RESPONSE
TO
SIGNIFICANT
PUBLIC
COMMENTS
ON
THE
PROPOSED
CLEAN
AIR
MERCURY
RULE
Received
in
response
to:

Proposed
National
Emission
Standards
for
Hazardous
Air
Pollutants;
and,
in
the
Alternative,
Proposed
Standards
of
Performance
for
New
and
Existing
Stationary
Sources:
Electric
Utility
Steam
Generating
Units
(
69
FR
4652;
January
30,
2004)

Supplemental
Notice
for
the
Proposed
National
Emission
Standards
for
Hazardous
Air
Pollutants;
and,
in
the
Alternative,
Proposed
Standards
of
Performance
for
New
and
Existing
Stationary
Sources:
Electric
Utility
Steam
Generating
Units
(
69
FR
12398;
March
16,
2004)

Proposed
National
Emission
Standards
for
Hazardous
Air
Pollutants;
and,
in
the
Alternative,
Proposed
Standards
of
Performance
for
New
and
Existing
Stationary
Sources,
Electric
Utility
Steam
Generating
Units:
Notice
of
Data
Availability
(
69
FR
69864;
December
1,
2004)

Docket
Number
OAR­
2002­
0056
3.0
PERFORMANCE
STANDARDS
FOR
COAL­
FIRED
ELECTRIC
UTILITY
STEAM
GENERATING
UNITS
US
Environmental
Protection
Agency
Emissions
Standards
Division
Office
of
Air
Quality
Planning
and
Standards
Research
Triangle
Park,
North
Carolina
27711
15
March
2005
i
General
Outline
1.0
INTRODUCTION
AND
BACKGROUND
2.0
APPLICABILITY
AND
SUBCATEGORIZATION
3.0
PERFORMANCE
STANDARDS
FOR
COAL­
FIRED
ELECTRIC
UTILITY
STEAM
GENERATING
UNITS
4.0
PERFORMANCE
STANDARDS
FOR
OIL­
FIRED
ELECTRIC
UTILITY
STEAM
GENERATING
UNITS
5.0
MERCURY
CAP­
AND­
TRADE
PROGRAM
6.0
MERCURY
EMISSIONS
MONITORING
7.0
IMPACT
ESTIMATES
8.0
COMPLIANCE
WITH
EXECUTIVE
ORDERS
AND
STATUTES
9.0
NODA
10.0
OTHER
Appendix
A
LIST
OF
COMMENTERS
3­
1
3.0
PERFORMANCE
STANDARDS
FOR
COAL­
FIRED
ELECTRIC
UTILITY
STEAM
GENERATING
UNITS
3.1
MERCURY
CONTROL
TECHNOLOGIES
Comment:

According
to
one
commenter
(
OAR­
2002­
0056­
2946),
although
there
may
be
no
single
technology
to
meet
the
needs
of
all
plants,
a
wide
set
of
solutions
are
available
for
each
subcategory.
The
commenter
listed
10
control
measures
ranging
from
coal
washing
to
switching
to
renewable
resources.

Response:

EPA
concurs
that
there
are
a
number
of
technologies
that
may
be
used
by
utility
units
to
reduce
Hg
emissions.

3.1.1
Availability
of
Mercury­
Specific
Control
Technologies
3.1.1.1
Current
Commercial
Availability
Comment:

Several
commenters
(
OAR­
2002­
0056­
1826,
­
2020,
­
2160,
­
2578,
­
2929,
­
2948,
­
3445,
­
3463,
­
3478,
­
3537)
stated
that
there
are
no
commercially
available
control
technologies
specifically
designed
for
Hg
emission
control
from
coal­
fired
power
plants.
Commenters
acknowledged
that
existing
control
technologies
used
to
control
PM,
SO2,
or
NOx
emissions
reduce
Hg
emissions
under
selected
conditions
and
significant
research
is
being
conducting
by
the
public
and
private
sectors
on
new
Hg
control
technologies.
Although
significant
research
is
underway
by
the
private
and
public
sectors,
before
commercial
availability
is
achieved,
additional
development
is
need
to
provide
for
new
technologies
that
account
for
variability
in
coal
content,
combustion
processes,
and
control
system
performance
under
different
kinds
of
firing
conditions.

One
commenter
(
OAR­
2002­
0056­
1814)
stated
that
Hg
is
a
naturally­
occurring
chemical
that
is
emitted
in
trace
amounts
when
coal
is
combusted.
The
low
concentrations
of
Hg
in
coal
make
capture
of
Hg
from
power
plants
very
difficult
and
subject
to
a
great
deal
of
variability.
Technologies
are
not
currently
available
that
are
specifically
designed
for
control
of
Hg
at
the
low
concentrations
emitted
by
power
plants
One
commenter
(
OAR­
2002­
0056­
2861)
stated
that
technology
is
not
ready
to
support
a
regulatory
program.
The
low
concentration
of
Hg
that
occurs
naturally
in
coal
makes
the
capture
of
Hg
from
the
flue
gas
of
coal­
fired
power
plants
very
difficult
and
subject
to
a
great
deal
of
3­
2
uncertainty
and
variability.
According
to
the
commenter,
there
currently
are
no
commercially
available
technologies
that
are
specifically
designed
to
control
the
very
low
concentrations
of
Hg
emitted
by
coal­
fired
power
plants.
The
commenter
stated
that
although
some
of
the
technologies
being
investigated
have
shown
some
promise,
there
are
still
many
unanswered
questions
regarding
the
level
of
reduction
that
can
reliably
be
achieved,
the
variables
that
will
affect
performance,
and
the
impacts
on
overall
plant
operation
and
maintenance.

One
commenter
(
OAR­
2002­
0056­
2899)
stated
that
for
a
technology
to
be
deemed
commercially
available,
it
must
be
able
to
control
Hg
emissions
from
power
plants
burning
different
coal
ranks
and
having
different
boiler
types
and
configurations;
a
few
isolated
tests
or
demonstrations
are
not
sufficient
to
conclude
that
a
technology
is
commercially
available.
A
technology
needs
to
be
installed
in
full­
scale
applications
at
a
number
of
sites
and
operated
over
extended
periods
of
time
before
it
can
be
viewed
as
commercially
available,
and
a
technology
is
not
commercially
available
just
because
a
vendor
is
willing
to
sell
it.
The
commenter
points
out
that
commercial
availability
requires
that
most
of
the
key
engineering
questions
about
the
technology
need
to
have
been
previously
resolved.
The
commenter
added
that
a
technology
is
not
commercially
available
if
one
installs
it
knowing
that
many
problems
will
need
to
be
resolved
as
part
of
the
installation
and
operation.

Response:

EPA
concurs
with
the
commenters
and
believes
that
Hg­
specific
control
technologies
are
not
yet
commercially
available.

Comment:

One
commenter
(
OAR­
2002­
0056­
3469)
stated
that
lack
of
control
and
monitoring
technology
impedes
speedy
compliance.
The
issues
relating
to
the
state
of
science
on
Hg
are
compounded
by
a
lack
of
technology
to
reliably
measure
or
control
Hg
emissions,
particularly
from
lignite­
fired
units,
to
justify
the
level
of
emission
reductions
proposed
by
EPA.
As
stated
by
the
Department
of
Energy
(
DOE),
"
Today,
there
is
no
commercially
available
technology
that
can
consistently
and
cost­
effectively
capture
Hg
from
coal­
based
power
plants."

Response:

EPA
concurs
that
Hg­
specific
control
technologies
are
not
currently
available.
However,
we
disagree
with
respect
to
Hg
monitoring
technologies
and
believe
that
such
systems
will
be
available
by
the
time
compliance
with
the
regulation
is
required.

Comment:

One
commenter
(
OAR­
2002­
0056­
3454)
stated
that
the
rapid
development
of
Hg
control
technologies
over
the
last
several
years
has
produced
a
number
of
technologies
that
are
available
for
the
implementation
of
a
national
Hg
control
regulation
for
coal­
and
oil­
fired
power
plants.
A
large
number
of
laboratory
tests
and
full­
scale
demonstrations
have
been
conducted
that
provide
3­
3
information
on
the
effectiveness
of
controls
for
various
coal
ranks
and
control
configurations.
Despite
the
current
lack
of
a
national
control
requirement
for
Hg,
a
number
of
options
are
commercially
available
while
others
are
still
in
the
development
and
testing
phases.

One
commenter
(
OAR­
2002­
0056­
3210)
stated
that
the
EPA
understates
the
availability
of
Hg
control
technology
because
it
failed
to
acknowledge
the
DOE/
National
Energy
Technology
Laboratory
Mercury
Control
Technology
Research
Program
on
coal­
fired
power
plants.

Response:

EPA
disagrees
with
the
commenters
about
the
availability
of
Hg­
specific
control
technologies
at
the
present
time.
EPA
is
fully
aware
of
the
DOE
research
program
cited
by
the
commenter.
The
limited,
but
increasing,
number
of
tests
have
not
yet
brought
the
technologies
to
the
level
of
demonstration
that
we
feel
necessary
to
be
considered
"
commercially
available"
and
the
basis
for
a
national
standard.

Comment:

One
commenter
(
OAR­
2002­
0056­
2247)
stated
that
sorbent
injection
technologies
should
be
considered
available
for
Hg.
Permits
have
been
issued
that
will
rely
on
sorbent
injection
technologies
such
as
ACI
(
MidAmerican
Energy,
Council
Bluffs
Unit
4,
PSD
permit
issued
by
Iowa;
and
Wisconsin
Public
Service
Corporation,
Weston
Unit
4,
issued
by
Wisconsin).
These
show
that
Hg
removal
technologies
capable
of
achieving
more
than
80
percent
control
are
available.

Many
commenters
stated
that
EPA
failed
to
consider
Hg
control
technologies
and
methods
that
are
currently
available
and
cost
effective.
EPA
must
consider
the
costs
and
environmental
effects
of
these
technologies,
such
as
ACI
and
other
sorbent
injection
systems,
coal
washing,
and
selective
catalytic
reduction
(
SCR).
New
units
can
design
these
into
their
control
systems
without
retrofit
problems.
EPA
should
also
consider
technologies
required
in
consent
decrees,
case­
by­
case
MACT
and
BACT
analyses,
State
regulations,
and
permit
data.

Response:

As
noted
earlier,
EPA
does
not
believe
that
Hg­
specific
control
technologies,
including
ACI,
are
commercially
available
for
nationwide
application
to
the
coal­
fired
utility
industry.
Installation
of
such
technologies
on
a
limited
number
of
units
(
e.
g.,
the
two
cited)
is
possible
and
will
serve
to
advance
the
technologies
such
that
they
are
widely
for
use
in
compliance
with
the
phase
II
cap.

3.1.1.2
Mercury
Control
Technology
Development
Time
Comment:

Many
commenters
(
OAR­
2002­
0056­
1471,
­
1608,
­
1636,
­
1667,
­
1773,
­
1777,
­
1791,
3­
4
­
1806,
­
1817,
­
1987,
­
2064,
­
2233,
­
2887,
­
2946,
­
3454,
­
3538)
disagreed
with
the
EPA's
conclusion
that
Hg­
specific
controls
for
electric
utility
power
plants
will
not
be
commercially
available
on
a
wide
scale
until
2010
or
later.
Other
commenters
agreed
with
EPA's
time
estimate
on
the
availability
of
Hg­
specific
controls
(
OAR­
2002­
0056­
1969,
­
3537,
­
3565).
Arguments
stated
by
various
commenters
disagreeing
with
EPA's
assessment
included
the
following.
Mercury
control
technologies
are
available
now.
The
EPA
disregarded
studies
on
emerging
Hg
control
technologies.
The
EPA's
own
numbers
and
other
studies
indicate
that
coal­
fired
plants
can
achieve
90
percent
reduction
regardless
of
the
type
of
plant
or
coal.
Field
testing
of
ACI
has
shown
90
percent
capture
of
Hg.
Units
equipped
with
scrubbers
and
fabric
filters
can
obtain
near
90
percent.
Studies
indicate
that
the
cost
of
these
controls
would
be
comparable
to
those
for
other
pollutants
and
EPA
disregards
these
studies
and
emerging
state
of
the
art
Hg
control
technologies.
The
EPA
did
not
provide
a
detailed
analysis
of
the
current
available
technologies.
Outside
of
the
U.
S.,
the
Berrenrath
275
MWe
and
the
Wachtberg
166
MWe
plants
in
Germany
operate
on
carbon
injection
technology
to
control
Hg.
What
is
contradictory
in
EPA's
analysis
is
that
they
used
ACI
in
their
cost
modeling
exercises
with
the
integrated
planning
model
(
IPM)
but
failed
to
recognize
this
technology
in
setting
the
level
of
Hg
reductions
for
the
emission
limits.

Several
commenters
(
OAR­
2002­
0056­
2873,
­
3449),
although
agreeing
that
ACI
technology
currently
is
not
commercially
available,
stated
that
this
technology
will
be
available
before
2010.
One
commenter
(
OAR­
2002­
0056­
3449)
stated
that
ACI
can
be
developed
and
widely
implemented
within
the
next
6
years.
A
second
commenter
stated
that
ACI
can
be
developed
and
widely
implemented
by
2008
to
2009.

Response:

EPA
disagrees
with
the
commenter's
assessment
regarding
the
time
that
it
will
take
for
ACI,
or
other
Hg­
specific
control
technologies,
to
become
commercially
available.
We
do
not
believe
that
these
technologies
are
available
now
for
wide­
spread
usage.
We
have
been
following
the
studies
of
such
technologies
closely
and
have
discussed
their
degree
of
development
with
vendors,
the
industry,
and
the
DOE.
No
utility
unit
has
operated
a
Hg­
specific
control
technology
full­
scale
for
longer
than
a
month
or
so.
Further,
the
technologies
have
not
been
fully
evaluated
on
all
coal
ranks
(
e.
g.,
Gulf
Coast
lignite),
even
under
short­
term
conditions.
In
addition,
other
aspects
of
the
use
of
Hg­
specific
control
technologies
(
e.
g.,
balance
of
plant,
waste
issues,
other
atmospheric
concerns)
have
not
been
fully
addressed.
Studies
continue
to
(
1)
evaluate
the
impact
on
the
coal­
fired
facility
as
a
whole
of
both
ACI
and
enhanced
ACI
(
e.
g.,
corrosion);
(
2)
assess
the
impact
on
the
fly
ash
of
the
ACI
or
enhanced
ACI
with
regard
to
its
reuse
and
disposal;
and
(
3)
study
the
other
atmospheric
emissions
that
may
result
from
use
of
ACI
or
enhanced
ACI
(
e.
g.,
brominated
dioxins
emitted
either
directly
or
formed
following
emission
to
the
atmosphere).
Based
on
these
tests,
on­
going
studies,
and
discussions,
we
do
not
believe
that
the
technologies
have
consistently
demonstrated
an
ability
to
reduce
Hg
emissions
by
90
percent
(
or
any
other
level)
for
an
extended
period
of
time
on
all
coal
ranks
and
all
boiler
types.
Use
of
sorbent
injection
technologies
for
Hg
removal
on
European
facilities
is
informative
but
does
not
serve
to
prove
the
technologies
on
U.
S.
facilities.
We
believe
that
the
cap­
and­
trade
approach
selected
for
the
final
regulation
is
the
best
method
for
encouraging
the
continued
development
of
these
technologies.
Use
of
sorbent
injection
in
the
3­
5
IPM
model
served
to
estimate
the
impact
of
these
Hg­
specific
control
technologies
in
the
outyears
of
the
cap­
and­
trade
program
and
was
based
on
EPA's
projections
that
such
technologies
would
be
available
after
2010.

Comment:

One
commenter
(
OAR­
2002­
0056­
2929)
stated
that
the
reliable,
cost­
effective
control
technologies
designed
specifically
for
capturing
Hg
have
not
yet
been
fully
developed
or
tested.
EPRI,
DOE,
and
EPA
have
conducted
extensive
research
and
development
(
R&
D)
programs
over
the
past
decade
with
the
objective
of
developing
cost­
effective
methods
for
reducing
power
plant
Hg
emissions.
Mercury
control
technology
capable
of
achieving
high
removal
rates
(
i.
e.,
greater
than
80
percent)
across
the
entire
industry
is
not
available.
Full­
scale
demonstrations
of
Hg
control
technologies
at
individual
power
plants
are
just
getting
underway.
It
will
take
at
least
2
or
3
years
to
complete
these
initial
demonstrations
and
evaluate
the
potential
effectiveness
of
possible
new
control
technologies.
And
then,
several
more
years
will
be
needed
before
these
technologies
can
be
considered
"
commercially
available."

One
commenter
(
OAR­
2002­
0056­
2160)
stated
that
programs
for
testing
new
technologies
such
as
ACI
have
been
conducted
for
only
short
run
times
as
opposed
to
the
long
running
times
needed
to
validate
a
technology
for
deployment
in
a
power
plant.

Response:

EPA
concurs
with
this
assessment
of
the
level
of
demonstration
of
Hg­
specific
control
technologies.

3.1.2
Mercury
Control
Technology
Transfer
from
Other
Industrial
Sectors
Comment:

One
commenter
(
OAR­
2002­
0056­
3454)
stated
that
the
air
pollution
control
industry
already
has
considerable
experience
with
the
implementation
of
Hg
controls
for
other
industrial
sectors.
Sorbent
injection
has
been
commercially
proven
to
augment
the
removal
of
Hg
in
waste­
to­
energy
plants.
Experience
controlling
Hg
emissions
has
been
gained
in
more
than
60
U.
S.
and
120
international
waste­
to­
energy
plants
which
burn
municipal
or
industrial
waste
or
sewage
sludge.
For
the
past
two
decades,
sorbent
injection
upstream
of
a
fabric
filter
has
been
successfully
used
for
removing
Hg
from
flue
gases
from
these
facilities.
Other
reagents
used
include
activated
carbon,
lignite
coke,
sulfur­
containing
chemicals,
or
combinations
of
these
compounds.
The
Hg
control
experience
gained
from
the
municipal
and
industrial
waste
combustors
demonstrates
that
the
air
pollution
control
industry
has
been
able
to
control
Hg
in
the
past
and
is
able
to
apply
their
expertise
to
the
electric
power
sector.

Response:

EPA
disagrees
that
experience
gained
through
use
of
Hg­
specific
control
technologies
on
3­
6
municipal
waste
combustors
(
MWC)
is
directly
transferrable
to
coal­
fired
utility
units.
As
noted
in
the
proposal
preamble,
this
results
from
differences
in
the
level
of
Hg
emissions
(
e.
g.,
Hg
emissions
from
a
controlled
MWC
unit
are
roughly
the
same
level
as
uncontrolled
Hg
emissions
from
a
coal­
fired
utility
unit)
and
differences
in
the
species
of
Hg
emitted
(
e.
g.,
because
of
the
Cl
content
of
the
waste
stream,
Hg
emissions
from
MWC
units
are
primarily
in
the
oxidized
form).
Mercury­
specific
control
experience
in
the
MWC
industry
was
the
basis
for
initiating
testing
on
coal­
fired
utility
units
but
not
as
the
basis
for
direct
transfer
of
results.

Comment:

Several
commenters
(
OAR­
2002­
0056­
2867,
­
3478)
stated
that
experience
with
application
of
Hg
control
technologies
on
waste
incinerators
cannot
be
applied
to
electric
utility
power
plants
because
of
process
differences
and
differences
in
the
fuel
assays.
Waste
incinerators
operate
at
much
lower
temperatures
which
are
not
as
much
of
a
hindrance
to
the
Hg
removal
process
as
the
higher
temperatures
that
are
typical
of
utility
power
plant
systems.
The
waste
incinerator
fuel
is
also
higher
in
Cl,
a
constituent
that
is
associated
with
higher
fractions
of
the
soluble
and
removable
form
of
Hg.

One
commenter
(
OAR­
2002­
0056­
2850)
stated
that
although
electric
utilities
that
burn
coal
have
measurable
Hg
emissions,
the
concentration
of
Hg
from
utilities
might
typically
run
only
1/
10th
that
of
the
control
limit
established
for
incinerators.
The
commenter
stated
that
this
low
concentration
makes
further
Hg
reductions
in
electric
utility
boiler
flue
gas
difficult
and
complicates
the
transfer
of
control
technologies
established
for
other
industries
such
as
MWC
to
the
utility
sector.

Response:

EPA
concurs
with
the
commenters'
assessment.

Comment:

One
commenter
(
OR­
2002­
0056­
4139)
disagreed
with
the
proposal
preamble
discussion
of
the
differences
between
solid
and
medical
waste
incinerators
and
coal­
fired
utility
units.
The
EPA
stated
that
greater
Hg
reductions
are
achieved
from
the
incinerators
compared
to
utility
units
because
of
waste
separation
techniques.
This
is
false
because
the
Hg
reductions
are
from
inlet
and
outlet
tests,
independent
from
waste
stream
separation.
Also,
EPA's
description
of
Hg
spikes
is
highly
unlikely.
Mercury
reductions
of
80
to
90
percent
are
achieved
even
after
good
waste
separation.

Response:

EPA's
discussion
in
the
proposal
preamble
relating
to
waste
separation
indicated
that
this
is
but
one
of
several
methods
by
which
MWC
units
may
achieve
high
levels
of
Hg
reduction.

3.1.3
Pre­
Combustion
Technologies
3­
7
Comment:

One
commenter
(
OAR­
2002­
0056­
3454)
stated
that
with
the
implementation
of
a
national
program,
multiple
control
options
including
pre­
combustion,
combustion
and
post
combustion
technologies
will
contribute
to
meeting
the
required
emission
reductions.
Coal
cleaning
as
well
as
coal
switching
are
examples
of
options
that
have
the
potential
to
reduce
Hg
emissions
prior
to
fuel
combustion.

Response:

EPA
concurs
with
this
comment.

Comment:

One
commenter
(
OAR­
2002­
0056­
1817)
stated
that
the
EPA
dismisses
switching
to
lower
Hg
coal
and
any
other
pre­
combustion
controls
but
has
no
problem
presenting
DOE's
variability
analysis
that
assumes
all
plants
can
switch
to
higher
Hg
coal.
However,
KFx
Corporation
is
currently
constructing
a
facility
for
pre­
combustion
treatment
of
subbituminous
coal
(
70
to
90
percent
Hg
removal).

Response:

EPA
has
not
"
dismissed"
the
use
of
fuel
switching,
lower­
Hg
coal,
or
any
pre­
combustion
control
technologies
as
compliance
options.
However,
EPA
believes
that
a
regulation
that
requires
a
facility
to
switch
fuels
to
achieve
compliance
results
is
an
"
unachievable"
standard.
We
acknowledge,
that
some
utilities
will
choose
to
switch
fuels
and,
in
fact,
our
IPM
modeling
predicts
some
minimal
amount
of
fuel
switching.
Technologies
such
as
that
developed
by
KFx
Corporation
could
be
used
at
the
discretion
of
the
utility.
With
regard
to
the
DOE
variability
analysis,
EPA
presented
their
analysis
as
being
yet
another
approach
to
handling
variability
and
sought
comment.
EPA
used
its
own
variability
analysis.

Comment:

One
commenter
(
OAR­
2002­
0056­
1952)
stated
that
studies
have
shown
pre­
combustion
beneficiation
of
western
coal,
including
lignite,
can
provide
Hg
reductions
of
up
to
70
percent.
Coal
as­
mined
contains
rock
and
minerals,
and
the
commenter
asserts
that
much
of
the
Hg
in
coal
is
typically
associated
with
these
non­
fuel
impurities.
Removal
of
these
impurities
using
proven,
commercial
coal
cleaning
technology
will
result
in
greater
than
60
percent
Hg
reduction
in
many
western
coals.
The
commenter
has
investigated
coals
from
the
Southwest,
Northern
Great
Plains,
the
Rocky
Mountains,
Gulf
Basin,
PRB,
and
Western
Canada
and
states
that,
with
the
exception
of
the
PRB,
Hg
reductions
were
substantial
using
simple
gravity
separation.
The
commenter
stated
that
coal
cleaning
provided
only
a
25
percent
reduction
of
Hg
in
PRB
coal.
Pre­
combustion
Hg
removal
should
be
investigated
by
EPA
as
a
preferred
technology
for
western
coal;
it
is
economical,
it
is
proven
technology,
and
it
reduces
other
key
pollutants
such
as
ash,
sulfur,
arsenic,
and
NOx.
The
commenter
also
stated
that
regulations
that
don't
encourage
3­
8
economical
pre­
combustion
Hg
reduction
will
actually
increase
the
pollution
from
coal­
fired
plants
in
two
ways.
First,
there
will
be
a
disincentive
to
provide
cleaner­
burning
coal
fuels.
Coal
buyers
will
be
attracted
to
cheaper,
dirtier
fuels.
The
commenter
stated
that
if
post­
combustion
clean­
up
is
the
only
technology
recognized
by
EPA,
power
plants
will
have
higher
emissions
of
pollutants
per
megawatt­
hour
(
MWh)
produced,
than
if
policy
encourages
burning
cleaner
coal.
The
commenter
asserts
that
although
we
may
have
lower
Hg
emissions,
we'll
have
more
solid
waste,
SO2,
NOx,
CO2,
and
arsenic.
Second,
natural
gas
prices
are
unlikely
to
return
to
levels
where
they
can
provide
low­
cost,
low­
emissions
electricity
for
the
U.
S.
market.
The
commenter
states
that
if
we
are
to
reduce
emissions
from
the
production
of
electricity,
we
must
implement
the
most
cost­
effective
technology
available.
The
commenter
notes
that
some
utility
clients
report
that
post­
combustion
Hg
removal
could
add
$
8
to
$
12
per
ton
to
the
cost
of
using
coal.
The
commenter
states
that
those
costs
will
be
directly
absorbed
by
U.
S.
industry,
impacting
American
products
and
services
from
aluminum
to
computer
server
farms.

One
commenter
(
OAR­
2002­
0056­
3478)
stated
that
fuel
processing
technologies
are
being
developed
to
remove
Hg
and
sulfur
from
the
coal
before
it
reaches
the
plant.
Washing
coal
is
one
method
that
has
been
used
on
the
higher
rank
coals
for
some
time.
Processes
are
being
developed
for
low
rank
coals
such
as
lignite
that
have
the
potential
of
reducing,
or
perhaps
eliminating,
the
requirement
for
post
combustion
equipment.
Some
technologies
incorporate
novel
ways
of
physical
screening
while
others
involve
heat
and
pressure
to
drive
off
pollutants.
Additional
work
will
be
required
on
coal
treatment
processes
to
complete
the
economics
of
these
processes.
Also,
in
fuel
processing,
it
may
not
be
practical
to
treat
all
of
the
coal
going
to
a
plant
because
of
the
large
amount
of
tonnage
involved.
The
commenter
believes
the
economics
of
coal
treatment
systems
would
be
greatly
enhanced
if
it
were
possible
to
treat
only
a
fraction
of
the
total
tonnage
consumed
by
a
unit.

Response:

Utility
units
are
free
to
utilize
any
means
available,
including
pre­
combustion
treatments,
to
achieve
compliance
with
the
standards.

3.1.4
Combustion
Technologies
Comment:

One
commenter
(
OAR­
2002­
0056­
2922)
stated
that
the
single
Hg
combustion
technology
that
has
been
investigated
to
control
Hg
has
been
demonstrated
only
on
a
pilot
scale
without
full­
scale
applications.
A
Hg
control
combustion
practice
has
been
investigated
by
GE­
EER
on
a
pilot­
scale
combustor
that
is
several
orders
of
magnitude
smaller
than
a
utility
boiler.
Essentially,
the
technique
achieves
high
loss
on
ignition
(
LOI)
by
combusting
the
fuel
initially
at
low
oxygen
concentrations
to
promote
the
formation
of
carbon
in
the
boiler
and
the
fly
ash.
GE­
EER
primarily
evaluated
the
Hg
removal
potential
for
low­
rank
coals
such
as
PRB
and
lignite.
The
vendor
claims
Hg
removal
rates
of
up
to
40
percent
for
low­
rank
coals,
although
its
own
data
seem
to
indicate
that
only
25
percent
removals
were
actually
achieved.
This
technology
goes
against
the
trend
in
the
utility
industry
whereby
burner
manufacturers
for
years
have
been
trying
to
3­
9
minimize
LOI
to
address
the
concern
of
the
utility
industry
that
high
carbon
levels
make
it
impossible
to
sell
fly
ash
as
an
additive
to
cement.
Although
the
GE­
EER
"
in­
situ"
carbon
formation
concept
for
Hg
removal
results
looks
interesting,
it
is
far
from
being
a
commercial
process.
At
this
stage
of
development
it
is
impossible
to
evaluate
its
true
costs.
For
example,
costs
cannot
be
evaluated
without
knowing
the
extent
to
which
this
technology
would
result
in
lost
income
from
the
inability
to
sell
fly
ash
with
high
LOI
levels
and
increased
disposal
costs
of
up
to
$
30
to
$
40
per
ton
for
fly
ash.
Finally,
this
technology
might
cause
the
radiant
and
convective
boiler
section
tubes
to
be
blanketed
with
carbon,
decreasing
boiler
efficiency
and
increasing
the
cost
of
electric
production.

Response:

EPA
is
not
mandating
use
of
any
technology
to
achieve
compliance
with
the
final
rule.
The
industry
is
free
to
use
any
means,
including
the
one
cited
by
the
commenter,
to
achieve
compliance
with
the
standards.

Comment:

One
commenter
(
OAR­
2002­
0056­
2889)
stated
that
EPA
did
not
adequately
consider
low­
NOx
burners
as
a
Hg
control
technology.
The
EPA
wrongly
characterized
this
system
as
poor
tuning
(
69
FR
12402).
Low­
NOx
burners
result
in
higher
levels
of
unburned
carbon
in
coal
ash,
and
are
a
mature
technology
required
in
the
Northeast
for
years
to
achieve
the
NOx
RACT.
In
Massachusetts,
units
at
the
Salem
Harbor
and
Mt.
Tom
Station
power
plants
are
averaging
83
to
87
percent
Hg
capture
in
coal
using
low­
NOx
burners
and
ESP
units.
The
EPA
should
recognize
the
possible
role
of
low
NOx
burners
in
helping
reduce
Hg
emissions.

One
commenter
(
OAR­
2002­
0056­
3449)
stated
that
the
best
Hg
control
technology
for
existing
coal­
fired
power
plants
is
use
of
fabric
filters
with
low­
NO
x
burners.
Rather
than
injecting
carbon
like
ACI,
the
low­
NO
X
burners
tend
to
generate
carbon
that
is
caught
by
the
bags
and
then
may
absorb
Hg.
Controls
in
use
today
at
power
plants
in
New
Jersey
to
reduce
emissions
of
SO2
and
PM
have
achieved
Hg
reductions
of
90
percent
or
more
(
scrubbers
and
fabric
filters
with
low­
NOx
burners
and
SCR
for
NOx).

Response:

EPA's
description
of
"
poorly
tuned
coal
burners"
in
the
supplemental
notice
did
not
refer
to
properly
installed
and
operated
low­
NOx
burners
as
the
commenter
states.
Rather,
the
discussion
was
directed
at
any
type
of
burner
that
had
not
been
properly
maintained
and
operated.
Low­
NOx
burners
are
in
wide­
spread
use
in
the
coal­
fired
utility
sector
and
could
be
a
part
of
any
utility's
compliance
strategy.
EPA
notes,
however,
that
use
of
low­
NOx
burners
on
low­
rank
coals
is
unlikely
to
result
in
significant
Hg
capture
due
to
the
low
levels
of
chlorine
in
the
coal.

3.1.5
Post­
Combustion
Technologies
3­
10
3.1.5.1
General
Comments
on
Hg
Control
Performance
Comment:

Many
commenters
stated
that
coal
plants
can
achieve
greater
than
90
percent
Hg
control
using
existing
technology
which
is
available
at
many
plants
(
e.
g.,
scrubbers,
fabric
filters,
and
SCR)
or
by
ACI.
ACI
is
commercially
available
today
and
technology
transfer
from
MWC
units
is
clearly
feasible.
Municipal
waste
combustors
with
fabric
filters
and
ACI
have
achieved
99
percent
Hg
control;
DOE
analyses
show
that
retrofitting
a
coal­
fired
boiler
with
ACI
and
fabric
filter
also
can
achieve
90
percent
control
with
low
capital
and
operating
costs.

Many
commenters
also
stated
that
the
emission
reductions
used
by
EPA
are
much
too
low
compared
to
what
is
technically
achievable
and
cost
effective.
Based
on
currently
available
control
technology,
existing
units
should
be
able
to
meet
at
least
80
percent
Hg
efficiency
for
subbituminous
coal
and
a
minimum
of
90
percent
for
bituminous
coal.

Response:

EPA
agrees
that
some
coal­
fired
units
have
exhibited
greater
than
90
percent
Hg
reductions
in
the
limited
test
data
available.
However,
not
all
units
have
been
able
to
achieve
this
level
of
control,
even
with
similar
control
technologies
installed.
As
noted
earlier,
EPA
disagrees
with
the
commenter's
assessment
regarding
the
commercial
availability
of
Hg­
specific
control
technologies
and
on
the
ability
to
transfer
the
technology
from
the
MWC
industry.

Comment:

One
commenter
(
OAR­
2002­
0056­
3210)
stated
that
based
on
the
ICR
III
data,
the
best
reductions
for
Hg
and
sulfur
an
be
achieved
with
wet
scrubbers
and
fabric
filters
or
spray
dryer
absorbers
with
fabric
filters.
Analysis
of
the
data
showed
that
in
the
8
states
surrounding
New
York,
fabric
filters
achieved
the
best
control
of
Hg,
followed
by
an
ESP
with
wet
scrubbers.
Municipal
waste
combustors
in
New
York
using
ACI
with
fabric
filters
achieve
90
percent
Hg
reduction
while
combustors
with
ACI
and
an
ESP
achieve
at
least
85
percent
reduction.

Response:

EPA
is
charged
with
establishing
a
standard
that
is
achievable
nationwide,
not
just
in
one
sector
of
the
nation.
As
noted
elsewhere
in
this
document,
EPA
has
reanalyzed
the
available
data
and
revised
the
new­
source
NSPS
limits
based,
in
part,
on
the
control
technologies
suggested
by
the
commenter.

Comment:

One
commenter
(
OAR­
2002­
0056­
4209)
agrees
that
optimizing
controls
for
NOx
and
SO2
can
reduce
Hg
from
60
to
over
90
percent.
3­
11
Response:

EPA
agrees
that
existing
controls
can
be
optimized
for
Hg
removal
and
believes
that
the
approach
taken
for
the
final
rule
will
provide
the
greatest
incentive
to
induce
early
applications
of
such
optimization.

Comment:

One
commenter
(
OAR­
2002­
0056­
2661)
stated
that
there
are
inherent
problems
in
a
Hg
control
philosophy
based
on
Hg
control
technologies
that
require
converting
elemental
Hg
to
a
form
potentially
more
harmful
to
human
health
for
the
purpose
of
Hg
emission
control
efficiency.
It
doesn't
make
sense
to
require
the
formation
of
a
potentially
more
harmful
species
of
Hg
in
order
to
remove
it
from
the
flue
gas
stream.

Response:

EPA
has
no
information
to
indicate
that
oxidized
Hg
is
any
more
harmful
to
human
health
than
is
the
elemental
form,
particularly
at
the
concentrations
found
in
the
atmosphere
(
i.
e.,
the
levels
found
in
the
atmosphere
are
significantly
lower
than
those
expected
from
a
Hg
spill
in
a
confined
space).
Oxidized
Hg
would
tend
to
deposit
closer
to
the
emission
source
than
elemental
Hg,
but
elemental
Hg
is
ultimately
transformed
to
oxidized
Hg
forms
in
the
atmosphere
and
subsequently
deposited.
Oxidizing
the
elemental
form
enhances
the
ability
of
many
control
technologies
to
remove
significant
levels
of
Hg
from
the
exhaust­
gas
stream.
EPA
believes
that
the
rule
will
reduce
the
risks
from
Hg,
rather
than
increase
them.

Comment:

One
commenter
(
OAR­
2002­
0056­
2422)
stated
that
the
EPA
paper,
"
Control
of
Mercury
Emissions
from
Coal­
Fired
Electric
Utility
Boilers,"
presents
a
narrow
and
misleading
view
of
the
Hg
capture
performance
of
conventional
SO2
and
particulate
control
technologies.
If
the
purpose
of
the
paper
was
to
communicate
what
is
and
is
not
known
about
Hg
control,
the
paper
should
have
discussed
the
limitations
of
the
data
from
which
conclusions
were
drawn,
the
variability
and
uncertainty
of
the
results
in
that
data,
the
performance
that
can
be
expected
over
a
range
of
coal
ranks,
the
confidence
intervals
for
those
estimates,
and
what
EPA
is
doing
to
improve
the
state
of
knowledge
on
the
effectiveness
of
conventional
as
well
as
advanced
control
systems.

Response:

The
referenced
paper
was
written
in
late
2003
and
was
based
on
data
available
at
that
time
on
Hg
capture
performance
of
conventional
SO2
and
particulate
control
technologies.
The
paper
was
intended
to
provide
a
brief
overview
of
the
state
of
Hg
controls
for
Hg
emissions
from
coal­
fired
utility
boilers
and
it
was
not
intended
to
provide
a
detailed
statistical
analysis
of
the
available
data.
Such
analyses
of
Information
Collection
Request
(
ICR)
data
have
been
conducted
by
EPA
and
are
in
the
docket.
The
paper
does
briefly
discuss
results
obtained
from
the
ICR
data.
However,
that
data
represented
a
wide
range
of
combinations
of
boilers,
coal
3­
12
types,
and
air
pollution
control
configurations.
As
mentioned
in
the
referenced
paper,
and
elsewhere,
the
ability
to
capture
Hg
in
the
PM
or
SO2
control
device
is
highly
dependent
upon
the
form
(
elemental,
oxidized,
or
particulate­
bound)
of
the
Hg.
The
form
of
Hg
in
the
flue
gas
is
dependent
upon
the
type
of
coal
being
burned,
the
combustion
conditions,
and
the
installed
air
pollution
control
configuration.
The
paper
does
discuss
the
variability
resulting
from
interactions
of
these
many
combinations.
For
example,
on
page
7,
the
final
paragraph
of
the
referenced
paper
notes
that
the
"
ICR
data
reflected
that
average
Hg
captures
ranged
from
29
percent
for
on
PC­
fired
ESP
plus
flue
gas
desulfurization
(
FGD)
unit
burning
subbituminous
coal
to
98
percent
in
a
PC­
fired
FF
plus
FGD
unit
burning
bituminous
coals."

Comment:

One
commenter
(
OAR­
2002­
0056­
2843)
stated
that
implementation
of
the
proposed
standards
would
require
new
plants
to
comply
with
levels
of
Hg
emissions
that
are
inconsistent
with
available
demonstrated
technology.
The
commenter
stated
that
there
are
no
creditworthy
suppliers
of
Hg
control
technology
in
a
position
to
provide
guarantees
of
performance
consistent
with
the
levels
required
under
the
rulemaking.
Absent
such
technology
guarantees
of
performance,
the
commenter
submits
that
only
a
small
portion
of
the
available
coal
resources
in
the
U.
S.,
particularly
those
in
the
PRB
in
Montana
and
Wyoming,
are
known
to
have
Hg
content
sufficiently
low
as
to
permit
operation
in
conjunction
with
commercially
available
air
pollution
control
device
technologies,
such
as
fabric
filters,
to
meet
the
requirements
of
the
rulemaking.
The
commenter
cites
for
example,
only
about
8
percent
of
PRB
subbituminous
coal
reserves
would
qualify
as
"
compliance
coal"
if
the
"
new
source"
criteria
proposed
by
the
EPA
is
adopted.

Response:

As
noted
later
in
this
document,
EPA
has
reanalyzed
its
new­
source
NSPS
limits.

3.1.5.2
Fabric
Filter
Hg
Control
Performance
Comment:

One
commenter
(
OAR­
2002­
0056­
2359)
stated
that
fabric
filter
technology
exists
today
that
can
reduce
Hg
emissions
by
72
percent
on
average
for
subbituminous
coal
and
up
to
92
percent
for
bituminous
coal.
Activated
carbon
injection
is
very
cost
effective
and
in
the
early
stages
of
full
scale
commercialization.
The
combination
of
ACI
and
fabric
filters
essentially
eliminate
problems
with
carbon
contamination
of
fly
ash
and
would
allow
for
the
beneficial
reuse
of
ash
in
concrete
and
other
products.

Response:

As
noted
above,
EPA
disagrees
on
the
degree
of
commercial
availability
level
of
ACI.
In
addition,
as
noted
later
in
this
document,
EPA
has
reanalyzed
its
new­
source
NSPS
limits.
EPA
agrees
that
the
use
of
a
supplemental
fabric
filter
with
ACI
will
allow
for
the
beneficial
reuse
of
fly
ash.
3­
13
3.1.5.3
ESP
Hg
Control
Performance
Comment:

One
commenter
(
OAR­
2002­
0056­
2259)
stated
that
the
his
company
installed
in
2001
a
pilot­
scale
wet
ESP
at
FirstEnergy's
Penn
Power's
Bruce
Mansfield
Plant
located
in
Shippingport,
PA.
The
ESP
uses
a
slipstream
of
flue
gas
from
the
exhaust
of
the
FGD
system
on
boiler
unit
No.
2,
which
has
a
rated
capacity
of
835
MW
and
burns
3
percent
sulfur
coal.
The
plant
installed
the
pilot
ESP
to
test
for
PM2.5
and
S03
mist
removal
as
a
potential
control
technology
to
reduce
visible
emissions.
Further
Hg
testing
was
performed
during
2003
under
an
award
from
DOE's
National
Energy
Technology
Laboratory.
The
tests
confirm
that
wet
ESP
technology
can
collect
PM2.5
and
sulfuric
acid
(
SO3)
mist
as
well
as
Hg
at
very
high
levels.
Particulate
and
oxidized
Hg
species
were
collected
with
greater
than
70
percent
efficiency
while
elemental
Hg
can
be
partially
oxidized,
in
the
range
of
18
percent
to
44
percent.
Successful
development
of
the
Plasma­
ESP
technology
will
also
allow
for
high
removal
efficiency
of
elemental
Hg
within
the
wet
ESP.
Therefore,
wet
ESP
technology
should
be
given
consideration
as
another
control
technique
that
offers
the
co­
benefits
of
capturing
PM2.5
and
SO3
with
little
pressure
drop
(<
1
inch
water
column),
low
power
consumption
(
1
kW/
MW).
and
no
additional
real
estate
if
mounted
on
top
of
the
FGD
system
or
retrofitted
within
a
dry
ESP.

One
commenter
(
OAR­
2002­
0056­
1842)
stated
that
the
Croll
Reynolds'
Plasma
Enhanced
ESP
technology
(
PEESPTM)
is
to
be
installed
at
Southern
Company's
Miller
plant.
In
this
pilot,
a
5,000
actual
cubic
foot
per
minute
(
acfm)
wet
ESP
will
be
installed
after
a
dry
ESP
to
test
for
PM2.5
and
SO3
and
Hg
removal
under
a
EPRI
funded
contract.
It
will
operate
in
an
unsaturated
flue
gas
environment
and
will
incorporate
the
PEESPTM
technology,
which
at
lab
scale
has
demonstrated
up
to
79
percent
elemental
Hg
control.
Buzz
Reynolds
says
that
successful
demonstration
of
the
Hybrid
dry­
wet
ESP
with
PEESPTM
could
offer
plants
burning
low
sulfur
coals
a
cost­
effective
option
to
that
of
injecting
activated
carbon
followed
by
fabric
filter.
Croll
Reynolds
claims
that
the
wet
ESP
approach
adds
less
than
one­
half
inch
pressure
drop,
requires
no
additional
real
estate
if
retrofitted
into
the
last
field
of
the
dry
ESP,
operates
at
low
power
(
1
kW/
l
MW),
has
no
impact
on
the
dry
ESP
fly
ash,
and
minimizes
the
handling
of
the
waste
by­
product
by
concentrating
the
Hg
in
the
WESP
slurry,
which
is
then
treated
in
a
recycle
system
where
the
Hg
is
precipitated
out
of
the
water.
The
Hg
by­
product
is
in
a
much
more
concentrated,
compact
form
for
easier
disposal
and
handling.

Response:

EPA
is
not
mandating
use
of
any
technology
to
achieve
compliance
with
the
final
rule.
The
industry
is
free
to
use
any
means,
including
the
one
cited
by
the
commenters,
to
achieve
compliance
with
the
standards.

Comment:

One
commenter
(
OAR­
2002­
0056­
2889)
stated
that
a
statement
by
DOE
in
a
Hg
control
technology
R&
D
fact
sheet
wrongly
dismisses
the
high
Hg
capture
efficiency
achieved
at
Brayton
3­
14
Point
as
an
"
unusual
ESP
configuration."
A
more
appropriate
reaction
is
that
an
ESP
can
be
used
with
ACI
to
achieve
high
Hg
removal
rates.
Salem
Harbor's
90
percent
Hg
removal
rate
is
also
portrayed
as
unusual
even
though
the
State
has
other
units
with
similar
particulate­
bound
Hg
fractions.
DOE's
characterization
only
serves
to
promote
as
lenient
a
control
level
as
possible
rather
than
building
on
the
strong
successes
their
funding
helped
document.

Response:

EPA
concurs
that
ESP
units
may
be
used
with
ACI
under
the
proper
conditions
to
effect
Hg
removal.

3.1.5.4
Wet
Scrubber
Hg
Control
Performance
Comment:

One
commenter
(
OAR­
2002­
0056­
3478)
stated
that
enhancing
gas
phase
oxidation
systems
warrant
further
investigation
to
reduce
Hg.
The
term
"
gas
phase
oxidation
systems"
refers
to
the
process
of
improving
the
ability
of
a
scrubber
to
capture
Hg
by
using
a
technology
to
oxidize
the
Hg.
Compounds
that
are
water­
soluble
are
"
scrubbed"
or
removed
from
the
flue
gas
into
the
scrubbing
liquid
and
removed
with
the
scrubber
sludge.
Thus,
an
existing
FGD
system
has
the
ability
to
remove
the
fraction
of
the
Hg
that
is
oxidized.

Response:

Systems
such
as
the
one
the
commenter
describes
are
included
in
the
DOE
program
and,
if
proved
successful,
would
be
available
as
compliance
options
by
industry
should
they
so
choose.

3.1.5.5
Sorbent
Injection
for
Hg
Control
Comment:

Several
commenters
(
OAR­
2002­
0056­
2871,
­
2889)
stated
that
ACI
is
commercially
available
and
widely
recognized
as
a
viable
control
for
Hg.
It
has
been
demonstrated
with
pilot
and
full­
scale
demonstration
projects
on
coal
and
has
been
used
for
over
10
years
on
other
large
combustion
projects.
States
are
now
requiring
it
on
new
coal­
fired
units
for
Hg
control.
The
EPA's
failure
to
consider
this
technology
is
inconsistent
with
its
past
approaches
for
developing
Hg
limit
for
combustion
sources
and
EPA
provides
no
justification
for
the
change.
In
previous
standards,
EPA
has
not
required
technologies
to
be
in
long­
term
us
to
be
considered
"
commercially
available"
and
to
be
evaluated
as
a
potential
control
method.
For
example,
EPA
proposed
NSPS
and
emission
guidelines
for
MWC
units
that
require
ACI
even
though
it
had
been
tested
at
only
two
facilities
(
and
went
beyond
the
floor
because
lower
emissions
were
achievable
at
low
costs).
The
EPA
also
evaluated
ACI
for
hazardous
waste
and
medical
waste
incinerators,
even
though
the
technology
was
rarely
used.
Sorbent
injection
technologies
such
as
ACI
have
been
demonstrated
to
achieve
significant
Hg
reductions
at
coal­
fired
power
plants
regardless
of
3­
15
coal
type;
Hg
control
above
90
percent
is
feasible
at
costs
similar
to
those
for
NOx
removal
(
Mercury
Emissions
from
Coal
Fired
Power
Plants,
NESCAUM,
October
2003).
State
and
local
agencies
are
using
these
studies
to
establish
permit
limits
for
new
boilers.
Wisconsin
is
preparing
to
permit
a
coal­
fired
unit
using
subbituminous
coal
at
83
percent
control
efficiency
for
Hg
(
Wisconsin
Public
Service
Company
Weston
Unit
4).
Iowa
has
issued
a
permit
for
a
facility
using
subbituminous
coal
requiring
1.7
lb
Hg/
TBtu
(
equivalent
to
an
83
percent
control
efficiency
for
operation
with
coal
from
the
source
with
the
highest
average
Hg
content
(
MidAmerican
Energy
Company
Council
Bluffs
Energy
Center).
One
of
these
units
has
commenced
construction
under
that
permit.
Therefore,
the
technology
is
in
commercial
use
and
must
be
considered
in
the
development
of
performance
standards.

One
commenter
(
OAR­
2002­
0056­
3454)
stated
that
Hg
specific
control
technologies
such
as
sorbent
injection
systems
have
been
demonstrated
at
full­
scale.
Multi­
pollutant
control
approaches
as
well
as
other
Hg
specific
technologies
have
also
demonstrated
significant
progress
and
will
provide
additional
low
cost,
innovative
approaches
to
Hg
control.
A
number
of
these
technologies,
including
sorbent
injection
systems
as
well
as
SCR
coupled
with
wet
FGD,
have
achieved
removal
rates
greater
than
90
percent
under
certain
circumstances.

One
commenter
(
OAR­
2002­
0056­
3449)
disagreed
that
there
are
no
commercially
available
control
technologies
specifically
designed
for
reducing
Hg
emissions
as
the
EPA
stated
in
the
rationale
for
the
proposed
subpart
Da
standards
(
p.
4691).
Activated
carbon
injection
is
commercially
available
today
for
Hg
control.
Ten
years
of
experience
with
ACI
on
MWC
incinerators
in
New
Jersey
show
that
technology
transfer
is
feasible.
Some
of
these
incinerators
achieve
99
percent
Hg
control
with
fabric
filters.
The
EPA
is
mistaken
to
discount
ACI
because
it
has
only
been
pilot
tested
or
short
term
demonstration
tested
at
full
scale
units,
and
has
not
been
in
long
term
use
at
any
coal
units.
It
will
be
used
long
term
if
required.
The
NESCAUM
report
on
full­
scale
demonstration
of
ACI
shows
that
90
percent
Hg
removal
is
feasible
with
costs
comparable
to
NOx
removal.
A
recently
issued
Iowa
permit
requires
ACI
from
a
proposed
bituminous
coal
plant.
DOE
pilot
studies
show
up
to
95
percent
control
for
both
bituminous
and
subbituminous
control
with
fabric
filter
and
ACI.
National
Energy
and
Gas
Transmission
Company's
Carneys
Point
and
Logan
Township
boilers
are
each
equipped
with
low­
NOx
burners,
SCR,
dry
scrubber,
and
fabric
filter
which
reduce
Hg
emissions
by
more
than
90
percent.

One
commenter
(
OAR­
2002­
0056­
3205)
stated
sorbent
injection
is
available
for
the
control
of
Hg
emissions.
Activated
carbon
injection
has
been
used
successfully
on
MWC
units
for
the
past
7
to
8
years
and
the
technology
has
been
successfully
demonstrated
in
several
full­
scale
tests,
including
the
recent
year­
long
test
at
Gaston.
Vendors
such
as
ADA­
ES
have
indicated
that
ACI
is
available
now
for
utility
units.
The
commenter
also
refers
to
an
Iowa
permit
requiring
ACI
at
a
new
MidAmerican
Energy
Council
Bluffs
plant.
Xcel
also
proposes
to
use
ACI
at
a
new
unit
at
the
Comanche
plant.
The
commenter
referred
to
the
definition
of
available
technology
in
EPA's
new
source
review
workshop
manual
..."
a
technology
is
considered
available
if
it
can
be
obtained
by
the
applicant
through
commercial
channels
or
is
otherwise
available
within
the
common
sense
meaning
of
the
term."
Activated
carbon
injection
has
clearly
reached
the
commercial
availability
stage
for
utility
units.
3­
16
One
commenter
(
OAR­
2002­
0056­
2819)
stated
that
ACI
is
one
of
several
commercially
available,
cost
effective
technologies
for
coal­
fired
boilers.
Activated
carbon
injection
systems
are
commercially
available
and
have
been
install
on
MWC
units.
Others
include
wet
ESP,
fly
ash
injection
systems,
SCR,
wet
and
dry
FGD
system,
and
fabric
filters.
West
ESP
and
fly
ash
injection
systems
are
already
in
use
on
coal­
fired
boilers
in
the
U.
S.,
Europe,
and
Japan.
This
data
was
presented
to
EPA.
Wet
ESP,
fly
ash
injection
systems,
SCR,
wet
and
dry
FGD
systems,
and
fabric
filters
have
been
commercially
available
and
installed
on
coal
and
oil­
fired
utility
boilers
for
many
years.

Several
commenters
(
OAR­
2002­
0056­
2873,
­
3210)
stated
that
full­
scale
demonstration
projects
have
been
conducted
and
are
on­
going
at
many
U.
S.
coal­
fired
power
plants
to
test
the
effectiveness
of
ACI
with
conventional
PM
controls
for
control
of
Hg
emissions.
According
to
the
commenters,
these
full­
scale
ACI
demonstrations
so
far
have
demonstrated
at
least
50
percent
Hg
removal
and
those
with
pre­
halogenated
sorbents
have
observed
as
much
as
95
percent.
The
E.
C.
Gaston
plant
burning
low
sulfur
bituminous
coal
achieved
90
percent
removal
using
carbon
injection
with
a
hotside
ESP
and
COHPAC
fabric
filter.
The
Brayton
Point
plant
burning
low
sulfur
bituminous
coal
achieved
90
percent
with
carbon
injection
and
a
coldside
ESP.
The
Pleasant
Prairie
plant
burning
subbituminous
coal
achieved
65
percent
using
ACI
with
a
coldside
ESP.
Gaston
showed
that
a
high
removal
rate
using
significantly
less
ACI
can
be
achieved
with
the
COHPAC
system
in
comparison
to
other
conventional
controls.
The
controls
apply
to
bituminous
and
subbituminous
coal.

One
commenter
(
OAR­
2002­
0056­
2575)
stated
that
the
EPA
improperly
rejected
ACI
or
sorbent
injection
systems
as
viable
Hg
control
technologies.
Much
research
shows
that
these
systems
are
highly
effective
(
80
to
90
percent
Hg
removal).
EPA
claims
that
carbon­
based
and
sorbent
injection
control
systems
are
not
currently
available
on
a
commercial
basis.
However,
in
a
separate
discussion
of
certain
carbon­
based
injection
system,
the
EPA
repeatedly
describes
them
as
commercially
available.
The
EPA
also
rejects
injection­
based
systems
because
they
have
not
been
installed
except
on
a
demonstration
basis
and
no
long­
term
data
are
available
to
indicate
performance
on
all
representative
coal
ranks.
EPA's
refusal
is
a
direct
violation
of
the
CAA
goals.
The
legislative
history
clearly
shows
that
Congress
intended
the
statute
to
be
technology­
forcing.
EPA's
agreement
that
it
cannot
force
the
industry
to
implement
specific
controls
until
the
industry
has
fully
implemented
the
same
controls
is
circular
logic
and
destroys
any
incentive
for
industry
to
develop
better
controls.

One
commenter
(
OAR­
2002­
0056­
2199)
stated
that
a
90
percent
Hg
reduction
using
ACI
is
feasible
based
on
a
2002
technical
report
by
the
Massachusetts
Department
of
Environmental
Protection.
Also,
DOE
tests
at
an
Alabama
plant
found
that
ACI
achieved
90
percent
Hg
reduction
at
a
very
low
cost
(
0.05
cents/
KWh).
Preliminary
tests
with
ACI
by
EPRI
achieved
90
percent
with
eastern
coal
ranks
and
60
to
70
percent
with
western
coal
ranks
at
costs
from
0.2­
0.3
cents/
KWh.

Response:

As
noted
earlier,
EPA
disagrees
with
the
commenters
about
the
availability
of
Hg­
3­
17
specific
control
technologies
at
the
present
time.
The
limited,
but
increasing,
number
of
tests
have
not
yet
brought
the
technologies
to
the
level
of
demonstration
that
we
feel
necessary
to
be
considered
"
commercially
available"
and
the
basis
for
a
national
standard
for
this
industry.
We
do
not
believe
that
these
technologies
are
available
now
for
wide­
spread
usage.
We
have
been
following
the
studies
of
such
technologies
closely
and
have
discussed
their
degree
of
development
with
vendors,
the
industry,
and
the
DOE.
No
utility
unit
has
operated
a
Hg­
specific
control
technology
full­
scale
for
longer
than
a
month
or
so.
Based
on
these
tests
and
discussions,
we
do
not
believe
that
the
technologies
have
demonstrated
an
ability
to
reduce
Hg
emissions
by
90
percent
(
or
any
other
level)
for
an
extended
period
of
time
on
all
coal
ranks
and
all
boiler
types.
We
believe
that
the
cap­
and­
trade
approach
selected
for
the
final
regulation
is
the
best
method
for
encouraging
the
continued
development
of
these
technologies.

Comment:

One
commenter
(
OAR­
2002­
0056­
1842)
stated
that
sodium
tetrasulfide
(
Na2S4)
technology
can
remove
elemental
as
well
as
ionic
(
oxidized)
forms
of
Hg.
Other
advantages
include:
the
fact
that
it
results
in
an
inert,
stable
reaction
product
(
cinnabar).
Sodium
tetrasulfide
is
a
liquid,
and,
thus,
is
easier
and
safer
to
handle
and
inject
than
powdered
activated
carbon
and
is
less
abrasive
than
activated
carbon.
Both
full
scale
and
pilot
plant
tests
have
demonstrated
that
the
Na2S4
process
is
both
a
technologically
and
an
economically
effective
approach
to
controlling
Hg
emissions
on
MWC
units.
Pilot
plant
and
short­
term
tests
have
verified
that
the
Na2S4
technology
alone
or
in
combination
with
activated
carbon
technologies
can
achieve
a
controlled
Hg
emission
rate
approaching
the
expected
regulatory
requirements
for
coal­
fired
boilers.
Longer
test
programs
are
planned
to
optimize
the
flue
gas
temperature
regime
and
Na2S4
dose
rate.
Because
the
efficiency
of
the
Na2S4
process
is
influenced
by
mass
transfer
rates,
the
technology
may
be
most
effective
on
facilities
equipped
with
fabric
filters
and
wet
FGD
systems
due
to
the
additional
retention
and
contact
time.

Response:

EPA
is
not
mandating
use
of
any
technology
to
achieve
compliance
with
the
final
rule.
The
industry
is
free
to
use
any
means,
including
the
one
cited
by
the
commenter,
to
achieve
compliance
with
the
standards.

Comment:

Several
commenters
(
OAR­
2002­
0056­
2887,
­
2946)
stated
that
in
the
proposal
notice
(
69
FR
4674)
the
EPA
presented
a
misleading
characterization
of
conclusions
from
the
NESCAUM
October
2003
report
"
Mercury
Emissions
from
Coal
Fired
Power
Plants."
The
NESCAUM
analyses
show
that
commercially
available
control
technologies,
as
well
as
rapidly
emerging
technologies,
are
capable
of
achieving
greater
than
90
percent
Hg
control.
Activated
carbon
injection
has
been
used
on
MWC
units
for
5
to
10
years,
are
routinely
achieving
greater
than
90
percent
Hg
control,
and
has
been
successfully
demonstrated
on
coal­
fired
electric
utility
generating
units
by
DOE.
The
commenters
requested
that
the
EPA
correct
the
preamble
statements
to
reflect
the
actual
conclusions
of
the
report.
3­
18
Response:

As
noted
elsewhere,
EPA
disagrees
with
the
commenter
on
the
availability
and
level
of
Hg
reduction
achievable
by
ACI.
We
apologize
for
any
misleading
characterization
of
the
NESCAUM
report
in
the
proposal
preamble.

Comment:

One
commenter
(
OAR­
2002­
0056­
3478)
stated
that
ACI
is
not
a
one­
size­
fits
all
control
technology;
it
is
highly
dependent
upon
boiler
exhaust
temperatures,
and
works
best
at
temperatures
below
300

F.
Because
the
commenter's
coal­
fired
boilers
experience
exhaust
temperatures
in
excess
of
350

F
to
over
400

F,
the
overall
removal
efficiency
of
vapor
phase
Hg
by
ACI
would
be
significantly
decreased.
The
commenter
states
that
to
achieve
a
desired
removal
rate
with
the
higher
back­
end
temperatures
will
require
significantly
more
activated
carbon
to
be
injected.
The
commenter
adds
that
recent
pilot
scale
testing
indicates
that
ACI
may
not
be
effective
at
all
at
temperatures
of
400

F
or
above.

Response:

EPA
is
aware
of
the
concerns
expressed
by
the
commenter.
It
is
concerns
such
as
these
that
factor
into
the
Agency's
decision
that
ACI
is
not
yet
a
commercially
available
technology
ready
for
universal,
wide­
spread
usage.

Comment:

One
commenter
(
OAR­
2002­
0056­
3478)
stated
that
ACI
is
already
used
in
water
and
wastewater
applications,
and
it
is
not
clear
that
a
new
significant
demand
in
production
for
use
in
ACI
controls
at
coal­
fired
power
plants
could
be
met.
The
commenter
added
that
the
added
demand
will
increase
the
price
of
activated
carbon,
changing
the
cost
effectiveness
of
this
technology.

Response:

EPA
also
has
concerns
about
the
short­
term
availability
of
activated
carbons
suitable
for
wide­
spread
usage
by
the
coal­
fired
utility
industry.

3.1.5.6
Other
Hg
Control
Technologies
Comment:

One
commenter
(
OAR­
2002­
0056­
1842)
listed
a
number
of
Hg
control
technologies
under
various
stages
of
development
that
should
be
considered
by
the
EPA
as
Hg
control
options.
These
included
the
following.
Powerspan
has
a
50
MWe
commercial
demonstration
unit
of
electro­
catalytic
oxidation
(
ECO)
technology
at
the
FirstEnergy
R.
E.
Burger
plant.
The
Mitsui
BF
activated
coke
system
is
used
in
full­
scale
installations
on
combustion
sources
in
Japan.
The
3­
19
Kentucky
Utilities
Ghent
generating
station
is
the
host
of
a
5
MW
slipstream
demonstration
of
the
Airborne
process
being
developed
by
Babcock
&
Wilcox
and
US
Filter
HPD
systems.
The
EPRI
and
Apogee
Scientific
have
been
developing
a
Hg
control
technology
called
MerCAP
(
Mercury
Control
via
Amalgamation
Process).
ADA
Technologies,
Inc.
and
CH2M­
Hill
are
developing
a
new
family
of
Hg
sorbents,
called
Amended
Silicates
J.
Nooter
Eriksen
and
EnviroScrub
are
offering
the
Pahlman
technology
with
multi­
pollutant
control
capabilities
including
Hg
removal.

One
commenter
(
OAR­
2002­
0056­
5316)
stated
that
the
Toxecon
II
J
technology
is
applicable
to
715
coal­
fired
plants
while
the
new
high
temperature
sorbents
cover
another
82
plants.
The
industry
can
reach
a
70
percent
Hg
removal
rate
by
2010
and
90
percent
by
2014
with
this
technology.
The
Toxecon
II
J
system
can
be
rapidly
deployed
because
it
takes
advantage
of
the
existing
ESP.
It
is
cost
effective
because
the
new
chemically­
enhanced
AC
sorbent
has
low
injection
rates.
In
addition,
the
plant
can
continue
to
sell
90
percent
of
its
fly
ash
for
use
in
concrete.

Response:

EPA
is
aware
of
these
technologies
and
aware
that
none
are
in
full­
scale
application.
We
believe
that
the
final
rule's
cap­
and­
trade
approach,
with
declining
caps
and
market
rewards
for
reductions
will
provide
the
impetus
necessary
to
bring
these
technologies
to
full
development.

3.1.5.7
Impact
of
Coal
Chlorine
Content
on
Hg
Control
Performance
Comment:

One
commenter
(
OAR­
2002­
0056­
2422)
stated
that
the
best
performing
technologies
for
Hg
removal
are
fabric
filters
(
with
or
without
scrubbing)
and
wet
scrubbers
with
(
cold­
or
hot­
side)
ESP
units.
The
Hg
removal
capability
of
these
technologies
is
found
to
be
correlated
with
coal
Cl
content.
The
performance
of
these
control
technologies
is
substantially
reduced
and
highly
variable
when
firing
coals
with
low
Cl
content.
Thus,
there
is
not
have
a
high
level
of
confidence
that
the
best
performing
technologies
will
reduce
Hg
emissions
to
a
significant
degree
when
units
fire
coals
of
relatively
low
Cl
content.
The
performance
of
other
emission
control
technologies
does
not
appear
to
be
sensitive
to
Cl
content.

Response:

EPA
has
based
its
emission
limits
on
the
performance
of
technologies
within
each
of
the
subcategories
and,
thus,
feels
the
situation
noted
by
the
commenter
has
been
addressed.

Comment:

One
commenter
(
OAR­
2002­
0056­
1842)
stated
that
the
addition
of
a
chloride
prescrubber
before
a
SO2
wet
scrubber
should
ensure
90
percent
Hg
removal
and
even
more.
A
number
of
European
waste­
to­
energy
plants
utilize
this
technology
and
achieve
90
percent
removal
(
combination
of
pre­
scrubber
and
scrubber).
There
is
no
reason
this
technology
needs
to
3­
20
be
modified
for
coal­
fired
power
plant
use.
The
chloride
pre­
scrubber
levels
the
playing
field
for
those
burning
PRB
coal.
No
matter
how
small
the
Cl
content
of
the
coal,
the
pre­
scrubber
captures
it
as
hydrochloric
acid
and
then
builds
the
concentration
to
the
needed
level.
The
chloride
scrubber
by
itself
should
provide
all
the
oxidation
necessary.
But
if
higher
oxidation
is
still
desired,
you
can
deposit
some
of
the
salts
or
the
acid
back
on
the
coal
feed
belt.
So
the
concern
that
low
sulfur,
low
Cl
coals
will
make
it
more
difficult
to
remove
Hg
is
eliminated
with
this
scheme.

Response:

EPA
is
not
mandating
use
of
any
technology
to
achieve
compliance
with
the
final
rule.
The
industry
is
free
to
use
any
means,
including
the
one
cited
by
the
commenter,
to
achieve
compliance
with
the
standards.

Comment:

One
commenter
(
OAR­
2002­
0056­
3517)
stated
that
EPA
has
hypothesized
that
Hg
removal
rates
are
influenced
by
the
amount
of
Cl
that
is
contained
in
the
coal.
The
commenter
notes
that
although
western
bituminous
coals
are
low
in
Hg
content
relative
to
other
coals,
they
are
also
very
low
in
Cl.
The
commenter
adds
that
there
is
evidence
that
these
coals
perform
very
well
with
current
technology
in
reducing
Hg,
despite
the
low
Cl
content.
According
to
the
commenter,
a
case
in
point
is
the
Intermountain
Plant
in
Utah,
which
burns
a
low
Hg,
low
Cl
Utah
bituminous
coal;
it
is
possible
that
the
Cl
content
may
be
a
surrogate
for
other
factors
that
influence
Hg
reduction
performance.

Response:

EPA
believes
that
the
Cl
content
is
but
one
of
many
factors
that
may
impact
on
Hg
removal
from
coal­
fired
utility
units.
The
performance
of
western
bituminous
coals
noted
by
the
commenter
is
one
factor
that
lead
EPA
to
subcategorize
all
bituminous
coals
together.

Comment:

One
commenter
(
OAR­
2002­
0056­
2944)
stated
that
the
different
average
Cl
levels
of
the
different
coal
ranks
was
used
as
the
justification
to
support
proposing
different
emission
limits
and
allocation
weightings.
However,
based
on
the
commenter's
analysis,
Cl
is
found
to
be
of
little
importance.
The
utility
industry
is
already
removing
about
30
percent
of
the
estimated
75
tons
of
Hg
coming
into
its
plants
annually
with
the
fuel.
The
commenter
further
stated
that
according
to
ICR
flue
gas
measurements,
this
Hg
is
primarily
removed
at
bituminous
coal
plants
as
soluble
oxidized
Hg
(
due
to
coal
Cl)
in
existing
SO2
scrubbers
and
as
absorbed
oxidized
and
elemental
Hg
on
unburned
carbon
captured
in
the
plants'
particulate
collectors
(
due
not
to
higher
Cl
in
bituminous
coal,
but
due
to
their
higher
unburned
carbon
levels,
which
result
from
less­
reactive
fly
ash).
(
The
commenter
notes
however,
that
only
20
percent
of
U.
S.
boilers
have
scrubbers.)
The
commenter
states
that
as
indicated
in
the
second
Hg­
content
plot
(
see
OAR­
2002­
0056­
2944),
if
all
the
bituminous
coals
are
adjusted
for
a
30
percent
(
post­
combustion)
reduction
in
their
Hg
3­
21
levels,
and
subbituminous
coals
see
no
corresponding
reduction,
the
Hg
(
emission)
distributions
for
these
two
fuels,
which
encompass
over
95
percent
of
U.
S.
coal
use,
are
amazingly
identical.
Therefore,
when
considering
coal
Hg,
Cl,
and
fly
ash
together,
without
any
specifically­
added
Hg
control
technology,
subbituminous
coals
are
currently
not
at
any
disadvantage
relative
to
bituminous
coals
with
respect
to
Hg
emission
limits.

Response:

EPA
is
not
alone
in
its
belief
that
the
Cl
content
of
coal
is
a
factor
in
the
level
of
Hg
removal
achievable.
It
is
also
realized
that
the
level
of
unburned
carbon
in
the
exhaust
gas
is
also
a
contributing
factor.
We
believe
that
the
final
rule
does
not
disadvantage
any
coal
rank.

3.1.5.8
Impact
of
SCR
for
NOx
Control
on
Hg
Control
Performance
Comment:

Several
commenters
(
OAR­
2002­
0056­
1969,
­
2830)
disagreed
with
EPA's
statement
in
the
proposal
preamble
that
although
no
full­
scale
lignite­
fired
SCR­
equipped
unit
has
been
tested
for
Hg
removal
it
is
possible
that
greater
Hg
removal
would
result
when
an
SCR
unit
was
applied
to
a
lignite­
fired
unit.
The
commenters
stated
that
testing
of
a
pilot­
scale
SCR
reactor
at
Coyote
Station
(
a
nominal
420
MWe
lignite­
fired
generating
facility
that
is
located
near
Beulah,
North
Dakota)
showed
that
the
SCR
technology
is
ineffective
in
oxidizing
Hg
and
that
the
saltation
of
calcium
and
sodium
ash
deposits
foul
the
catalyst
rendering
the
SCR
technology
ineffective
for
NOx
control.
The
installation
was
in
conjunction
with
a
study
entitled
"
Impact
of
SCR
Catalyst
on
Mercury
Oxidation
in
Lignite­
Fired
Combustion
Systems"
that
conducted
by
the
Energy
and
Environmental
Research
Center
located
in
Grand
Forks,
North
Dakota.

One
commenter
(
OAR­
2002­
0056­
3514)
stated
that
currently,
no
technology
has
been
shown
to
be
effective
in
capturing
Hg
from
lignite
coals.
Lignite
and
other
low­
rank
western
coals
face
additional
obstacles
that
do
not
affect
other
ranks
of
coals,
specifically
higher
ash,
lower
Cl,
and
higher
elemental
Hg
content.
According
to
the
commenter,
these
factors
make
it
impossible,
at
least
currently,
to
attain
the
removal
percentages
being
achieved
with
other
coals.
The
commenter
noted
that
EPA
refers
to
SCR,
intended
for
NOx
reduction,
as
an
option
that
could
also
significantly
increase
the
oxidation
of
Hg
in
the
flue
gas
to
improve
capture.
The
commenter
stated
that
although
this
may
have
been
shown
to
work
for
certain
coal
ranks,
it
has
been
shown
that
the
SCR
blinds
almost
immediately
in
lignite
applications
in
recent
large
scale
testing
conducted
by
the
University
of
North
Dakota
Energy
and
Environmental
Research
Center.
Thus,
in
establishing
Hg
removal
goals
and
limits,
the
commenter
believes
EPA
must
consider
that
SCR
is
not
a
viable
option
for
lignite.

Response:

EPA
still
believes
that
SCR
installations
on
lignite­
fired
units
will,
with
further
development,
provide
improved
Hg
removal.
Any
improvement
will
provide
yet
another
means
for
such
units
to
effect
compliance
with
the
final
rule.
However,
the
use
of
SCR
units
on
lignite­
3­
22
fired
units
was
not
included
in
the
analyses
that
led
to
the
final
emission
limits
for
this
subcategory.

3.1.6
Analysis
of
ICR
Hg
Emission
Data
Comment:

Many
commenters
(
OAR­
2002­
0056­
1675,
­
1677,
­
1680,
­
1692,
­
1762,
­
2160,
­
2422,
­
2535,
­
2818,
­
2876,
­
3198,
­
3478,
­
3534,
­
3565)
stated
that
the
ICR
Part
III
data
are
not
appropriate
for
establishing
any
regulatory
standard
because
of
the
deficiencies
in
the
quantity,
quality,
and
accuracy
of
this
data
set.
Reasons
cited
by
commenters
include
the
following.
The
ICR
emissions
data
fail
to
meet
generally
accepted
limits
of
experimental
accuracy
and
precision.
The
data
set
includes
estimates
of
negative
Hg
removal,
incomplete
data,
failure
to
close
the
material
balance
in
the
overall
accounting
for
Hg
input
and
output,
and
low
precision.
The
80­
plant
ICR
sample
data
provide
an
unrepresentative
snapshot
of
emissions
from
a
limited
number
of
facilities
because
the
data
include
emissions
from
the
use
of
a
limited
number
of
fuel
types
over
a
limited
period
of
time.
The
wide
variability
of
coals
and
process
conditions
is
not
accounted
for
in
the
ICR
sample
data.
The
units
chosen
by
EPA
for
Hg
emissions
sampling
in
the
ICR
program
are
unrepresentative
of
the
coal­
fired
power
plants
in
the
U.
S.
The
companies
that
performed
the
tests
had
inadequate
experience
with
the
required
test
methodology.
The
data
are
affected
by
a
bias
in
testing
conditions,
because
the
testing
was
done
during
high­
load
and
steady­
state
operations.
The
data
were
gathered
using
a
test
method
that
is
very
different
from
what
is
proposed
for
compliance
demonstration
under
the
rule
and
no
effort
has
been
made
to
translate
the
proposed
standards
that
were
developed
from
the
data
to
the
basis
of
the
test
methods
proposed
for
compliance
demonstration.
The
reported
coal
rank
used
to
classify
some
of
the
units
tested
was
incorrect
or
did
not
accurately
reflect
the
blending
of
coals
from
different
ranks.
The
selection
of
the
units
chosen
by
the
EPA
for
testing
is
skewed
toward
wet­
and
dryscrubbed
units
which
are
more
likely
to
show
lower
emissions
than
the
majority
of
plants,
which
are
inscribed.

In
contrast,
Commenter
OAR­
2002­
0056­
5535
stated
that
EPA's
ICR
III
dataset
is
more
than
adequate
to
support
establishment
of
stringent
standards.
The
industry
commenters
opposing
the
ICR
data
set
identify
nothing
in
the
language
of
the
CAA
that
requires
that
the
dataset
comprehensively
account
for
emissions
information
from
the
industry
as
a
whole,
provided
the
data
allow
EPA
to
make
a
reasonable
estimate
of
performance
of
the
top
12
percent
of
units.
(
Sierra
Club
v.
EPA,
167
F.
3d
658,
662
(
D.
C.
Cir.
1999).
The
D.
C.
Circuit
has
further
observed
that
EPA
typically
has
wide
latitude
in
determining
the
extent
of
data­
gathering
necessary
to
solve
a
problem.
(
Cement
Kiln
Recycling
Coalition
v.
EPA,
255
F.
3d
855,
867
(
D.
C.
Cir.
2001)
("
CKRC")
(
quoting
Sierra
Club,
167
F.
3d
at
662)).
It
is
only
when
the
model
or
dataset
chosen
bears
"
no
rational
relationship
to
the
reality
it
purports
to
represent"
that
a
court
will
interfere
with
the
agency's
exercise
of
its
discretion.
(
Columbia
Falls
Aluminum
Co.
v.
EPA,
139
F.
3d
914,923
(
D.
C.
Cir.
1998)).
Putting
aside
the
legal
requirements,
EPA
also
thoroughly
debunked
the
factual
basis
underlying
the
industry
claims
that
the
ICR
database
is
too
weak
to
use
for
standard­
setting.
Industry
stakeholders
first
raised
this
issue
in
2001
during
the
Utility
Working
Group
process.
At
that
time,
EPA
presented
their
analysis
of
both
the
coal
sampling
data
and
the
3­
23
emissions
testing
data.
With
respect
to
the
fuel
analyses,
EPA
concluded
that
the
"
data
are
sufficient
to
use
in
the
development
of
MACT
standards."
For
the
emissions
tests,
the
agency
undertook
what
they
described
as
an
"[
e]
xtensive
quality
assurance
effort."
After
examining
individual
test
data,
excluding
invalid
data
and
examining
data
points
for
potential
outliers,
EPA
found
no
reason
to
exclude
any
of
the
complete
datasets
as
outliers.
As
with
the
fuel
analyses,
EPA
concluded
the
"[
s
]
tack
test
analyses
data
are
sufficient
to
use
in
the
development
of
MACT
standards."

Response:

EPA
believes
that
the
data
are
adequate
with
which
to
establish
appropriate
emission
limits
for
the
industry.
EPA
agrees
with
some
of
the
comments
made
but
not
with
the
conclusions.
For
example,
EPA
made
no
attempt
to
conduct
a
material
balance
around
the
utility
unit,
this
not
being
necessary
to
establish
an
emission
standard.
Units
showing
negative
removals
are,
obviously,
not
among
the
better
controlled
units
and,
thus,
were
not
used
in
establishing
the
emission
limit.
The
80
units
tested,
although
seemingly
limited
in
number,
represent
a
larger
data
set
than
available
for
other
CAA
section
111
or
section
112
regulatory
efforts.
The
matrix
of
unit
types
to
be
tested
was
subject
to
public
notice
and
comment
prior
to
being
sent
to
the
industry.
The
resulting
mix
and
number
of
units
is
a
compromise
between
the
greater
number
of
units
that
could
have
been
tested
as
inferred
by
the
commenters
and
the
cost
of
such
testing.
EPA
reported
the
rank
of
coal
used
during
the
testing
based
on
what
the
companies
involved
provided
to
EPA.
EPA
did
not
specify
the
load
to
be
maintained
during
testing
but
concurs
that
testing
of
this
type
is
generally
undertaken
during
periods
of
steady­
state
operation
to
minimize
the
problems
associated
with
evaluating
test
results
obtained
during
periods
of
fluctuating
operation.
However,
we
feel
that
the
incorporation
of
variability
in
the
analyses
adequately
addresses
this
issue.
The
testing
runs
were
conducted
sequentially,
so
source
variation
in
emissions
is
present
from
run
to
run.
Therefore,
no
measurement
of
sampling
precision
is
possible
as
this
would
have
required
the
use
of
paired
sampling
trains
at
all
sites.
The
test
contractors
utilized
by
the
industry
are
among
those
regularly
employed
in
such
activities
and,
thus,
are
familiar
with
both
the
industry
as
well
as
the
various
EPA
Reference
Methods.
The
Ontario­
Hydro
method
used,
although
requiring
attention
to
the
details
of
the
procedures,
utilizes
much
of
the
same
sampling
equipment
as
does
other
EPA
Reference
Methods
more
widely
utilized
by
the
test
contractors.
Further,
attention
to
the
details
of
the
test
methodology
is
not,
or
should
not
be,
anything
different
from
such
contractor's
performance
of
any
emissions
test.
The
proposed
continuous
Hg
measurements
are
for
gaseous
Hg
only;
most
of
the
Hg
measured
through
the
Ontario­
Hydro
method
was
also
determined
to
be
gaseous.
EPA
performed
some
comparisons
of
the
data
obtained
through
manual
vs.
continuous
monitoring
for
those
sites
at
which
the
continuous
monitors
were
evaluated
and
believes
that
the
12­
month
rolling
average
format
chosen
adequately
reflects
an
appropriate
translation
of
the
data.

Comment:

One
commenter
(
OAR­
2002­
0056­
2535)
retested
in
2003
some
of
the
power
plants
burning
Wyoming
PRB
coal
included
in
the
ICR
Part
III
data
sets.
The
re­
testing
methods
used
at
these
plants
were
consistent
with
the
methodologies
and
protocols
used
in
the
EPA
ICR
III
3­
24
testing.
Irrespective
of
the
distribution
of
the
Hg
species
at
the
APCD
inlet,
the
outlet
stream
contains
mostly
elemental
Hg.
Both
the
ICR
and
the
newly
acquired
data
are
directionally
consistent
but
have
significant
variation
due
to
coal
Hg
content
and
operational
variability.
This
corroborates
the
earlier
observations
that
data
variability
is
an
issue.
Hence,
any
regulatory
standards
or
guidelines
must
account
for
the
variability,
specifically
in
the
case
of
subbituminous
coal
due
to
its
higher
fraction
of
elemental
Hg
exiting
the
furnace.

Response:

EPA
concurs
that
variability
must
be
accounted
for
in
any
emission
limits.
We
believe
that
the
final
emission
limits
adequately
address
the
commenter's
concerns.

Comment:

One
commenter
(
OAR­
2002­
0056­
3560)
stated
the
ICR
data
collection
effort
appears
to
have
been
done
on
a
dry
basis.
This
introduces
minor
error
when
the
actual
testing
is
done
including
moisture
on
an
as­
received
basis,
but
the
impact
on
the
regulation
in
lb/
MWh
may
be
more
significant.
This
issue
has
not
been
addressed
by
EPA.

Response:

EPA
provided
the
data
on
a
dry
basis
for
consistency
and
ease
of
use
of
the
data
because
some
data
were
reported
by
the
companies
on
a
wet
basis
and
some
on
a
dry
basis.
We
do
not
believe
that
this
will
have
a
significant
impact
on
the
rulemaking.

Comment:

One
commenter
(
OAR­
2002­
0056­
2422)
stated
that
based
on
the
commenter's
analysis
of
the
ICR
Part
III
data,
no
statistically
significant
differences
can
be
detected
in
the
Hg
removal
performance
among
the
three
configurations
of
fabric
filter
controls
alone
or
combined
with
wet
or
dry
scrubbers.
Similarly,
no
statistically
significant
differences
can
be
detected
in
the
Hg
removal
performance
among
cold­
and
hot­
side
ESPs
combined
with
wet
scrubbers.

Response:

EPA
concurs
with
the
comment.

Comment:

One
commenter
(
OAR­
2002­
0056­
5564)
provided
additional
information
on
the
ICR
fuel
sampling
to
show
that
the
amount
of
Hg
was
significantly
understated
in
Gulf
Coast
lignites
because
of
the
test
method
(
ASTM
D3684)
used
in
the
analyses.
When
the
analytical
lab
switched
to
ASTM
D
6414,
the
Hg
levels
essentially
tripled.
Method
3684,
which
most
Gulf
Coast
lignite
plants
used,
is
not
accurate
for
lignite
with
high
levels
of
Hg.
3­
25
Response:

EPA
is
aware
of
the
issue
but
believes
that
the
value
is
limited
in
that
the
final
emission
limits
were
based
on
Hg
emissions
to
the
atmosphere
rather
than
on
any
calculation
based
on
the
Hg
content
of
the
coal
being
used.
EPA
reserves
the
right
to
revisit
this
issue
during
normal
reviews
of
the
NSPS
but
believes
that
the
revised
Hg
emission
limits
adequately
address
the
commenter's
concerns.

3.1.7
Cross­
Media
Impacts
Comment:

Two
commenters
(
OAR­
2002­
0056­
2008,
­
3478
)
stated
that
the
implementation
of
ACI
Hg
controls
could
potentially
impact
the
sale
of
combustion
byproducts,
eliminating
an
income
stream
for
utility
companies
and
increasing
expenses
for
permanent
ash
disposal.
One
of
the
commenters
(
OAR­
2002­
0056­
2008)
stated
that
the
largest
market
segment
for
coal
combustion
by­
product
(
e.
g,
fly
ash)
use
is
the
construction
materials
market.
Fly
ash
is
used
as
a
replacement
for
Portland
cement
in
concrete
production
and
other
cementitious
based
applications.
The
commenter
stated
that
the
severe
detrimental
influence
exhibited
by
activated
carbon
on
airentrained
concrete
was
shown
in
a
presentation
provided
at
the
American
Coal
Council
Mercury
and
Multi­
Emission
Compliance:
Strategies
and
Tactics
for
New
and
Existing
Coal
Plants
Symposium,
Irving,
Texas,
March
24­
25,
2004.
The
reported
findings
indicated
that
the
addition
of
activated
carbon
in
amounts
of
less
than
one
percent
could
render
fly
ash
unusable
for
concrete
applications.
The
laboratory
findings
are
consistent
with
reports
from
large­
scale
demonstration
projects
such
as
the
one
conducted
by
ADA­
ES
at
WE
Energies'
Pleasant
Prairie
Power
Plant.
In
that
study,
powdered
activated
carbon
was
used
as
the
Hg
sorbent.
Although
the
activated
carbon
removed
Hg
from
the
flue
gas
stream
during
the
test
program,
it
also
contaminated
the
fly
ash,
darkening
the
light­
colored
material
and
making
it
unusable
for
air
entrained
concrete.
The
commenter
stated
that
the
ACI
process
would
not
only
cause
the
plant
to
lose
a
source
of
revenue
through
lost
fly
ash
sales,
but
lead
to
additional
disposal
cost.
The
commenter
reported
that
these
combined
issues
were
estimated
to
be
valued
at
$
5,000,000
per
year
(
R.
Peatier;
Mercury
removal
standards
are
coming;
Where's
the
technology?,
Power,
May
2003
p
40).

Response:

EPA
agrees
that
the
use
of
ACI
can
impact
on
the
usability
and
disposal
of
fly
ashes
from
coal­
fired
utility
units.
However,
we
believe
that
means
are
available
that
minimize
this
impact
(
e.
g.,
use
of
a
polishing
fabric
filter
following
an
ESP;
the
ESP
to
capture
the
majority
of
the
"
clean"
fly
ash
for
re­
use
and
the
fabric
filter
to
capture
the
activated
carbon
injected
between
the
two
units).
Further,
sorbents
are
under
development
and
testing
that
do
not
cause
the
same
degradation
with
air­
entrained
concretes
that
are
posed
by
activated
carbon.

3.2
EMISSIONS
LIMITATIONS
3.2.1
General
3­
26
Comment:

One
commenter
(
OAR­
2002­
0056­
2443)
stated
that
establishing
nationwide
emission
limits
is
not
justifiable
given
the
wide
variability
in
coal
properties
(
e.
g.,
Hg
content,
Cl
content),
plant
operating
practices,
and
the
uncertainty
about
the
chemistry
of
Hg
speciation
and
its
control.

Response:

EPA
believes
that
its
use
of
subcategories
in
establishing
the
final
emission
limits
adequately
addresses
the
commenter's
concern.

Comment:

One
commenter
(
OAR­
2002­
0056­
2422)
stated
that
the
EPA's
proposed
new
source
standards
are
not
based
on
the
"
best
controlled
similar
source"
using
a
worst­
case
operating
scenario.
New
coal­
fired
units
are
not
uniform
in
design;
coal
properties
and
other
factors
can
significantly
affect
plant
designs.
Current
bituminous
PC
plant
designs
typically
incorporate
a
wet
scrubber
for
SO2
control,
an
ESP
or
fabric
filter
for
particulate
control,
and
an
SCR
for
NOx
reduction.
New
plants
designed
for
PRB
coal
will
likely
be
dry
scrubbed,
have
a
fabric
filter,
and
some
advanced
form
of
NOx
control
such
as
SCR.
As
noted
previously,
dry
scrubbed
plants
with
fabric
filters
obtain
virtually
no
Hg
reduction.
An
SCR
or
other
form
of
NOx
control
may
aid
in
the
reduction
of
Hg,
but
there
are
no
data
in
EPA's
ICR
database
on
which
to
base
a
sound
decision
on
the
effectiveness
of
NOx
controls
in
reducing
Hg
emissions
from
either
eastern
or
western
coals.

Response:

EPA
concurs
with
the
commenter's
concerns
but
believes
that
because
the
final
emission
limits
were
based
on
the
performance
or
permitted
levels
of
current
controls
within
each
subcategory,
the
commenter's
concerns
have
been
addressed
adequately.

Comment:

One
commenter
(
OAR­
2002­
0056­
2843)
recommended
that
no
standard
be
promulgated
unless
existing
control
technology
can
be
demonstrated
to
be
able
to
attain
and
sustain
the
standard
over
a
wide
range
of
coals
and
for
a
long
period
of
time.
The
commenter
stated
that
this
is
true
whether
the
emission
reduction
is
to
be
accomplished
by
a
requirement
under
either
CAA
section
111
or
section
112.
The
commenter
believes
that
unless
achievable
control
technology
is
available
at
the
outset,
the
construction
of
new
coal­
fired
facilities
will
be
improbable.
Further,
the
commenter
believes
that
no
demonstrated
technology
exists
that
is
capable
of
affecting
the
levels
of
emissions
reduction
which
would
be
required
under
either
of
EPA's
proposed
rulemaking.
Therefore
imposition
of
either
proposed
approach
would
make
it
extremely
difficult,
if
not
impossible,
to
construct
new
coal­
fired
plants.
The
commenter
specifically
cites
that
the
latest
DOE
solicitation
for
a
full­
scale
demonstration
of
Hg
reduction
technologies
on
a
scrubbed
unit
burning
PRB
coal
will
not
be
concluded
until
2005.
The
first
such
tests
being
conducted
at
3­
27
the
commenter's
Holcomb
1
unit
will
not
be
concluded
until
late
summer
2004.

Response:

As
noted
above,
EPA
concurs
with
the
commenter's
concerns
but
believes
that
because
the
final
emission
limits
were
based
on
the
performance
or
permitted
levels
of
current
controls
within
each
subcategory,
the
commenter's
concerns
have
been
addressed
adequately.
Further,
as
noted
later
in
this
document,
EPA
has
reanalyzed
the
data
and
revised
the
NSPS
emission
limit
for
new
sources.

Comment:

One
commenter
(
OAR­
2002­
0056­
2916)
stated
that
CAA
section
111(
a)(
1)
requires
that
the
level
of
emissions
reductions
for
a
new
source
must
reflect
a
level
of
performance
of
a
technology
that
has
been
put
to
practice
in
a
number
of
commercial
applications
using
a
number
of
coal
ranks
in
order
to
meet
the
test
of
being
"
adequately
demonstrated."
The
commenter
stated
that
there
are
currently
no
commercially
available
technologies
that
are
designed
to
control
Hg
from
coal­
fired
power
plants
to
the
levels
proposed.
The
EPA
must
reconsider
and
revise
the
proposed
NSPS
limits
in
light
of
the
wide
range
of
uncertainty
concerning
the
performance
and
future
availability
of
commercial
Hg
control
technology.
In
order
to
support
the
emissions
levels
and
time
frames
set
forth
in
the
proposed
rulemaking,
the
commenter
believes
that
the
EPA
and
DOE
must
make
certain
that
sufficient
funds
are
provided
to
complete
the
required
R&
D
to
fully
develop
and
commercially
demonstrate
advanced
Hg
control
technologies.
The
commenter
stated
that
Hg
emission
reductions
that
are
required
before
the
technology
has
been
fully
developed
will
lead
to
significantly
increased
costs,
to
likely
fuel
switching
from
coal
to
natural
gas,
and
to
possible
disruption
of
the
nation's
electricity
supply.

Response:

EPA
might
argue
the
commenter's
discussion
of
what
section
111(
a)(
1)
requires
but
agrees
with
the
commenter
that
Hg­
specific
control
technologies
are
not
yet
commercially
available.
Further,
as
noted
later
in
this
document,
EPA
has
reanalyzed
the
data
and
revised
the
NSPS
emission
limit
for
new
sources.

Comment:

One
commenter
(
OAR­
2002­
0056­
2331)
stated
that
to
reduce
the
possibility
of
overly
stringent
source
limits
and
resultant
fuel
switching,
the
EPA
should
set
the
new
source
standards
on
demonstrated,
commercially
viable
technologies
as
provided
in
CAA
section
111(
a).
The
EPA's
proposal
to
set
standards
based
on
"
emerging"
pollution
control
technology
introduces
unnecessary
uncertainty
in
the
viability
of
all
fuel
sources
for
future
generation
of
electricity.

Response:

As
noted
later
in
this
document,
EPA
has
reanalyzed
the
data
and
revised
the
NSPS
3­
28
emission
limit
for
new
sources.
We
do
not
believe
that
the
final
emission
limits
have
been
based
on
"
emerging"
technology
but,
rather,
that
the
format
of
the
standards
selected
will
allow
for
the
full
development
of
such
technologies.

Comment:

One
commenter
(
OAR­
2002­
0056­
1852)
stated
that
utilities
should
be
allowed
the
greatest
flexibility
in
switching
and
blending
fuels
to
meet
an
emissions
standard.
Therefore,
it
is
important
that
the
final
rule
be
designed
to
allow
for
inclusion
of
pre­
combustion
controls
as
a
viable
compliance
strategy.
For
many
electric
generating
facilities,
pre­
combustion
Hg
removal
can
be
more
cost­
effective
than
post­
combustion
removal,
as
pre­
combustion
methods
control
the
Hg
while
it
is
in
a
more
concentrated
and
contained
form,
permitting
significant
savings
in
waste
disposal
volumes
and
costs.
For
older
facilities
in
particular,
for
which
retrofits
would
be
extremely
costly,
using
fuel
that
has
been
cleaned
and
upgraded
on
a
pre­
combustion
basis
offers
the
most
cost­
effective
compliance
method.

One
commenter
(
OAR­
2002­
0056­
1760)
stated
the
final
rule
should
incorporate
pollution
prevention
strategies
to
remove
Hg
prior
to
release
to
the
air.

Response:

EPA
believes
that
pre­
combustion
removal
of
Hg
is
a
viable
option
available
to
the
industry
under
the
final
rules.

Comment:

One
commenter
(
OAR­
2002­
0056­
2331)
stated
that
EPA
must
ensure
that
any
new
source
emission
limits
are
economically
attainable
for
electric
utilities
and
will
not
lead
to
facilities
switching
to
natural
gas
for
base
load
electrical
generation.

Some
commenters
(
OAR­
2002­
0056­
1692,
­
1768)
stated
that
Hg
standards
must
be
technically
achievable
for
all
types
of
coal­
based
electric
generation
sources.

Response:

EPA
believes
that
its
final
rule
is
founded
on
the
cost
requirements
of
section
111,
is
technically
achievable
for
all
types
of
coal­
based
electric
generation
sources,
and
will
not
lead
to
fuel
switching.
Note,
however,
that
utilities
are
free
to
comply
with
the
final
emission
limits
in
any
manner
they
choose.

Comment:

One
commenter
(
OAR­
2002­
0056­
2843)
stated
that
circulating
fluidized
bed
technology
sources
behave
quite
differently
than
pulverized
coal
sources
and
should
not
be
used
to
determine
either
emission
limit
levels
or
allowance
allocations.
3­
29
Response:

As
noted
earlier,
with
the
exception
of
coal
refuse­
fired
units
(
which
is
a
result
of
the
coal
rank
rather
than
of
the
boiler
type),
the
data
did
not
suggest
that
FBC
units
(
including
CFB
units)
emitted
Hg
any
differently
than
other
boiler
types
and,
therefore,
no
subcategory
specifically
for
FBC
units
was
established.

3.2.2
Regulated
Pollutants
Comment:

One
commenter
(
OAR­
2002­
0056­
2219)
stated
that
the
proposed
rules
fail
to
address
speciation
of
Hg.
Mercury
takes
different
forms
(
ionic,
elemental,
particulate)
depending
on
the
rank
of
coal
burned.
Although
ionic
and
particulate
Hg
can
be
controlled
by
existing
technology,
additional
controls
are
needed
for
elemental
Hg.
This
is
particularly
important
in
the
case
of
a
cap­
and­
trade
program.

One
commenter
(
OAR­
2002­
0056­
2067)
stated
that
there
is
a
lack
of
effective
removal
technology
for
elemental
Hg,
which
is
prevalent
in
Wyoming
PRB
coal.
Without
existing
technology,
it
is
unfair
to
the
purported
neutral
treatment
of
coal
ranks
to
require
removal
of
elemental
Hg
(
which
is
more
prevalent
in
subbituminous
coal)
before
2010.
This
is
especially
true
for
those
power
plant
units
that
have
existing
scrubbers
in
place.

Response:

The
rank
of
coal
burned
impacts
the
relative
amounts
of
each
of
the
three
primary
species
of
Hg
emitted;
all
coals
emit
some
of
each
species.
The
final
rule
is
based
on
the
performance
and
permitted
levels
of
existing
technologies
for
new
sources.
For
existing
sources,
adequate
time
is
provided
before
the
Phase
II
cap
is
in
place
to
allow
for
the
development
of
the
promising
Hg­
specific
control
technologies
that
will
effectively
capture
the
elemental
Hg.

3.2.3
Format
of
Standards
3.2.3.1
General
Comment:

One
commenter
(
OAR­
2002­
0056­
3449)
objected
to
the
format
of
the
proposed
standard
because
the
EPA
failed
to
select
a
format
that
best
addresses
variability.
It
is
wrong
for
a
contaminant
where
variability
of
the
concentration
of
the
contaminant
in
a
fuel
is
an
important
consideration.
The
appropriate
format
was
used
in
the
MWC
standards
and
the
NSPS
for
SO2
emissions
from
coal­
fired
plants.
Both
of
these
rules
have
a
combination
standard
(
X
micrograms
per
cubic
meter
or
W
percent
control
for
the
MWC
units)
and
Y
lb/
Btu
or
Z
percent
control
for
the
NSPS.
This
format
allows
the
concentration
limit
to
be
based
on
the
average
level
of
the
constituent
because
the
percent
reduction
limit
can
be
used
for
situations
with
the
constituent
3­
30
level
is
much
higher.
The
logical
way
to
structure
a
combination
standard
is
to
base
the
lb/
TBtu
(
or
lb/
MWh)
on
the
median
case
and
to
base
the
percent
reduction
on
the
worst
coal
case.
This
ensures
that
real
reductions
occur
for
the
median
coal
and
the
worst
case
coal
can
still
be
burned
with
good
control.

One
commenter
(
OAR­
2002­
0056­
3406)
stated
that
because
of
variations
in
plant
design
and
the
coal
ranks
used,
the
commenter
recommended
that
the
standard
for
all
plants
be
a
combination
of
a
reduction
in
emission
rate
or
an
emission
rate,
whichever
is
less
restrictive
(
e.
g.,
X
percent
reduction
or
Y
lb/
TBtu).
The
commenter
stated
that
this
is
consistent
with
the
approach
used
in
Connecticut
and
proposed
in
other
States.

Response:

EPA
disagrees
that
a
percent
reduction
format,
or
a
combination
format
that
includes
percent
reduction,
is
appropriate
for
this
rulemaking
because
of
the
difficulty
in
determining
where
the
percent
reduction
should
be
assessed.
Further,
EPA
has
proposed
to
eliminate
the
percent
reduction
portion
of
the
subpart
Da
emission
limits
for
SO2
emissions
from
coal­
fired
power
plants.

3.2.3.2
Percent
Reduction
Format
Comment:

Many
commenters
(
OAR­
2002­
0056­
2054,
­
2067,
­
2068,
­
2160,
­
2224,
­
2422,
­
2634,
­
2661,
­
2827,
­
2867,
­
2922,
­
3200,
­
3403,
­
3514,
­
3432,
­
3565,
­
4891)
recommended
that
the
EPA
add
a
"
percent
reduction"
alternative
for
emission
limit
standards,
based
on
the
Hg
in
the
coal
supplied
to
the
boiler
and
the
Hg
in
the
stack.
Affected
units
would
have
the
option
of
meeting
an
emissions
rate
limit
or
a
removal
efficiency
requirement.
Reasons
for
adopting
a
percent
reduction
format
include
such
an
approach
is
appropriate
given
the
variability
between
and
among
units
and
the
differences
in
coal
characteristics
among
coal
within
a
given
rank.
This
would
allow
units
to
burn
higher
Hg
content
coals
by
removing
Hg
to
the
greatest
extent
possible.
In
addition
to
providing
a
realistic
option
for
units
that
would
result
in
significant
Hg
reductions,
this
approach
ensures
that
existing
coal
reserves
remain
a
viable
fuel
source.
Such
an
option
would
insure
that
one
coal
is
not
favored
over
another.

One
commenter
(
OAR­
2002­
0056­
2160)
recommended
that
standards
should
have
an
alternative
standard
to
emission
limits
based
on
a
percent
reduction
from
the
raw
coal
as
mined.
This
alternative
would
provide
some
relief
for
coals
with
unusually
high
Hg
content
while
still
achieving
meaningful
emission
reduction.

According
to
Commenter
OAR­
2002­
0056­
5535,
industry's
suggested
alternative
percent­
reduction
format
is
inappropriate
because
unit
operators
can
control
pollutant
input
levels.
In
the
MACT
standards
for
the
brick
and
structural
clay
products
and
clay
ceramics
manufacturing
industries,
EPA
allowed
units
to
meet
either
an
emissions
rate
or
a
percent
reduction
standard
for
hydrogen
fluoride
(
HF)
and
hydrogen
chloride
(
HCl),
but
did
not
provide
3­
31
the
alternative
approach
for
PM
(
a
surrogate
for
HAP
metals).
EPA's
reasons
for
declining
to
provide
an
alternative,
percent­
reduction
standard
for
PM
are
equally
applicable
here:
EPA
decided
not
to
finalize
a
percent
reduction
alternative
because
"
a
percent
reduction
standard
rewards
those
facilities
that
have
high
inlet
PM
loadings...[
a
situation
different]
from
the
percent
reduction
standards
for
HF
and
HCl
because
facilities
do
not
typically
have
options
for
reducing
the
uncontrolled
levels
of
HF
and
HCl."
In
other
words,
a
percent
reduction
alternative
is
appropriate
when
the
input
levels
of
the
HAP
in
question
are
outside
the
control
of
the
operator.
When,
as
here,
there
are
options
available
to
reduce
input
levels
of
the
HAP
being
regulated,
however,
a
percent
reduction
standard
has
the
perverse
effect
of
rewarding
those
operators
who
do
not
take
prophylactic
steps
to
reduce
input
levels.
For
the
coal­
fired
electric
generating
industry,
as
we
demonstrated
in
our
initial
comments,
there
are
a
variety
of
pre­
combustion
techniques
B
such
as
coal
washing
B
that
reduce
input
levels
of
Hg
and
other
HAP
from
all
coal
types.
Allowing
an
alternative
percent
reduction
approach
would
reward
operators
who
do
not
use
such
techniques
and
approaches.

Response:

EPA
continues
to
believe
that
a
percent
reduction
format
is
not
appropriate
for
this
rulemaking.
As
noted
in
the
proposal
preamble,
in
order
to
accommodate
pre­
combustion
Hg
control
technologies,
a
percent
reduction
format
would
require
tracking
the
Hg
concentrations
in
the
coal
basically
from
the
mine
to
the
stack,
and
not
just
before
and
after
the
control
device(
s)
and
could
be
difficult
to
implement.
We
believe
that
this
would
require
an
inordinate,
and
possibly
unworkable,
recordkeeping
effort.
We
believe
that
the
subcategorization
approach
and
revised
emission
limits
being
finalized
will
address
the
commenter's
concerns.

Comment:

One
commenter
(
OAR­
2002­
0056­
2243)
stated
that
the
Hg
limits
should
be
set
as
a
minimum
percent
removal
in
place
of
a
specific
emission
limit.
This
is
consistent
with
EPA's
earlier
efforts
at
SO2
control.

Response:

As
noted
earlier,
EPA
has
proposed
to
eliminate
the
percent
reduction
portion
of
the
subpart
Da
emission
limits
for
SO2
emissions
from
coal­
fired
power
plants
and
does
not
believe
that
a
percent
reduction
format
is
appropriate
for
this
rulemaking.

Comment:

One
commenter
(
OAR­
2002­
0056­
3288)
supported
an
emission
rate
limit
rather
than
a
percent
reduction
requirement.
A
percent
reduction
format
will
likely
result
in
higher
overall
emissions
which
would
end
up
being
more
costly
for
consumers
and
would
create
a
bias
against
cost
effective,
environmentally­
preferable
subbituminous
coal.
An
emissions
limit
will
result
in
the
highest
level
of
overall
reduction
at
a
lower
cost
to
consumers,
avoid
massive
disruptions
to
the
coal
industry,
and
encourage
continued
development
of
effective
pre­
combustion
technologies.
3­
32
Response:

EPA
concurs
that
an
emission
rate
limit
is
more
appropriate
than
an
emission
reduction
requirement.

3.2.3.3
Output­
based
Format
Comment:

One
commenter
(
OAR­
2002­
0056­
3435)
recommended
that
the
EPA
establish
input­
based
standards
for
Hg
control.
Although
the
output
format
promotes
energy
efficiency,
this
is
not
the
purpose
of
a
standard
for
protection
of
public
health
and
the
environment.
The
EPA
makes
an
economic
argument
for
an
output­
based
format
(
69
FR
4699)
which
contradicts
the
purpose
of
promoting
efficiency
through
emission
standards.
The
output­
based
limit
uses
an
assumed
efficiency
and
will
be
based
on
output
energy.
Using
output
energy
(
gross
or
net)
can
introduce
error
in
representing
actual
emissions
because
of
the
variability
in
assuming
efficiency
and
the
introduction
of
other
variabilities
inherent
to
the
standard,
especially
compared
to
heat
input
determinations.
Mercury
is
present
in
the
gas
in
such
trace
amounts
that
the
most
stringent
measurement
standard
of
the
emission
rate
should
be
used.
A
lb/
TBtu
standard
has
less
error
than
a
lb/
MWh
standard
and
will
better
represent
emission
levels.
Because
heat
input
is
already
required
for
the
Acid
Rain
Program,
this
format
is
already
standard
and
places
no
additional
burden
on
the
plants.

Response:

EPA
believes
that
an
output­
based
standard
is
consistent
with
the
intent
of
section
111
and
will
serve
to
protect
the
public
health
and
the
environment,
as
well
as
promote
energy
efficiency.
Further,
EPA
has
revised,
or
is
in
the
process
of
revising,
subpart
Da
to
place
the
emission
limits
for
PM,
NOx,
and
SO2
in
an
output­
based
format.

Comment:

One
commenter
(
OAR­
2002­
0056­
2161)
recommended
that
the
EPA
provide
the
maximum
encouragement
for
energy
efficiency
by
promulgating
a
standard
based
upon
lb/
MWh­
net
instead
of
lb/
MWh­
gross.
The
true
efficiency
of
a
coal­
fired
unit
is
based
upon
how
much
of
its
energy
is
available
after
reduction
by
internal
station
power
consumption,
measured
as
MWh­
net.
Therefore,
if
EPA's
goal
is
to
help
encourage
increased
energy
efficiency
with
the
Hg
standard,
the
most
effective
way
to
do
this
is
to
utilize
the
net
production.

One
commenter
(
OAR­
2002­
0056­
3449)
recommended
changing
the
proposed
format
for
the
output
based
standards
from
lb/
MWh
"
gross"
to
lb/
MWh
"
net"
to
encourage
efficiency.
A
net
standard,
like
the
one
in
their
State,
should
lead
to
lower
emissions
from
electricity
productions.

Response:
3­
33
EPA
agrees
with
the
commenters
that
using
the
"
net"
output
would
more
adequately
address
energy
losses
within
the
utility
station.
However,
our
intent
is
to
encourage
existing
units
to
utilize
the
output­
based
format
also.
Therefore,
we
believe
that
the
lb/
MWh­
gross
format
is
more
appropriate
for
this
rulemaking
because
implementation
on
existing
units
could
require
significant
and
costly
additional
monitoring
and
reporting
systems
because
the
energy
output
that
is
used
for
internal
components
(
and
not
sent
to
the
grid)
cannot
be
accounted
for
by
simply
installing
another
meter.
EPA
agrees
that
new
units
could
accommodate
the
lb/
MWh­
net
format
but
we
do
not
want
to
institute
a
dual
set
of
formats
for
the
same
industry
and
the
implementation
and
compliance
problems
that
would
result.

Comment:

Two
commenters
(
OAR­
2002­
0056­
3406,
­
5445)
supported
the
use
of
an
output­
based
standard
because
this
approach
rewards
efficiency
and
allows
the
market
to
make
decisions
about
fuel
choices
rather
than
favoring
one
type
of
generation
over
another.
The
commenters
also
supported
the
proposed
use
of
gross,
as
opposed
to
net,
plant
energy
output.
Commenter
OAR­
2002­
0056­
3406
stated
that
gross
energy
output
is
the
amount
of
energy
generated
before
internal
energy
consumption
and
losses
are
considered.
Net
electricity
generation
is
the
amount
of
energy
that
is
delivered
to
the
energy
grid
after
taking
into
account
internal
consumption
losses.
The
commenter
notes
that
those
losses
can
be
significant,
and
can
actually
increase
with
the
operation
of
emission
controls
such
as
SCR
and
scrubber
units.
The
commenter
concluded
that
the
use
of
net
plant
energy
output
would
penalize
a
power
plant
that
installed
additional
control
equipment,
which
the
commenter
takes
to
be
contrary
to
the
intent
of
the
rule.

One
commenter
(
OAR­
2002­
0056­
1969)
stated
that
EPA
has
suggested
that
the
output­
based
standard
be
calculated
on
gross
rather
than
net
energy
output
basis.
According
to
the
commenter,
a
net
energy
output
based
standard
is
certainly
the
most
comprehensive
and
is
able
to
capture
energy
efficiency
improvements
for
the
entire
plant.
The
commenter
states,
however,
that
EPA
is
correct
in
concluding
that
calculation
of
emissions
on
a
net
energy
output
basis
is
a
more
complex
task.
The
commenter
asserts
that
although
that
concern
may
not
be
the
sole
reason
to
exclude
a
net
energy
output
standard,
a
gross
energy
based
standard
is
able
to
capture
most
efficiency
improvement
projects
and
is
less
burdensome
to
administer.
The
commenter
supports
an
output­
based
standard
based
on
a
gross
energy
output
basis.

Response:

EPA
concurs
with
the
commenters.
However,
we
note
that
a
practice
of
overall
energy
efficiency
would
also
look
to
utilizing
more
efficient
equipment
on
the
SCR
and
scrubber
units,
although
we
do
not
believe
that
use
of
lb/
MWh­
net
is
appropriate
here.

Comment:

Two
commenters
(
OAR­
2002­
0056­
1969,
­
2850)
supported
the
option
of
either
an
input­
based
or
a
gross
output­
based
standard
for
existing
units
as
long
as
the
mathematical
relationship
between
the
two
standards
is
equitable.
The
output­
based
standard
offers
a
3­
34
regulatory
incentive
to
improve
unit
efficiency.
Any
output­
based
standard
should
give
consideration
to
average
unit
efficiency
subcategorized
for
unit
type
so
that
differences
in
installed
design
can
be
reflected
when
establishing
Hg
control
stringency.
An
output­
based
standard
should
not
be
periodically
revised
for
existing
units
because
doing
so
would
discourage
energy
efficiency
as
a
compliance
option
for
existing
sources.

Response:

EPA
believes
that
the
conversions
used
in
developing
the
final
emission
limits
are
equitable
and
appropriate.
However,
EPA
does
not
believe
that
it
is
appropriate
to
average
unit
efficiency
subcategory­
by­
subcategory
at
this
time.

Comment:

One
commenter
(
OAR­
2002­
0056­
4132)
objected
to
using
an
output­
based
emission
limit
format
for
old
or
new
sources.
Output­
based
emission
standards
are
not
desirable.
First,
they
draw
in
complexities
like
gross
electrical
output,
net
electrical
output,
and
disassociated
monitoring
systems.
Secondly,
they
deal
poorly
with
systems
that
choose
to
use
steam
for
a
variety
of
auxiliary
functions,
because
they
may
have
a
corresponding
loss
of
electrical
output.
Thirdly,
for
facilities
involved
with
some
level
of
cogeneration,
complex
and
unnecessary
accounting
regimes
are
required.
Input­
based
emission
standards
work
well,
and
should
be
available
to
all
emission
units.

Response:

EPA
disagrees
with
the
commenter.
Although
output­
based
formats
do
require
different
considerations
than
input­
based
formats,
all
formats
involve
a
certain
amount
of
complexity.
The
final
rule
addresses
the
issues
related
to
cogeneration
units
in
a
manner
similar
to
that
done
under
subpart
Da
for
NOx
emissions
from
such
units.
EPA
continues
to
believe
that
outputbased
emission
limits
encourage
energy
efficiency,
are
consistent
with
other
Agency
actions
on
subpart
Da,
and
are
appropriate
for
this
rulemaking.

Comment:

One
commenter
(
OAR­
2002­
0056­
3210)
disagreed
with
the
method
EPA
used
to
convent
input­
based
limits
to
output­
based
limits.
The
commenters
stated
that
the
EPA
should
establish
output
emission
limits
for
Hg
using
actual
emission
data,
not
a
calculated
value
from
a
heat
inputbased
standard.

Response:

The
conversion
used
by
EPA
was
based
on
that
used
in
the
subpart
Da
NOx
revisions,
which
were
based
on
data
received
from
new
facilities.

Comment:
3­
35
Several
commenters
(
OAR­
2002­
0056­
3210,
­
1474,
­
2721,
­
3437,
3459)
disagreed
with
the
power
plant
efficiency
values
EPA
used
to
convent
input­
based
limits
to
output­
based
limits.
The
commenters
stated
that
the
plant
percent
efficiencies
used
by
the
EPA
are
too
low.

One
commenter
(
OAR­
2002­
0056­
3437)
used
the
1999
National
Electric
Data
System
to
estimate
the
efficiency
of
the
69
coal­
fired
units
greater
than
25
MW
that
are
potentially
affected
by
the
proposed
rule.
The
average
efficiency
is
34
percent
for
existing
units.
The
commenter
provided
data
showing
that
new
units
can
achieve
efficiencies
significantly
greater
than
35
percent
with
IGCC
units
operating
at
42
to
47
percent
efficiency
and
certain
pulverized
units
achieving
40
percent
efficiency.

Several
commenters
opposed
the
use
of
35
percent
efficiency
as
the
baseline
efficiency
for
new
units.
One
commenter
(
OAR­
2002­
0056­
1474)
stated
that
the
baseline
efficiency
should
be
set
at
35
percent
to
move
power
plants
toward
higher
efficiencies.
Commenter
OAR­
2002­
0056­
2721
disagreed
that
35
percent
efficiency
is
an
appropriate
baseline
for
all
new
units.
The
commenter
stated
this
may
be
a
good
assumption
for
higher
quality
fuels
but
not
for
the
low
rank
fuels.
According
to
Commenter
OAR­
2002­
0056­
3459,
EPA
used
a
35
percent
baseline
efficiency
for
new
units
and
did
not
provide
any
support
for
their
assumption.
The
commenter
stated
that
the
EIA
assumes
that
a
new
scrubbed
coal
plant
with
SCR
will
have
an
efficiency
of
38
to
40
percent.
For
new
IGCC
units,
the
EIA
assumes
42.5
percent
efficiency.

Response:

EPA
is
unclear
about
the
reference
to
"
69
coal­
fired
units...
that
are
potentially
affected
by
the
proposed
rule."
Only
new
units
would
be
impacted
under
the
section
111
approach.
EPA
used
data
from
the
EIA
(
OAR­
2002­
0056­
0017)
that
provided
average
coal­
fired
power
plant
efficiencies
over
the
period
1935
to
1996
for
all
boilers
and
fuels.
The
35
percent
value
chosen
is
higher
than
that
achieved
in
all
but
2
of
those
years.

3.2.4
Numerical
Emission
Limits
3.2.4.1
General
Comment:

Several
commenters
(
OAR­
2002­
0056­
2843,
­
2897,
­
2911,
­
3324)
stated
concerns
that
the
proposed
emission
limits
for
new
sources
are
unjustifiably
stringent
citing
general
reasons
including
the
control
technologies
needed
to
comply
with
the
standards
have
not
been
adequately
demonstrated,
the
controls
are
too
costly
to
implement,
and
the
standards
would
prevent
the
use
of
much
of
the
coal
resources
in
the
U.
S.

Many
commenters
(
OAR­
2002­
0056­
1692,
­
1804,
­
2068,
­
2224,
­
2243,
­
2264,
­
2365,
­
2431,
­
2661,
­
2835,
­
2891,
­
2898,
­
2907,
­
2948,
­
3200,
­
3403,
­
3432,
­
3514,
­
3517,
­
3560)
stated
concerns
that
the
proposed
emission
limits
would
adversely
impact
the
construction
of
new
coal­
fired
power
plants
in
the
U.
S.
Reasons
cited
by
the
commenters
include
the
following.
The
3­
36
proposed
Hg
emission
standards
are
at
levels
that
are
not
be
achievable
with
currently
available
technology
except
for
the
lowest
Hg
content
coals.
This
would
preclude
the
ability
of
new
units
to
combust
coal
from
many
seams
that
have
high
Hg
content
levels.
The
proposed
limits
fail
to
account
for
variability
in
the
Hg
content
from
coals
mined
from
a
given
seam.
Also,
no
vendors
of
control
technology
are
willing
to
guarantee
Hg
removal
at
the
rates
needed
to
achieve
the
proposed
emission
levels.
No
company
would
make
a
large
capital
investment
in
a
new
plant
if
performance
guarantees
to
meet
required
environmental
standards
were
not
available.
Financial
institutions
will
be
very
wary
of
participating
in
projects
that
are
given
emission
limits
that
cannot
be
guaranteed
by
equipment
suppliers
and
whose
limits
will
be
difficult
to
verify.
Additionally,
if
facilities
are
forced
to
use
alternative
coal
sources,
it
could
dramatically
increase
the
cost
of
the
fuel
and
decrease
the
economic
viability
of
the
units,
also
impacting
the
decision
to
construct
the
unit.

Several
commenters
(
OAR­
2002­
0056­
1952,
­
2331,
­
2560,
­
2725,
­
2833,
­
2897,
­
3200,
­
3257)
stated
concerns
that
the
proposed
emission
limits
would
require
base
load
electric
utility
generating
units
to
switch
to
firing
natural
gas
to
ensure
compliance
with
the
standards.
Commenters
stated
that
forced
fuel
switching
from
coal
is
unacceptable
as
a
national
energy
policy.
It
would
adversely
impact
natural
gas
supplies
and
costs
to
other
natural
gas
users.
It
is
important
that
EPA
set
emission
rates
that
maintain
coal
as
a
major
fuel
source
option
in
a
diversified
national
energy
program.

One
commenter
(
OAR­
2002­
0056­
2210)
stated
that
the
EPA's
proposed
limits
for
new
sources,
under
either
a
MACT
or
cap­
and­
trade
(
NSPS)
approach,
are
unduly
stringent
and
would
preclude
the
use
of
many
U.
S.
coals
B
bituminous,
subbituminous
and
lignite.
Unrealistic
new
source
limits
could
present
an
insurmountable
barrier
to
the
construction
of
new,
low­
cost
coal
powered
generation,
conflicting
with
the
Administration's
energy
policies
favoring
the
development
of
all
forms
of
domestic
energy.
The
proposed
emission
limits
for
new
plants
need
to
reflect
the
emission
performance
that
can
be
expected
from
different
coal
ranks
at
plants
equipped
with
state­
of­
the­
art
emission
controls,
and
must
ensure
that
all
U.
S.
coals
may
be
utilized
at
such
new
plants.
The
U.
S.
can
ill
afford
to
create
artificial
barriers
to
the
development
and
use
of
its
largest
domestic,
fossil
energy
resource.

One
commenter
(
OAR­
2002­
0056­
2160)
stated
that
any
revised
rules
should
be
fuel
neutral
(
equitable
for
all
geographic
regions
and
coal
ranks),
reduction
requirements
should
be
based
on
reasonable
estimates
of
when
technology
will
be
available
to
meet
the
limits
(
and
consider
economic
and
time
constraints),
and
take
into
account
the
effect
of
the
inherent
variability
of
coals
with
respect
to
Hg
content,
combustion
characteristics,
and
control
system
performance.

Several
commenters
(
OAR­
2002­
0056­
1175,
­
1658,
­
1781,
­
1783,
­
1848,
­
1861,
­
1863,
­
2333,
­
2924)
stated
that
the
EPA's
proposed
limits
are
more
stringent
than
those
recommended
by
industry
as
part
of
the
workgroup
recommendations.
The
proposed
limits
do
not
reflect
the
level
of
control
that
is
technically
achievable.

Response:
3­
37
As
stated
in
the
preamble,
EPA
has
re­
analyzed
the
data
collected
in
the
1999
ICR
and
examined
the
Hg
limits
issued
in
recently
issued
permits
to
establish
new
source
NSPS
Hg
emissions
limits
for
five
subcategories
of
Utility
Units.
Based
on
these
findings,
EPA
believes
that
the
revised
new­
source
NSPS
Hg
emission
limits
are
reflective
of
the
level
of
Hg
control
that
is
currently
technically
achievable
for
these
subcategories.

Comment:

Several
commenters
(
OAR­
2002­
0056­
2871,
­
2889)
stated
that
the
proposed
limits
are
flawed
because
EPA
failed
to
consider
all
available
technologies.
Activated
carbon
injection
is
commercially
available
and
widely
recognized
as
a
viable
control
for
Hg.
It
has
been
demonstrated
with
pilot
and
full­
scale
demonstration
projects
on
coal
and
has
been
used
for
over
10
years
on
other
large
combustion
projects.
States
are
now
requiring
it
on
new
coal­
fired
units
for
Hg
control.
EPA's
failure
to
consider
this
technology
is
inconsistent
with
its
past
approaches
for
developing
Hg
limit
for
combustion
sources
and
EPA
provides
no
justification
for
the
change.

One
commenter
(
OAR­
2002­
0056­
2108)
stated
there
is
not
adequate
justification
for
not
examining
more
control
technologies/
options
in
setting
the
emission
limits
for
new
sources.

Response:

EPA
disagrees
with
the
commenters.
As
noted
earlier,
EPA
does
not
believe
that
ACI,
or
any
other
Hg­
specific
control
technology,
has
been
adequately
demonstrated
under
the
criteria
of
section
111
to
be
considered
viable
control
options
for
new
sources
under
this
rulemaking.
However,
we
do
believe
that
the
cap­
and­
trade
approach
being
taken
will
allow
such
technologies
the
necessary
time
to
be
fully
proven
for
widespread
commercial
installation.

Comment:

One
commenter
(
OAR­
2002­
0056­
2422)
stated
that
the
EPA's
proposed
limits
for
new
sources
must
be
revised
to
fully
account
for
variability
in
the
performance
of
the
"
best
performing"
unit,
regardless
of
whether
it
imposes
an
emissions
limit
or
a
"
cap­
and­
trade"
program.
Also,
the
proposed
emission
limits
for
new
plants
need
to
reflect
the
emission
performance
that
can
be
expected
from
different
coal
ranks
at
plants
equipped
with
state­
of­
theart
emission
controls,
and
must
ensure
that
all
U.
S.
coals
may
be
utilized
at
such
new
plants.
The
U.
S.
can
ill
afford
to
create
artificial
barriers
to
the
development
and
use
of
its
largest
domestic
energy
resource.

Response:

Although
use
of
the
"
best
performing
unit"
is
not
applicable
under
CAA
section
111,
EPA
believes
that
the
reanalysis
noted
earlier
adequately
addresses
the
concerns
related
to
variability,
use
of
different
coal
ranks,
and
emission
controls
noted
by
the
commenters.

Comment:
3­
38
One
commenter
(
OAR­
2002­
0056­
2441)
stated
that
the
EPA
ignored
their
workgroup
position
on
limits
for
existing
and
new
sources
that
would
allow
the
use
of
all
coals.
The
proposed
limits
for
western
subbituminous
and
lignite
coal
are
substantially
less
stringent
that
the
limits
recommended
by
the
industry
workgroup
participants,
the
proposed
limit
for
eastern
bituminous
coal
is
9
percent
more
stringent.
More
than
half
of
eastern
bituminous
coal
would
require
greater
than
75
percent
removal
to
meet
the
limit
(
beyond
EPA's
estimate
of
50
to
70
percent)
removal
capability
for
current
technologies).
In
contrast,
the
proposed
limit
for
western
subbituminous
coal
gives
PRB
coal
a
free
ride
in
that
most
of
62
percent
of
these
coals
could
meet
the
limit
without
any
controls.
This
would
encourage
cherry
picking
of
western
coal
that
could
be
sold
without
the
need
to
reduce
emissions.
This
preferential
treatment
would
invite
massive
fuel
switching
to
western
coal
and,
thus,
a
massive
shift
of
coal
production
from
eastern
to
western
states.
This
would
have
a
disastrous
economic
impact
on
coal
mining.
For
existing
sources,
84
to
100
percent
of
the
bituminous
coal
mined
in
eastern
states
could
not
be
used
under
the
NSPS
limits,
even
at
80
percent
removal.

Response:

EPA
believes
that
the
emission
limits
developed
through
the
reanalysis
of
the
data
will
address
the
concerns
noted
by
the
commenter
and
provide
equitable
treatment
for
all
coal
ranks.

Comment:

One
commenter
(
OAR­
2002­
0056­
2068)
stated
that
the
EPA
should
set
a
separate
emission
limit
for
fluidized­
bed
combustors.

Response:

As
noted
earlier
in
this
document,
EPA
does
not
believe
that
the
data
justify
a
separate
subcategory
for
FBC
units
and,
thus,
no
separate
emission
limit
has
been
established
for
FBC
units
utilized
in
the
bituminous,
subbituminous,
and
lignite
subcategories.

Comment:

One
commenter
(
OAR­
2002­
0056­
2913)
stated
that
the
proposed
limits
based
on
the
use
of
wet
limestone
scrubbers
for
Hg
control
do
not
include
an
allowance
for
the
Hg
content
of
the
limestone
used
as
the
sorbent
material.

Response:

Because
the
emission
limits
established
are
based
on
Hg
testing
conducted
at
the
stack
(
i.
e.,
following
any
limestone
injection
into
the
wet­
or
dry­
scrubber),
EPA
believes
that
any
Hg
contained
in
the
limestone
has
been
accounted
for
in
the
revised
emission
limits.

Comment:
3­
39
One
commenter
(
OAR­
2002­
0056­
2108)
stated
that
new
units
should
be
required
to
reduce
Hg
emissions
by
90
percent
regardless
of
fuel
type.

One
commenter
(
OAR­
2002­
0056­
3449)
recommended
that
emissions
limits
for
all
new
units
(
regardless
of
the
rank
of
coal)
be
determined
on
a
case­
by­
case
basis
and
be
no
higher
than
the
proposed
limit
for
new
bituminous
coal
units,
along
with
a
90
percent
control
option
to
address
high
Hg
coals.

Response:

As
noted
earlier,
EPA
does
not
believe
that
a
percent
reduction
format,
or
combination
format
including
percent
reduction,
is
appropriate
for
this
rulemaking.
Further,
EPA
believes
that
it
is
consistent
with
the
criteria
of
CAA
section
111
to
establish
the
emission
limits
for
each
subcategory
based
on
information
available
for
each
subcategory
which
is
the
procedure
followed
for
this
rulemaking.
Nor
does
EPA
believe
that
CAA
section
111
allows
for
a
case­
bycase
determination
of
new­
source
NSPS
emission
limits
as
suggested
by
the
commenter.

Comment:

Commenter
OAR­
2002­
0056­
2922
stated
that
it
is
not
clear
why
such
a
strange
mix
of
units
is
required
throughout
the
proposed
rules.
The
commenter
found
lb/
TBtu,
10­
6
lb/
MWh,
ounces,
tons,
MMBtu,
etc.
For
example,
it
does
not
make
sense
for
the
State
allocations
to
be
done
in
factional
tons
while
the
unit
allocations
are
in
ounces.
Why
not
use
ounces
for
both?

Response:

EPA
agrees
with
the
commenter
and
has
standardized
the
units
of
measure
as
much
as
possible
in
the
final
rules.

3.2.4.2
Approach
to
Setting
New­
source
Limits
Comment:

Two
commenters
(
OAR­
2002­
0056­
2422,
­
2862)
stated
that
the
EPA
proposed
the
same
numerical
limits
for
new
source
MACT
under
CAA
section
112
and
the
alternative
NSPS
under
CAA
section
111.
Under
section
112,
the
new
source
MACT
limit
should
"
not
be
less
stringent
than
the
emission
control
that
is
achieved
in
practice
by
the
best
controlled
similar
source."
Under
section
111,
NSPS
should
"
reflect
the
degree
of
emission
limitation
and
the
percentage
reduction
achievable
through
application
of
the
best
technological
system
of
continuous
emission
reduction
(
taking
into
consideration
the
cost
of
achieving
such
emission
reduction,
any
nonair
quality
health
and
environmental
impact
and
energy
requirements)."
Limits
under
both
sections
of
the
CAA
begin
with
an
assessment
of
what
limit
is
achievable
in
practice
with
the
best
available
controls,
but
the
NSPS
goes
on
to
consider
cost,
energy
use
and
non­
air
impacts.
Accordingly,
it
is
inappropriate
and
inconsistent
with
the
CAA
for
the
EPA
to
establish
an
NSPS
requirement
based
on
an
analysis
undertaken
pursuant
to
the
requirements
of
CAA
section
112.
3­
40
Response:

EPA
agrees
with
the
commenters
who
indicated
that
the
new­
source
NSPS
limits
were
not
established
in
a
manner
consistent
with
the
requirements
of
CAA
section
111.
We
have,
therefore,
re­
analyzed
the
information
collection
request
(
ICR)
data
collected
in
1999,
and
examined
the
Hg
limits
in
recently
issued
permits.
Based
on
this
refined
analysis,
we
have
arrived
at
the
following
new­
source
NSPS
Hg
emission
limits
for
the
five
subcategories:

Bituminous
units:
0.0026
ng/
J
(
21
x
10­
6
lb/
MWh);
Subbituminous
units:
­
wet
FGD
units
0.0055
ng/
J
(
42
x
10­
6
lb/
MWh);
­
dry
FGD
units:
0.0103
ng/
J
(
78
x
10­
6
lb/
MWh);
Lignite
units:
0.0183
ng/
J
(
145
x
10­
6
lb/
MWh);
Coal
refuse
units:
0.00017
ng/
J
(
1.4
x
10­
6
lb/
MWh);
IGCC
units:
0.0025
ng/
J
(
20
x
10­
6
lb/
MWh).

Documentation
for
this
re­
analysis
may
be
found
in
the
e­
docket
(
OAR­
2002­
0056).

To
establish
the
revised
new­
source
limits,
EPA
re­
examined
the
1999
ICR
data
which
includes
an
estimate
of
the
Hg
removal
efficiency
for
the
suite
of
emission
controls
in
use
on
each
unit
tested.
The
EPA
focused
primarily
on
the
1999
ICR
data
because
it
is
the
only
test
data
for
a
large
number
of
Utility
Units
employing
a
variety
of
control
technologies
currently
available
to
the
Agency
and
because
there
is
very
limited
permit
data
for
new
or
projected
facilities
from
which
to
determine
existing
Hg
emission
limits.
(
The
EPA
has
historically
relied
on
permit
data
in
establishing
new­
source
NSPS
limits
because
it
believes
that
such
limits
reasonably
reflect
the
actual
performance
of
the
unit.)
We
analyzed
the
performance
of
currently
installed
control
technologies
in
the
respective
subcategories
in
an
effort
to
identify
a
best
adequately
demonstrated
system
of
emission
reduction,
also
referred
to
as
a
best
demonstrated
control
technology
(
BDT),
for
each
subcategory.
To
do
this,
we
determined
the
combination
of
control
technologies
that
a
new
unit
would
install
under
the
current
NSPS
to
comply
with
the
emissions
standards
for
PM,
SO2,
and
NOx.
Based
on
the
available
data,
units
using
these
combinations
of
controls
had
the
highest
reported
control
efficiency
for
Hg
emissions.
Thus,
we
determined
that
BDT
for
each
subcategory
of
units
is
a
combination
of
controls
that
would
generally
be
installed
to
control
PM
and
SO2
under
the
NSPS.
For
bituminous
units,
BDT
is
a
combination
of
a
fabric
filter
and
a
FGD
(
wet
or
dry)
system.
For
subbituminous
units,
BDT
was
determined
to
be
dependent
on
water
availability.
For
subbituminous
units
located
in
the
western
U.
S.
that
may
face
potential
water
restriction
and,
thus,
do
not
have
the
option
of
using
a
wet
FGD
system
for
SO2
control,
BDT
is
a
combination
of
either
a
fabric
filter
with
a
spray
dryer
absorber
(
SDA)
system
or
an
ESP
with
a
SDA
system.
For
subbituminous
units
that
do
not
face
such
potential
water
restrictions,
BDT
is
a
fabric
filter
in
combination
with
a
wet
FGD
system.
For
lignite
units,
BDT
is
either
a
fabric
filter
and
SDA
system
or
an
ESP
with
a
wet
FGD
system.

To
determine
the
appropriate
achievable
Hg
emission
level
for
each
coal
type,
a
statistical
analysis
was
conducted.
Specifically,
the
Hg
emissions
limitation
achievable
for
each
3­
41
coal
type
was
determined
based
on
the
highest
reported
annual
average
Hg
fuel
content
for
the
coal
rank
being
controlled
by
the
statistically­
calculated
control
efficiency
for
the
BDT
determined
for
that
fuel
type.
The
control
efficiency
for
BDT
was
calculated
by
determining
the
90th
percentile
confidence
level
using
the
one­
sided
z­
statistics
test
(
i.
e.,
the
Hg
removal
efficiency,
using
BDT,
estimated
to
be
achieved
90
percent
of
the
time).
The
data
used
consisted
of
stack
emission
measurements
(
pounds
Hg
per
trillion
Btu,
lb
Hg/
TBtu)
for
each
unit,
the
average
fuel
Hg
content
for
the
fuel
being
burned
by
that
unit
during
the
test
(
parts
per
million,
ppm),
and
the
highest
average
annual
fuel
Hg
content
reported
for
any
unit
in
the
coal
rank.
Because
the
Hg
emissions
from
any
control
system
is
a
linear
function
of
the
inlet
Hg
(
i.
e.,
Hg
fuel
content),
assuming
a
constant
control
efficiency,
the
reported
highest
annual
average
inlet
Hg
was
adjusted
to
determine
the
potential
maximum
Hg
emissions
that
would
be
emitted
if
BDT
was
employed.
The
calculated
90th
percentile
confidence
limit
control
reduction
for
each
subcategory,
based
on
the
calculated
highest
annual
average
uncontrolled
Hg
emissions,
in
lb
Hg/
TBtu,
for
the
subcategory
was
determined
to
be
the
new
source
emission
limit.
Finally,
the
new
source
limit
for
IGCC
units
and
its
justification
remains
unchanged
from
the
limit
proposed
in
January
2004
(
69
FR
4652).

EPA
also
evaluated
recent,
available
permit
Hg
levels
for
comparison
with
the
limits
presented
above.
EPA
does
not
believe
that
the
use
of
permit
Hg
limits
is
appropriate
for
independently
establishing
new­
source
NSPS
emission
limits
because
of
the
limited
number
of
permits
issued
with
Hg
emission
levels
and
the
limited
experience
of
both
State
permitting
authorities
and
the
industry
itself
with
establishing
appropriate
permit
conditions.
However,
comparison
of
the
available
permit
limits
with
those
developed
by
EPA
is
a
valid
"
reality
check"
on
the
appropriateness
of
EPA's
limits.
Available
permits
on
bituminous­
fired
units
have
Hg
emission
limits
ranging
from
approximately
20
x
10­
6
lb/
MWh
to
39
x
10­
6
lb/
MWh;
those
for
subbituminous­
fired
units
range
from
11
x
10­
6
lb/
MWh
to
126
x
10­
6
lb/
MWh.
Considering
the
limited
number
of
permits
and
the
limited
experience
in
developing
appropriate
Hg
limits
for
those
permits,
EPA
believes
that
its
final
new­
source
NSPS
Hg
emission
limits
are
in
reasonable
agreement
with
these
permits.
Insufficient
permit
information
is
available
to
do
a
similar
comparison
for
lignite­
and
coal
refuse­
fired
units
but
we
have
used
the
same
analytic
procedure
for
these
subcategories.

Further,
EPA
concurs
with
those
commenters
who
indicated
that
we
had
overstated
the
variability
in
the
context
of
the
proposed
CAA
section
111
NSPS
limits
by
using
both
a
rigorous
statistical
analysis
and
a
12­
month
rolling
average
for
compliance.
Therefore,
for
the
final
rule,
while
we
have
retained
the
12­
month
rolling
average
for
compliance,
we
have
used
the
annual
average
fuel
Hg
content
in
the
ICR
data
to
establish
the
NSPS
limits.
Given
the
favorable
comparison
with
the
available
permit
data,
we
believe
that
variability
has
been
adequately
addressed.
Documentation
for
the
new­
source
limits
is
provided
in
"
Statistical
Analysis
of
Mercury
Test
Data
to
Determine
BDT
for
Mercury"
(
OAR­
2002­
0056­
6192).

3.2.4.3
Bituminous
Coal­
fired
Units
Comment:
3­
42
Two
commenters
(
OAR­
2002­
0056­
2160,
­
3199)
stated
that
in
the
supplemental
notice,
the
EPA
stated
that
50
to
70
percent
Hg
removal
technologies
may
be
commercially
available
after
2010
which
could
address
emissions
from
bituminous
coal.
The
proposed
emission
limits
would
require
about
94
percent
reduction
from
the
average
bituminous
coal.
This
is
higher
than
EPA's
assessment
of
the
Hg­
specific
control
technologies
which
would
be
available
at
the
time
of
implementation.
The
standards
should
be
based
on
the
emission
reduction
that
is
achievable
at
the
time
of
implementation.
Reductions
should
be
based
on
realistic
estimates
of
when
technology
will
be
available
and
include
consideration
of
the
economic
and
time
constraints
in
meeting
the
limits.

One
commenter
(
OAR­
2002­
0056­
3445)
stated
that
the
control
level
required
of
new
sources
under
either
of
EPA's
proposed
regulatory
approaches
would
make
it
nearly
impossible
to
build
new
bituminous
coal­
fired
power
plants.

One
commenter
(
OAR­
2002­
0056­
2862)
stated
EPA's
proposed
emission
limit
for
new
bituminous
units
contradicts
the
Agency's
findings
about
achievable
Hg
reductions
and
would
prevent
the
use
of
many
coals.
If
the
Hg
emission
standard
for
new
bituminous
coal
units
is
set
at
the
proposed
0.6
lb/
TBtu
emission
rate,
coals
from
many
regions
of
the
country
could
not
be
used
in
new
coal­
fired
plants
because
Hg
removal
in
excess
of
90
percent
would
be
required.
This
would
eliminate
billions
of
tons
of
coal
from
the
nation's
energy
supply.
The
commenter
included
an
analysis
that
the
commenter
stated
demonstrated
that
the
majority
of
bituminous
coal
supplies
available
to
utilities
in
the
Midwest
would
be
unable
to
achieve
the
proposed
emission
standard.
The
bituminous
coals
used
in
the
analysis
represent
typical
coals
(
coals
from
West
Virginia,
Pennsylvania,
Kentucky,
Illinois
and
Colorado)
that
a
new
unit
in
the
Midwest
would
burn.

Response:

As
noted
above,
EPA
has
reanalyzed
the
data
and
revised
the
new­
source
NSPS
Hg
emission
limits
which
should
address
the
commenter's
concerns.

Comment:

One
commenter
(
OAR­
2002­
0056­
2064)
stated
that
the
proposed
limits
for
bituminous
coal
are
above
the
levels
that
are
technically
achievable
and
cost
effective.
For
bituminous
coal,
proposed
limit
would
require
a
77
percent
reduction
but
this
is
well
below
the
average
90
percent
control
demonstrated
by
fabric
filters.
This
State
recently
permitted
a
plant
for
90
percent
removal
for
bituminous
coal
using
a
fabric
filter
and
wet
FGD.
These
controls
are
applicable
to
existing
and
new
units.

Response:

As
noted
elsewhere,
EPA
has
reanalyzed
the
data
and
revised
the
new­
source
NSPS
Hg
emission
limits
based
on
the
use
of
current
technologies.

3.2.4.4
Subbituminous
Coal­
fired
Units
3­
43
Comment:

One
commenter
(
OAR­
2002­
0056­
2535)
stated
that
the
Wyoming
PRB
subbituminous
coal,
which
is
the
most
widely
used
subbituminous
coal,
should
not
be
used
to
establish
Hg
emission
limits
for
all
subbituminous
coal­
fired
plants.
Instead,
Colorado,
Montana,
and
New
Mexico
subbituminous
coals
should
be
used.
These
coals
are
typically
higher
in
caloric
(
Btu)
content,
and
resemble
a
bituminous
coal.
Wyoming
PRB
coal
grades
out
as
a
Subbituminous
C
coal,
while
most
other
western
subbituminous
coals
grade
out
as
Subbituminous
A
(
according
to
ASTM
standards).
For
the
proposed
limit,
an
analysis
by
the
National
Mining
Association
estimates
that
41
percent
of
subbituminous
coals
would
not
be
able
to
meet
the
limit
with
any
degree
of
confidence
due
to
the
high
variability
in
Hg
content
of
the
coal.

Response:

EPA
has
used
the
data
and
information
available,
including
permit
information,
to
revise
the
new­
source
NSPS
Hg
emission
limits
for
all
of
the
subcategories,
including
subbituminous.
We
believe
that
these
revised
limits
will
accurately
reflect
the
level
of
control
expected
in
each
subcategory.
EPA
does
not
understand
why
PRB
coal
should
be
excluded
from
this
analysis,
particularly
given
the
fact
(
acknowledged
by
the
commenter)
that
it
is
the
most
widely
used
subbituminous
coal.

Comment:

One
commenter
(
OAR­
2002­
0056­
3437)
opposed
the
proposed
emission
limit
for
subbituminous
coal.
The
proposed
limit
would
require
little
or
no
control
at
some
power
plants
in
Indiana
that
use
subbituminous
coal
or
a
blend
of
bituminous
and
subbituminous
coal.
This
disparity
also
can
cause
bituminous
units
to
switch
to
subbituminous
coal
or
a
blend
of
the
two,
which
would
increase
Hg
emissions
above
1999
levels.

One
commenter
(
OAR­
2002­
0056­
3449)
stated
that
the
proposed
limit
for
subbituminous
coal­
fired
units
is
three
times
higher
that
the
proposed
limit
for
bituminous
coal.
The
proposed
limit
is
so
high
that
it
would
result
in
little,
if
any,
Hg
reductions.
The
ICR
data
shows
that
the
proposed
limit
is
about
the
average
Hg
content
of
subbituminous
coal,
assuming
all
Hg
in
the
coal
is
emitted.
With
co­
benefits
of
existing
controls,
over
80
percent
of
subbituminous
coal
is
likely
to
be
burned
without
any
additional
control.
About
two­
thirds
of
the
subbituminous
coals
have
Hg
content
less
than
5.8
lb/
TBtu
(
the
proposed
limit).
Assuming
a
30
percent
co­
benefit
of
minimal
existing
controls,
this
results
in
equivalent
Hg
content
of
more
than
about
8.3
lb/
TBtu
for
which
added
control
would
be
needed.
Only
about
15
percent
of
subbituminous
coal
is
above
this
level.
And,
when
long
term
averaging
is
considered,
even
fewer
subbituminous
coal­
burning
units
are
likely
to
required
controls.
Even
if
no
units
switch
from
bituminous
to
subbituminous
coal,
Western
states
will
obtain
little
or
no
Hg
reduction.
If
widespread
switches
to
subbituminous
coal
occur,
the
East
will
have
much
higher
Hg
emissions
than
EPA
projects.

Response:
3­
44
EPA
has
used
the
data
and
information
available,
including
permit
information,
to
revise
the
new­
source
NSPS
Hg
emission
limits
for
all
of
the
subcategories,
including
subbituminous.
We
believe
that
these
revised
limits
will
accurately
reflect
the
level
of
control
expected
in
each
subcategory.

Comment:

Several
commenters
(
OAR­
2002­
0056­
2064,
­
2198)
stated
that
the
proposed
limits
for
subbituminous
coal­
fired
units
are
too
high.
In
Wisconsin,
only
one
plant
in
the
State
(
the
largest
unit
with
the
highest
emission
rate)
would
be
required
to
reduce
emissions
at
all;
this
plant
would
need
to
reduce
emissions
by
40
percent.
Plants
firing
subbituminous
coal
should
be
capable
of
achieving
50
to
83
percent
removal
from
fuel
input
based
on
use
of
a
fabric
filter
alone.
One
State
recently
permitted
one
coal­
fired
plant
at
83
percent
removal
for
subbituminous
coal
(
1.70
lb/
Btu)
using
a
fabric
filter
(
dry
FGD
system)
with
sorbent
injection.
This
control
equipment
is
applicable
to
existing
and
new
units.

Response:

As
noted
above,
EPA
has
used
the
data
and
information
available,
including
permit
information,
to
revise
the
new­
source
NSPS
Hg
emission
limits
for
all
of
the
subcategories,
including
subbituminous.
We
believe
that
these
revised
limits
will
accurately
reflect
the
level
of
control
expected
in
each
subcategory.
Further,
it
has
also
been
noted
that
EPA
does
not
believe
that
ACI
is
a
commercially
available
technology
upon
which
a
CAA
section
111
standard
can
be
established.

Comment:

One
commenter
(
OAR­
2002­
0056­
2535)
stated
that
the
AES
Hawaii
power
plant
should
not
be
used
to
set
the
subbituminous
emission
limit
for
two
reasons.
First,
the
plant
is
an
FBC
unit,
which
relies
on
a
fundamentally
different
combustion
process
and
is
not
representative
of
the
type
of
plant
that
burns
subbituminous
coal
in
the
48
contiguous
States.
Second,
the
plant
burns
Indonesian
subbituminous
coal,
which
is
also
not
representative
of
the
subbituminous
coal
burned
in
the
48
contiguous
States.
In
addition
to
the
coal
not
being
representative,
EPA
must
recognize
that
coal
is
our
largest
reserve
of
domestic
fossil
fuel,
and
should
not
be
using
a
foreign
coal
to
set
a
domestic
standard.
This
goes
against
EPA's
stated
principle
of
not
considering
fuel
switching
as
a
viable
method
for
setting
a
MACT
floor.
Use
of
Indonesian
coal
to
set
the
MACT
floor
will
result
in
more
domestic
coal
being
displaced
from
use
in
domestic
coal­
fired
power
plants.

Response:

The
AES
Hawaii
facility
was
one
of
nine
subbituminous­
fired
units
used
in
establishing
the
revised
new
source
NSPS
emission
limits.
EPA
did
not
feel
that
it
would
be
appropriate
to
exclude
this
unit
from
the
reanalysis
because
(
1)
at
least
one
new
subbituminous­
fired
unit
has
received
a
permit
based
on
the
use
of
FBC
technology
in
Utah
(
and,
thus,
FBC
technology
is
representative
of
the
type
of
plant
that
could
burn
subbituminous
coal
in
the
contiguous
48
3­
45
States),
and
(
2)
Indonesian
coal
was
reported
to
be
used
by
at
least
two
other
utility
units
in
1999
(
and,
thus,
could
be
used
by
other
units
in
the
U.
S.).

Comment:

One
commenter
(
OAR­
2002­
0056­
3437)
requested
that
EPA
examine
the
recent
State
MACT/
BACT
decisions
that
have
required
ACI.
In
particular,
Iowa
set
a
case­
by­
case
MACT
limit
equal
to
1.7
lb/
TBtu
based
on
ACI
with
83
percent
control
efficiency
using
PRB
coal.

Response:

As
noted
earlier,
EPA
does
not
believe
that
Hg­
specific
control
technologies,
including
ACI,
are
commercially
available
for
nationwide
application
to
the
coal­
fired
utility
industry.
Installation
of
such
technologies
on
a
limited
number
of
units
(
e.
g.,
the
one
cited)
is
possible
and
will
serve
to
advance
the
technologies
such
that
they
are
widely
for
use
in
compliance
with
the
phase
II
cap.

3.2.4.5
Lignite­
fired
Units
Comment:

One
commenter
(
OAR­
2002­
0056­
3469)
supported
the
proposed
limits.
The
commenter
stated
that
despite
flaws
in
the
ICR
data
used
to
determine
the
emission
limits,
they
supported
the
emission
limits
for
existing
units
of
9.2
lb/
TBtu
for
Fort
Union
lignite­
fired
units.

Another
commenter
(
OAR­
2002­
0056­
2115)
supported
the
proposed
limit
for
lignite
because
it
is
the
same
level
that
would
be
required
under
the
Clear
Skies
proposal.
An
unattainable
Hg
limit
would
result
in
a
practical
ban
on
lignite
as
fuel.
Texas
mines
over
40
million
tons
of
lignite
coal
per
year
for
use
as
power
plant
fuel.

Response:

EPA
concurs
that
the
standards
must
be
achievable
by
all
coal
ranks.

Comment:

One
commenter
(
OAR­
2002­
0056­
3514)
stated
that
currently,
no
technology
has
been
shown
to
be
effective
in
capturing
Hg
from
lignite
coals.
Although
EPA
has
proposed
for
lignite
more
reasonable
emission
limits,
it
still
has
not
been
shown
that
these
levels
can
be
met.
Lignite
and
other
low­
rank
western
coals
face
additional
obstacles
that
do
not
affect
other
coals,
namely
higher
ash,
lower
Cl
and
higher
elemental
Hg
content.
Accordingly,
these
factors
make
it
impossible,
at
least
currently,
to
attain
the
removal
percentages
being
achieved
with
other
coals.

One
commenter
(
OAR­
2002­
0056­
2054)
stated
that
the
EPA
has
not
proposed
standards
for
new
lignite­
fired
units
on
a
level
of
performance
that
is
"
achievable"
by
a
unit
that
is
"
similar"
3­
46
to
most
new
lignite­
fired
units.
The
commenter
stated
that
the
highest
Hg
removal
rate
of
any
lignite­
fired
plant
in
the
ICR
data
was
21
percent,
and
the
plant
that
achieved
this
removal
rate,
Stanton
Station,
is
a
relatively
small,
older
plant
that
is
definitely
not
"
similar"
to
a
new
lignite­
fired
unit.
According
to
the
commenter,
the
agency
did
not
base
its
new
unit
standard
on
performance,
but
rather
on
the
lowest
Hg
coal;
the
result
of
this
basis
is
to
eliminate
the
vast
majority
of
lignite
reserves
from
any
new
units.
The
commenter
asserts
that
the
choice
of
best
performing
unit
should
comply
with
the
direction
given
by
the
DC
Circuit
Court
in
National
Lime
Association
v.
EPA,
627_
F.
2d
416,
431n.
46(
1980).
The
commenter
stated
that
this
best
performing
unit
is
based,
according
to
the
court,
on
a
level
of
performance
that
can
be
achieved
"
under
the
most
adverse
circumstance
which
can
reasonably
be
expected
to
occur."
Therefore,
the
"
best
controlled"
source
must
be
taken
into
account
to
predict
emissions
from
any
reasonable
situation,
including
different
lignites.

Response:

As
noted
above,
EPA
has
revised
the
new­
source
NSPS
Hg
emission
limits
based
on
a
reanalysis
of
the
available
information,
including
permits.
We
believe
that
the
revised
emission
limits
address
the
commenters'
concerns.

Comment:

One
commenter
(
OAR­
2002­
0056­
3478)
requested
that
any
emission
limits
for
lignitefired
power
plants
include
Gulf
Coast
lignite
as
a
separate
subcategory.
Based
upon
their
evaluation
of
the
tested
units,
the
commenter
requested
that
EPA
set
any
limit
for
Gulf
Coast
lignite­
fired
units
at
a
rate
no
less
than
28
lb/
TBtu.
The
commenter
stated
that
if
EPA
does
not
establish
a
separate
subcategory
for
Gulf
Coast
lignite
with
a
higher
standard,
then
a
percent
reduction
option
and/
or
a
"
safety
net"
must
be
offered
so
that
units
in
Texas,
Louisiana,
and
Mississippi
can
continue
to
utilize
locally
mined
lignite
as
fuel.

Two
commenters
(
OAR­
2002­
0056­
2915,
­
3463)
stated
that
Gulf
Coast
lignite
cannot
achieve
the
reductions
required
to
meet
the
proposed
standard
for
lignite
Because
it
was
set
with
units
firing
North
Dakota
lignite.
One
commenter
(
OAR­
2002­
0056­
2915)
stated
that
coal­
fired
utility
units
already
face
emissions
control
requirements
that
are
duplicative,
contradictory,
costly,
and
complex,
and
create
enormous
uncertainty
for
future
investment
and
that
adding
Hg
emissions
regulations
will
create
even
greater
challenges
for
coal­
fired
utility
units,
and
this
is
especially
true
for
Gulf
Coast
lignite
because
its
unique
physical
composition
makes
reductions
in
Hg
emissions
from
utility
units
firing
it
very
difficult
to
achieve
and
more
difficult
to
achieve
than
for
non­
lignite
coal­
fired
utility
units.

One
commenter
(
OAR­
2002­
0056­
4891)
added
that
the
proposed
Hg
rule
rules
should
be
revised
to
better
recognize
the
high
concentration
and
type
of
Hg
present
in
Gulf
Coast
lignite
and
the
difficulties
associated
with
controlling
its
Hg
emissions
as
compared
to
Hg
emissions
from
other
coal
ranks.
The
commenter
stated
that
without
significant
changes
to
the
proposed
Hg
rule
to
lessen
compliance
costs,
it
is
likely
that
most
Gulf
Coast
lignite
mines
and
some
Gulf
Coast
lignite­
fired
power
plants
will
ultimately
be
forced
to
close.
According
to
the
commenter,
power
3­
47
grid
stability
would
be
compromised
without
the
generation
capacity
provided
by
Gulf
Coast
lignite­
fired
power
plants,
resulting
in
the
potential
for
frequent
and
sustained
power
outages
that
would
undermine
economic
stability
and
prospects
for
economic
growth.

One
commenter
(
OAR­
2002­
0056­
3478)
believed
the
Hg
content
in
Gulf
Coast
lignite
is
higher
than
the
ICR
data
indicate,
a
more
accurate
analytical
method
for
Cl
in
coal
demonstrates
that
the
Cl
content
in
Gulf
Coast
lignite
is
much
lower.
The
commenter
also
believed
this
revised
information
on
Cl
content
may
explain
why
lignite
combustion
results
in
a
significant
percentage
of
elemental
Hg
being
emitted.

One
commenter
(
OAR­
2002­
0056­
2929)
expressed
concern
that
facilities
burning
Texas
lignite
will
be
unable
to
comply
with
the
proposed
Hg
emission
limits
because
the
best
performing
lignite
units
all
fire
cleaner­
burning
North
Dakota
lignite.

Response:

As
noted
elsewhere,
EPA
does
not
see
a
basis
at
this
time
for
further
subcategorizing
lignite
coals.

Comment:

Several
commenters
expressed
concern
over
the
impacts
resulting
from
the
stringent
rules
for
lignite
coals.
The
commenters
(
OAR­
202­
0056­
1692,
­
2915,
­
3510,
­
3543,
­
4891)
were
concerned
that
stringent
rules
for
lignite
coal
would
result
in
fuel
switching
and
have
negative
impacts
on
lignite­
burning
units.
One
commenter
(
OAR­
2002­
0056­
2915)
stated
that
EPA
must
ensure
that
the
Hg
rule
does
not
disadvantage
coal,
especially
Gulf
Coast
lignite,
because
doing
so
would
aggravate
the
already
precarious
natural
gas
supply
and
price
situation.
The
commenter
stated
that
if
the
Hg
rule
was
to
even
slightly
decrease
the
dependence
on
coal,
the
natural
gas
supply
and
the
price
problems
would
increase.
According
to
the
commenter,
it
is
estimated
that
forced
replacement
of
coal
with
natural
gas
as
fuel
in
electric
generation
would
increase
the
demand
for
natural
gas
by
about
35
percent
and
would
increase
natural
gas
prices
by
about
33
percent.
According
to
one
commenter
(
OAR­
2002­
0056­
3510),
natural
gas
is
not
available
to
utilities
in
the
winter
when
it
is
apportioned
to
residential
users.
One
commenter
(
OAR­
2002­
0056­
1692)
stated
that
the
proposed
lignite
limit
of
9.2
lb/
Btu
is
so
stringent
that
it
would
preclude
many
southern
lignite
coals
from
future
use
and
would
promote
the
use
of
natural
gas,
especially
in
smaller
plants,
where
the
high
cost
of
controls
may
not
be
justified
given
the
anticipated
life
of
the
plant.
According
to
another
commenter
(
OAR­
2002­
0056­
3543),
the
current
rule
structure
could
cause
generators
to
switch
coal
ranks,
primarily
from
lignite
and
subbituminous
to
bituminous
coal
with
resultant
economic
impacts.
Without
a
higher
limit
for
Gulf
Coast
lignite,
commenters
(
OAR­
2002­
0056­
3510,
­
4891)
stated
that
their
State
economies
will
suffer
because
lignite
is
an
important
fuel
in
their
State.
One
commenter
(
OAR­
2002­
0056­
2915)
stated
that
utility
units
designed
to
burn
lignite
cannot
easily,
quickly,
or
cheaply
switch
to
burn
other
fuel
types.
According
to
the
commenter,
lignite's
low
heat
content
and
its
other
properties
would
require
time
consuming
and
expensive
alterations
to
allow
them
to
burn
non­
lignite
fuels.
The
commenter
further
stated
that
lignite­
fired
utility
units
are
often
3­
48
parties
to
long­
term
contracts
to
purchase
the
lignite;
therefore,
even
if
such
utility
units
could
no
longer
burn
lignite,
they
would
still
be
required
to
purchase
it
pursuant
to
any
such
long­
term
contracts.
The
commenter
also
added
that
Gulf
Coast
lignite­
fired
utility
units
in
Texas
are
located
on
the
property
from
which
the
lignite
is
mined
and
that
for
many
such
units,
rail
lines
that
could
be
used
to
transport
other
types
of
fuel
to
the
site
would
have
to
be
constructed.

Response:

As
noted
elsewhere,
EPA
has
reanalyzed
the
new­
source
NSPS
limits
and
believes
that
the
emission
limit
being
finalized
is
achievable
and
appropriate.

Comment:

Several
comments
addressed
the
control
of
Hg
from
lignite­
fired
units.
Three
commenters
(
OAR­
2002­
0056­
3327,
­
3469,
­
4191)
expressed
concern
regarding
the
availability
of
proven
control
measures.
According
to
one
commenter
(
OAR­
2002­
0056­
3469),
the
lack
of
availability
on
the
market
of
proven
cost­
effective
control
and
monitoring
equipment
will
make
compliance
with
the
proposed
regulations
difficult
for
utilities,
particularly
those
burning
lignite.
The
commenter
supported
proposals
which
give
utilities
flexibility
in
how
they
implement
controls
and
comply
with
the
regulations
and
which
provide
incentives
to
comply
as
quickly
as
practical
and
make
the
most
cost­
effective
investments
that
provide
the
largest
emissions
reductions.
Thus,
the
commenter
supported
EPA's
proposal
to
allow
utilities
to
use
facility­
wide
averaging,
system­
wide
averaging,
and
use
12­
month
rolling
averages
to
calculate
emissions
and
demonstrate
compliance.
One
commenter
(
OAR­
2002­
0056­
3327)
was
concerned
that
despite
the
progress
made
in
their
State
and
efforts
to
continue
identifying
new
technologies
to
control
emissions
from
coal­
fired
power
plants,
the
imposition
of
the
proposed
requirements
will
force
the
closure
of
lignite­
fired
power
plants
prior
to
the
time
that
effective
emissions
control
technology
can
be
developed
and
made
commercially
available.

One
commenter
(
OAR­
2002­
0056­
3478)
cited
lignite
properties
in
addition
to
monitoring
technology
as
the
main
hindrance
to
pollution
control
companies
providing
a
Hg
removal
guarantee.
Citing
the
high
degree
of
elemental
Hg
remaining
in
flue
gases
of
lignite
as
well
as
lignite's
tendency
to
have
relatively
high
total
Hg
content,
the
commenter
believed
they
have
valid
concerns
that
Hg
control
for
lignite­
fired
boilers
will
be
more
difficult
and
costly
than
for
bituminous
coal­
fired
boilers.
According
to
the
commenter,
coal
analysis
from
one
mine
indicated
that
a
59
­
76
percent
reduction
in
Hg
emissions
would
be
required.
The
commenter
also
reports
that
similarly,
if
they
examine
the
70
percent
lignite/
30
percent
PRB
data,
a
weighted
average
limit
of
8.18
lbs/
TBtu
would
have
to
be
met.
According
to
the
commenter,
one
pollution
control
company
does
have
experience
with
an
ACI
system
supplier
that
have
given
guarantees
of
50
percent
removal
for
PRB
coals
at
a
carbon
consumption
rate
equivalent
to
an
expenditure
of
~$
5M
per
year
and
subject
to
very
specific
restrictions
and
very
limited
liability,
but
no
such
guarantees
to
date
have
been
given
for
lignite­
fired
plants.
The
commenter
further
stated
that
the
PRB
guarantees
to
date
have
been
predicated
upon
availability
of
necessary
quantities
of
suitable
activated
carbon,
total
amount
of
Hg
entering
the
system,
and
averaging
period
allowed
to
meet
guarantees.
According
to
the
commenter,
Hg
control
technologies
are
highly
coal
and
3­
49
boiler/
AQCS
configuration
dependent,
not
to
mention
the
issues
with
test
accuracy
when
measuring
Hg
with
a
concentration
six
orders
of
magnitude
less
than
SO2.
The
commenter
stated
that
it
will
only
be
after
multiple
demonstrations
have
been
completed
before
all
the
anomalies
are
sorted
out
in
order
for
suppliers
to
take
on
the
risk
of
Hg
removal
guarantees.

Response:

EPA
believes
that
the
regulatory
approach
being
taken
will
address
the
commenters'
concerns,
particularly
with
regard
to
the
flexibility
afforded
to
a
company.
The
flexibility
afforded
by
the
cap­
and­
trade
approach
will
preclude
any
concerns
about
having
to
arbitrarily
close
coal­
fired
utility
units
and
provide
the
time
necessary
to
fully
develop
the
emerging
Hgspecific
control
technologies.
Further,
EPA
believes
that
reliable,
cost­
effective
Hg
monitoring
systems
are
available
and
will
be
further
refined
by
the
time
utilities
must
be
in
compliance
with
the
revised
standards.

Comment:

One
commenter
(
OAR­
2002­
0056­
3478)
stated
that
regulations
to
control
SO2
and
NOx
will
require
the
installation
of
pollution
controls
that
will
also
capture
the
forms
of
Hg
that
tend
to
deposit
nearby.
The
commenter
stated
that,
based
on
testing
of
SCR
performance
for
Hg
cobenefits
on
a
lignite
facility
in
North
Dakota,
SCR
will
not
provide
much,
if
any,
Hg
co­
benefit
reduction.
According
to
the
commenter,
SCR
technology
is
ineffective
in
oxidizing
Hg
and
that
the
saltation
of
calcium
and
sodium
ash
deposits
fouls
the
catalyst
rendering
the
SCR
technology
ineffective
for
NOx
control.

Response:

EPA
believes
that
the
cap­
and­
trade
approach
being
taken
will
address
the
commenter's
concerns.
For
the
new­
source
NSPS
Hg
emission
limits,
EPA
has
not
assumed
any
removal
contribution
by
SCR
units
on
lignite
coal.

Comment:

One
commenter
(
OAR­
2002­
0056­
4891)
stated
that,
given
the
lack
of
scientific
evidence
linking
health
impacts
to
Gulf
Coast
lignite­
fired
power
plant
Hg
emissions
and
its
insignificant
contribution
to
Hg
emissions
relative
to
other
sources
and
the
global
Hg
emissions
pool,
there
is
no
present
justification
for
a
regulation
with
as
significant
an
economic
impact
as
the
proposed
Hg
rule.

Response:

EPA
sees
no
basis
for
excluding
Gulf
Coast
lignite
from
the
revised
standards.

Comment:
3­
50
One
commenter
(
OAR­
2002­
0056­
3398)
stated
that
North
Dakota
lignite
has
a
lower
Cl
content
than
subbituminous
or
bituminous
coal
and
that
Hg
control
from
lignite
is
much
more
difficult,
warranting
a
higher
emission
limit.

Response:

EPA
concurs
that
lignite
coal
exhibits
unique
combustion
and
control
characteristics
and,
as
such,
has
placed
lignite
in
a
separate
subcategory.

Comment:

Commenter
OAR­
2002­
0056­
5535
disagreed
that
the
best­
performing
lignite
units
fire
North
Dakota
lignite
and
that
North
Dakota
lignite
is
significantly
different
from
other
lignite.
First,
the
commenter's
analysis
of
the
best­
performing
units
indicates
that
TXU's
TNP­
One
unit,
which
burns
Texas
lignite
coal,
is
the
best­
performing
lignite
unit.
The
commenter
used
EPA's
methodology
to
estimate
Hg
emissions
for
every
coal
shipment
fired
by
TNP­
One.
When
these
estimates
are
averaged,
the
average
annual
emission
rate
is
1.29
lb/
TBtu
B
the
best
performance
of
any
lignite­
fired
unit.
Second,
although
it
is
true
that
Texas
lignite
has
a
higher
ash
content
than
North
Dakota
lignite
and
that
facilities
firing
Texas
lignite
are
among
the
biggest
Hg
emitters
in
the
U.
S.,
none
of
these
facilities
has
opted
to
participate
in
any
of
the
DOE­
sponsored
emissions
tests
aimed
at
evaluating
Hg
control
technologies.
The
Monticello
plant,
which
fires
Texas
lignite,
is
scheduled
to
be
tested
during
the
Phase
II
DOE
tests
(
mid­
2005),
but
the
test
plan
excludes
the
most
promising
technology
for
lower
ranks
coals
B
halogenated
activated
carbon
sorbents.
Consequently,
it
will
not
be
possible
to
compare
Monticello's
performance
with
that
of
North
Dakota
facilities,
which
have
been
tested
with
these
sorbents,
achieving
Hg
emission
reductions
in
excess
of
90
percent.
In
addition,
the
commenter
noted
that
one
facility
firing
Texas
lignite
B
the
Big
Brown
plant
B
has
been
operating
a
COHPAC
baghouse
for
a
number
of
years.
This
small
add­
on
fabric
filter
is
the
key
component
of
EPRI's
patented
TOXECON
process,
whereby
activated
carbon
is
injected
upstream
of
the
COHPAC.
Tests
of
this
configuration
on
low
sulfur
bituminous
coal
resulted
in
Hg
capture
averaging
86
percent
over
a
19­
week
period.
Use
of
a
COHPAC
with
a
halogenated
sorbent
could
result
in
very
high
Hg
capture
B
even
with
Texas
lignite
B
but,
unfortunately,
the
test
will
not
include
this
configuration.
If
EPA
were
to
establish
a
more
lenient
standard
for
Texas
lignites,
it
would
have
harmful
environmental
and
health
consequences.
One
outcome
of
a
more
lenient
standard
for
these
facilities
is
that
they
will
continue
to
emit
Hg
in
huge
amounts.
A
second
potential
outcome
of
a
higher
emission
rate
B
if
EPA
adopts
its
ill­
advised
trading
scheme
B
is
that
these
facilities
might
decide
to
reduce
their
Hg
emissions
using
the
most
promising
technologies,
and
then
to
bank
and
sell
a
large
number
of
Hg
allowances,
thereby
allowing
other
polluters
to
avoid
controls.
Thus,
convincing
EPA
that
they
are
unable
to
control
their
Hg
emissions
(
in
the
absence
of
any
data
substantiating
that
assertion)
is
clearly
in
the
financial
interest
of
these
companies
and
against
the
interests
of
public
health
and
welfare.

Response:

As
noted
earlier,
EPA
continues
to
believe
that
placing
lignite
in
a
separate
subcategory
3­
51
is
warranted
but
that
further
subcategorization
into
Fort
Union
and
Gulf
Coast
lignites
is
not
necessary.
The
revised
new­
source
NSPS
Hg
limits
incorporate
data
from
both
types
of
lignite
and,
thus,
are
believed
to
be
representative
and
appropriate.
Further,
as
noted
earlier,
EPA
does
not
believe
that
Hg­
specific
control
technologies
are
currently
available
for
use
as
the
basis
of
a
national
Hg
standard.
We
believe
that
the
declining
cap
under
the
cap­
and­
trade
approach
being
finalized
will
ensure
both
development
of
the
emerging
Hg­
specific
control
technologies
and
continued
Hg
emission
reductions
by
all
utility
units
in
the
most
efficient
manner.

3.2.4.6
Coal
Refuse­
fired
Units
Comment:

One
commenter
(
OAR­
2002­
0056­
2068)
requested
that
the
EPA
not
set
emission
limits
for
new
and
existing
coal
refuse­
fired
plants
so
as
to
ensure
that
any
limit
is
achievable
and
takes
into
account
the
wide
variability
within
this
important
fuel
supply.

Response:

The
current
subpart
Db
emission
limits
for
PM,
SO2,
and
NOx,
are
applicable
to
coal
refuse­
fired
units
(
with
an
existing
definition
of
"
coal
refuse")
and
EPA
sees
no
basis
to
exclude
such
units
from
the
Hg
emission
limits.
EPA
believes
that
the
revised
new­
source
NSPS
Hg
emission
limits
for
these
sources
are
achievable
and
appropriate.

Comment:

Three
commenters
(
OAR­
2002­
0056­
1766,
­
2162,
­
5495)
opposed
the
proposed
emission
limits
for
coal
refuse­
fired
units.
The
proposed
limits
are
more
than
five
times
more
stringent,
on
a
lb/
TBtu
basis,
than
the
proposed
limits
for
the
next
most
stringently­
regulated
coal­
fired
source
category.
The
EPA
selected
this
proposed
standard
based
upon
limited
data
from
only
two
waste
coal­
fired
sources.
Such
an
insignificant
amount
of
data
is
an
insufficient
basis
upon
which
to
promulgate
emission
limits.
Also,
the
EPA
did
not
appropriately
consider
the
variability
inherent
in
the
waste
coal
fuel
source.
Specifically,
the
characteristics
of
waste
coal
vary
to
a
much
greater
extent
than
other
coal
ranks.
Finally,
the
commenters
stated
that
in
addition
to
being
inequitable
and
based
on
inadequate
data,
the
proposed
emission
limit
for
waste
coal­
fired
sources
may
not
be
achievable.
Commenter
OAR­
2002­
0056­
5495
thought
the
measured
emissions
from
the
test
data
used
to
set
the
limit
were
abnormally
low
due
to
a
variety
of
factors.

Response:

The
revised
emission
limits
(
as
noted
earlier)
for
coal
refuse­
fired
units
are
based
on
data
from
units
within
the
coal
refuse
subcategory.
EPA
believes
that
the
stringency
of
the
limits
accurately
reflects
the
performance
on
Hg
emissions
of
controls
used
on
such
units.
No
data
were
provided
during
the
public
comment
period
that
refuted
the
relative
levels
of
control
achievable
by
coal
refuse­
fired
units
as
evidenced
by
the
Hg
emission
limits
established
in
the
rule.
Further,
EPA
did
consider
the
variability
inherent
in
the
fuel
source
in
arriving
at
the
final
3­
52
emission
limits.
EPA
disagrees
with
the
commenters
regarding
the
level
of
variability
found
in
coal
refuse
related
to
other
coal
ranks.
Some
constituents
(
e.
g.,
ash,
Btu
content)
do
exhibit
wider
variability,
as
would
be
expected
given
the
nature
of
the
fuel
source.
However,
other
constituents
(
e.
g.,
sulfur,
Hg)
exhibit
similar
or
less
variability
than
do
other
coal
ranks.

Comment:

One
commenter
(
OAR­
2002­
0056­
2261)
stated
that
the
emissions
limits
for
coal
refuse­
fired
units
must
be
consistent
with
the
current
levels
of
Hg
emissions
from
each
of
these
sources
to
ensure
that
all
coal
refuse­
fired
sources
could
comply
with
such
levels,
notwithstanding
the
inherently
variable
characteristics
of
the
waste
coal
source.
The
commenter
recommended
that
any
such
limit
should
be
reflective
of
a
90
percent
reduction
in
Hg,
based
upon
the
Hg
content
in
the
coal
refuse
prior
to
combustion,
as
measured
by
inlet
and
outlet
concentrations
evaluated
during
biennial
performance
testing.
The
commenter
stated
that
such
a
limit
would
be
consistent
with
the
effective
Hg
control
achieved
by
coal
refuse
sources.

Response:

EPA
sees
no
basis
for
providing
coal
refuse­
fired
units
a
different
compliance
approach
than
for
other
subcategories.
Therefore,
the
12­
month
rolling
average
and
continuous
monitoring
requirements
have
been
maintained
in
the
revised
standards.
The
revised
standards
are
believed
reflective
of
the
expected
performance
of
coal
refuse­
fired
units.

Comment:

One
commenter
(
OAR­
2002­
0056­
3560)
stated
that
the
EPA
did
not
gather
information
concerning
a
non­
CFB
unit
that
burns
coal
refuse.
Therefore,
the
proposed
Hg
emission
limit
for
coal­
refuse
units
cannot
be
justifiably
applied
to
a
cyclone
unit
burning
at
least
25
percent
coal
refuse
and
the
rest
of
the
fuel
input
is
essentially
bituminous
coal.

One
commenter
(
OAR­
2002­
0056­
2826)
points
out
that
their
member
electric
cooperative
burns
Illinois
basin
bituminous
waste
coal.
Along
with
their
electric
cooperative,
the
commenter
believes
that
EPA's
data
from
units
burning
waste
coals
and
the
EPA's
related
analysis
are
neither
complete
nor
representative
of
the
emission
characteristics
of
this
coal
rank.
The
commenter
adds
that
there
is
apparently
no
information
in
the
EPA
database
for
a
cyclone
unit,
such
as
their
member
cooperative's
Unit
4,
which
burns
a
coal
waste
product
as
a
significant
portion
of
its
fuel
input.
The
commenter
respectfully
requests
that
EPA
address
the
above
issues
as
it
finalizes
its
proposed
rule.

One
commenter
(
OAR­
2002­
0056­
2261)
observed
that
the
proposed
rules
establish
Hg
emission
rates
based
on
rank
of
coal,
including
waste
coal,
being
burned.
The
commenter
does
not
believe
that
the
rates
for
existing
and
new
units
are
reflective
of
either
new
or
existing
technology
used
to
burn
waste
coal.
According
to
the
commenter,
EPA
must
consider:

(
1)
the
difference
and
variation
in
the
chemical
quality
of
waste
coal
from
different
coal
3­
53
fields
in
different
parts
of
the
country;

(
2)
the
technology
used
to
clean
the
coal
producing
the
waste
coal;

(
3)
the
percent
SO2
reduction
required
at
the
different
waste
coal
plants
and
its
impact
on
heat
input
and
Hg
emissions;
and
(
4)
NOx
controls
being
implemented.

Response:

Such
data
as
noted
by
the
commenters
(
e.
g.,
non­
CFB
units
burning
coal
refuse;
data
on
coal
refuse
from
different
parts
of
the
country)
were
not
made
available
to
EPA
during
the
public
comment
period.
Units
co­
firing
coal
refuse
and
other
coal
ranks
would
be
subject
to
the
provisions
applicable
to
units
that
blend
coal
ranks.
EPA
believes
that
future
units
constructed
to
combust
coal
refuse
will
be
similar
in
nature
to
existing
units
and,
thus,
that
the
final
emission
limits
are
reflective
of
units
that
will
combust
coal
refuse.

3.2.4.7
IGCC
Units
Comment:

One
commenter
(
OAR­
2002­
0056­
2721)
stated
there
has
not
been
a
full­
scale
demonstration
of
sorbent
bed
technology
on
IGCC
units
with
lignite
or
subbituminous
coal.
The
commenter
noted
that
the
process
addressed
by
EPA
in
the
preamble
for
the
proposed
rule
is
for
an
industrial
facility
firing
bituminous
coal
and
producing
a
synthetic
gas
(
syngas)
that
is
cooled
to
about
100

F.
According
to
the
commenter,
not
all
IGCC
units
have
a
gas
stream
with
a
temperature
that
low.
The
commenter
stated
that
sorbent
beds
do
not
work
when
temperatures
are
several
hundred
degrees.
The
commenter
believes
there
is
no
justifiable
basis
to
use
this
technology
for
setting.
performance
standards.

Response:

EPA
agrees
with
the
commenter
that
no
IGCC
unit
utilizing
syngas
produced
from
subbituminous
or
lignite
coals
has
been
operated
with
a
carbon
bed
for
Hg
removal.
However,
IGCC
units
have
been
operated
on
subbituminous
and
lignite
coals
in
the
U.
S.
(
OAR­
2002­
0056­
5684).
Application
of
the
carbon
bed
to
IGCC
units
burning
such
coals
would
not
present
any
additional
technical
obstacles.
EPA
also
concurs
that
the
optimal
temperature
for
carbon
bed
utilization
is
around
100

F.
However,
EPA
disagrees
with
the
commenter's
inference
that
this
temperature
is
not
found
in
some
IGCC
applications.
The
DOE
conducted
a
feasibility
study
to
evaluate
the
cost
of
removal
of
Hg
from
IGCC
units
(
OAR­
2002­
0056­
5685).
This
study
found
that
the
optimal
location
for
the
carbon
bed
was
between
the
fuel
gas
cooling/
knockout
unit
and
the
acid
gas
removal
unit
(
and
prior
to
the
combustion
turbine),
a
location
likely
to
be
found
in
any
new
IGCC
installation.
This
location
affords
temperatures
close
to
the
optimal
100

F.
Therefore,
EPA
believes
that
carbon
bed
technologies
are
3­
54
appropriate
for
use
on
new
IGCC
installations.

Comment:

One
commenter
(
OAR­
2002­
0056­
3459)
stated
that
ACI
should
be
the
basis
of
the
standard
for
IGCC
units.
DOE
has
concluded
Hg
controls
are
available
and
applicable
to
IGCC
units.
The
technology
is
already
commercially
demonstrated
to
remove
greater
than
90
percent
Hg
removal
and
was
specifically
applicable
to
gasification
systems
using
high­
temperature
slagging
gasifiers
and
bituminous
coal,
which
includes
both
of
the
IGCC
plants
currently
in
operation.
EPA
must
establish
emission
limits
that
reflect
at
least
90
percent
reduction
in
Hg
emissions
from
IGCC
units.
Based
on
commenter's
analysis,
the
rate
should
be
0.49
lb/
TBtu
or
3.9
x
10­
6
lb/
MWh
(
output­
based
standard
based
on
42.5
percent
efficiency
and
conversion
factor
for
mass/
1012
Btu
to
mass/
MWh
at
42.5
percent
efficiency
is
8
x
10­
6
TBtu/
MWh).

Two
commenters
(
OAR­
2002­
0056­
1852,
­
2160)
opposed
separate
limits
for
new
or
existing
IGCC
units.
According
to
one
commenter
(
OAR­
2002­
0056­
2160),
Hg
capture
from
syngas
has
been
proven
to
be
readily
available
and
inexpensive
at
the
Eastman
Chemical
coal
gasification
plant
in
Kingsport,
TN.
There
is
no
reason
to
establish
separate
and
overly
lenient
limits
for
the
two
existing
U.
S.
plants
or
for
any
new
plants.

Response:

The
existing
units
will
be
covered
under
the
cap­
and­
trade
portion
of
the
final
CAA
section
111
rulemaking
and,
thus,
will
be
subject
to
their
respective
State's
Hg
budget.
Any
new
units
would
be
subject
to
the
more
restrictive
new
source
NSPS
Hg
emission
limit
which
is
based
on
the
use
of
a
carbon
bed
as
at
the
Tennessee
facility.

Comment:

One
commenter
(
OAR­
2002­
0056­
4139)
stated
that
the
WEST
Associates
statistical
model
should
not
have
been
applied
in
developing
the
emission
limits
for
IGCC
units.
Both
Cl
and
Hg
would
be
removed
separately
from
the
coal
as
part
of
the
coal
gasification
process.
Therefore,
Cl
and
Hg
could
not
interact
as
they
are
not
present
at
the
same
quantitative
levels
as
in
the
combustion
process
of
coal­
fired
boilers.
This
leads
to
an
artificially
high
limit
for
new
sources.

Response:

Analysis
of
variability
is
appropriate
for
IGCC
units.
It
is
true
that
Hg
and
Cl
would
be
removed
separately
in
an
IGCC
process;
however,
based
on
information
available
to
EPA
(
OAR­
2002­
0056­
5685),
the
Hg
would
likely
be
removed
prior
to
the
removal
of
the
Cl,
and,
thus,
the
Hg­
Cl
interaction
presumed
in
the
current
statistical
analysis
would
still
be
valid.

3.2.4.8
Blended
Fuel­
fired
Units
3­
55
Comment:

One
commenter
(
OAR­
2002­
0056­
2198)
stated
that
the
proposed
limits
for
subbituminous
coal
are
less
stringent
than
those
for
bituminous
coal.
A
unit
that
previously
burned
bituminous
coal
could
choose
to
blend
or
switch
to
subbituminous
coal
and
emit
more
Hg
proportional
to
the
ratio
of
subbituminous
to
bituminous
coal.
Because
many
units
can
burn
either
of
the
two
ranks
of
coal,
this
commenter
does
not
support
this
approach.
The
EPA
should
require
sources
that
burn
a
blend
of
coal
to
be
subject
to
the
more
stringent
limit
regardless
of
the
coal
or
coal
mixture
being
burned
at
any
time.

One
commenter
(
OAR­
2002­
0056­
2889)
stated
the
proposed
rule
would
allow
units
that
blends
coal
ranks
to
average
the
standards
for
those
coal.
Facilities
must
compute
the
weighted
average
limit
based
on
the
proportion
of
energy
input
(
Btu)
contributed
by
each
coal
rank
burned
during
the
compliance
period.
However,
the
rule
does
not
specify
how
this
is
to
be
done.
Because
blending
is
typically
done
with
a
bulldozer,
the
quantity
of
each
fuel
is
not
determined
with
any
precision.
To
avoid
inaccuracies
inherent
in
computing
a
limit
for
blended
coal,
the
rule
should
require
the
facility
to
meet
the
most
stringent
standard
of
the
fuels
combusted.

One
commenter
(
OAR­
2002­
0056­
3449)
stated
that
limits
for
blended
coals
should
not
be
prorated.
Under
the
proposal,
50/
50
blending
of
subbituminous
coal
with
bituminous
coal
would
increase
allowable
emissions
for
a
facility
previously
burning
only
bituminous
coal
by
almost
2
times.
This
type
of
coal
blending
is
already
popular
for
reducing
SO2
and
NOx
emissions;
the
additional
benefit
of
lower
Hg
emissions
coupled
with
a
higher
limit
would
further
increase
the
incentive.
The
fact
that
coal
blending
provides
the
benefit
of
a
higher
limit
is
a
clear
rationale
that
EPA
should
adopt
one
limit
for
subbituminous
and
bituminous
coal.
At
the
very
least,
the
lowest
applicable
standard
should
apply
to
blended
coal,
not
a
prorated
higher
standard.
The
commenter's
experience
demonstrates
that
blended
coal
does
not
need
a
higher
limit.
Blending
should
be
encouraged
to
reduce
emissions,
not
increase
allowable
emissions.

Response:

EPA
believes
that
it
is
appropriate
to
use
the
prorated
approach
as
this
is
consistent
with
that
already
included
in
subpart
Da.

Comment:

One
commenter
(
OAR­
2002­
0056­
2247)
disagreed
with
EPA's
description
of
the
importance
of
coal
rank
at
units
such
that
fuel
switching
will
not
occur.
Given
the
demands
of
further
NOx
and
SO2
reductions,
fuel
switching
will
continue,
the
EPA's
proposal
to
proportion
the
Hg
emissions
limit
between
subbituminous
and
bituminous
(
or
between
subbituminous
and
lignite)
will
result
in
emissions
increases
for
the
unit
that
switches.
Rather
than
apportioning
the
limit
between
the
percent
of
fuel
burned,
the
limit
should
apply
for
the
fuel
used
in
the
majority.
Thus,
any
facility
using
up
to
50
percent
bituminous
coal
with
subbituminous
coal
should
meet
a
limit
for
bituminous
coal.
Because
subbituminous
coal
has
lower
overall
Hg
content
and
bituminous
appears
to
have
sufficient
Cl
or
oxidize
Hg
for
downstream
capture,
a
unit
blending
3­
56
coal
should
not
have
difficulty
meeting
the
limit.

Response:

EPA
believes
that
it
is
appropriate
to
use
the
prorated
approach
as
this
is
consistent
with
that
already
included
in
subpart
Da.

Comment:

One
commenter
(
OAR­
2002­
0056­
2897)
believed
blending
of
different
coal
ranks
can
also
be
readily
accommodated
under
a
rule
that
includes
subcategorization.
Of
the
two
options
presented,
the
commenter
recommended
that
the
EPA
proceed
with
a
weighted
average
standard,
as
this
will
be
less
susceptible
to
"
gaming."

One
commenter
(
OAR­
2002­
0056­
2922)
supported
EPA's
decision
to
utilize
a
blended
emissions
limit
rather
than
attempt
to
establish
a
separate
subcategory
for
blended
fuel
units,
or
to
classify
a
unit
based
on
the
predominant
coal
it
combusts.

One
commenter
(
OAR­
2002­
0056­
2900)
urged
EPA
to
retain
the
five
subcategories
identified
in
the
proposal.
The
commenter
stated
that
the
failure
to
do
so
would
have
an
immediate
impact
on
the
balance
of
coal
ranks
burned
in
the
U.
S.
and
would
jeopardize
the
nation's
fuel
diversity.
For
the
same
reasons,
the
commenter
supported
the
Agency's
proposed
approach
of
addressing
units
burning
blended
coal
by
weighting
the
applicable
Hg
limit
according
to
the
amounts
of
the
different
coals
that
are
burned.

One
commenter
(
OAR­
2002­
0056­
2922)
supported
EPA's
decision
to
utilize
a
blended
emissions
limit
rather
than
attempt
to
establish
a
separate
subcategory
for
blended
fuel
units,
or
to
classify
a
unit
based
on
the
predominant
coal
it
combusts.

Response:

EPA
concurs
with
the
weighted
average
approach
endorsed
by
the
commenters.

Comment:

One
commenter
(
OAR­
2002­
0056­
2900)
supported
the
EPA's
proposed
approach
for
a
unit
burning
a
blend
of
coals
and
a
supplemental
fuel
that
the
supplemental
fuel
would
not
be
taken
into
account
for
purposes
of
determining
the
unit­
specific
emission
limit
that
the
unit
must
meet.
However,
the
supplemental
fuel's
heat
input
and
Hg
emissions
would
be
considered
in
determining
the
unit's
compliance
with
the
emission
limit.
The
commenter
requested
that
the
regulatory
text
clearly
reflects
EPA's
preamble
language
on
this
issue.
In
addition,
the
commenter
requested
that
EPA
clarify
that
units
burning
a
single
coal
rank
and
a
supplemental
fuel
would
be
treated
the
same
as
units
burning
a
coal
blend
and
a
supplemental
fuel.
That
is,
the
supplemental
fuel
would
not
be
considered
in
determining
the
emission
limit
to
which
the
unit
is
subject.
The
commenter
explains
the
unit
would
be
subject
to
the
emission
limit
for
the
coal
rank
3­
57
it
is
combusting;
however,
for
compliance
purposes,
the
heat
input
and
Hg
emissions
from
the
supplemental
fuel
would
be
taken
into
account.

Response:

EPA
believes
that
the
final
regulatory
language
is
clear.

Comment:

One
commenter
(
OAR­
2002­
0056­
3517)
pointed
out
that
there
has
been
little
in­
depth
study
of
plants
that
burn
coal
blends,
and
how
that
might
impact
or
benefit
Hg
removal.
The
commenter
stated
that
a
case
in
point
is
the
Valmont
Plant
in
Colorado
which
burns
a
blend
of
low
Hg,
low
Cl
bituminous
and
subbituminous
coal,
both
of
which
are
mined
in
Colorado.
According
to
the
commenter,
although
it
is
recognized
that
these
coals
are
low
in
Hg,
there
is
still
significant
Hg
reduction
that
occurs.
The
commenter
request
that
EPA
seek
further
information
on
coal
blending
as
a
potential
option
in
addressing
Hg
reductions
between
the
close
of
the
comment
period
and
the
issuance
of
the
final
rule.
The
commenter
stated
they
would
like
to
retain
the
option
to
provide
additional
information
on
this
topic
as
it
becomes
available.

Response:

The
impact
of
intentional
coal
blending
for
the
purpose
of
Hg
removal
is
being
investigated
under
the
DOE
Hg
research
program.
EPA
believes
that
this
approach
is
yet
another
option
that
facilities
may
have
in
achieving
compliance
with
the
final
emission
limits.

Comment:

Two
commenters
(
OAR­
2002­
0056­
1848,
­
2108)
expressed
concern
about
compliance
burden
on
sources
that
blend
coal
and
the
State
agencies
that
regulate
them.
According
to
the
commenters,
there
are
no
industry­
wide
blending
procedures
and
the
lack
of
specificity
will
lead
to
an
inaccurate
accounting
of
emissions.

Response:

EPA
believes
that
the
States
are
familiar
with
the
weighted
average
approach
being
used
in
the
final
rule
as
it
is
currently
a
part
of
subpart
Da
for
other
pollutants.

3.2.4.9
Cogeneration
Units
Comment:

One
commenter
(
OAR­
2002­
0056­
2906)
stated
that
the
proposed
emission
rate
calculation
for
cogeneration
units
appears
to
unfairly
penalize
these
units
for
sales
of
any
electric
power
less
than
the
full
generation
capacity,
contrary
to
the
EPA's
stated
intent
to
advance
the
application
of
cogeneration
facilities
and
thereby
improve
the
nation's
energy
efficiency
and
3­
58
achieve
greenhouse
gas
emission
intensity
reductions.
In
both
the
proposed
rule
and
preamble,
EPA
applies
the
18
CFR
292.205
efficiency
methodology
to
cogeneration
facilities
(
implied
to
be
limited
to
solid
fuel­
fired
facilities
because
gas­
fired
units
are
not
included
in
the
rule
applicability)
(
69
FR
4696
and
69
FR
4762).
Application
of
that
methodology
appears
to
penalize
those
cogeneration
facilities
that
sell
only
a
portion
of
their
total
net
electricity
generation
output
to
the
grid.
Irrespective
of
the
comments
relating
to
the
need
for
annual
net
sales
of
electricity
to
the
grid,
the
EPA
approach
would
restrict
the
total
output
of
energy
in
the
denominator
to
only
that
electricity
sold
to
the
grid
plus
one­
half
of
the
net
steam
output
of
the
unit,
assuming
that
any
energy
input
that
is
not
utilized
through
electricity
sales
is
used
as
steam
output
(
69
FR
4762).
However,
this
penalizes
a
facility
for
using
any
electricity
generated
in
the
cogeneration
facility
within
the
manufacturing
facility.
Many
cogeneration
facilities
are
located
within
manufacturing
plants
that
use
most,
or
all
of
the
generated
electricity.
The
cogeneration
unit
only
sells
to
the
grid
the
excess
power
on
an
as­
available
basis
in
order
to
maintain
optimum
overall
system
efficiency.
For
example,
the
methodology
in
the
proposed
rule
was
applied
to
a
typical
coal­
fired
cogeneration
facility.
At
a
constant
Hg
lb/
hr
emission
rate,
constant
heat
input
to
the
boiler,
and
constant
electricity
generation
rate,
the
calculated
emission
rate
on
a
lb/
MWh
basis
would
vary
with
the
quantity
of
electricity
sold
to
the
grid.
For
this
example,
the
lb/
MWh
calculated
emission
rate
would
be
70
percent
higher
when
selling
25
MWe
to
the
grid
than
when
selling
100
percent
of
generated
electricity
to
the
grid.
This
equation
unfairly
penalizes
cogeneration
facilities.
The
EPA
needs
to
provide
full
consideration
of
the
complexities
of
cogeneration
units
when
trying
to
develop
and
utilize
output
based
emission
limits.
An
equitable
and
workable
solution
would
be
to
follow
past
EPA
practice
in
establishing
emissions
standards
and
allow
cogeneration
facilities
the
ability
to
use
input
based
emission
limits
and
calculations.
With
this
approach,
the
boiler,
fuels,
and
emissions
controls
will
determine
compliance
without
the
apparent
emission
rate
being
unfairly
skewed
by
the
portion
of
electricity
sold
to
the
grid.
The
EPA
should
establish
emissions
standards
that
encourage
installation
and
operation
of
highly
efficient
cogeneration
facilities,
and
recognize
their
inherent
variability
in
design
and
operating
profiles
versus
typical
single
use
electric
utility
units.

Response:

The
commenter
appears
to
believe
that,
for
a
cogeneration
unit
classified
as
a
utility
generating
unit,
reducing
the
percentage
of
electricity
sold
to
the
grid
increases
the
emission
rate
for
regulatory
purposes.
The
commenter
provided
no
calculations
to
support
this
contention,
but,
in
any
event,
we
believe
the
comment
to
be
incorrect.
If
a
unit's
entire
rated
output
is
sold
to
the
grid,
the
unit
would
be
charged
with
emissions
equivalent
to
full
load.
If
the
unit
sells
half
of
its
rated
output
to
the
grid
and
the
other
half
is
used
internally
(
either
as
steam
or
as
electricity),
the
unit
would
be
charged
only
75
percent
of
its
entire
emissions
output
(
all
of
the
emissions
from
the
50
percent
sold
to
the
grid
plus
half
of
the
emissions
from
the
remaining
energy
used
internally,
or
25
percent,
which
totals
75
percent.

The
boiler
emits
Hg
for
all
the
coal
burned,
whether
the
output
is
sold
or
used
internally.
EPA's
proposed
rule
was
formulated
to
be
consistent
with
State
implementation
plan
(
SIP)
rules
for
NOx
emissions
under
the
provisions
of
the
Acid
Rain
program
and
with
the
revised
NOx
emissions
limits
under
subpart
Da.
3­
59
Comment:

One
commenter
(
OAR­
2002­
0056­
2913)
stated
that
the
proposed
output­
based
limits
should
be
modified
to
take
into
account
the
Hg
emissions
resulting
from
the
combustion
of
fuel
for
co­
generating
steam
for
uses
other
than
electricity
production.

Response:

The
commenter
implies
that
no
emissions
limit
exists
for
fuel
burned
to
produce
steam
not
used
for
electricity
sold
to
the
grid.
This
assertion
is
incorrect.
Emissions
from
fuel
used
for
cogenerating
steam
are
generated
at
the
same
time
as
those
emissions
resulting
from
fuel
used
for
electricity
generation.
However,
for
purposes
of
determining
the
"
output"
from
the
unit,
credit
for
the
steam
is
given
a
50
percent
credit,
versus
100
percent
credit
for
electricity.
As
stated
earlier,
this
policy
is
consistent
with
SIP
rules
for
NOx
emissions
under
the
provisions
of
the
Acid
Rain
program
and
with
the
revised
NOx
emissions
limits
under
subpart
Da.

3.2.5
Emissions
Limit
Averaging
Period
Comment:

Many
commenters
(
OAR­
2002­
0056­
1803,
­
1969,
­
2067,
­
2365,
­
2535,
­
2634,
­
2661,
­
2721,
­
2827,
­
2867,
­
2900,
­
2918,
­
2922,
­
3403,
­
3432,
­
3444,
­
3463,
­
3478,
­
3509,
­
3513,
­
3514,
­
3539,
­
4891)
supported
EPA's
proposal
to
determine
compliance
with
emissions
standards
based
on
a
12­
month
averaging
period.
Reasons
cited
by
the
commenters
included:

(
1)
the
large
variability
in
coal
Hg
content.
plant
operations,
and
control
technology
performance;

(
2)
Hg
is
not
an
acute
health
hazard
and
concerns
about
Hg
arise
from
long­
term
chronic
exposure;
and
(
3)
the
compliance
period
would
provide
greater
certainty
that
units
will
consistently
meet
the
limits,
particularly
given
that
operational
and
material­
related
variability
beyond
the
control
of
the
owner/
operator
can
impact
emission
levels.

Response:

EPA
concurs
with
this
comment.

Comment:

Several
commenters
(
OAR­
2002­
0056­
1969,
­
2260,
­
2721,
­
2830,
­
2835,
­
2850,
­
2918,
­
3449)
supported
using
a
12­
month
averaging
period
but
disagreed
with
using
the
proposed
averaging
method
for
computing
the
12­
month
average
based
on
averaging
monthly
average.
Instead,
the
commenters
recommended
that
the
averaging
method
be
revised
to
a
simple
average
3­
60
of
all
valid
hourly
data
from
the
previous
12
months.
The
commenters
stated
that
this
revision
would
equally
weight
all
valid
data
over
the
12­
month
period.
The
proposed
rolling
monthly
average
calculation
method
would
place
disproportionate
weighting
on
hour
values
in
months
that
include
extended
periods
of
lower
loads,
load
following,
or
other
operational
variances.

Response:

To
address
the
commenters'
concerns,
the
final
rule
requires
the
12­
month
rolling
averages
to
be
computed
on
a
weighted
basis.
The
rule
requires
valid
Hg
emissions
data
to
be
obtained
for
at
least
75
percent
of
the
unit
operating
hours
in
each
month
in
which
the
unit
operates.
For
each
operating
month,
a
monthly
average
Hg
emission
rate
is
calculated,
which
weights
all
of
the
hours
of
valid
data
equally.
However,
when
the
12­
month
rolling
average
is
calculated,
each
monthly
average
Hg
emission
rate
is
weighted
according
to
the
number
of
valid
hours
of
data
collected
in
that
month.
This
ensures
that
the
Hg
emission
rate
for
a
month
with
few
unit
operating
hours
is
not
counted
the
same
as
the
emission
rate
for
a
month
in
which
the
unit
is
in
continuous
operation.
For
any
month
in
which
less
than
75
percent
of
the
Hg
emission
data
is
captured,
the
rule
requires
a
substitute
Hg
emission
rate
to
be
reported,
and
in
the
rolling
average,
the
substitute
emission
rate
is
weighted
according
to
the
number
of
unit
operating
hours
in
that
month.

Comment:

One
commenter
(
OAR­
2002­
0056­
3210)
objected
to
the
12­
month
average
compliance
determination
because
it
further
weakens
the
proposed
standards.
The
purpose
of
the
12­
month
average
is
to
adjust
for
variability
in
the
process,
fuel
source,
etc.
However,
variability
is
already
overstated
in
the
proposed
emission
limits
resulting
in
many
units
being
able
to
avoid
controls.
The
12­
month
average
would
be
acceptable
if
the
proposed
standards
were
significantly
more
stringent.

Response:

EPA
disagrees
that
the
12­
month
rolling
average
format
weakens
the
standards.
Further,
as
noted
above,
the
new­
source
emission
limits
have
been
revised
which
may
also
address
the
comment.

Comment:

Two
commenters
(
OAR­
2002­
0056­
2835,
­
2922)
noted
that
40
CFR
60.50a(
h)(
1),
which
covers
calculation
of
mass
Hg
emissions
from
Hg
CEMS
and
Method
324,
calls
for
calculating
the
"
arithmetic
average
of
all
weekly
emission
rates
for
[
Hg]
for
the
12
successive
calendar
months."
Subsequent
subsections
refer
to
calculation
of
Hg
mass
emissions
"
over
a
month"
from
CEMS
and
over
the
"
emission
rate
period"
from
Method
324.
It
is
not
clear
why
40
CFR
60.50a(
h)(
1)
refers
to
calculation
of
weekly
rates
or
how
those
rates
fit
into
the
more
specific
calculations.
The
EPA
needs
to
correct
this
discrepancy.
3­
61
Response:

Proposed
Method
324
has
been
renamed
as
appendix
K
to
40
CFR
part
75
and
the
method
and
regulatory
text
has
been
clarified.

Comment:

One
commenter
(
OAR­
2002­
0056­
2889)
stated
that
a
correction
is
needed
is
40
CFR
60.45a(
a).
This
section
refers
to
a
12­
month
rolling
average
and
conflicts
with
40
CFR
60.45(
a)(
5)
which
refers
to
a
monthly
limit.

Response:

The
commenter
is
correct
in
that
the
compliance
monitoring
period
is
based
on
a
12­
month
rolling
average.
However,
in
order
to
arrive
at
this
average,
monthly
averages
must
be
established
on
a
continuous
basis.
Thus,
EPA
believes
that
40
CFR
60.45(
a)(
5)
is
correct
as
stated.

3.2.6
Emissions
Averaging
Comment:

Several
commenters
(
OAR­
2002­
0056­
2634,
­
2830,
­
2835)
requested
that
the
EPA
include
facility­
wide
averaging
in
the
section
111
Emission
Guidelines
for
existing
sources
as
an
additional
compliance
alternative.
States
should
be
encouraged
to
allow
such
flexible
compliance
alternatives
if
states
decline
to
adopt
a
section
111
trading
program,
if
that
option
is
selected
by
the
EPA.
By
including
this
type
of
flexibility
mechanism,
the
EPA
will
ensure
that
those
facilities
located
in
States
opting
out
of
the
trading
program
will
retain
some
degree
of
flexibility
when
complying
with
the
requirements
of
the
Emission
Guidelines.
Similarly,
facility­
wide
emissions
averaging
provides
a
flexible
compliance
alternative
to
a
cap­
and­
trade
program
in
the
event
that
neither
cap­
and­
trade
option
can
be
authorized
under
the
statute.
Varying
operational
modes
or
combination
of
systems,
e.
g.,
wet/
dry
scrubber,
ESP
or
fabric
filter,
could
be
employed
to
provide
the
greatest
potential
to
economically
reduce
Hg
emissions
to
meet
compliance
requirements.

Response:

States
and
Tribes
are
free
to
allocate
their
Hg
budgets
as
they
see
fit,
whether
they
participate
in
the
nationwide
trading
program
or
not,
as
long
as
the
reductions
are
achieved
from
coal­
fired
Utility
Units.
We
believe
that
this
will
address
the
commenter's
concerns.

Comment:

Several
commenters
(
OAR­
2002­
0056­
2900,
­
3432)
supported
allowing
facilities
with
both
industrial
boiler
units
and
coal­
fired
utility
units
to
opt
the
industrial
boiler
units
into
the
electric
utility
rule
for
purposes
of
meeting
the
emissions
standard.
The
commenter
believes
a
3­
62
final
rule
should
allow
affected
facilities
with
both
industrial
boilers
and
coal­
fired
utility
units
the
compliance
flexibility
to
meet
one
Hg
emission
limit
through
facility­
wide
emissions
averaging.

Response:

Industrial
boilers
that
do
not
meet
the
definition
of
an
"
electric
utility
steam
generating
unit"
under
either
40
CFR
part
60,
subpart
Da
or
HHHH
will
not
be
subject
to
the
final
rule.
Therefore,
facility­
wide
emissions
averaging
between
such
units
and
Utility
Units
will
not
be
allowed.

Comment:

One
commenter
(
OAR­
2002­
0056­
2922)
supported
EPA's
proposal
to
allow
emissions
averaging
as
a
compliance
option
for
two
or
more
coal­
fired
units,
including
blended
coal
units,
that
are
located
at
a
single
contiguous.
However,
the
commenter
suggested
the
following
clarifications
to
provide
for
smooth
implementation
of
averaging
plans.
First,
one
situation
under
which
sources
might
wish
to
utilize
averaging
is
where
two
or
more
units
utilize
a
common
stack.
Common
stack
monitoring
is
allowed
under
the
general
provisions
as
long
as
the
"
monitoring
is
sufficient
to
demonstrate
compliance
with
the
relevant
standard."
However,
the
proposed
rules
do
not
make
clear
under
what
provision
such
units
should
report
and
whether
EPA
expects
sources
to
submit
averaging
plans
for
such
units.
The
EPA
should
revise
the
rule
to
address
that
point.

One
commenter
(
OAR­
2002­
0056­
3398)
recommended
extension
of
emissions
averaging
to
all
units
under
common
control
within
a
State
to
add
flexibility.

Two
commenters
(
OAR­
2002­
0056­
1608,
­
2922)
recommended
that
a
Multi­
Source
Averaging
Plan
(
MAP)
to
meet
the
Hg
emission
standards
be
based
on
the
existing
CAA
Title
IV
NOx
program.
Although
this
approach
for
Hg
averaging
would
benefit
larger
utility
systems,
it
is
of
far
less
benefit
to
small
systems.
Instead,
it
could
be
altered
to
allow
averaging
among
different
owners
and
operators.
The
CAA
Title
V
permit
program
could
serve
to
ensure
multi­
source
compliance
after
a
MAP
is
approved
by
EPA.
Additionally,
the
MAP
approach
could
be
extended
across
state
lines
as
appropriate,
as
is
done
in
the
CAA
Title
IV
NOx
program.

Response:

EPA
believes
that
the
cap­
and­
trade
approach
being
finalized
will
adequately
address
the
comment.

3.3
VARIABILITY
Comment
EPA
used
a
similar
variability
methodology
to
calculate
the
new­
source
NSPS
limits
as
it
did
to
select
the
MACT
floor
limits;
the
only
difference
is
that
it
did
not
apply
the
inter­
variability
3­
63
analysis.
Thus,
several
of
the
commenters'
concerns
also
apply
to
the
selection
of
new
source
NSPS
limits.

Many
commenters
explained
that
EPA
improperly
used
a
short­
term
worst­
case
analysis
to
develop
a
long­
term
standard
(
12­
month
rolling
average).
EPA
chose
the
12­
month
rolling
average
emission
limit
format
and
then
applied
the
industry's
variability
method
to
account
for
coal
composition.
The
EPA's
own
variability
analysis
explained
that
it
was
inappropriate
to
apply
its
variability
analysis
where
a
long­
term
compliance
period
is
allowed.
Commenter
OAR­
2002­
0056­
2878
stated
that,
given
the
long­
term
format,
there
is
no
need
for
the
industry's
variability
analysis;
the
12­
month
averaging
time
provides
more
than
enough
buffer
to
address
the
worst
foreseeable
circumstances.

Commenter
OAR­
2002­
0056­
3459
stated
that
although
EPA
acknowledged
that
one
method
for
dealing
with
variability
was
the
length
of
the
compliance
period,
EPA
did
not
assess
that
option.
Instead,
EPA
added
an
annual
averaging
time
on
top
of
is
inflated
variability
approach
and
attempted
to
justify
this
double
counting
by
stating
that
Hg
poses
a
chronic
and
not
acute
health
risk.
Whether
or
not
this
justification
is
warranted,
EPA
neglected
the
effect
of
using
a
long­
term
standard
on
the
stringency
of
the
standard.
And
because
EPA
proposed
to
determine
compliance
using
a
long­
term
average,
the
compliance
status
of
the
unit
will
be
unaffected
by
short­
term
fluctuations
in
the
coal
characteristics
of
coal
shipments
and
control
equipment.

Commenter
OAR­
2002­
0056­
2920
stated
that
EPA
must
specifically
explain
why
the
12­
month
averaging
period
is
necessary
and
appropriate.

After
applying
the
industry's
variability
analysis,
EPA
calculated
the
emission
rate
over
the
full
range
of
coal
compositions
presumed
to
be
used
and
sorted
those
emissions
to
obtain
a
cumulative
frequency
distribution
and
selected
limits
based
on
the
97.5
percentile
(
compared
to
the
95
percent
in
the
WEST
analysis).
Many
commenters
(
e.
g.,
OAR­
2002­
0056­
2920)
noted
that
EPA
provided
no
rationale
for
selection
of
the
97.5
percentile.
This
violates
the
CAA,
is
inconsistent
with
EPA's
own
guidance
and
past
practices,
and
improperly
results
in
emission
limits
that
are
many
times
higher
than
appropriate
or
allowed
under
State
permits.
Commenter
OAR­
2002­
0056­
3459
recommended
that
EPA
use
the
mean
calculated
emissions
from
annual
coal
data
rather
than
the
97.5
percent
upper
confidence
limit
of
the
mean.
The
arithmetic
mean
is
consistent
with
the
12­
month
rolling
average
and
consistent
with
EPA
policy.

Commenter
OAR­
2002­
0056­
3449
criticized
EPA's
failures
to
consider
data
on
management
of
Hg
variability
from
operating
facilities.

Two
commenters
(
OAR­
2002­
0056­
2920,
­
3449)
criticized
in
detail
EPA's
equations
based
on
coal
Cl
content.
EPA
has
not
established
a
valid
statistical
relationship
between
Hg
removal
and
coal
Cl
content.

State
(
e.
g.,
OAR­
2002­
0056­
2660,
­
2823,
­
2889,
­
3210,
­
3449)
and
environmental
group
(
e.
g.,
OAR­
2002­
0056­
3459)
commenters
asserted
that
EPA
has
not
established
a
valid
statistical
relationship
between
Hg
removal
and
coal
Cl
content.
Analyses
conducted
by
WEST
and
DOE
3­
64
demonstrate
that
there
is
indeed
a
valid
correlation,
particularly
with
respect
to
units
equipped
with
a
fabric
filter/
spray
dryer
combination.
To
the
extent
coal
rank
is
indicative
of
Cl
content,
coal
rank
may
be
an
important
factor
with
respect
to
the
Hg
removal
fraction
as
the
New
Jersey
Department
of
Environmental
Protection
suggests.
However,
the
New
Jersey
Department
of
Environmental
Protection
incorrectly
asserts
that
EPA
should
have
used
raw
data
rather
than
average
Cl
concentrations
to
develop
its
Equation
(
5).
EPA's
approach
properly
relies
on
average
values
because
they
provide
less
uncertainty
as
compared
to
raw
data
alone.

Commenter
OAR­
2002­
0056­
5498
provided
detailed
supplemental
comments
that
address
criticisms
of
EPA/
DOE's
variability
analysis.
Their
comments
explain
why
EPA's
analysis
is
both
consistent
with
the
CAA
as
interpreted
by
the
D.
C.
Circuit
and
scientifically
sound.

One
commenter
(
OAR­
2002­
0056­
2835)
stated
that
the
variability
factors
used
by
the
EPA
in
the
proposal
are
appropriate.

Response:

EPA
continues
to
believe
that
accounting
for
variability
is
required
in
the
establishment
of
national
emission
standards.
However,
EPA
concurs
with
those
commenters
who
indicated
that
we
had
overstated
the
variability
by
using
both
a
rigorous
statistical
analysis
and
a
12­
month
rolling
average
for
compliance.
Our
revised
analysis
of
the
data
and
new­
source
selection
procedures
are
described
elsewhere
in
this
document.
We
believe
that
this
adequately
addresses
the
variability.

3.4
COAL
ANALYSIS
Comment:

One
commenter
(
OAR­
2002­
0056­
3546)
stated
that
the
proposed
rule
does
not
specify
protocols
for
determining
rank
classification
of
the
coal
burned
in
a
unit.
The
EPA
needs
to
propose
a
methodology
for
establishing
and
reporting
coal
rank
classification
for
determining
which
of
the
emission
limits
is
applicable
to
a
given
unit.
The
commenter
asked
if
coal
ranks
will
be
based
on
coal
samples
taken
at
the
mine
or
upon
delivery
point
at
the
power
plant,
and
who
is
ultimately
accountable
for
conducting
the
ASTM
coal
rank
tests;
the
supplier
or
the
power
plant
owner/
operator.

Response:

It
is
the
owner/
operator
that
is
ultimately
responsible
for
compliance
with
the
final
rules.
How
he/
she
chooses
to
comply,
however,
is
not
specified.
EPA
believes
that
facilities
are
currently
contracting
for
a
certain
rank
of
coal
with
specific
properties
best
suited
for
the
given
boiler
and,
thus,
are
well
aware
of
the
rank
of
coal
being
utilized.
EPA
believes
that
reliance
on
the
ASTM
coal
rank
classification
scheme
is
appropriate
for
this
rule
as
it
is
a
commonly
accepted
means
of
ranking
coals.
3­
65
Comment:

One
commenter
(
OAR­
2002­
0056­
1969)
stated
that
the
proposed
rule
does
not
specify
fuel
measurement/
sampling
method
required
to
determine
the
Btu
input
contributed
by
each
coal
rank.
The
commenter
recommended
that
EPA
should
make
it
clear
that
sources
can
use
procedures
already
in
place
at
the
source
for
recording
fuel
type
and
monitoring
fuel
consumption.

Response:

The
final
rule
does
not
require
fuel
measurement
or
sampling.
Further,
there
are
no
specifications
for
fuel
consumption
monitoring
procedures
so
the
facility
may
continue
to
use
procedures
already
in
place.

3.5
NOTIFICATION,
RECORDKEEPING,
AND
REPORTING
3.5.1
Recordkeeping
Comment:

One
commenter
(
OAR­
2002­
0056­
3543)
stated
that
regardless
of
the
approach
taken,
the
requirements
for
emissions
testing
and
recordkeeping
must
be
sufficient
to
provide
data
for
development
of
TMDL.
These
requirements
will
provide
a
rich
source
of
data
and
should
not
be
weakened.

Response:

EPA
has
not
weakened
its
monitoring
or
recordkeeping
requirements.

Comment:

Several
commenters
(
OAR­
2002­
0056­
2634,
­
2718,
­
2922)
stated
that
the
preamble
indicates
that
sources
will
be
required
to
maintain
monthly
records
of
types
of
fuel
burned,
total
fuel
usage,
and
fuel
heating
value,
but
these
requirements
do
not
appear
in
the
proposed
rule.
EPA
should
add
a
provision
for
recording
those
values
consistent
with
existing
company
practices.

Response:

EPA
believes
that
the
final
rule
addresses
the
commenter's
concerns.

3.5.2
Notifications
and
Reporting
Comment:
3­
66
Several
commenters
(
OAR­
2002­
0056­
2634,
­
2718,
­
2922)
stated
that
section
63.10030(
e)
states
that
a
"
Notification
of
Compliance
Status"
must
be
submitted
for
each
"
performance
test"
or
"
initial
compliance
demonstration"
as
specified
in
section
63.10007.
Section
63.10007,
however,
only
covers
performance
testing
for
"
oil­
fired"
units.
Initial
compliance
demonstrations
for
coal­
fired
units
are
addressed
in
section
63.10009.
If
EPA
intends
a
"
Notification
of
Compliance
Status"
to
be
submitted
by
coal­
fired
units
following
the
first
12­
month
period,
a
reference
to
section
63.10009
should
be
added.
If
EPA
does
not
intend
for
coal­
fired
units
to
submit
that
notice,
the
reference
to
the
"
initial
compliance
demonstration"
should
be
removed
or
clarified.

Several
commenters
(
OAR­
2002­
0056­
2634,
­
2718,
­
2922)
stated
that
section
63.10030(
a)
requires
compliance
with
many
notices
in
the
general
provisions,
including
the
"
Notification
of
Compliance
Status"
in
section
63.9(
h).
That
requirement
is
confusing
given
that
section
63.10030(
e)
sets
out
requirements
for
"
Notification
of
Compliance
Status"
that
are
narrower
than
those
in
section
63.9(
h)
(
e.
g.,
section
63.10030(
e)
only
requires
compliance
with
section
63.9(
h)(
2)(
ii)).
EPA
should
review
these
provisions
and
address
the
inconsistencies
and
overlapping
requirements
to
better
explain
to
sources
what
is
required
in
each
applicable
notice
or
report.
Section
63.10030(
a)
also
requires
compliance
with
section
63.6(
h)(
4)
and
(
5).
However,
according
to
Table
4,
those
provisions,
which
relate
to
opacity
and
visible
emissions
observations
are
not
applicable.
As
a
result,
they
should
be
removed
from
section
63.10030(
a).

Response:

Part
60
does
not
have
a
requirement
for
a
Notification
of
Compliance
Status.
The
proposed
amendments
to
subpart
Da
require
for
Hg
and
Ni
emissions
that
the
performance
test
data
from
the
initial
and
subsequent
performance
test
and
from
the
performance
evaluation
of
continuous
monitors
be
submitted
to
the
Administrator.
Subpart
Da
also
requires
semiannual
reports
indicating
whether:
(
1)
The
required
continuous
monitoring
system
calibration,
span,
and
drift
checks
or
other
periodic
audits
have
or
have
not
been
performed
as
specified.
(
2)
The
data
used
to
show
compliance
was
or
was
not
obtained
in
accordance
with
approved
methods
and
procedures
of
this
part
and
is
representative
of
plant
performance.
(
3)
The
minimum
data
requirements
have
or
have
not
been
met:
or,
the
minimum
data
requirements
have
not
been
met
for
errors
that
were
unavoidable.
(
4)
Compliance
with
the
standards
has
or
has
not
been
achieved
during
the
reporting
period.
Therefore,
subpart
Da
with
the
proposed
amendments
would
require
that
performance
tests
and
semiannual
compliance
reports
be
submitted
for
both
oil­
and
coal­
fired
units.
Regarding
the
reference
to
the
requirement
to
comply
with
section
63.6(
h)(
4)
and
(
5),
63.6
(
h)
deals
with
compliance
with
opacity
and
visible
emission
standards.
Section
60.48a
in
subpart
Da
with
the
proposed
amendments
does
not
have
any
compliance
provisions
for
opacity
and
visible
emission
standards.
Therefore
this
comment
does
not
apply
to
the
part
60
standards.

Comment:

Several
commenters
(
OAR­
2002­
0056­
2634,
­
2718,
­
2922)
stated
that
section
63.10009(
d)(
4)
requires
reporting
of
the
12­
month
rolling
average
Hg
emissions
rate
in
the
"
first
3­
67
semi­
annual
compliance
report."
If
the
initial
semi­
annual
report
will
be
submitted
before
12
months
of
data
have
been
collected,
as
section
63.10031
requires,
it
is
not
possible
to
report
a
12­
month
rolling
average.
EPA
should
remove
this
requirement
and
clarify
how
and
when
results
are
to
be
reported
(
e.
g.,
first
in
the
initial
"
Notification
of
Compliance
Status"
and
thereafter
in
the
next
semi­
annual
report).

Response:

This
comment
does
apply
to
reporting
of
compliance
with
the
Hg
emission
limit.
Compliance
cannot
be
determined
until
12
months
of
data
are
available.
This
has
been
addressed
in
MACT
standards
with
emission
rates
that
are
12­
month
rolling
averages,
such
as
subpart
SSSS
(
the
NESHAP
for
Surface
Coating
of
Metal
Coil,
see
68
FR
12591,
March
17,
2003
for
correction
notice)
as
follows
for
new
affected
sources:
(
1)
The
initial
compliance
period
begins
immediately
upon
start­
up
or
by
(
data
of
publication
of
the
final
rule
in
the
FR)
and
ends
on
the
last
day
of
the
12th
month
following
the
compliance
date.
If
the
compliance
date
falls
on
any
day
other
than
the
first
day
of
a
month,
then
the
initial
compliance
period
extends
through
that
month
plus
the
next
12
months.
(
2)
The
first
semiannual
reporting
period
begins
1
day
after
the
end
of
the
initial
compliance
period
described
in
(
1)
that
applies
to
your
affected
source
and
ends
6
months
later.
Comment:

Several
commenters
(
OAR­
2002­
0056­
2634,
­
2718,
­
2922)
stated
that
table
2
in
the
proposed
rule
would
require
each
coal­
fired
unit
subject
to
a
limit
in
section
63.9990
must
demonstrate
"
initial
compliance"
by
establishing
"
a
site
specific
[
Hg]
limit
according
to
the
procedures
in
section
63.10009
and
reporting
the
limit
in
your
notification
of
compliance
status."
This
articulation
of
the
initial
compliance
demonstration
is
not
consistent
with
the
rules.
Many
units
do
not
establish
site­
specific
limits,
and
it
is
not
clear
how
simply
reporting
a
limit
establishes
compliance.
Section
63.10009
does
not
call
for
reporting
of
a
limit,
but
rather
calculation
of
a
12­
month
rolling
average.
Also,
there
is
no
requirement
in
section
63.10030(
e),
addressing
"
notification
of
compliance
status,"
to
report
the
applicable
limit
Response:

This
does
not
apply
to
the
proposed
amendments
to
subpart
Da.
The
Hg
compliance
provisions
added
as
paragraph
(
m)
to
Section
60.48a
specify
how
Hg
emissions
will
be
calculated
using
data
measured
as
specified
in
Section
60.49a.
There
is
no
mention
of
establishing
a
site­
specific
Hg
limit.

3.6
COMPLIANCE
DATES
Comment
One
commenter
(
OAR­
2002­
0056­
3449)
disagreed
with
the
EPA's
preamble
statement
that
overly
ambitious
Hg
mandates
in
the
near
term
could
actually
hamper
innovation
toward
more
cost
effective
and
less
costly
technologies
(
69
FR
4687).
It
is
more
likely
that
EPA's
3­
68
minimal
reductions
over
the
next
decade
would
hamper
innovation
and
improvement
of
public
health.
The
sooner
ACI
with
fabric
filters
or
other
control
combinations
are
required,
the
sooner
costs
will
drop.
The
costs
are
reasonably
now
and
much
less
than
the
costs
of
Hg
poisoning.

Response:

EPA
stands
by
its
position
that
Hg­
specific
control
technologies
are
not
yet
commercially
available
and
that
the
regulatory
approach
being
finalized
is
the
best
approach
to
both
effect
significant
SO2,
NOx,
AND
Hg
emission
reductions
while
also
encouraging
the
further
development
of
the
emerging
Hg­
specific
technologies.

Comment:

One
commenter
(
OAR­
2002­
0056­
1969)
stated
the
concern
that
the
monitoring
and
recording
technology
has
not
evolved
to
the
level
of
reliability
necessary
to
collect
continuous
Hg
emissions
data
for
compliance
purposes
and
to
report
those
results
consistent
with
EPA's
proposed
requirements.

Response:

EPA
believes
that
the
monitoring
and
recording
technology
is
available
and
reliable
at
this
time
sufficient
to
show
compliance
with
the
final
emission
limits.
However,
further
developments
are
sure
to
ensue
in
the
coming
years
such
that
the
commenter's
concerns
will
be
alleviated.

Comment:

Several
commenters
(
OAR­
2002­
0056­
2634,
­
2718,
­
2922)
stated
that
under
proposed
compliance
date
provisions
are
confusing
and
conflicting.
Under
40
CFR
63.9983(
a)
and
63.10008(
d),
new
units
must
(
1)
install
and
operate
monitors,
and
(
2)
comply
with
the
"
emissions
limitations
and
work
practice
standards"
upon
the
later
of
publication
of
the
final
rule,
or
startup.
Under
Section
63.10005,
sources
then
have
180
days
after
the
date
for
compliance
with
"
emissions
limitations
and
work
practice
standards"
in
Section
63.9983,
to
complete
all
performance
tests,
selection
of
operating
parameters,
and
monitoring
equipment
performance
evaluations.
Under
Section
63.9983(
b)
and
63.10008(
d),
existing
units
must
(
1)
install
and
operate
monitors,
and
(
2)
comply
with
"
emissions
limitations"
by
3
years
after
the
final
rule
is
published.
Section
63.10005
states
that
performance
tests,
operating
limits
and
monitoring
equipment
performance
evaluations
also
must
be
conducted
by
the
compliance
date
in
Section
63.9983
(
i.
e.,
3
years
from
publication
of
the
final
rule),
but
done
so
according
to
the
applicable
provisions
in
section
63.7(
a)(
2),
which
(
unlike
section
63.9983)
allows
180
days
from
the
compliance
date
for
performance
testing
(
i.
e.,
3
years
and
180
days
after
publication).
For
both
new
units
and
existing
units,
however,
section
63.10009(
d)
states
that
"
compliance
monitoring"
must
begin
on
the
"
effective
date
of
this
subpart."
For
new
units,
the
requirement
to
"
comply"
with
"
emissions
limitations
and
work
practice
standards"
on
the
date
of
publication
of
the
final
rule
and
180
days
before
the
deadline
for
performance
testing,
Section
63.10005
makes
no
sense.
3­
69
Sources
cannot
be
expected
to
comply
without
a
performance
test
or
monitoring
system
in
place
to
establish
compliance.
For
existing
units,
the
rules
also
are
in
conflict
as
to
whether
an
additional
180
days
is
allowed
for
performance
testing,
selection
of
operating
parameters,
and
monitoring
equipment
performance
evaluations.
If
the
additional
180
days
is
provided,
the
deadline
for
compliance
must
also
be
extended.
Moreover,
for
Hg,
these
provisions
also
fail
to
recognize
that
sources
cannot
establish
compliance
with
the
Hg
emissions
limitation
until
12
months
after
the
monitoring
system
evaluation
has
been
completed
and
the
required
12
months
of
compliance
data
have
been
collected.
For
both
new
and
existing
units,
the
requirement
to
begin
"
compliance
monitoring"
on
the
effective
date
of
the
rule,
also
makes
no
sense.
It
conflicts
with
the
provisions
in
section
63.9983
establishing
"
publication"
(
not
the
effective
date)
as
the
triggering
point,
ignores
the
fact
that
some
new
units
may
not
even
have
started­
up,
and
ignores
the
additional
180
days
that
are
supposed
to
be
provided
under
section
63.10005.
The
EPA
should
follow
the
model
in
Part
75
and
establish
a
deadline
for
applicability
of
the
subpart
and
then
a
single
deadline
for
installation,
operation,
and
evaluation
of
monitoring
systems
and
for
performance
testing
and
selection
of
operating
parameters.
For
new
units,
the
deadline
for
applicability
of
the
rule
would
be
the
later
of
publication
or
unit
startup.
For
existing
units,
the
applicability
date
would
be
3
years
after
the
date
of
publication
of
the
rule.
The
deadline
for
installation,
operation,
and
certification
of
monitoring
systems
and
for
performance
testing
and
selection
of
operating
parameters
(
i.
e.,
the
point
when
"
compliance
monitoring"
is
begun)
would
be
180
days
later.
The
deadlines
for
establishing
compliance
with
the
Hg
standard
should
be
the
end
of
the
initial
12­
month
compliance
period.
At
the
time
of
the
demonstration
of
compliance
for
the
initial
12­
month
period,
sources
would
be
deemed
to
be
in
compliance
for
the
prior
12
months.
As
a
result,
they
would
at
that
time
have
met
the
statutory
deadline
for
compliance.

Response:

Although
the
commenters
cited
concerns
with
the
proposed
MACT
standard,
which
is
not
being
finalized,
EPA
believes
that
their
concerns
may
also
have
been
valid
for
the
proposed
subpart
Da
revisions.
Section
60.8
Performance
tests
of
the
General
Provisions
to
part
60
requires
that
within
60
days
after
achieving
the
maximum
production
rate
at
which
the
affected
facility
will
be
operated,
but
not
later
than
180
days
after
initial
startup
of
such
facility...
the
owner
or
operator
of
such
facility
shall
conduct
performance
test(
s)
and
furnish
the
Administrator
a
written
report
of
the
results
of
such
performance
test(
s).
Therefore,
a
new
facility
has
180
days
after
initial
startup
or
date
of
publication
of
the
final
rule
to
complete
and
report
the
results
of
the
initial
performance
test.
The
timing
of
the
initial
compliance
period
and
required
reporting
should
be
as
follows:
(
1)
The
initial
compliance
period
begins
upon
submitting
the
report
of
the
initial
performance
test
to
the
Administrator,
but
no
later
than
180
days
after
start­
up
or
(
data
of
publication
of
the
final
rule
in
the
FR)
and
ends
on
the
last
day
of
the
12th
month
following
the
compliance
date.
If
the
compliance
date
falls
on
any
day
other
than
the
first
day
of
a
month,
then
the
initial
compliance
period
extends
through
that
month
plus
the
next
12
months.
(
2)
The
first
semiannual
reporting
period
begins
1
day
after
the
end
of
the
initial
compliance
period
described
in
(
1)
that
applies
to
your
affected
source
and
ends
6
months
later.
