SIDS
concluded
that:

$
MEK
did
not
appear
to
present
a
significant
risk
of
adverse
reproductive
or
developmental
effects.

$
MEK
had
a
low
toxicity
following
acute
oral,
dermal,
and
inhalation
exposure
studies
although
contact
with
the
eyes,
skin,
or
respiratory
tract
did
produce
irritation.

$
There
were
no
significant
signs
of
toxicity
after
repeated
inhalation
exposure
at
high
concentrations
and
MEK
has
not
been
shown
to
have
any
neurotoxic
potential.

$
Human
volunteers
who
underwent
exposure
did
not
demonstrate
any
significant
effects
other
than
minor
irritation
and
sensory
effects
at
relatively
high
levels.

$
MEK
was
not
genotoxic
and
was
not
likely
to
be
carcinogenic.

$
MEK
has
not
been
shown
to
produce
skin
sensitization.

One
commenter
(
82)
asserted
that
the
petitioner=
s
dispersion
modeling
resulted
in
conservative
estimates
because
the
1996
National
Toxics
Inventory
(
NTI)
data
used
for
the
analysis
is
recognized
as
a
conservative
estimate
of
speciated
emissions,
and
because
the
1996
NTI
did
not
account
for
reduced
MEK
use
over
the
last
few
years.
Therefore,
the
commenter
stated,
through
using
the
NTI,
all
the
significant
MEK
emission
sources
were
identified
and
evaluated.
Additionally,
the
modeled
ambient
levels
(
1
mg/
m3)
at
the
facility
entrance
and
not
residential
areas
are
one­
tenth
of
the
proposed
RfC
(
15
mg/
m3).
Also,
the
petitioner
(
108)
determined
that
actual
exposures
to
MEK
would
be
far
less
than
modeled
exposures,
as
MEK
has
a
lower
solvent
emission
per
unit
of
surface
area
compared
to
many
non­
ketone
alternatives,
and
a
smaller
amount
of
MEK
is
needed
to
perform
the
same
function.

Comment:
One
commenter
(
108)
concurred
with
the
use
of
the
Cox
et
al.
(
1975)
study
as
an
appropriate
basis
for
the
reference
dose
(
RfD).
However,
the
commenter
noted
that
the
IRIS
summary
for
MEK
posted
in
1993
considered
1,771
mg/
kg/
day
from
Cox
et
al.
(
1975)
to
be
a
NOAEL,
whereas
this
dose
level
was
considered
to
be
a
lowest­
observed­
adverse­
effect
level
(
LOAEL)
in
the
external
review
draft
of
the
2003
RfC
based
on
reduced
pup
weight
observed
in
the
F1A
litter,
but
not
in
F1B
or
F2
litters.
In
addition,
the
commenter
observed
that
as
part
of
the
application
of
the
benchmark
dose
(
BMD)
methodology
to
this
study,
EPA
selected
a
5
percent
decrease
in
mean
pup
or
fetal
body
weight
per
litter
as
the
benchmark
response
(
BMR)
because
it
fell
within
the
range
of
experimental
dose
levels.
The
commenter
disagreed
with
the
selection
of
the
BMR
because
5
percent
is
less
than
one
standard
deviation
from
the
mean
and
the
effect
is
not
observed
at
1,771
mg/
kg/
day
in
F1B
and
F2
litters.
The
commenter
suggested
using
10
percent
instead,
which
results
in
an
effective
dose
(
ED10)
of
1,756
mg/
kg/
day
and
a
95
percent
lower
bound
on
the
ED
(
LED10)
of
1,314
mg/
kg/
day.
8
Response:
In
the
2003
IRIS
assessment,
the
summary
of
the
Cox
et
al.
(
1975)
study
was
revised
such
that
a
NOAEL
and
LOAEL
for
the
study
as
a
whole
were
identified.
Because
the
body
weight
reductions
compared
to
control
in
the
low­
and
mid­
dose
groups
in
F1A
pups
were
not
observed
in
subsequent
generations
(
i.
e.,
FIB
and
F2),
a
NOAEL
and
LOAEL
for
the
study
as
a
whole
of
1
percent
(
1,771
mg/
kg­
day)
and
2
percent
(
3,122
mg/
kg­
day),
respectively,
were
presented.

One
standard
deviation
as
a
BMR
serves
as
a
recommended
point
of
comparison
across
assessments,
but
is
generally
considered
the
last
choice
as
the
basis
for
the
BMR
if
there
is
no
other
basis
for
a
biologically
relevant
degree
of
change
in
response.
In
the
case
of
pup
body
weight
data
from
the
Cox
et
al.
(
1975)
study,
there
is
no
specific
decrement
that
is
generally
regarded
as
indicative
of
a
biologically
relevant
response.
EPA
considered
the
use
of
one
standard
deviation
as
the
BMR.
Using
data
from
Cox
et
al.
(
1975),
one
standard
deviation
from
the
control
mean
resulted
in
BMDs
that
corresponded
to
body
weights
9
to
26
percent
below
the
control
mean,
values
generally
outside
the
range
of
experimental
data.
Because
an
aim
in
BMD
modeling
is
to
select
a
BMD
within
the
range
of
observation,
other
measures
of
the
BMR
were
examined.
A
5
percent
reduction
in
fetal/
pup
body
weight
relative
to
the
control
was
a
response
rate
that
fell
within
the
range
of
experimental
dose
levels
in
the
Cox
et
al.
(
1975)
study,
and
consequently
was
selected
as
the
BMR.
In
addition,
an
ED10
and
LED10
for
each
endpoint
were
estimated
as
a
consistent
point
of
comparison
across
chemicals,
as
recommended
in
the
Benchmark
Dose
Technical
Guidance
Document
(
U.
S.
EPA,
2000).
Justification
for
the
selection
of
a
BMR
of
5
percent
is
provided
in
the
final
IRIS
assessment
for
MEK
(
Section
5.1.2.1
of
the
Toxicological
Review).
See
the
IRIS
toxicological
review
for
the
complete
discussion
on
this
topic:
http://
www.
epa.
gov/
iris/
subst/
0071.
htm.

Comment:
One
commenter
(
108)
stated
that
the
EPA
did
not
explain
why
a
10
percent
extra
risk
for
misaligned
sternebrae
was
relevant
or
why
it
should
not
be
considered
a
constraint
on
the
BMD
modeling
outcome.

Response:
A
10
percent
extra
risk
for
misaligned
sternebrae
was
selected
as
the
benchmark
response
since
the
model
and
data
are
most
consistent
in
this
range
of
the
data
set.
Also,
EPA=
s
Benchmark
Dose
Technical
Guidance
Document
(
U.
S.
EPA,
2000)
recommends
estimation
of
a
10%
BMR
for
a
point
of
consistent
comparison
across
chemicals.
Further,
the
nested
model
did
not
provide
a
useful
estimate
of
the
lower
bound
for
a
BMR
of
5
percent
for
this
particular
data
set
(
the
lower
bound
on
the
BMC
was
estimated
as
essentially
zero).

Comment:
One
commenter
(
108)
contends
that
the
BMD
is
too
restrictive
because
it
is
based
on
misaligned
sternebrae
studies,
and
believes
that
decreased
fetal
birth
weight
is
a
more
scientifically
sound
basis
for
determining
the
LOAEL,
NOAEL,
and
RfC.
9
Response:
EPA
considered
the
use
of
both
reduced
fetal
weight
and
incidence
of
misaligned
sternebrae
as
the
basis
for
the
RfC
point
of
departure,
taking
into
account
external
peer
reviewers=
comments
and
further
analysis
of
the
data
from
Schwetz
et
al.
(
1991).
Points
that
support
the
use
of
fetal
weight
rather
than
misaligned
sternebrae
included
the
fact
that
the
incidence
of
misaligned
sternebrae
was
highly
nonlinear
with
dose,
and
the
incidence
showed
a
high
degree
of
variability.
The
full
laboratory
report
of
this
developmental
toxicity
study
(
Mast
et
al.,
1989)
also
reported
a
positive
trend
(
not
statistically
significant)
for
reduced
ossification
of
the
sternebrae,
raising
the
possibility
that
misaligned
sternebrae
may
reflect
a
more
general
growth
delay.
None
of
the
external
peer
reviewers,
however,
offered
an
opinion
that
misaligned
sternebrae
was
not
a
biologically
significant
endpoint
of
toxicity.
Selection
of
the
critical
effect
(
i.
e.,
reduced
fetal
weight
or
incidence
of
misaligned
sternebrae)
was
further
deliberated
during
Agency
consensus
review.
In
general,
the
arguments
for
using
fetal
weight
as
the
more
supportable
critical
effect
were
not
found
to
be
sufficiently
compelling.
Thus,
the
decision
was
made
to
use
the
more
health
protective
endpoint,
misaligned
sternebrae,
as
the
critical
effect.

Comment:
One
commenter
(
108)
stated
that
EPA
did
not
present
adequate
scientific
justification
for
applying
a
duration
adjustment
to
the
inhalation
developmental
toxicity
study
and,
at
the
very
least,
the
additional
conservatism
added
by
the
application
of
this
factor
should
be
explicitly
recognized.
The
commenter
pointed
to
the
draft
Toxicological
Review
that
indicated
that
MEK
was
rapidly
absorbed,
distributed,
and
metabolized,
suggesting
that
the
duration
adjustment
may
be
inappropriate.

Response:
Duration
adjustment
of
the
exposure
concentrations
in
the
developmental
study
of
MEK
(
Schwetz
et
al.,
1991)
was
performed
consistent
with
the
EPA
Risk
Assessment
Forum
RfD/
RfC
Technical
Panel
report,
"
A
Review
of
the
Reference
Dose
and
Reference
Concentration
Processes"
(
U.
S.
EPA,
2002).
The
report
recommends
that
procedures
for
adjusting
to
continuous
exposure
based
on
the
product
of
concentration
and
time
be
used
as
a
default
for
inhalation
developmental
toxicity
studies
as
it
is
for
other
health
effects
from
inhalation
exposure.
While
the
recommendation
is
based
on
evidence
that
shows
that
some
agents
cause
developmental
toxicity
more
as
a
function
of
peak
concentration,
the
effects
of
other
agents
are
related
to
area­
under­
the­
curve
(
AUC).
The
latter
is
true
even
of
some
developmental
toxicants
with
a
short
half­
life.
In
the
absence
of
data
that
support
peak
concentration
or
AUC
as
more
closely
correlated
with
developmental
toxicity,
EPA=
s
2002
review
document
recommends
duration
adjustment
as
the
more
health­
protective
default
procedure.
As
noted
in
the
Toxicological
Review
of
MEK,
because
the
data
are
insufficient
to
argue
convincingly
for
either
peak
exposure
level
or
AUC
as
the
most
appropriate
metric,
the
more
health­
protective
procedure
(
duration
adjustment)
was
applied
as
a
policy
matter.

Comment:
The
petitioner
commented
on
EPA's
interpretation
of
the
Cavender
et
al.
(
1983)
study.
They
stated
that
EPA
regarded
5,000
ppm
in
a
90­
day
inhalation
study
as
the
LOAEL
based
on
reduced
weight
gain,
increased
liver
weight,
and
decreased
brain
weight.
The
commenter
stated
that
this
was
inconsistent
with
the
IRIS
database
where
EPA
indicated
that
a
10
change
in
liver
weight
may
not
be
conclusively
caused
by
MEK
inhalation.
The
petitioner
recommended
that
5,000
ppm
be
the
NOAEL.

Response:
In
the
2003
IRIS
assessment,
EPA
gave
further
consideration
to
the
biological
significance
of
the
findings
in
the
5,000
ppm
animals
in
the
Cavender
et
al.
(
1983)
study,
specifically
the
organ
weight
findings.
Although
the
decrease
in
brain
weight
in
female
high­
dose
animals
is
of
some
concern,
EPA
agrees
that
this
effect,
in
the
absence
of
corresponding
histopathology
and
functional
abnormalities,
cannot
be
clearly
characterized
as
being
of
toxicological
relevance.
In
light
of
these
uncertainties,
characterization
of
the
effects
associated
with
the
5,000
ppm
exposure
level
as
adverse,
use
of
that
level
as
a
LOAEL,
and
the
use
of
middose
group
(
2,518
ppm)
as
a
NOAEL
were
dropped.

4.2
Opposition
Comment:
Three
commenters
(
83,
128,
120)
suggested
that
the
actual
emissions
of
MEK
may
result
in
environmental
concentrations
below
the
RfC,
but
allowable
emissions
would
not.
This
means
that
should
the
emissions
reach
allowable
limits,
then
the
concentrations
of
MEK
will
be
above
the
RfC.
One
commenter
provided
an
example
of
a
facility
that
emits
500
tons
per
year
(
tpy)
of
MEK
but
is
permitted
to
emit
up
to
2,200
tpy.
The
commenter
states
that
a
simple
screening
model
run
(
most
likely
similar
to
the
tier
1
or
tier
2
analysis
submitted
by
the
petitioner)
of
this
facility
at
the
allowable
emission
rate
predicts
24­
hour
peak
concentrations
to
be
about
75
mg/
m3,
which
is
above
the
maximum
predicted
24­
hour
average
concentration
of
10
mg/
m3
that
EPA
cited
in
the
preamble.

Response:
The
maximum
offsite
24­
hr
MEK
concentration
for
the
worst­
case
facility
in
the
petition
as
predicted
by
the
Industrial
Source
Complex
Short
Term
3
(
ISCST3)
model
was
10
mg/
m3.
The
maximum
annual
concentration
was
1.2
mg/
m3.
This
facility
emits
about
500
tpy
MEK.
The
maximum
offsite
concentration
occurs
within
a
few
hundred
meters
of
the
facility.

The
commenters
provided
limited
information
on
the
facility
that
has
the
potential
to
emit
2,200
tpy.
EPA
contacted
the
commenter
in
order
to
understand
how
they
estimated
the
value
of
75
mg/
m3.
EPA
was
told
that
the
SCREEN3
model
used
to
estimate
this
concentration.
However,
EPA
was
unable
to
obtain
the
modeling
runs
which
would
contain
important
model
input
data
(
e.
g.,
stack
heights
and
distances
from
stacks
to
fence
lines).
From
the
comment,
EPA
does
know
that
the
maximum
offsite
concentration
for
this
facility
as
predicted
by
the
SCREEN3
model
was
75
mg/
m3
for
a
24­
hr
average
and
1.1
mg/
m3
for
an
annual
average.
If
this
facility
were
modeled
with
a
more
refined
dispersion
model,
such
as
the
ISCST3
model,
EPA
would
expect
impacts
that
are
considerably
lower
than
those
predicted
with
the
more
conservative
SCREEN3
model.
Most
likely,
the
maximum
offsite
concentration
for
the
facility
would
be
much
closer
to
10
mg/
m3
for
a
24­
hr
average
near
the
facility,
and
well
below
1
mg/
m3
for
the
annual
average.
EPA
would
suspect
that
the
facility
to
which
the
commenter
refers
has
much
better
dispersion
characteristics
than
the
petitioner=
s
worst­
case
facility,
which
had
a
very
low
stack
and
11
nearby
fenceline.

Comment:
Three
commenters
(
83,
104,
120)
stated
that
EPA
failed
to
meet
the
CAA
deadline
(
18
months)
for
adding
or
deleting
a
substance
from
the
HAP
list,
instead
taking
78
months
total.
Therefore,
the
commenters
believed
the
1994
Toxic
Release
Inventory
(
TRI)
data
used
in
the
assessment
were
not
appropriate
and
that
current
TRI
data
should
have
been
used.
These
commenters
also
contended
that
the
calculations
in
the
petition
did
not
consider
potential
increases
in
MEK
use
once
MEK
is
delisted,
and
that
EPA
should
base
its
decision
to
delist
MEK
on
emission
levels
and
locations
expected
after
delisting.

Response:
EPA
interprets
the
CAA
to
require
consideration
of
current
emissions.
It
is
not
appropriate
to
make
a
decision
on
what
can
only
be
speculative
emissions.
EPA
states
in
the
final
rule
to
delist
caprolactam
(
61
FR
30816,
June
18,
1996)
that
AEPA
does
not
interpret
Section
112(
b)(
3)(
C)
to
require
consideration
of
hypothetical
emissions
from
facilities
that
might
be
constructed
in
the
future.
The
logical
consequence
of
such
an
expansive
construction
would
be
that
no
substance
could
ever
be
delisted,
due
to
the
hypothetical
possibility
of
some
future
facility
that
has
uncontrolled
emissions
large
enough
to
cause
adverse
effects.
In
the
event
some
future
facility
has
uncontrolled
caprolactam
emissions
great
enough
to
change
the
conclusion
of
the
current
EPA
risk
assessment,
EPA
can
revisit
its
decision
to
delist
caprolactam
at
that
time."
It
is
not
the
case,
however,
that
EPA
can
never
take
potential
increases
in
emissions
into
account.
For
example,
such
consideration
is
appropriate
where
EPA
has
information
regarding
specific
facilities,
such
as
the
information
it
considered
in
denying
the
methanol
delisting
petition.

Using
similar
logic
in
this
case,
EPA
does
not
interpret
CAA
section
112
(
b)(
3)(
C)
to
require
consideration
of
hypothetical
emissions
from
facilities
that
might
be
constructed
in
the
future,
nor
projections
of
increases
in
emissions
from
existing
facilities.

There
are
several
reasons
why
EPA
does
not
expect
that
increases
in
emissions
of
MEK
will
cause
health
or
environmental
concerns.
With
regard
to
increased
emissions
themselves,
EPA
believes
that
such
increases
will
be
limited
by
good
housekeeping
practices
which
are
designed
to
save
product.
Methyl
ethyl
ketone
is
an
effective
solvent,
but
one
that
evaporates
readily.
Employing
techniques
to
prevent
wasting
the
product
also
results
in
decreased
emissions.

Due
to
the
health­
protective
nature
of
the
analysis
upon
which
the
decision
to
delist
is
based,
EPA
concludes
that
the
potential
risks
from
outdoor
exposures
to
MEK
are
overestimated.
It
is
unlikely
that
future
emissions
increases
will
result
in
unacceptable
risk.
For
example,
the
petitioner
based
the
risk
assessment
on
1994
TRI
total
air
emissions
of
MEK,
which
were
628
tpy
for
the
worst­
case
facility.
This
facility=
s
modeled
annual
average
concentration
is
only
20
percent
of
the
RfC.
This
facility
could
increase
emissions
significantly
before
the
concentration
would
be
above
a
level
of
concern.
The
highest­
emitting
facility
in
the
2003
TRI
emits
638
tpy
of
MEK,
only
slightly
higher
than
the
1994
TRI
emissions
for
the
worst­
case
facility.
12
In
addition,
the
national
trend
in
MEK
emissions
is
distinctly
downward.
Comparing
the
1994
and
2003
TRI
MEK
emissions
data
for
the
100
highest­
emitting
facilities
indicates
that
air
emissions
have
decreased
by
approximately
20
percent
during
that
nine
year
period.

The
risk
assessment
was
based
on
a
maximum
off­
site
concentration.
The
assessment
did
not
consider
the
amount
of
time
people
would
be
at
that
location,
or
other
factors
that
address
the
amount
of
exposure
faced
by
actual
individuals.
Further,
this
maximum
concentration
was
located
at
the
entrance
to
a
facility
in
an
industrial
park.
The
probability
that
an
individual
would
live
at
this
location
in
the
future
is
extremely
low.

Given
the
low
hazard
presented
by
the
worst­
case
facility,
the
health­
protective
nature
of
the
analysis,
and
the
overall
downward
trend
of
MEK
emissions
over
the
last
several
years,
EPA
believes
that
emissions
of
MEK
may
not
reasonably
be
anticipated
to
cause
adverse
human
health
effects.

The
preamble
to
the
proposed
rule
discussed
the
March
30,
1998,
Federal
Register
notice
(
63
FR
15195)
in
which
EPA
issued
a
Denial
of
Petition
entitled
"
Methyl
Ethyl
Ketone;
Toxic
Chemical
Release
Reporting;
Community
Right­
to­
Know."
The
denial
was
in
response
to
a
petition
from
the
Ketones
Panel
of
the
Chemical
Manufacturers
Association
(
CMA)
that
requested
the
deletion
of
MEK
from
the
list
of
chemicals
reportable
under
section
313
of
the
Emergency
Planning
and
Community
Right­
To­
Know
Act
of
1986
(
EPCRA)
and
section
6607
of
the
Pollution
Prevention
Act.

The
American
Chemistry
Council
filed
suit
challenging
EPA's
decision
in
the
United
States
District
Court
for
the
District
of
Columbia.
Subsequently,
the
court
granted
summary
judgment
in
favor
of
EPA
(
American
Chemistry
Council
v.
Whitman,
309
F.
Supp.
2d
111
(
D.
D.
C.
2004)).
On
appeal,
the
Court
of
Appeals
for
the
District
of
Columbia
Circuit
reversed
the
lower
court's
decision,
vacating
the
lower
court's
decision,
and
directed
the
district
court
to
issue
an
order
to
"
direct
EPA
to
delete
MEK
from
the
TRI"
(
406
F.
3d
738,
742
(
D.
C.
Cir.
2005)).
The
circuit
court
issued
its
mandate
on
June
13,
2005
and,
accordingly,
on
June
30,
2005,
EPA
issued
a
final
rule
(
70
FR
37698)
revising
the
EPCRA
section
313
list
of
reportable
chemicals
in
40
CFR
372.65
to
delete
MEK.

The
deletion
of
MEK
from
the
EPCRA
section
313
list
eliminates
the
main
source
of
data
EPA
uses
to
track
MEK
emissions.
However,
there
are
other
data
sources
available
to
estimate
MEK
emissions,
including
market
research
data
on
MEK
production,
import,
export,
and
consumption.
Consumption
of
MEK
should
provide
an
adequate
surrogate
for
emissions
to
determine
whether
significant
increases
in
emissions
are
occurring.
If
data
indicate
that
MEK
emissions
are
increasing
significantly,
EPA
has
the
option
to
add
MEK
back
on
the
HAP
list.

Comment:
One
commenter
(
120)
suggested
that
the
risk
was
not
adequately
identified
because
the
industry
was
not
studied
comprehensively
enough
to
determine
chronic
exposure.
13
Response:
In
order
to
determine
the
risks
from
emissions
of
MEK,
the
petitioner
used
the
1994
TRI
as
the
basis
of
an
emissions
inventory
intended
to
quantify
annual
emissions
of
MEK,
to
identify
and
locate
emissions
sources,
and
to
acquire
some
facility­
specific
emissions
information.
The
1994
TRI
shows
that
there
are
over
2,000
sources
with
reported
emissions
of
MEK.
The
petition
states
that
over
85
percent
of
these
facilities
(
approximately
1,700)
emit
25
tpy
or
less.
The
petition
also
states
that
approximately
800
facilities
emit
between
10
and
200
tpy,
and
27
facilities
emit
200
tpy
or
more.
In
addition
to
using
the
1994
TRI,
the
petitioner
queried
a
subset
of
individual
sources
to
obtain
site­
specific
source,
release,
and
facility
information
for
the
purpose
of
conducting
more
detailed
risk
assessments.
EPA
has
determined
that
this
approach
to
establishing
reasonable
worst­
case
exposures
to
MEK
emissions
is
an
adequate
basis
upon
which
to
base
a
decision
to
delist
MEK.
EPA
states
in
the
preamble
to
the
proposed
rule
that
it
does
not
interpret
CAA
section
112(
b)(
3)(
C)
to
require
absolute
certainty
that
a
pollutant
will
not
cause
adverse
effects
on
human
health
or
the
environment
before
it
may
be
deleted
from
the
list.
The
use
of
the
terms
Aadequate@
and
Areasonably@
indicate
that
EPA
must
weigh
the
potential
uncertainties
and
likely
significance.
In
this
case,
the
uncertainty
in
the
predicted
exposure
levels
is
biased
toward
protecting
public
health.
Therefore,
EPA
concludes
that
delisting
MEK
is
appropriate.

Comment:
Several
commenters
(
87,
91,
104,
120)
contended
that
chronic
effects
of
MEK
had
not
been
adequately
studied
or
evaluated,
and
that
the
delisting
was
not
supported
by
new
or
compelling
scientific
evidence.
One
commenter
(
87)
requested
that
EPA
conduct
long­
term
health
effects
studies.
Additionally,
two
of
the
commenters
(
87,
104)
stated
that
there
were
no
lifetime­
chronic
studies
included,
no
studies
evaluating
developmental
effects,
nor
studies
concerning
reproductive
toxicity.
These
commenters
also
asserted
that
there
were
no
multigenerational
studies
included,
and
the
evaluation
of
the
carcinogenic
potential
was
not
adequate.

Response:
EPA's
RfC
methodology
(
U.
S.
EPA,
1994)
does
not
always
require
a
complete
database
in
order
to
develop
an
RfC;
however,
the
database
must
at
least
meet
minimum
data
requirements.
For
MEK,
A.
.
.
confidence
in
the
database
is
medium
.
.
..
@
(
U.
S.
EPA,
2003).
AThe
subchronic
study
by
Cavender
et
al.
(
1983)
satisfies
the
minimum
inhalation
database
requirements
for
derivation
of
an
RfC.@
(
U.
S.
EPA,
2003).

In
the
case
where
there
are
enough
quality
data
with
which
to
set
an
RfC,
but
where
the
database
is
less
than
complete,
EPA
adds
a
database
uncertainty
factor
to
account
for
the
lack
of
data.
For
MEK,
that
factor
is
10.
EPA
acknowledges
the
lack
of
a
chronic
toxicity
bioassay
and
an
inhalation
multigeneration
reproductive
toxicity
study
(
an
oral
multigeneration
is
available),
but
notes
that
contrary
to
the
commenters=
statements,
the
developmental
toxicity
of
MEK
has
been
well
studied.

The
MEK
RfC
is
an
estimate
(
with
uncertainty
spanning
perhaps
an
order
of
magnitude)
14
of
a
daily
inhalation
exposure
to
the
human
population
(
including
sensitive
subgroups)
that
is
likely
to
be
without
appreciable
risk
of
deleterious
noncancer
effects
during
a
lifetime.
Because
maximum
expected
ambient
air
concentrations
are
well
below
the
RfC,
EPA
does
not
expect
adverse
noncancer
effects
to
result.
In
addition,
the
health­
protective
nature
of
the
assessment
described
above
adds
to
EPA's
confidence
that
no
adverse
health
effects
will
occur
from
ambient
exposures
to
MEK.

Comment:
One
commenter
(
104)
asserted
that
the
appropriate
averaging
time
for
assessing
the
potential
for
adverse
developmental
effects
to
occur
is
the
24­
hour
average
not
an
annual
average.
The
commenter
held
that
evaluating
developmental
toxicity
on
a
24­
hour
basis
is
supported
by
EPA
guidelines
for
evaluating
developmental
risk.
This
issue
was
also
raised
by
the
Voluntary
Children's
Chemical
Evaluation
Program
(
VCCEP)
review
panel
as
they
considered
the
information
industry
submitted
on
MEK
and
children=
s
health.

Response:
EPA
agrees
with
the
commenter
that
potential
concern
for
developmental
effects
from
short­
term
exposures
should
be
addressed,
and
EPA
did
this.
With
regard
to
the
use
of
endpoint­
specific
reference
values,
EPA=
s
review
of
the
RfD/
RfC
processes
recommended
against
the
use
of
endpoint­
specific
reference
values,
and
instead
recommended
that
durationspecific
reference
values
be
derived
in
consideration
of
the
full
range
of
adverse
effects.

Comment:
One
commenter
(
120)
remarked
that
EPA
did
not
take
into
account
all
routes
of
exposure
to
MEK
and,
therefore,
did
not
adequately
identify
the
risk.

Response:
MEK
is
neither
bioaccumulative
nor
persistent.
It
has
a
half­
life
of
approximately
9
days.
The
releases
of
MEK
to
air
are
unlikely
to
result
in
elevated
concentrations
in
surface
water,
ground
water,
or
the
food
supply.
Therefore,
the
route
of
exposure
EPA
is
concerned
with
is
direct
inhalation
of
MEK
released
to
the
ambient
air.
For
this
reason,
inhalation
was
the
focus
of
the
analysis.
The
petitioner
also
assessed
the
potential
for
risks
due
to
ingestion
of
water
contaminated
with
MEK.
In
both
cases,
the
risks
were
below
a
level
of
concern.

Comment:
One
commenter
(
99)
asserted
that
the
risk
assessment
did
not
fully
address:
(
1)
other
solvents
released
from
stationary
and
area
sources
of
MEK,
(
2)
actual
ambient
concentrations
near
stationary
and
area
sources
(
only
modeled
concentrations
were
used),
and
(
3)
the
human
health
effects
within
the
facilities
as
opposed
to
fenceline
ambient
concentrations.

Response:
The
maximum
annual
average
air
concentration
resulting
from
emissions
of
MEK
is
not
expected
to
exceed
an
HQ
of
0.2.
This
value,
which
is
20
percent
of
the
RfC,
is
quite
low.
EPA
believes
that
there
is
a
large
enough
margin
of
exposure
to
preclude
a
need
to
address
any
other
emitted
HAP
that
may
affect
the
same
target
organ
as
MEK.

The
petitioner
did
not
monitor
ambient
air
around
actual
MEK­
emitting
facilities.
Such
an
effort
would
not
add
to
the
analysis,
or
change
EPA=
s
conclusion
with
regard
to
delisting.
This
is
15
because
the
maximum
monitored
concentration
EPA
found
in
the
U.
S.
was
over
two
orders
of
magnitude
below
the
maximum
modeled
concentration,
and
because
the
modeling
conducted
was
designed
to
over­
estimate
ambient
concentrations.
For
example,
the
model
assumed
that
individuals
are
continuously
exposed
to
the
maximum
modeled
concentrations
of
MEK
in
air
for
70
years,
and
EPA
used
the
maximum
annual
average
concentration
as
a
surrogate
for
long­
term
exposure.
Also,
the
model
used
1994
emission
rates
which
are
significantly
higher
than
current
emissions.
EPA
believes
that
the
health­
protective
air
dispersion
modeling
performed
as
part
of
the
petition
and
described
in
detail
in
the
proposed
rule
resulted
in
higher
concentrations
than
would
monitoring
around
facilities.

EPA
cannot
consider
the
health
effects
of
emissions
within
facility
boundaries.
That
is
the
purview
of
the
Occupational
Safety
and
Health
Administration.

Comment:
One
commenter
(
102)
recommended
that
a
comparative
analysis
with
the
1998
Office
of
Pollution
Prevention
and
Toxics
(
OPPT)
assessment
(
located
in
the
docket)
be
done
to
fully
assess
the
risks
of
MEK.

Response:
EPA
agrees
with
the
comment,
and
EPA
conducted
a
comparison
of
the
1998
OPPT
assessment
and
the
assessment
in
the
proposal
to
delist
MEK.

The
assessment
presented
in
the
petition
to
delist
MEK
estimated
a
maximum
annual
average
MEK
concentration
of
1.2
mg/
m3.
It
used
the
ISCST3
model,
which
is
a
refined
air
dispersion
model
that
predicts
an
annual
average
by
averaging
8,760
hours
of
actual
meteorological
data.
The
ISCST3
model
predicted
a
maximum
24­
hour
average
MEK
concentration
of
10
mg/
m3.

The
1998
OPPT
study
estimated
maximum
24­
hour
average
concentrations
of
100­
200
mg/
m3.
It
used
a
screening
model
similar
to
the
SCREEN3
model
and
predicted
1­
hour
average
concentrations
under
defined
meteorological
conditions
with
the
assumption
that
the
receptor
is
always
directly
downwind
from
the
source.
Such
screening
model
runs
typically
result
in
high
air
concentrations
as
compared
to
the
ISCST3
model.
EPA
would
expect
the
difference
in
concentrations
to
be
as
high
as
a
factor
of
10.
In
addition,
the
OPPT
study
applied
a
multiplicative
factor
to
predict
typical
(
5),
stagnant
(
10),
and
maximum
(
60)
acute
impacts.
Thus,
the
difference
between
the
two
model
results
can
be
attributed
to
the
multiplicative
factors
and
differences
between
a
refined
and
screening
model.

5.0
Miscellaneous
Comments
Comment:
One
commenter
(
92)
recommended
that
EPA
not
wait
for
the
formal
IRIS
review
of
MEK
or
the
VCCEP
results
to
make
a
final
decision
regarding
delisting
of
MEK,
as
there
was
enough
evidence
to
delist
MEK
without
the
additional
information.
Another
commenter
(
99)
asserted
that
if
the
new
IRIS
RfC
is
different
from
the
proposed
RfC,
then
the
16
petition
should
be
reconsidered
and
an
additional
public
comment
period
should
be
allowed
giving
the
public
an
opportunity
to
comment
on
EPA
=
s
decision.
This
commenter
also
stated
that
the
results
of
the
VCCEP
should
be
concluded
before
the
comments
on
the
delisting
are
due.

Response:
Regarding
the
first
comment,
EPA
waited
to
make
a
final
decision
to
delist
MEK
until
the
2003
IRIS
RfC
was
determined
and
until
the
information
submitted
by
industry
under
the
VCCEP
was
reviewed
in
case
the
results
of
each
of
these
processes
altered
our
decision
to
remove
MEK
from
the
HAP
list.

Regarding
the
second
comment,
EPA
considers
an
additional
comment
period
unnecessary
for
a
number
of
reasons.
First,
EPA
explicitly
solicited
comment
on
the
effect
of
a
difference
between
the
prospective
RfC
and
the
RfC
resulting
from
the
completed
IRIS
assessment.
EPA
specifically
requested
comments
on
the
decision
in
light
of
potential
values
for
the
RfC
of
9
mg/
m3,
3
mg/
m3
and
1
mg/
m3.
The
2003
RfC
of
5
mg/
m3
is
in
the
middle
of
the
range
upon
which
EPA
solicited
comment.
Second,
while
the
2003
RfC
is
lower
than
the
prospective
RfC,
the
result
of
this
change
was
only
to
increase
the
HQ
for
the
maximum
annual
average
ambient
exposure
from
0.1
to
0.2
(
20
percent
of
the
RfC).
This
HQ
is
well
below
a
level
of
concern.

In
addition,
EPA
judges
that
the
exposures
to
MEK
of
actual
persons
living
in
the
immediate
vicinity
of
an
MEK
emission
source
would
more
typically
be
at
least
a
factor
of
2
to
10
less
than
the
predicted
maximum
ambient
concentration
presented
in
the
petition
of
1
mg/
m3.
This
is
because
the
concentration
of
MEK
declines
very
rapidly
as
the
plume
disperses,
and
the
analysis
showed
that
people
do
not
live
at
the
point
of
maximum
concentration.
Therefore,
actual
exposed
individuals
would
be
subject
to
MEK
concentrations
less
than
1
mg/
m3.
If
EPA
were
to
replace
the
maximum
ambient
concentration
with
a
more
realistic
exposure
scenario,
it
would
yield
an
HQ
less
than
0.2.
Based
on
the
current
information,
and
given
the
conservative
nature
of
the
parameters
used
to
estimate
the
maximum
exposure,
and
because
the
petition
and
subsequent
analyses
characterize
the
vast
majority
of
MEK
exposures
from
stationary
sources,
EPA
concludes
that
by
applying
the
RfC
of
5
mg/
m3,
potential
ambient
exposures
to
MEK
may
not
reasonably
be
anticipated
to
cause
adverse
human
health
effects.

With
respect
to
the
results
of
the
VCCEP,
EPA
found
it
unnecessary
to
extend
the
public
comment
period
until
after
the
review
of
the
industry­
submitted
information
was
complete.
This
is
because
the
industry
provided
no
new
information
to
EPA
that
was
not
already
available.
Therefore,
there
was
no
new
information
upon
which
to
solicit
comments.

Comment:
Several
commenters
(
87,
91,
104,
120)
noted
that
the
interactions
with
nhexane
and
other
ketones
have
not
been
sufficiently
investigated
should
the
MEK
emissions
increase.
These
commenters
stated
that
MEK
interactions
with
n­
hexane
have
been
shown
to
increase
neurotoxicity
of
n­
hexane.
17
Response:
EPA
stated
in
the
preamble
to
the
proposed
rule
that
MEK
has
been
shown
to
potentiate
the
neurotoxicity
of
other
solvents
in
experiments
with
laboratory
animals
when
both
MEK
and
the
other
solvent
are
present
in
high
concentrations.
EPA
also
stated
that
studies
of
occupationally­
exposed
populations
(
as
reviewed
by
Noraberg
and
Alien­
Soborg,
2000)
provide
some
evidence
of
possible
interactions
in
humans.
EPA
reviewed
the
occupational
epidemiology
literature
in
more
depth
during
the
development
of
the
final
RfC
for
MEK.
These
findings
are
summarized
in
the
Toxicological
Review
for
MEK
(
http://
www.
epa.
gov/
iris/
toxreviews/
0071­
tr.
pdf,
section
4.4.4).
Available
occupational
studies
involving
multiple
chemical
exposures
do
not
provide
information
adequate
to
clearly
establish
an
interaction
between
MEK
and
other
neurotoxic
solvents
in
humans.
In
studies
suggesting
a
potential
interaction,
neurotoxicity
has
been
observed
only
in
workplace
populations
exposed
to
solvent
mixtures
where
reported
MEK
air
concentrations
reached
levels
at
or
above
the
Threshold
Limit
Value
(
TLV)
(
200
ppm
or
590
mg/
m3).
EPA
concluded
that
the
concerns
for
chemical
interactions
are
especially
diminished
at
the
low
levels
seen
in
this
assessment:
less
than
1
mg/
m3
for
chronic
exposures,
10
mg/
m3
for
24­
hour
exposures
and
25
mg/
m3
for
a
1­
hour
exposure.
These
exposures
are
all
well
below
the
reversible
effects
level
of
590
mg/
m3.
Therefore,
EPA
does
not
expect
possible
potentiation
of
n­
hexane
by
MEK
at
the
low
environmental
concentrations
that
would
be
associated
with
industrial
releases.

Comment:
One
commenter
(
91)
was
concerned
that
MEK
was
detected
by
the
National
Health
and
Nutrition
Examination
Survey
in
biomonitoring
programs.

Response:
EPA
acknowledged
in
the
preamble
to
the
proposed
rule
that
MEK
has
been
reported
to
be
found
in
blood.
EPA
also
stated
that
the
data
indicated
the
source
of
the
MEK
is
likely
a
by­
product
of
normal
human
metabolism,
and
it
is
reasonable
to
expect
it
did
not
result
from
an
air
exposure
to
MEK
at
the
concentrations
seen
in
the
ambient
air.

Comment:
One
commenter
(
99)
requested
that
EPA
consider
the
role
of
MEK
as
an
ozone
precursor
in
deciding
the
petition.

Response:
EPA
stated
in
the
preamble
to
the
proposed
rule
that
it
was
inappropriate
to
consider
the
role
of
MEK
as
an
ozone
precursor
because
the
Adual
structure
(
differentiating
between
HAP
and
criteria
pollutants/
precursors)
would
lose
its
significance
if
EPA
were
to
include
substances
on
the
HAP
list
solely
as
a
result
of
their
contribution
to
concentrations
of
criteria
air
pollutants.@
Specifically,
the
structure
of
the
CAA
is
best
protected
by
including
compounds
on
the
HAP
list
only
where
such
inclusion
is
warranted
based
upon
the
HAP
noncriteria
pollutant
related
effects.
This
interpretation
is
supported
by
the
following
prohibition
related
to
listing
of
new
HAP
contained
in
CAA
section
112(
b)(
2):
ANo
air
pollutant
which
is
listed
under
section
7408(
a)
of
this
title
[
the
criteria
pollutant
list]
may
be
added
to
the
list
under
this
section,
except
that
the
prohibition
of
this
sentence
shall
not
apply
to
any
pollutant
which
independently
meets
the
listing
criteria
of
this
paragraph
and
is
a
precursor
to
a
pollutant
which
is
listed
under
section
7408(
a)
.
.
..@
18
Comment:
One
commenter
(
99)
stated
that
decisions
to
list
or
delist
are
governed
by
the
precautionary
principle.
The
commenter
stated
that,
Ain
considering
whether
a
petitioner
has
met
the
heavy
burden
of
demonstrating
that
a
substance
should
be
removed
from
the
hazardous
air
pollutant
list,
the
precautionary
principle
requires
that
EPA
resolve
uncertainty
in
favor
of
more
protection,
not
less.
The
recognition
of
uncertainty
in
the
listing
and
delisting
process
does
not
give
EPA
to
delist
a
chemical
based
on
incomplete
and
outdated
information
as
it
has
proposed
to
do
with
MEK.@

Response:
EPA
does
not
believe
it
is
appropriate
to
require
that
all
uncertainty
be
resolved
in
favor
of
not
delisting.
Such
a
requirement
of
absolute
certainty
is
inconsistent
with
EPA's
interpretation
of
the
requirement
that
to
delist
a
HAP,
EPA
must
determine
that
there
are
Aadequate
data
on
the
health
and
environmental
effects
of
the
substance
to
determine
that
emissions,
ambient
concentrations,
bioaccumulation
or
deposition
of
the
substance
may
not
reasonably
be
anticipated
to
cause
any
adverse
effect
to
human
health
or
adverse
environmental
effects.@
As
explained
in
denying
the
petition
to
delist
methanol,
EPA
does
"
not
interpret
CAA
section
112(
b)(
3)(
C)
to
require
absolute
certainty
that
the
pollutant
will
not
cause
adverse
effects
on
human
health
.
.
.
before
it
may
be
deleted
from
the
list.
The
use
of
the
terms
>
adequate=
and
>
reasonably=
indicate
that
EPA
must
weigh
the
potential
uncertainties
and
their
likely
significance.@
(
See
66
FR
21929
­
21930,
May
2,
2001.)
For
the
reasons
explained
above,
EPA
determined
that
this
burden
has
been
met
here.

Comment:
One
commenter
(
87)
asserted
that
EPA
has
not
adequately
considered
the
odor
problems
associated
with
MEK.
The
commenter
stated
that
odors
can
cause
neurological
problems
such
as
fatigue,
dizziness,
headache,
and
nausea
resulting
in
a
diminished
quality
of
life.
The
commenter
also
stated
that
odor
thresholds
for
MEK
have
been
reported
in
the
range
of
6­
250
mg/
m3,
and
the
estimates
presented
in
the
proposed
rule
for
a
1­
hour
maximum
concentration
near
MEK
sources
is
25
mg/
m3,
which
is
within
the
range
of
the
reported
odor
thresholds.
The
commenter
also
suggested
that
EPA
recognize
that
the
risk
to
sensitive
individuals
could
increase
after
delisting.

Response:
While
EPA
does
not
expressly
consider
odor
as
a
health
endpoint,
EPA
considers
the
physiological
effects
of
chemical
exposures,
including
the
neurological
effects
that
the
commenter
described.
In
the
proposed
rule,
EPA
stated
the
following,
@
The
IRIS
assessment
of
MEK
states
that
at
present,
there
is
no
convincing
experimental
evidence
that
MEK
is
neurotoxic
.
.
.
other
than
possibly
inducing
CNS
(
central
nervous
system)
depression
at
high
exposure
levels.@
The
IRIS
documentation
shows
that
no
peripheral
neurohistopathological
changes
were
reported
in
rats
exposed
continuously
to
3,320
mg/
m3
of
MEK
for
up
to
5
months
(
Saida
et
al.,
1976).
No
treatment­
related
central
or
peripheral
neurohistopathology
was
observed
in
rats
exposed
for
90
days
(
6
hours/
day,
5
days/
week)
at
concentrations
of
MEK
as
high
as
14,865
mg/
m3,
even
among
animals
in
animal
tissues
specifically
prepared
and
examined
for
neurohistopathology
(
Cavender
et
al.,
1983).
Also,
ten
of
ten
rats
exposed
to
MEK
at
17,700
19
mg/
m3
and
higher
for
8
hours/
day,
7
days/
week,
died
in
the
seventh
week
of
exposure
without
neurological
symptoms
or
histopathology@
(
Altenkirch
et
al.,
1978).

Regarding
sensitive
individuals,
EPA
could
not
identify
any
specific
data
that
address
the
potential
differences
in
susceptibility
to
adverse
effects
from
MEK
exposure.
In
the
MEK
Toxicological
Review
in
support
of
the
IRIS
assessment,
EPA
did
note
that
AThe
potential
exists
for
increased
susceptibility
to
neurotoxicity,
hepatotoxicity,
and
renal
toxicity
following
exposure
to
MEK
in
combination
with
certain
other
solvents
.
.
..@
The
potentiating
effects
of
MEK
on
the
toxicity
of
other
solvents
have
only
been
demonstrated
at
relatively
high
exposure
concentrations
(
200
­
1,000
ppm
or
590
­
2950
mg/
m3).

Comment:
One
commenter
(
65)
recommended
changing
the
hazardous
waste
regulations
that
apply
to
MEK
as
follows:
remove
MEK
as
a
listed
toxicity
characteristic
in
40
CFR
261.64;
remove
MEK
as
a
toxic
constituent
in
part
261,
appendix
VIII;
and
remove
MEK
from
the
F005
listing,
but
it
may
be
appropriate
to
add
it
to
F2003
listing.

Response:
EPA
was
petitioned
under
CAA
section
112(
b)(
3)
to
remove
MEK
from
the
CAA
section
112
HAP
list.
This
is
the
only
action
under
consideration
as
part
of
the
final
rule.

6.0
References
Altenkirch,
H.,
G.
Stoltenburg,
and
H.
M.
Wagner.
(
1978)
Experimental
studies
on
hydrocarbon
neuropathies
induced
by
methyl­
ethyl­
ketone
(
MEK).
J
Neurol
219:
159­
70.

Cavender,
F.
L.,
H.
W.
Casey,
H.
Salem,
et
al.
(
1983)
A
90­
day
vapor
inhalation
toxicity
study
of
methyl
ethyl
ketone.
Fundam
Appl
Toxicol
3(
4):
264­
70.

Cox,
G.
E.,
D.
E.
Bailey,
and
K.
Morgareidge.
(
1975)
Toxicity
studies
in
rats
with
2­
butanol
including
growth,
reproduction
and
teratologic
observations.
Food
and
Drug
Research
Laboratories,
Inc.
Waverly,
NY.
Report
No.
91MR
R
1673.

Deacon,
MM,
Pilny,
MD,
John,
JA,
et
al.
(
1984)
Embryo­
and
Fetotoxicity
of
Inhaled
Methyl
Ethyl
Ketone
in
Rats.
Toxicol
Appl
Pharmacol
59:
620­
22.

Dick,
R.
B.,
E.
F.
Krieg
Jr.,
J.
Setzer,
B.
Taylor.
(
1992)
Neurobehavioral
effects
from
acute
exposures
to
methyl
isobutyl
ketone
and
methyl
ethyl
ketone.
Fundam
Appl
Toxicol
19(
3):
453­
73.

Lowengart,
R.
A.,
J.
M.
Peters,
C.
Cicioni,
et
al.
(
1987)
Childhood
leukemia
and
parents'
occupational
and
home
exposures.
J
Natl
Canc
Inst
79(
1):
39­
46.

Noraberg
J,
Arlien­
Søborg
P.
2000.
Neurotoxicology
21(
3):
409­
18.
20
OECD.
(
1997)
SIDS
Screening
Information
Assessment
Report
for
Methyl
Ethyl
Ketone.
(
SIAM
6
1997)

Saida,
K.,
J.
R.
Mendell,
and
H.
S.
Weiss.
(
1976)
Peripheral
nerve
changes
induced
by
methyl
n­
butyl
ketone
and
potentiation
by
methyl
ethyl
ketone.
J
Neuropath
Exp
Neurol
35(
3):
205­
25.

Schwetz,
B.
A.,
Leong,
BKJ,
Gehring,
PJ
(
1974)
Embryo­
and
Fetotoxicity
of
Inhaled
Carbon
Tetrachloride,
1,1­
dicloroethane
and
Methyl
Ethyl
Ketone
in
Rats.
Toxicol
Appl
Pharmacol
28:
452­
64.

Schwetz,
B.
A.,
T.
J.
Mast,
R.
J.
Weigel,
et
al.
(
1991)
Developmental
toxicity
of
inhaled
methyl
ethyl
ketone
in
mice.
Fundam
Appl
Toxicol
16:
742­
8.

U.
S.
EPA.
(
1992)
Integrated
Risk
and
Information
System.
Online
at:
http://
www.
epa.
gov/
iris/
subst/
0071.
htm
U.
S.
EPA.
(
1994)
Methods
for
Derivation
of
Inhalation
reference
Concentrations
and
Application
of
Inhalation
Dosimetry.
EPA/
600/
8­
90/
066F.
Online
at
http://
cfpub2.
epa.
gov/
ncea/
raf/
recordisplay.
cfm?
deid=
71993
U.
S.
EPA.
(
1999)
Guidelines
for
Carcinogen
Risk
Assessment
[
Review
Draft].
Risk
Assessment
Forum.
NCEA­
F­
0644
http://
www.
epa.
gov/
ncea/
raf/
cancer/
htm
U.
S.
EPA.
(
2000)
Benchmark
Dose
Technical
Guidance
Document
[
External
Review
Draft].
Online
at:
http://
cfpub2.
epa.
gov/
ncea/
raf/
recordisplay.
cfm?
deid=
20871
U.
S.
EPA.
(
2002)
A
Review
of
the
Reference
Dose
and
Reference
Concentration
Processes.
EPA/
630/
P­
02/
002F.
Online
at:
http://
cfpub2.
epa.
gov/
ncea/
raf/
recordisplay.
cfm?
deid=
55365
U.
S.
EPA
(
2003)
Toxicological
Review
of
Methyl
Ethyl
Ketone.
EPA
635/
R­
03/
009
Online
at:
http://
www.
epa.
gov/
iris
Woo,
Y­
T.,
D.
Lai,
J.
L.
McLain,
et
al.
(
2002)
Use
of
mechanism­
based
structure­
activity
relationships
analysis
in
carcinogenic
potential
ranking
for
drinking
water
disinfection
byproducts
Environ
Health
Persp
110
(
Suppl.
1):
75­
8.

Zimmermann,
FK,
Mayer,
VM,
Scheel,
I,
et
al.
(
1985)
Acetone,
methyl
ethyl
ketone,
ethyl
acetate,
acetonitrile
and
other
polar
aprotic
solvents
are
strong
inducers
of
aneuploidy
in
Saccharomyces
cerevisiae.
Mutat
Res
149(
3):
339­
351.
21
7.0
Commenters
Commenters
on
the
MEK
Delisting
Proposal
Document
Number
I
Author
Company/
Affiliation
0057
Anonymous
Unspecified
0058
Pat
Foust
IPS
Corporation
0059
Anonymous
Quality/
Environmental
Systems
Assistant
from
Minnesota
Unspecified
sheet
metal
fabrication,
welding,
and
painting
company
0060
J.
D.
Robinson,
Director
National
Environmental
Health
&
Safety
Council
(
NEHSC)

0061
Richard
Winn,
MS
CSP
Safety
Consulting
Services
0062
Sandra
Berg,
President
CEO
Ellis
Paint
Company
0063
Scott
Davidson,
President
Johnson
Laminating
&
Coating,
Inc.

0064
Donald
Kugelberg,
Technical
Director
Johnson
Laminating
&
Coating,
Inc.

0065
Benjamin
Garth
Studebaker,
CSP
Videojet
Technologies,
Inc.

0068
Don
Dickens,
Manager
Halton
Chemical,
Inc.

0069
Edward
W.
Kifer,
Director
of
R&
D
Ranbar
Electrical
Materials
0070
W.
Pat
Foust,
Chief
Operating
IPS
Weldon
22
Document
Number
I
Author
Company/
Affiliation
Officer
0071
Harold
Henderson,
President
Trail
Chemical
Corporation
0072
Marc
Germain,
Eng.
Technical
Service
Supervisor
Brenntag
Canada
Inc.

0073
Jeffrey
Julis,
Technical
Director
EMCO
Chemical
Distributors,
Inc.

0074
Thuc
Nguyen,
Technical
Services
Manager
Univar
Canada
Ltd.

0075
Naresh
Patel,
President
IPS
Corporation
0076
Joseph
Lucas,
President
Inland
Technology,
Incorporated
0077
Richard
Hatfield,
Director
of
Environmental
Affairs
The
Haartz
Corporation
0079
Line
Cormier,
Environmental
Coordinator
Peinture
Can­
Lak,
Inc.

0080
Ram
Singhal,
Director,
Regulatory
&
Government
Relations
Flexible
Packaging
Associations
0081
Deborah
Chapin,
Kodak
Park
Environmental
Services
Eastman
Kodak
Company
0082
Rich
Raiders,
Manager,
Clean
Air
Act
Programs
ATOFINA
Chemicals,
Inc.

0083
Robert
Colby,
Chair
ALAPCO
Air
Toxics
Committee
&
Lloyd
Eagan,
Chair
STAPPA/
ALAPCO
23
Document
Number
I
Author
Company/
Affiliation
STAPPA
Air
Toxics
Committee
0084
Hexcel
Corporation
Hexcel
Corporation
0085
DuPont
Performance
Coatings
DuPont
Performance
Coatings
0086
Jeremy
Tovey,
CHMM,
HSE
Manager
Akzo
Nobel
Coatings,
Inc.

0087
Catherine
Witherspoon,
Executive
Officer
&
Joan
Denton,
Ph.
D.,
Director
California
Environmental
Protection
Agency,
State
of
California
0088
Dwayne
Meyer,
Director
Technical
Resources
Carboline
Company
0089
Laura
Schubert,
EH&
S
Technician
US
Paint
Corporation
0090
Marcia
Kinter,
VP
Government
Affairs
The
Graphic
Arts
Coalition
0091
P.
Michael
Peck,
Director
of
Solvent
Technology
Dynamold
Solvents,
Inc.

0092
Desi
Chari,
Senior
Environmental
Manager
Rohm
and
Haas
Company
0093
Geoffrey
Cullen,
Director
of
Government
Relations
The
Can
Manufacturers
Institute
(
CMI)

0094
John
Festa,
Ph.
D.,
Senior
Scientist
American
Forests
&
Paper
Association
24
Document
Number
I
Author
Company/
Affiliation
0096
Identical
to
120
0097
Suzie
Brindle,
Program
Organizer
Clean
Water
Action
0098
James
Witkowski,
Air
Subcommittee
Chair
South
Carolina
Chamber
of
Commerce,
Environmental
Technical
Committee
0099
Vicki
Patton,
Senior
Attorney
Environmental
Defense
100
Frank
Black
Cloud
Spirit
Lake
Tribe
101
Larry
Metzman
Quaker
City
Chemicals,
Inc.

102
Henry
Weltman
Consultant,
Retired
Senior
Engineering
Specialist
from
General
Dynamics/
Fort
Worth
Division
103
Roy
Cobb,
Jr.,
Senior
Environmental
Counsel
Smurfit­
Stone
Container
Corporation
104
Northeast
States
for
Coordinated
Air
Use
Management
(
NESCAUM)
NESCAUM
105
Identical
to
104
106
Michael
Cudahy,
Technical
Manager
FRY
Technology
107
Eric
Clark,
Government
Relations
Synthetic
Organic
Chemical
Manufacturers
Association
(
SOCMA)

108
Courtney
Price,
VP,
Chemstar,
The
The
American
Chemistry
25
Document
Number
I
Author
Company/
Affiliation
American
Chemistry
Council
Council
Ketones
Panel
(
Petitioner)

109
The
American
Chemistry
Council
Ketones
Panel
The
American
Chemistry
Council
Ketones
Panel
(
Petitioner)

110
The
American
Chemistry
Council
Ketones
Panel
The
American
Chemistry
Council
Ketones
Panel
(
Petitioner)

111
Identical
to
93
112
R.
Scott
Thomas,
Corporate
Director,
Environmental
Affairs
The
Sherwin­
Williams
Company
113
Valerie
Ughetta,
Director
of
Stationary
Sources
Alliance
of
Automobile
Manufacturers
(
The
Alliance)

114
Allen
Stegman,
Corporate
Manger
of
Environmental
Affairs
The
Valspar
Corporation
115
Fred
Scholer,
Director
of
R&
D
Hercules
Chemical
Company
Inc.

116
Identical
to
112
117
Identical
to
99
118
Identical
to
97
119
Phillip
Pesola,
Executive
VP
&
Hordan
Dern,
Delta
Laboratories,
Inc.
26
Document
Number
I
Author
Company/
Affiliation
Engineering
Manager
120
Jennifer
Sass,
Senior
Scientist
&
Jon
Devine,
Jr.,
Senior
Attorney
Natural
Resources
Defense
Council
121
Linda
Greer
&
John
Devine
Natural
Resources
Defense
Council
122
Identical
to
103
124
Alison
Keane
&
David
Darling
National
Paint
&
Coatings
Association
(
NPCA)

126
Daniel
Peterson,
Jr.,
VP
Operations
Koala
Corporation
127
Larry
Brandenburger,
VP
R&
D
Valspar
Corporation
128
Joann
Held,
Chief
Bureau
of
Air
Quality
Evaluation
NJ
Dept
of
Environmental
Protection
129
Samuel
Wolfe,
Assistant
Commissioner
Environmental
Regulation,
State
of
NJ
130
Technology
Transfer
Network
National
Air
Toxics
Assessment
Technology
Transfer
Network
National
Air
Toxics
Assessment
131
Andre
Hozer,
President
DURO­
LAK
INC
I
Document
number
in
rulemaking
docket
OAR­
2003­
0028
