Redline­
strikeout
highlighting
changes
made
during
OMB
review
1
Currently,
we
are
not
aware
of
any
preheater/
precalciner
kiln
that
vents
its
alkali
bypass
gases
through
a
separate
stack.

123
VIII.
How
Did
EPA
Determine
the
Proposed
Emission
Standards
for
Hazardous
Waste
Burning
Cement
Kilns?
In
this
section,
the
basis
for
the
proposed
emission
standards
is
discussed.
See
proposed
§
63.1204A1220.
The
proposed
emission
limits
apply
to
the
kiln
stack
gases,
in­
line
kiln
raw
mill
stack
gases
if
combustion
gases
pass
through
the
in­
line
raw
mill,
and
kiln
alkali
bypass
stack
gases
if
discharged
through
a
separate
stack.
1
The
proposed
standards
for
existing
and
new
cement
kilns
that
burn
hazardous
waste
are
summarized
in
the
table
below:

PROPOSED
STANDARDS
FOR
EXISTING
AND
NEW
CEMENT
KILNS
Hazardous
Air
Pollutant
or
Surrogate
Emission
Standard1
Existing
Sources
New
Sources
Dioxin
and
furan
0.20
ng
TEQ/
dscm;
or
0.40
ng
TEQ/
dscm
and
control
of
flue
gas
temperature
not
to
exceed
400
°
F
at
the
inlet
to
the
particulate
matter
control
device.

Mercury2
64
ug/
dscm
35
ug/
dscm
Particulate
Matter
65
mg/
dscm
(
0.028
gr/
dscf)
13
mg/
dscm
(
0.0058
gr/
dscf)

Semivolatile
metals3
4.0
x
10­
4
lb/
MMBtu
6.2
x
10­
5
lb/
MMBtu
Low
volatile
metals3
1.4
x
10­
5
lb/
MMBtu
1.4
x
10­
5
lb/
MMBtu
Hydrogen
chloride
and
chlorine
gas4
110
ppmv
or
the
alternative
emission
limits
under
§
63.1215
78
ppmv
or
the
alternative
emission
limits
under
§
63.1215
Hydrocarbons:
kilns
without
bypass5,
6
20
ppmv
(
or
100
ppmv
carbon
monoxide)
5
Greenfield
kilns:
20
ppmv
(
or
100
ppmv
carbon
monoxide
and
50
ppmv7
hydrocarbons).
All
others:
20
ppmv
(
or
100
ppmv
carbon
monoxide)
5
Hydrocarbons:
kilns
with
bypass;
main
stack6,
8
No
main
stack
standard
50
ppmv7
Redline­
strikeout
highlighting
changes
made
during
OMB
review
2
Even
though
all
sources
have
recently
demonstrated
compliance
with
the
interim
standards,
the
dioxin/
furan
data
in
our
data
base
preceded
the
compliance
demonstration.
This
explains
why
we
have
emissions
data
that
are
higher
than
the
interim
standard.

124
Hydrocarbons:
kilns
with
bypass;
bypass
duct
and
stack5,
6,
8
10
ppmv
(
or
100
ppmv
carbon
monoxide)
10
ppmv
(
or
100
ppmv
carbon
monoxide)

Destruction
and
removal
efficiency
For
existing
and
new
sources,
99.99%
for
each
principal
organic
hazardous
constituent
(
POHC).
For
sources
burning
hazardous
wastes
F020,
F021,
F022,
F023,
F026,
or
F027,
however,
99.9999%
for
each
POHC.

1
All
emission
standards
are
corrected
to
7%
oxygen,
dry
basis.
If
there
is
a
separate
alkali
bypass
stack,
then
both
the
alkali
bypass
and
main
stack
emissions
must
be
less
than
the
emission
standard.
2
Mercury
standard
is
an
annual
limit.
3
Standards
are
expressed
as
mass
of
pollutant
stack
emissions
attributable
to
the
hazardous
waste
per
million
British
thermal
unit
heat
input
of
the
hazardous
waste.
4
Combined
standard,
reported
as
a
chloride
(
Cl(­))
equivalent.
5
Sources
that
elect
to
comply
with
the
carbon
monoxide
standard
must
demonstrate
compliance
with
the
hydrocarbon
standard
during
the
comprehensive
performance
test.
6
Hourly
rolling
average.
Hydrocarbons
reported
as
propane.
7
Applicable
only
to
newly­
constructed
cement
kilns
at
greenfield
sites
(
see
64
FR
at
52885).
The
50
ppmv
standard
is
a
30­
day
block
average
limit.
8
Measurement
made
in
the
bypass
sampling
system
of
any
kiln
(
e.
g.,
alkali
bypass
of
a
preheater/
precalciner
kiln;
midkiln
gas
sampling
system
of
a
long
kiln).

A.
What
Are
the
Proposed
Standards
for
Dioxin
and
Furan?
We
are
proposing
to
establish
standards
for
existing
and
new
cement
kilns
that
limit
emissions
of
dioxin
and
furans
to
either
0.20
ng
TEQ/
dscm
or
0.40
ng
TEQ/
dscm
and
control
of
flue
gas
temperature
not
to
exceed
400
°
F
at
the
inlet
to
the
particulate
matter
control
device.
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
Dioxin
and
furan
emissions
for
existing
cement
kilns
are
currently
limited
by
§
63.1204(
a)(
1)
to
0.20
ng
TEQ/
dscm
or
0.40
ng
TEQ/
dscm
and
control
of
flue
gas
temperature
not
to
exceed
400
°
F
at
the
inlet
to
the
particulate
matter
control
device.
This
standard
was
promulgated
in
the
Interim
Standards
Rule
(
See
67
FR
at
6796,
February
13,
2002).
Since
promulgation
of
the
September
1999
final
rule,
we
have
obtained
additional
dioxin/
furan
emissions
data.
We
now
have
compliance
test
emissions
data
representing
maximum
emissions
for
all
but
one
cement
kiln
that
burns
hazardous
waste.
The
compliance
test
dioxin/
furan
emissions
in
our
data
base
range
from
approximately
0.004
to
20
ng
TEQ/
dscm.
2
Redline­
strikeout
highlighting
changes
made
during
OMB
review
125
Cement
kilns
control
dioxin
by
quenching
kiln
gas
temperatures
so
that
gas
temperatures
at
the
inlet
to
the
particulate
matter
control
device
are
below
the
range
of
optimum
dioxin/
furan
formation.
To
identify
the
MACT
floor,
we
evaluated
the
compliance
test
emissions
data
associated
with
the
most
recent
test
campaign
using
the
Emissions
Approach
described
in
Part
Two,
Section
VI.
C
above.
The
calculated
floor
is
0.22
ng
TEQ/
dscm,
which
considers
emissions
variability.
These
best
performing
sources
controlled
inlet
temperatures
to
the
particulate
matter
control
device
from
380
°
­
475
°
F.
Although
some
best
performing
sources
had
inlet
temperatures
to
the
particulate
matter
control
device
within
the
optimum
temperature
range
(
i.
e.,
>
400
°
F)
for
formation
of
dioxin/
furan,
their
emissions
were
lower
than
other
non­
best
performing
sources.
Our
data
base
shows
that
these
other
non­
best
performing
sources,
when
operating
within
a
temperature
range
up
to
475
°
F,
had
emissions
of
dioxin/
furan
as
high
as
1.2
ng
TEQ/
dscm.
We
cannot
explain
why
some
sources
emit
dioxin/
furan
at
significantly
lower
levels
than
other
sources
operating
at
similar
control
device
inlet
temperatures.
As
noted
earlier,
there
are
many
uncertainties
and
imperfectly
understood
complexities
relating
to
dioxin/
furan
formation.
The
data
generally
support
the
relationship
between
inlet
temperature
to
the
particulate
matter
control
device
and
dioxin/
furan
emissions:
When
inlet
temperatures
are
below
the
optimum
range
of
formation,
dioxin/
furan
emissions
are
lower.
However,
the
converse
may
not
hold:
When
inlet
temperatures
are
within
the
optimum
range
of
formation,
dioxin/
furan
emissions
may
or
may
not
be
higher
(
but
in
most
cases
are
higher).
Moreover,
we
are
concerned
that
a
floor
level
of
0.22
ng
TEQ/
dscm
is
not
replicable
by
all
sources
using
temperature
control
because
we
have
emissions
data
from
sources
operating
below
the
optimum
temperature
range
of
dioxin/
furan
formation
that
is
higher
than
the
calculated
floor
level
of
0.22
ng
TEQ/
dscm.
As
a
result
of
this
concern,
we
would
identify
the
floor
level
as
0.22
ng
TEQ/
dscm
or
controlling
the
inlet
temperature
to
the
particulate
matter
control
device.
Allowing
a
source
to
comply
with
a
temperature
limit
alone,
however,
absent
a
numerical
dioxin/
furan
emission
limit,
is
less
stringent
than
the
current
interim
standard
of
0.20
ng
TEQ/
dscm,
or
0.40
ng
TEQ/
dscm
and
control
of
flue
gas
temperature
not
to
exceed
400
°
F
at
the
inlet
to
the
particulate
matter
control
device.
The
current
interim
standard
is
a
regulatory
limit
that
is
relevant
in
identifying
the
floor
level
because
it
fixes
a
level
of
performance
for
the
source
category.
Given
that
all
sources
are
achieving
this
interim
standard
and
that
the
interim
standard
is
judged
as
more
stringent
than
the
calculated
MACT
floor,
the
dioxin/
furan
floor
level
can
be
no
less
stringent
than
the
current
regulatory
limit.
We
are,
therefore,
proposing
the
dioxin/
furan
floor
level
as
0.20
ng
TEQ/
dscm
or
0.40
ng
TEQ/
dscm
and
control
of
flue
gas
temperature
not
to
exceed
400
°
F
at
the
inlet
to
the
particulate
matter
control
device.
This
emission
level
is
being
achieved
by
all
sources
because
it
is
the
current
required
interim
standard.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
evaluated
activated
carbon
injection
as
beyond­
the­
floor
control
for
further
reduction
of
dioxin/
furan
emissions.
Activated
carbon
has
been
demonstrated
for
controlling
dioxin/
furans
in
various
combustion
applications.
However,
currently
no
cement
kiln
that
burns
hazardous
waste
uses
activated
carbon
injection.
We
evaluated
a
beyond­
the­
floor
level
of
0.10
ng
TEQ/
dscm,
which
represents
a
75%
reduction
in
dioxin/
furan
emissions
from
the
floor
level.
We
Redline­
strikeout
highlighting
changes
made
during
OMB
review
3
Under
the
exemption
from
hazardous
waste
status
in
§
261.4(
b)(
8),
cement
kiln
dust
is
not
currently
classified
as
a
hazardous
waste.

126
selected
this
level
because
it
represents
a
level
that
is
considered
routinely
achievable
with
activated
carbon
injection.
In
addition,
we
assumed
for
costing
purposes
that
cement
kilns
needing
activated
carbon
injection
to
achieve
the
beyond­
the­
floor
level
would
install
the
activated
carbon
injection
system
after
the
existing
particulate
matter
control
device
and
add
a
new,
smaller
baghouse
to
remove
the
injected
carbon
with
the
adsorbed
dioxin/
furan.
We
chose
this
costing
approach
to
address
potential
concerns
that
injected
carbon
may
interfere
with
cement
kiln
dust
recycling
practices.
The
national
incremental
annualized
compliance
cost
for
cement
kilns
to
meet
this
beyondthe
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
21
million
and
would
provide
an
incremental
reduction
in
dioxin/
furan
emissions
beyond
the
MACT
floor
controls
of
3.4
grams
TEQ
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
evaluated
to
estimate
the
impacts
between
activated
carbon
injection
and
controls
likely
to
be
used
to
meet
the
floor
level.
We
estimate
that
this
beyond­
the­
floor
option
would
increase
the
amount
of
solid
waste3
generated
by
7,800
tons
per
year
and
would
require
sources
to
use
an
additional
2.6
million
kW­
hours
per
year
beyond
the
requirements
to
achieve
the
floor
level.
The
costs
associated
with
these
impacts
are
accounted
for
in
the
national
annualized
compliance
cost
estimates.
Therefore,
based
on
these
factors
and
costs
of
approximately
$
6.2
million
per
additional
gram
of
dioxin/
furan
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
use
of
activated
carbon
injection.
3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
Dioxin
and
furan
emissions
for
new
cement
kilns
are
currently
limited
by
§
63.1204(
b)(
1)
to
either
0.20
ng
TEQ/
dscm
or
0.40
ng
TEQ/
dscm
and
control
of
flue
gas
temperature
not
to
exceed
400
°
F
at
the
inlet
to
the
particulate
matter
control
device.
This
standard
was
promulgated
in
the
Interim
Standards
Rule
(
See
67
FR
at
6796).
The
calculated
MACT
floor
for
new
sources
would
be
0.21
ng
TEQ/
dscm,
which
considers
emissions
variability.
This
is
an
emission
level
that
the
single
best
performing
source
identified
by
the
Emissions
Approach
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
conditions
identical
to
the
test
conditions
during
which
the
emissions
data
were
obtained.
As
discussed
for
existing
sources,
we
are
concerned
that
a
floor
level
of
0.21
ng
TEQ/
dscm
would
not
be
reproducible
by
all
sources
using
temperature
control
because
we
have
emissions
data
from
sources
operating
below
the
optimum
temperature
range
of
dioxin/
furan
formation
that
is
higher
than
the
calculated
floor
level
of
0.21
ng
TEQ/
dscm.
As
a
result
of
this
concern,
we
would
identify
the
MACT
floor
as
0.21
ng
TEQ/
dscm
or
controlling
the
inlet
temperature
to
the
particulate
matter
control
device.
Allowing
a
source
to
comply
with
a
temperature
limit
alone,
however,
absent
a
numerical
dioxin/
furan
emission
limit,
is
less
stringent
than
the
current
interim
standard
of
0.20
ng
TEQ/
dscm,
or
0.40
ng
TEQ/
dscm
and
control
of
flue
gas
temperature
not
to
exceed
400
°
F
at
the
inlet
to
the
particulate
matter
control
device.
The
current
interim
standard
is
a
regulatory
limit
Redline­
strikeout
highlighting
changes
made
during
OMB
review
4
An
alternative
mercury
standard
is
available
for
existing
cement
kilns
whereby
a
source
can
elect
to
comply
with
a
hazardous
waste
maximum
theoretical
emissions
concentration
or
MTEC
of
mercury
of
120
ug/
dscm.
MTEC
is
a
term
to
compare
metals
and
chlorine
feedrates
across
sources
of
different
sizes.
MTEC
is
defined
as
the
metals
or
chlorine
feedrate
divided
by
the
gas
flow
rate
and
is
expressed
in
units
of
ug/
dscm.

127
that
is
relevant
in
identifying
the
floor
level
because
it
fixes
a
level
of
performance
for
new
cement
kilns.
Given
that
all
sources
are
achieving
this
interim
standard
and
that
the
interim
standard
is
judged
as
more
stringent
than
the
calculated
MACT
floor,
the
dioxin/
furan
floor
level
can
be
no
less
stringent
than
the
current
regulatory
limit.
We
are,
therefore,
proposing
the
dioxin/
furan
floor
level
as
0.20
ng
TEQ/
dscm
or
0.40
ng
TEQ/
dscm
and
control
of
flue
gas
temperature
not
to
exceed
400
°
F
at
the
inlet
to
the
particulate
matter
control
device.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
We
evaluated
activated
carbon
injection
as
beyond­
the­
floor
control
for
further
reduction
of
dioxin/
furan
emissions.
We
evaluated
a
beyond­
the­
floor
level
of
0.10
ng
TEQ/
dscm,
which
represents
a
75%
reduction
in
dioxin/
furan
emissions
from
the
floor
level.
We
selected
this
level
because
it
represents
a
level
that
is
considered
routinely
achievable
with
activated
carbon
injection.
In
addition,
we
assumed
for
costing
purposes
that
a
new
cement
kiln
will
install
the
activated
carbon
injection
system
after
the
existing
particulate
matter
control
device
and
add
a
new,
smaller
baghouse
to
remove
the
injected
carbon
with
the
adsorbed
dioxin/
furan.
The
incremental
annualized
compliance
cost
for
a
new
cement
kiln
to
meet
this
beyond­
the­
floor
level,
rather
than
comply
with
the
floor
level,
would
be
approximately
$
1.0
million
and
would
provide
an
incremental
reduction
in
dioxin/
furan
emissions
of
approximately
0.17
grams
TEQ
per
year,
for
a
cost­
effectiveness
of
$
5.8
million
per
gram
of
dioxin/
furan
removed.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
not
significant
factors.
For
these
reasons,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
activated
carbon
injection
for
new
cement
kilns.
Therefore,
we
are
proposing
the
standard
as
0.20
ng
TEQ/
dscm
or
0.40
ng
TEQ/
dscm
or
control
of
flue
gas
temperature
not
to
exceed
400
°
F
at
the
inlet
to
the
particulate
matter
control
device.
B.
What
Are
the
Proposed
Standards
for
Mercury?
We
are
proposing
to
establish
standards
for
existing
and
new
cement
kilns
that
limit
emissions
of
mercury
to
64
and
35
ug/
dscm,
respectively.
If
we
were
to
adopt
these
standards,
then
sources
would
comply
with
the
limit
on
an
annual
basis
because
the
standards
are
based
on
normal
emissions
data.
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
Mercury
emissions
for
existing
cement
kilns
are
currently
limited
to
120
ug/
dscm
by
§
63.1204(
a)(
2).
4
This
standard
was
promulgated
in
the
Interim
Standards
Rule
(
See
67
FR
at
6796).
None
of
the
cement
kilns
burning
hazardous
waste
use
a
dedicated
control
device
to
remove
mercury
from
the
gas
stream;
however,
kilns
control
the
feed
concentration
of
mercury
in
the
hazardous
waste.
We
have
emissions
data
for
all
sources.
All
of
these
data
are
best
classified
as
from
Redline­
strikeout
highlighting
changes
made
during
OMB
review
5
Given
that
we
only
have
normal
feedrate
and
emissions
data
for
mercury
for
cement
kilns,
we
do
not
believe
it
is
appropriate
to
establish
a
hazardous
waste
thermal
emissionsbased
standard.
We
prefer
to
establish
emission
standards
under
the
hazardous
waste
thermal
emissions
format
using
compliance
test
data
because
the
metals
feedrate
information
from
compliance
tests
that
we
use
to
apportion
emissions
to
calculate
emissions
attributable
to
hazardous
waste
are
more
reliable
than
feedrate
data
measured
during
testing
under
normal,
typical
operations.

128
normal
operations,
although,
as
explained
below,
there
is
a
substantial
range
within
these
data.
For
most
sources,
we
have
normal
emissions
data
from
more
than
one
test
campaign.
The
normal
mercury
stack
emissions
in
our
data
base
range
from
less
than
2
to
118
ug/
dscm.
These
emissions
are
expressed
as
mass
of
mercury
(
from
all
feedstocks)
per
unit
volume
of
stack
gas.
To
identify
the
MACT
floor,
we
evaluated
all
normal
emissions
data
using
the
SRE/
Feed
Approach.
We
considered
normal
emissions
data
from
all
test
campaigns.
5
For
example,
one
source
in
our
data
base
has
normal
emissions
data
for
three
different
testing
campaigns:
1992,
1995,
and
1998.
Under
this
approach
we
would
consider
the
emissions
data
from
the
three
separate
years
or
campaigns.
We
believe
this
approach
better
captures
the
range
of
average
emissions
for
a
source
than
only
considering
the
most
recent
normal
emissions.
Given
that
no
cement
kilns
burning
hazardous
waste
use
a
control
device
which
captures
mercury
from
the
flue
gas
stream,
for
purposes
of
this
analysis
we
assumed
all
sources
achieved
a
SRE
of
zero.
The
effect
of
this
assumption
is
that
the
sources
with
the
lowest
mercury
concentrations
in
the
hazardous
waste
were
identified
as
the
best
performing
sources.
The
calculated
floor
is
64
ug/
dscm,
which
considers
emissions
variability,
based
on
a
hazardous
waste
maximum
theoretical
emissions
concentration
(
MTEC)
of
26
ug/
dscm.
This
is
an
emission
level
that
the
average
of
the
best
performing
sources
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
We
estimate
that
this
emission
level
is
being
achieved
by
59%
of
sources
and
would
reduce
mercury
emissions
by
0.23
tons
per
year.
If
we
were
to
adopt
such
a
floor
level,
we
are
proposing
that
sources
comply
with
the
limit
on
an
annual
basis
because
it
is
based
on
normal
emissions
data.
Under
this
approach,
compliance
would
not
be
based
on
the
use
of
a
total
mercury
continuous
emissions
monitoring
system
because
these
monitors
have
not
been
adequately
demonstrated
as
a
reliable
compliance
assurance
tool
at
cement
kiln
sources.
Instead,
a
source
would
maintain
compliance
with
the
mercury
standard
by
establishing
and
complying
with
short­
term
limits
on
operating
parameters
(
e.
g.,
for
pollution
control
equipment
and
annual
limits
on
maximum
total
mercury
feedrate
in
all
feedstreams)
on
an
annual
basis.
We
did
not
use
the
stack
emissions
data
of
preheater/
precalciner
kilns
in
the
floor
analysis
because
we
believe
the
mercury
emissions
are
biased
low
when
the
in­
line
raw
mill
is
on­
line
and
biased
high
when
the
in­
line
raw
mill
is
off­
line.
(
See
earlier
discussion
on
why
we
are
proposing
not
to
subcategorize
hazardous
waste
burning
cement
kilns
for
mercury
between
wet
process
kilns
and
preheater/
precalciner
kilns
with
in­
line
raw
mills.)
For
either
case,
we
believe
the
normal
Redline­
strikeout
highlighting
changes
made
during
OMB
review
6
Cement
Kiln
Recycling
Coalition
is
a
trade
organization
that
represents
cement
companies
that
burn
hazardous
wastes
as
a
fuel.
CKRC
also
represents
companies
that
manage
and
market
hazardous
waste
fuels
used
in
cement
kilns.

7
For
two
cement
facilities,
the
mercury
concentration
data
are
only
available
on
a
monthly­
averaged
basis.

129
mercury
data
are
not
representative
of
average
emissions
and,
therefore,
not
appropriate
to
include
in
the
floor
analysis.
We
request
comment
on
this
data
handling
decision.
In
the
September
1999
final
rule,
we
acknowledged
that
a
cement
kiln
using
properly
designed
and
operated
MACT
control
technologies,
including
controlling
the
levels
of
metals
in
the
hazardous
waste,
may
not
be
capable
of
achieving
a
given
emission
standard
because
of
mineral
and
process
raw
material
contributions
that
might
cause
an
exceedance
of
the
emission
standard.
To
address
this
concern,
we
promulgated
a
provision
that
allows
kilns
to
petition
for
alternative
standards
provided
they
submit
site­
specific
information
that
shows
raw
material
hazardous
air
pollutant
contributions
to
the
emissions
prevent
the
source
from
complying
with
the
emission
standard
even
though
the
kiln
is
using
MACT
control.
See
§
63.1206(
b)(
10).

In
June
2003,
the
Cement
Kiln
Recycling
Coalition
(
CKRC)
6
submitted
to
EPA
information
on
actual
mercury
concentrations
in
the
hazardous
waste
burn
tanks
of
all
14
cement
facilities
for
a
three
year
period
covering
1999
to
2001.
In
general,
the
information
shows
the
mercury
concentration
(
in
parts
per
million)
in
the
hazardous
waste
for
each
burn
tank.
7
In
total,
Redline­
strikeout
highlighting
changes
made
during
OMB
review
8
Data
from
three
of
the
facilities
had
a
significant
number
of
individual
measurements
reported
as
not
detectable
and
also
had
relatively
high
analysis
detection
limits
(
compared
to
levels
achieved
by
other
cement
plants).
The
detection
limit
for
most
cement
kilns
was
typically
0.1
ppm
or
less.
For
purposes
of
today's
preamble
discussion,
the
measurements
from
these
three
cement
plants
are
excluded
from
the
data
characterization
conclusions.

9
USEPA,
"
Draft
Technical
Support
Document
for
HWC
MACT
Replacement
Standards,
Volume
III:
Selection
of
MACT
Standards
and
Technologies,"
March
2004,
Chapter
23.

130
approximately
20,000
mercury
burn
tank
concentration
data
points
are
included
in
CKRC's
submission.
8
The
data
show
that
approximately
50%
of
the
individual
burn
tank
measurements
are
0.6
ppmw
or
less,
75%
are
less
than
1.1
ppmw,
88%
are
less
than
2
ppmw,
and
97%
of
all
burn
tank
measurements
are
less
than
5
ppmw.
For
a
hypothetical
wet
process
cement
kiln
that
gets
50%
of
its
required
heat
input
from
hazardous
waste,
a
hazardous
waste
with
a
mercury
concentration
of
0.6
ppmw
equates
approximately
to
an
uncontrolled
(
i.
e.,
a
system
removal
efficiency
of
zero)
stack
gas
concentration
of
24
ug/
dscm.
This
estimated
stack
gas
concentration,
of
course,
does
not
include
contributions
to
emissions
from
other
mercurycontaining
feedstocks
including
raw
materials
and
fossil
fuels.
Mercury
concentrations
of
1.1,
2,
and
5
ppmw
in
the
hazardous
waste
equate
to
uncontrolled
stack
gas
concentrations
of
approximately
43,
79,
and
196
ug/
dscm.
9
We
compared
the
concentration
of
mercury
in
the
hazardous
waste
associated
with
the
normal
emissions
data
in
our
data
base
to
the
3­
year
historical
burn
tank
concentration
data
to
estimate
whether
the
normal
data
in
our
data
base
 
the
basis
of
today's
proposed
floor
of
64
ug/
dscm
 
are
likely
to
represent
the
high
end,
low
end,
or
close
to
average
emissions.
Mercury
feed
concentration
information
is
not
available
for
every
test
condition;
however,
the
mercury
concentrations
in
the
hazardous
waste
burned
by
the
best
performing
sources
during
the
tests
that
generated
the
normal
emissions
ranged
from
0.1
to
0.44
ppmw.
For
the
best
performing
sources
comprising
the
MACT
pool
for
which
we
can
make
a
comparison,
it
appears
that
the
normal
concentrations
in
the
hazardous
waste
during
testing
represent
the
low
end
(
15th
percentile
or
less)
of
average
mercury
concentrations.
We
invite
comment
on
whether
the
normal
emissions
data
in
our
data
base
are
representative
of
average
emissions
in
practice
and
whether
evaluating
the
data
to
identify
a
floor
level
is
appropriate.
In
addition,
we
request
comment
on
how
to
identify
a
floor
level
using
the
3­
year
hazardous
waste
mercury
concentration
data.
One
potential
approach
would
be
to
establish
a
hazardous
waste
feed
concentration
standard
expressed
in
ppmw.
To
identify
a
floor
level
expressed
as
a
hazardous
waste
feed
concentration
in
ppmw,
we
identified
and
evaluated
the
3­
year
historical
burn
tank
concentration
data
of
the
five
best
performing
facilities
(
those
sources
with
the
lowest
mean
concentration
considering
variability).
The
calculated
alternative
floor
level
is
2.2
ppmw
in
the
hazardous
waste.
To
put
this
in
context
for
a
hypothetical
wet
process
cement
kiln
that
gets
50%
of
its
required
heat
input
from
hazardous
waste,
a
mercury
concentration
of
2.2
Redline­
strikeout
highlighting
changes
made
during
OMB
review
10
USEPA,
"
Draft
Technical
Support
Document
for
HWC
MACT
Replacement
Standards,
Volume
III:
Selection
of
MACT
Standards
and
Technologies"
Standards",
March
2004,
Chapter
23.

11
USEPA,
"
Draft
Technical
Support
Document
for
HWC
MACT
Replacement
Standards,
Volume
III:
Selection
of
MACT
Standards
and
Technologies"
Standards",
March
2004,
Chapter
23.

131
ppmw
in
the
hazardous
waste
equates
approximately
to
an
uncontrolled
stack
gas
concentration
of
86
ug/
dscm.
10
This
estimated
stack
gas
concentration,
of
course,
does
not
include
contributions
to
emissions
from
other
mercury­
containing
feedstocks
such
as
raw
materials
and
fossil
fuels.
If
we
were
to
adopt
such
an
approach,
we
would
require
sources
to
comply
with
the
feed
concentration
standard
on
a
short
term
basis
(
e.
g.,
12
hour
average).
We
also
invite
comment
on
whether
we
should
judge
an
annual
limit
of
64
ug/
dscm
as
less
stringent
than
either
the
current
emission
standard
of
120
ug/
dscm
or
the
hazardous
waste
MTEC
of
mercury
of
120
ug/
dscm
for
cement
kilns
(
so
as
to
avoid
any
backsliding
from
a
current
level
of
performance
achieved
by
all
sources,
and
hence,
the
level
of
minimal
stringency
at
which
EPA
could
calculate
the
MACT
floor).
In
order
to
comply
with
the
current
emission
standard,
generally
a
source
must
conduct
manual
stack
sampling
to
demonstrate
compliance
with
the
mercury
emission
standard
and
then
establish
a
maximum
mercury
feedrate
limit
based
on
operations
during
the
performance
test.
Following
the
performance
test,
the
source
complies
with
a
limit
on
the
maximum
total
mercury
feedrate
in
all
feedstreams
on
a
12­
hour
rolling
average
(
not
an
annual
average).
Alternatively,
a
source
can
elect
to
comply
with
a
hazardous
waste
MTEC
of
mercury
of
120
ug/
dscm
that
would
require
the
source
to
limit
the
mercury
feedrate
in
the
hazardous
waste
on
a
12­
hour
rolling
average.
The
floor
level
of
64
ug/
dscm
proposed
today
would
allow
a
source
to
feed
more
variable
mercury­
containing
feedstreams
(
e.
g.,
a
hazardous
waste
with
an
mercury
MTEC
greater
than
120
ug/
dscm)
than
the
current
12­
hour
rolling
average
because
today's
proposed
floor
level
is
an
annual
limit.
For
example,
we
estimated
a
hazardous
waste
MTEC
for
each
burn
tank
measurement
associated
with
the
3­
year
historical
concentration
data
submitted
by
CKRC.
We
found
that
approximately
5%
of
burn
tank
measurements
would
exceed
a
hazardous
waste
MTEC
of
120
ug/
dscm,
including
sources
upon
which
the
proposed
floor
is
based.
11
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
identified
three
potential
beyond­
the­
floor
techniques
for
control
of
mercury:
(
1)
activated
carbon
injection;
(
2)
control
of
mercury
in
the
hazardous
waste
feed;
and
(
3)
control
of
mercury
in
the
raw
materials
and
auxiliary
fuels.
For
reasons
discussed
below,
we
are
not
proposing
a
beyond­
the­
floor
standard
for
mercury.
Use
of
Activated
Carbon
Injection.
We
evaluated
activated
carbon
injection
as
beyondthe
floor
control
for
further
reduction
of
mercury
emissions.
Activated
carbon
has
been
demonstrated
for
controlling
mercury
in
several
combustion
applications;
however,
currently
no
cement
kiln
that
burns
hazardous
waste
uses
activated
carbon
injection.
Given
this
lack
of
Redline­
strikeout
highlighting
changes
made
during
OMB
review
12
USEPA,
"
Draft
Technical
Support
Document
for
HWC
MACT
Replacement
Standards,
Volume
V:
Emission
Estimates
and
Engineering
Costs",
March
2004,
Chapter
4.

132
experience
using
activated
carbon
injection,
we
made
a
conservative
assumption
that
the
use
of
activated
carbon
injection
will
provide
70%
mercury
control
and
evaluated
a
beyond­
the­
floor
level
of
19
ug/
dscm.
In
addition,
for
costing
purposes
we
assumed
that
cement
kilns
needing
activated
carbon
injection
to
achieve
the
beyond­
the­
floor
level
would
install
the
activated
carbon
injection
system
after
the
existing
particulate
matter
control
device
and
add
a
new,
smaller
baghouse
to
remove
the
injected
carbon
with
the
adsorbed
mercury.
We
chose
this
costing
approach
to
address
potential
concerns
that
injected
carbon
may
interfere
with
cement
kiln
dust
recycling
practices.
The
national
incremental
annualized
compliance
cost
for
cement
kilns
to
meet
this
beyondthe
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
16.8
million
and
would
provide
an
incremental
reduction
in
mercury
emissions
beyond
the
MACT
floor
controls
of
0.41
tons
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
evaluated
to
estimate
the
impacts
between
activated
carbon
injection
and
controls
likely
to
be
used
to
meet
the
floor
level.
We
estimate
that
this
beyond­
the­
floor
option
would
increase
the
amount
of
solid
waste
generated
by
4,400
tons
per
year
and
would
require
sources
to
use
an
additional
21
million
kW­
hours
per
year
beyond
the
requirements
to
achieve
the
floor
level.
The
costs
associated
with
these
impacts
are
accounted
for
in
the
national
annualized
compliance
cost
estimates.
Therefore,
based
on
these
factors
and
costs
of
approximately
$
41
million
per
additional
ton
of
mercury
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
activated
carbon
injection.
Feed
Control
of
Mercury
in
the
Hazardous
Waste.
We
also
evaluated
a
beyond­
the­
floor
level
of
51
ug/
dscm,
which
represents
a
20%
reduction
from
the
floor
level.
We
chose
a
20%
reduction
as
a
level
representing
the
practicable
extent
that
additional
feedrate
control
of
mercury
in
hazardous
waste
(
beyond
feedrate
control
that
may
be
necessary
to
achieve
the
floor
level)
can
be
used
and
still
achieve
modest
emissions
reductions.
12
The
national
incremental
annualized
compliance
cost
for
cement
kilns
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
3.7
million
and
would
provide
an
incremental
reduction
in
mercury
emissions
beyond
the
MACT
floor
controls
of
180
pounds
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
also
evaluated.
Therefore,
based
on
these
factors
and
costs
of
approximately
$
42
million
per
additional
ton
of
mercury
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
feed
control
of
mercury
in
the
hazardous
waste.
Feed
Control
of
Mercury
in
the
Raw
Materials
and
Auxiliary
Fuels.
Cement
kilns
could
achieve
a
reduction
in
mercury
emissions
by
substituting
a
raw
material
containing
lower
levels
of
mercury
for
a
primary
raw
material
with
a
higher
level.
We
believe
that
this
beyond­
the­
floor
option
would
be
even
less
cost­
effective
than
either
of
the
options
discussed
above,
however.
Given
that
sources
are
sited
near
the
supply
of
the
primary
raw
material,
transporting
large
quantities
of
an
alternate
source
of
raw
materials
is
likely
to
be
cost­
prohibitive,
especially
Redline­
strikeout
highlighting
changes
made
during
OMB
review
133
considering
the
small
expected
emissions
reductions
that
would
result.
We
also
considered
whether
fuel
switching
to
an
auxiliary
fuel
containing
a
lower
concentration
of
mercury
would
be
an
appropriate
control
option
for
sources.
Given
that
most
cement
kilns
burning
hazardous
waste
also
burn
coal
as
a
fuel,
we
considered
switching
to
natural
gas
as
a
potential
beyond­
the­
floor
option.
We
are
concerned
about
the
availability
of
natural
gas
to
all
cement
kilns
because
natural
gas
pipelines
are
not
available
in
all
regions
of
the
United
States.
See
68
FR
1673.
Moreover,
even
where
pipelines
provide
access
to
natural
gas,
supplies
of
natural
gas
may
not
be
adequate.
For
example,
it
is
common
practice
in
cities
during
winter
months
(
or
periods
of
peak
demand)
to
prioritize
natural
gas
usage
for
residential
areas
before
industrial
usage.
Requiring
cement
kilns
to
switch
to
natural
gas
would
place
an
even
greater
strain
on
natural
gas
resources.
Consequently,
even
where
pipelines
exist,
some
sources
may
not
be
able
to
use
natural
gas
during
times
of
limited
supplies.
Thus,
natural
gas
may
not
be
a
viable
control
option
for
some
sources.
Therefore,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
limiting
mercury
in
the
raw
material
feed
and
auxiliary
fuels.
For
the
reasons
discussed
above,
we
propose
not
to
adopt
a
beyond­
the­
floor
standard
for
mercury
and
propose
to
establish
the
emission
standard
for
existing
cement
kilns
at
64
ug/
dscm.
If
we
were
to
adopt
such
a
standard,
we
are
proposing
that
sources
comply
with
the
standard
on
an
annual
basis
because
it
is
based
on
normal
emissions
data.
3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
Mercury
emissions
from
new
cement
kilns
are
currently
limited
to
120
ug/
dscm
by
§
63.1204(
b)(
2).
New
cement
kilns
can
comply
with
an
alternative
mercury
standard
that
limits
the
hazardous
waste
maximum
theoretical
emissions
concentration
or
MTEC
of
mercury
of
120
ug/
dscm.
This
standard
was
promulgated
in
the
Interim
Standards
Rule
(
See
67
FR
at
6796).
The
MACT
floor
for
new
sources
for
mercury
would
be
35
ug/
dscm,
which
considers
emissions
variability,
based
on
a
hazardous
waste
MTEC
of
5.1
ug/
dscm.
This
is
an
emission
level
that
the
single
best
performing
source
identified
with
the
SRE/
Feed
Approach
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
conditions
identical
to
the
test
conditions
during
which
the
emissions
data
were
obtained.
As
for
existing
sources,
we
assumed
all
sources
equally
achieved
a
SRE
of
zero.
The
effect
of
this
assumption
is
that
the
single
source
with
the
lowest
mercury
concentration
in
the
hazardous
waste
was
identified
as
the
best
performing
source.
We
also
invite
comment
on
whether
we
should
judge
an
annual
limit
of
35
ug/
dscm
as
less
stringent
than
either
the
current
emission
standard
of
120
ug/
dscm
or
the
hazardous
waste
MTEC
of
mercury
of
120
ug/
dscm
for
cement
kilns
(
so
as
to
avoid
any
backsliding
from
a
current
level
of
performance
achieved
by
all
sources).
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
We
identified
the
same
three
potential
beyond­
the­
floor
techniques
for
control
of
mercury:
(
1)
use
of
activated
carbon;
(
2)
control
of
mercury
in
the
hazardous
waste
feed;
and
(
3)
control
of
the
mercury
in
the
raw
materials
and
auxiliary
fuels.
Use
of
Activated
Carbon
Injection.
We
evaluated
activated
carbon
injection
as
beyondthe
floor
control
for
further
reduction
of
mercury
emissions.
We
made
a
conservative
assumption
that
the
use
of
activated
carbon
injection
will
provide
70%
mercury
control
and
evaluated
a
beyond­
the­
floor
level
of
11
ug/
dscm.
The
incremental
annualized
compliance
cost
for
a
new
Redline­
strikeout
highlighting
changes
made
during
OMB
review
13
A
greenfield
cement
kiln
is
a
kiln
constructed
at
a
site
where
no
cement
kiln
previously
existed;
however,
a
newly
constructed
or
reconstructed
cement
kiln
at
an
existing
site
would
not
be
considered
as
a
greenfield
cement
kiln.

134
cement
kiln
to
meet
this
beyond­
the­
floor
level,
rather
than
comply
with
the
floor
level,
would
be
approximately
$
1.0
million
and
would
provide
an
incremental
reduction
in
mercury
emissions
of
approximately
88
pounds
per
year.
We
also
estimate
that
this
option
would
increase
the
amount
of
solid
waste
generated
by
400
tons
per
year
and
would
require
sources
to
use
an
additional
1.9
million
kW­
hours
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
are
accounted
for
in
the
national
annualized
compliance
cost
estimates.
Therefore,
based
on
these
factors
and
costs
of
$
23
million
per
ton
of
mercury
removed,
we
are
not
proposing
a
beyond­
thefloor
standard
based
on
activated
carbon
injection
for
new
cement
kilns.
Feed
Control
of
Mercury
in
the
Hazardous
Waste.
We
also
believe
that
the
expense
for
further
reduction
in
mercury
emissions
based
on
further
control
of
mercury
concentrations
in
the
hazardous
waste
is
not
warranted.
A
beyond­
the­
floor
level
of
28
ug/
dscm,
which
represents
a
20%
reduction
from
the
floor
level,
would
result
in
little
additional
mercury
reductions.
For
similar
reasons
discussed
above
for
existing
sources,
we
conclude
that
a
beyond­
the­
floor
standard
based
on
controlling
the
mercury
in
the
hazardous
waste
feed
would
not
be
justified
because
of
the
costs
coupled
with
estimated
emission
reductions.
Feed
Control
of
Mercury
in
the
Raw
Materials
and
Auxiliary
Fuels.
Cement
kilns
could
achieve
a
reduction
in
mercury
emissions
by
substituting
a
raw
material
containing
lower
levels
of
mercury
for
a
primary
raw
material
with
a
higher
level.
For
a
new
source
at
an
existing
cement
plant,
we
believe
that
this
beyond­
the­
floor
option
would
not
be
cost­
effective
due
to
the
costs
of
transporting
large
quantities
of
an
alternate
source
of
raw
materials
to
the
cement
plant.
Given
that
the
plant
site
already
exists
and
sited
near
the
source
of
raw
material,
replacing
the
raw
materials
at
the
plant
site
with
lower
mercury­
containing
materials
would
be
the
source's
only
option.
For
a
new
cement
kiln
constructed
at
a
new
site
 
a
greenfield
site13
 
we
are
not
aware
of
any
information
and
data
from
a
source
that
has
undertaken
or
is
currently
located
at
a
site
whose
raw
materials
are
low
in
mercury
which
would
consistently
decrease
mercury
emissions.
Further,
we
are
uncertain
as
to
what
beyond­
the­
floor
standard
would
be
achievable
using
a
lower,
if
it
exists,
mercury­
containing
raw
material.
Although
we
are
doubtful
that
selecting
a
new
plant
site
based
on
the
content
of
metals
in
the
raw
material
is
a
realistic
beyond­
the­
floor
option
considering
the
numerous
additional
factors
that
go
into
such
a
decision,
we
solicit
comment
on
whether
and
what
level
of
a
beyond­
the­
floor
standard
based
on
controlling
the
level
of
mercury
in
the
raw
materials
is
appropriate.
We
also
considered
whether
fuel
switching
to
an
auxiliary
fuel
containing
a
lower
concentration
of
mercury
would
be
an
appropriate
control
option
for
sources.
We
considered
using
natural
gas
in
lieu
of
a
fossil
fuel
such
as
coal
containing
higher
concentrations
of
mercury
as
a
potential
beyond­
the­
floor
option.
As
discussed
for
existing
sources,
we
are
concerned
about
the
availability
of
the
natural
gas
infrastructure
in
all
regions
of
the
United
States
and
believe
that
using
natural
gas
would
not
be
a
viable
control
option
for
all
new
sources.
Therefore,
we
are
not
Redline­
strikeout
highlighting
changes
made
during
OMB
review
14
This
standard
equates
approximately
to
a
stack
gas
concentration
level
of
0.030
gr/
dscf
for
wet
process
kilns
and
0.040
gr/
dscf
for
preheater/
precalciner
kilns.
The
conversion
varies
by
process
type
because
the
amount
of
flue
gas
generated
per
ton
of
raw
material
feed
varies
by
process
type.

135
proposing
a
beyond­
the­
floor
standard
based
on
limiting
mercury
in
the
raw
material
feed
and
auxiliary
fuels.
Therefore,
we
propose
a
mercury
standard
of
35
ug/
dscm
for
new
sources.
If
we
were
to
adopt
such
a
standard,
we
are
proposing
that
sources
comply
with
the
standard
on
an
annual
basis
because
it
is
based
on
normal
emissions
data.
C.
What
Are
the
Proposed
Standards
for
Particulate
Matter?
We
are
proposing
to
establish
standards
for
existing
and
new
cement
kilns
that
limit
emissions
of
particulate
matter
to
65
mg/
dscm
(
0.028
gr/
dscf)
and
13
mg/
dscm
(
0.0058
gr/
dscf),
respectively.
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
Particulate
matter
emissions
for
existing
cement
kilns
are
currently
limited
to
0.15
kilograms
of
particulate
matter
per
megagram
dry
feed14
and
20%
opacity
by
§
63.1204(
a)(
7).
This
standard
was
promulgated
in
the
Interim
Standards
Rule
(
See
67
FR
at
6796).
The
particulate
matter
standard
is
a
surrogate
control
for
the
metals
antimony,
cobalt,
manganese,
nickel,
and
selenium
in
the
hazardous
waste
and
all
HAP
metals
in
the
raw
materials
and
auxiliary
fuels
which
are
controllable
by
particulate
matter
control.
All
cement
kilns
control
particulate
matter
with
baghouses
and
electrostatic
precipitators.
We
have
compliance
test
emissions
data
representing
maximum
emissions
for
all
cement
kiln
sources.
For
most
sources,
we
have
compliance
test
emissions
data
from
more
than
one
compliance
test
campaign.
Our
data
base
of
particulate
matter
stack
emission
concentrations
range
from
0.0008
to
0.063
gr/
dscf.
To
identify
the
floor
level,
we
evaluated
the
compliance
test
emissions
data
associated
with
the
most
recent
test
campaign
using
the
Air
Pollution
Control
Technology
Approach.
The
calculated
floor
is
65
mg/
dscm
(
0.028
gr/
dscf),
which
considers
emissions
variability.
This
is
an
emission
level
that
the
average
of
the
best
performing
sources
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
We
estimate
that
this
emission
level
is
being
achieved
by
44%
of
sources
and
would
reduce
particulate
matter
emissions
by
43
tons
per
year.
We
are
also
proposing
to
delete
the
current
opacity
standard
in
conjunction
with
revisions
to
the
compliance
assurance
requirements
for
particulate
matter
for
cement
kilns.
These
proposed
compliance
assurance
amendments
include
requiring
a
cement
kiln
source
using
a
baghouse
to
comply
with
the
same
bag
leak
detection
system
requirements
that
are
currently
applicable
to
all
other
hazardous
waste
combustors
(
see
§
63.1209(
m)).
A
cement
kiln
source
using
an
ESP
has
the
option
either
to
(
1)
use
a
particulate
matter
emissions
detector
as
a
process
monitor
in
lieu
of
complying
with
operating
parameter
limits,
as
we
are
proposing
for
all
other
hazardous
waste
combustor
sources;
or
(
2)
establish
site­
specific,
enforceable
operating
parameter
limits
that
are
Redline­
strikeout
highlighting
changes
made
during
OMB
review
15
We
did
not
evaluate
a
beyond­
the­
floor
standard
based
on
fuel
substitution
because
particulate
matter
emissions
from
cement
kilns
are
primarily
entrained
raw
material,
not
ash
contributed
by
the
hazardous
waste
fuel.
There
is,
therefore,
no
correlation
between
particulate
matter
emissions
and
the
level
of
ash
in
the
hazardous
waste.

136
linked
to
the
automatic
waste
feed
cutoff
system.
See
Part
Three,
Section
III
for
a
discussion
of
the
proposed
changes.
We
also
request
comment
on
whether
the
particulate
matter
standard
should
be
expressed
on
a
concentration
basis
(
as
proposed
today)
or
on
a
production­
based
format.
A
concentrationbased
standard
is
expressed
as
mass
of
particulate
matter
per
dry
standard
volume
of
gas
(
e.
g.,
mg/
dscm
as
proposed
today)
while
a
production­
based
standard
is
expressed
as
mass
of
particulate
matter
emitted
per
mass
of
dry
raw
material
feed
to
the
kiln
(
e.
g.,
the
format
of
the
interim
standard).
We
evaluated
the
compliance
test
production­
based
data
associated
with
the
most
recent
test
campaign
to
determine
what
the
floor
level
would
be
under
this
approach.
The
calculated
floor
would
be
0.10
kilograms
of
particulate
matter
per
megagram
dry
feed.
We
note
that
a
concentration
format
can
be
viewed
as
penalizing
more
energy
efficient
kilns,
which
burn
less
fuel
and
produce
less
kiln
exhaust
gas
per
megagram
of
dry
feed.
This
is
because
with
a
concentration­
based
standard
the
more
energy­
efficient
kilns
would
be
restricted
to
a
lower
level
of
particulate
matter
emitted
per
unit
of
production.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
evaluated
improved
particulate
matter
control
to
achieve
a
beyond­
the­
floor
standard
of
32
mg/
dscm
(
0.014
gr/
dscf),
which
is
a
50%
reduction
from
MACT
floor
emissions.
15
For
an
existing
source
that
needs
a
significant
reduction
in
particulate
matter
emissions,
we
assumed
and
estimated
costs
for
a
new
baghouse
to
achieve
the
beyond­
the­
floor
level.
If
little
or
modest
emissions
reductions
were
needed,
then
improved
control
was
costed
as
design,
operation,
and
maintenance
modifications
of
the
existing
particulate
matter
control
equipment.
The
national
incremental
annualized
compliance
cost
for
cement
kilns
to
meet
this
beyondthe
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
4.8
million
and
would
provide
an
incremental
reduction
in
particulate
matter
emissions
beyond
the
MACT
floor
controls
of
385
tons
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
evaluated
to
estimate
the
impacts
between
further
improvements
to
control
particulate
matter
and
controls
likely
to
be
used
to
meet
the
floor
level.
We
estimate
that
this
beyond­
thefloor
option
would
increase
the
amount
of
solid
waste
generated
by
385
tons
per
year
and
would
require
sources
to
use
an
additional
15
million
kW­
hours
per
year
beyond
the
requirements
to
achieve
the
floor
level.
The
costs
associated
with
these
impacts
are
accounted
for
in
the
national
annualized
compliance
cost
estimates.
Therefore,
based
on
these
factors
and
costs
of
approximately
$
12,400
per
additional
ton
of
particulate
matter
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
improved
particulate
matter
control.
3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
Particulate
matter
emissions
from
new
cement
kilns
are
currently
limited
to
0.15
kilograms
of
particulate
matter
per
megagram
dry
feed
and
20%
opacity
by
§
63.1204(
b)(
7).
This
standard
Redline­
strikeout
highlighting
changes
made
during
OMB
review
137
was
promulgated
in
the
Interim
Standards
Rule
(
See
67
FR
at
6796).
The
MACT
floor
for
new
sources
for
particulate
matter
would
be
13
mg/
dscm
(
0.0058
gr/
dscf),
which
considers
emissions
variability.
This
is
an
emission
level
that
the
single
best
performing
source
identified
with
the
Air
Pollution
Control
Technology
Approach
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
operating
conditions
identical
to
the
test
conditions
during
which
the
emissions
data
were
obtained.
We
are
also
proposing
to
delete
the
current
opacity
standard
in
conjunction
with
revisions
to
the
compliance
assurance
requirements
for
particulate
matter
for
cement
kilns.
See
Part
Three,
Section
III
for
details.
As
discussed
for
existing
sources,
we
also
request
comment
on
whether
the
particulate
matter
standard
should
be
expressed
on
a
concentration
basis
or
on
a
production­
based
format.
We
evaluated
the
compliance
test
production­
based
data
associated
with
the
most
recent
test
campaign
to
determine
what
the
floor
level
would
be
under
this
approach.
The
calculated
floor
would
be
0.028
kilograms
of
particulate
matter
per
megagram
dry
feed.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
We
evaluated
improved
emissions
control
based
on
a
state­
of­
the­
art
baghouse
using
a
high
quality
fabric
filter
bag
material
to
achieve
a
beyond­
the­
floor
standard
of
6.7
mg/
dscm
(
0.0029
gr/
dscf).
This
reduction
represents
a
50%
reduction
in
particulate
matter
emissions
from
MACT
floor
levels.
The
incremental
annualized
compliance
cost
for
a
new
cement
kiln
to
meet
this
beyond­
the­
floor
level,
rather
than
comply
with
the
floor
level,
would
be
approximately
$
0.38
million
and
would
provide
an
incremental
reduction
in
particulate
matter
emissions
of
approximately
2.6
tons
per
year.
We
estimate
that
this
beyond­
the­
floor
option
would
increase
the
amount
of
solid
waste
generated
by
less
than
6
tons
per
year
and
would
require
sources
to
use
an
additional
1.8
million
kW­
hours
per
year
beyond
the
requirements
to
achieve
the
floor
level.
The
costs
associated
with
these
impacts
are
accounted
for
in
the
national
annualized
compliance
cost
estimates.
Therefore,
based
on
these
factors
and
costs
of
approximately
$
61,400
per
additional
ton
of
particulate
matter
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
improved
particulate
matter
control
for
new
cement
kilns.
Therefore,
we
propose
a
particulate
matter
standard
of
13
mg/
dscm
for
new
sources.
D.
What
Are
the
Proposed
Standards
for
Semivolatile
Metals?
We
are
proposing
to
establish
standards
for
existing
cement
kilns
that
limit
emissions
of
semivolatile
metals
(
cadmium
and
lead,
combined)
to
4.0
x
10­
4
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste.
The
proposed
standard
for
new
sources
is
6.2
x
10­
5
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste.
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
Semivolatile
metals
emissions
from
existing
cement
kilns
are
currently
limited
to
330
ug/
dscm
by
§
63.1204(
a)(
3).
This
standard
was
promulgated
in
the
Interim
Standards
Rule
(
See
67
FR
at
6796).
Cement
kilns
control
emissions
of
semivolatile
metals
with
baghouses
or
electrostatic
precipitators
and/
or
by
controlling
the
feed
concentration
of
semivolatile
metals
in
the
hazardous
waste.
We
have
compliance
test
emissions
data
representing
maximum
emissions
for
all
cement
kiln
sources.
For
most
sources,
we
have
compliance
test
emissions
data
from
more
than
one
Redline­
strikeout
highlighting
changes
made
during
OMB
review
16
USEPA,
"
Draft
Technical
Support
Document
for
HWC
MACT
Replacement
Standards,
Volume
III:
Selection
of
MACT
Standards
and
Technologies,"
March
2004,
Chaper
23.

138
compliance
test
campaign.
Semivolatile
metal
stack
emissions
range
from
approximately
1
to
2,800
ug/
dscm.
These
emissions
are
expressed
as
mass
of
semivolatile
metals
(
from
all
feedstocks)
per
unit
volume
of
stack
gas.
Hazardous
waste
thermal
emissions
range
from
3.0
x
10­
6
to
3.7
x
10­
3
lbs
per
million
Btu.
Hazardous
waste
thermal
emissions
represent
the
mass
of
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste.
Lead
was
the
most
significant
contributor
to
semivolatile
emissions
during
compliance
test
conditions.
To
identify
the
MACT
floor,
we
evaluated
the
compliance
test
emissions
data
associated
with
the
most
recent
test
campaign
using
the
SRE/
Feed
Approach.
The
calculated
floor
is
4.0
x
10­
4
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste,
which
considers
emissions
variability.
This
is
an
emission
level
that
the
average
of
the
best
performing
sources
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
We
estimate
that
this
emission
level
is
being
achieved
by
81%
of
sources
and
would
reduce
semivolatile
metals
emissions
by
1
ton
per
year.
To
put
the
proposed
floor
level
in
context
for
a
hypothetical
wet
process
cement
kiln
that
gets
50%
of
its
required
heat
input
from
hazardous
waste,
a
thermal
emissions
level
of
4.0
x
10­
4
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste
equates
approximately
to
a
stack
gas
concentration
of
180
ug/
dscm.
This
estimated
stack
gas
concentration
does
not
include
contributions
to
emission
from
other
semivolatile
metals­
containing
materials
such
as
raw
materials
and
fossil
fuels.
The
additional
contribution
to
stack
emissions
of
semivolatile
metals
in
an
average
raw
material
and
coal
is
estimated
to
range
as
high
as
20
to
50
ug/
dscm.
Thus,
for
the
hypothetical
wet
process
cement
kiln
the
thermal
emissions
floor
level
of
4.0
x
10­
4
lbs
semivolatile
metals
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste
is
estimated
to
be
less
than
230
ug/
dscm,
which
is
less
than
the
current
interim
standard
of
330
ug/
dscm.
Given
that
comparing
the
proposed
floor
level
to
the
interim
standard
requires
numerous
assumptions
(
as
just
illustrated)
including
hazardous
waste
fuel
replacement
rates,
heat
input
requirements
per
ton
of
clinker,
concentrations
of
semivolatile
metals
in
the
raw
material
and
coal,
and
system
removal
efficiency,
we
have
a
more
detailed
analysis
in
the
background
document.
16
Our
detailed
analysis
indicates
the
proposed
floor
level
is
as
least
as
stringent
as
the
interim
standard
(
so
as
to
avoid
any
backsliding
from
a
current
level
of
performance
achieved
by
all
cement
kilns,
and
hence,
the
level
of
minimal
stringency
at
which
EPA
could
calculate
the
MACT
floor).
Thus,
we
conclude
that
a
dual
standard
 
the
semivolatile
metals
standard
as
both
the
calculated
floor
level,
expressed
as
a
hazardous
waste
thermal
emissions
level,
and
the
current
interim
standard
 
is
not
needed
for
this
standard.
In
the
September
1999
final
rule,
we
acknowledged
that
a
cement
kiln
using
properly
Redline­
strikeout
highlighting
changes
made
during
OMB
review
139
designed
and
operated
MACT
control
technologies,
including
controlling
the
levels
of
metals
in
the
hazardous
waste,
may
not
be
capable
of
achieving
a
given
emission
standard
because
of
mineral
and
process
raw
material
contributions
that
might
cause
an
exceedance
of
the
emission
standard.
To
address
this
concern,
we
promulgated
a
provision
that
allows
kilns
to
petition
for
alternative
standards
provided
that
they
submit
site­
specific
information
that
shows
raw
material
hazardous
air
pollutant
contributions
to
the
emissions
prevent
the
source
from
complying
with
the
emission
standard
even
though
the
kiln
is
using
MACT
control.
See
§
63.1206(
b)(
10).
If
we
were
to
adopt
the
semivolatile
(
and
low
volatile)
metals
standard
using
a
thermal
emissions
format,
then
there
would
be
no
need
for
these
alternative
standard
provisions
for
semivolatile
metals
(
since,
as
explained
earlier,
that
standard
is
based
solely
on
semivolatile
metals
contributions
from
hazardous
waste
fuels).
Therefore,
we
would
delete
the
provisions
of
§
63.1206(
b)(
10)
as
they
apply
to
semivolatile
(
and
low
volatile)
metals.
We
invite
comment
on
this
approach.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
identified
three
potential
beyond­
the­
floor
techniques
for
control
of
semivolatile
metals:
(
1)
improved
particulate
matter
control;
(
2)
control
of
semivolatile
metals
in
the
hazardous
waste
feed;
and
(
3)
control
of
the
semivolatile
metals
in
the
raw
materials
and
fuels.
For
reasons
discussed
below,
we
are
not
proposing
a
beyond­
the­
floor
standard
for
semivolatile
metals.
Improved
Particulate
Matter
Control.
Controlling
particulate
matter
also
controls
emissions
of
semivolatile
metals.
Our
data
show
that
all
cement
kilns
are
already
achieving
greater
than
98.6%
system
removal
efficiency
for
semivolatile
metals,
with
most
attaining
99.9%
removal.
Thus,
additional
control
of
particulate
matter
are
likely
to
result
in
only
modest
additional
reductions
of
semivolatile
metals
emissions.
We
evaluated
a
beyond­
the­
floor
level
of
2.0
x
10­
4
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste,
which
represents
a
50%
reduction
in
emissions
from
MACT
floor
levels.
The
national
incremental
annualized
compliance
cost
for
cement
kilns
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
2.7
million
and
would
provide
an
incremental
reduction
in
semivolatile
metals
emissions
beyond
the
MACT
floor
controls
of
1.2
tons
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
evaluated
to
estimate
the
impacts
between
further
improvements
to
control
particulate
matter
and
controls
likely
to
be
used
to
meet
the
floor
level.
We
estimate
that
this
beyond­
the­
floor
option
would
increase
the
amount
of
solid
waste
generated
by
300
tons
per
year
and
would
also
require
sources
to
use
an
additional
5.7
million
kW­
hours
of
energy
per
year
to
achieve
the
floor
level.
The
costs
associated
with
these
impacts
are
accounted
for
in
the
national
annualized
compliance
cost
estimates.
Therefore,
based
on
these
factors
and
costs
of
approximately
$
2.3
million
per
additional
ton
of
semivolatile
metals
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
improved
particulate
matter
control.
Feed
Control
of
Semivolatile
Metals
in
the
Hazardous
Waste.
We
also
evaluated
a
beyond­
the­
floor
level
of
3.2
x
10­
4
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste,
which
represents
a
20%
reduction
from
the
floor
level.
We
chose
a
20%
reduction
as
a
level
representing
the
practicable
extent
that
additional
feedrate
control
of
semivolatile
metals
in
hazardous
waste
can
be
used
and
still
achieve
Redline­
strikeout
highlighting
changes
made
during
OMB
review
140
appreciable
emissions
reductions.
The
national
incremental
annualized
compliance
cost
for
cement
kilns
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
0.30
million
and
would
provide
an
incremental
reduction
in
semivolatile
metals
emissions
beyond
the
MACT
floor
controls
of
0.36
tons
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
evaluated
and
are
included
in
the
national
compliance
cost
estimates.
Therefore,
based
on
these
factors
and
costs
of
approximately
$
0.84
million
per
additional
ton
of
semivolatile
metals
removed,
we
are
not
proposing
a
beyond­
thefloor
standard
based
on
feed
control
of
semivolatile
metals
in
the
hazardous
waste.
Feed
Control
of
Semivolatile
Metals
in
the
Raw
Materials
and
Auxiliary
Fuels.
Cement
kilns
could
achieve
a
reduction
in
semivolatile
metal
emissions
by
substituting
a
raw
material
containing
lower
levels
of
lead
and/
or
cadmium
for
a
primary
raw
material
with
higher
levels
of
these
metals.
We
believe
that
this
beyond­
the­
floor
option
would
even
be
less
cost­
effective
than
either
of
the
options
discussed
above,
however.
Given
that
cement
kilns
are
sited
near
the
primary
raw
material
supply,
acquiring
and
transporting
large
quantities
of
an
alternate
source
of
raw
materials
is
likely
to
be
cost­
prohibitive.
Therefore,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
limiting
semivolatile
metals
in
the
raw
material
feed.
We
also
considered
whether
fuel
switching
to
an
auxiliary
fuel
containing
a
lower
concentration
of
semivolatile
metals
would
be
an
appropriate
control
option
for
sources.
Given
that
most
cement
kilns
burning
hazardous
waste
also
burn
coal
as
a
fuel,
we
considered
switching
to
natural
gas
as
a
potential
beyond­
the­
floor
option.
For
the
same
reasons
discussed
for
mercury,
we
judge
a
beyond­
thefloor
standard
based
on
fuel
switching
as
unwarranted.
For
the
reasons
discussed
above,
we
propose
to
establish
the
emission
standard
for
existing
cement
kilns
at
4.0
x
10­
4
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste.
3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
Semivolatile
metals
emissions
from
new
cement
kilns
are
currently
limited
to
180
ug/
dscm
by
§
63.1204(
b)(
3).
This
standard
was
promulgated
in
the
Interim
Standards
Rule
(
See
67
FR
at
6796).
The
MACT
floor
for
new
sources
for
semivolatile
metals
would
be
6.2
x
10­
5
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste,
which
considers
emissions
variability.
This
is
an
emission
level
that
the
single
best
performing
source
identified
with
the
SRE/
Feed
Approach
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
conditions
identical
to
the
test
conditions
during
which
the
emissions
data
were
obtained.
To
put
the
proposed
floor
level
in
context
for
a
hypothetical
wet
process
cement
kiln
that
gets
50%
of
its
required
heat
input
from
hazardous
waste,
a
thermal
emissions
level
of
6.2
x
10­
5
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste
equates
approximately
to
a
stack
gas
concentration
of
80
ug/
dscm,
including
contributions
from
typical
raw
materials
and
coal.
Thus,
for
the
hypothetical
wet
process
cement
kiln
the
thermal
emissions
floor
level
of
6.2
x
10­
5
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste
is
estimated
to
be
less
than
the
current
interim
standard
for
new
sources
of
180
ug/
dscm.
Redline­
strikeout
highlighting
changes
made
during
OMB
review
141
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
We
identified
the
same
three
potential
beyond­
the­
floor
techniques
for
control
of
semivolatile
metals:
(
1)
improved
control
of
particulate
matter;
(
2)
control
of
semivolatile
metals
in
the
hazardous
waste
feed;
and
(
3)
control
of
semivolatile
metals
in
the
raw
materials
and
fuels.
Improved
Particulate
Matter
Control.
Controlling
particulate
matter
also
controls
emissions
of
semivolatile
metals.
We
evaluated
improved
control
of
particulate
matter
based
on
a
state­
of­
the­
art
baghouse
using
a
high
quality
fabric
filter
bag
material
as
beyond­
the­
floor
control
for
further
reductions
in
semivolatile
metals
emissions.
We
evaluated
a
beyond­
the­
floor
level
of
2.5
x
10­
5
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste.
The
incremental
annualized
compliance
cost
for
a
new
cement
kiln
with
an
average
gas
flow
rate
to
meet
this
beyond­
the­
floor
level,
rather
than
to
comply
with
the
floor
level,
would
be
approximately
$
0.38
million
and
would
provide
an
incremental
reduction
in
semivolatile
metals
emissions
of
approximately
144
pounds
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
evaluated
and
are
included
in
the
cost
estimates.
For
these
reasons
and
costs
of
$
5.3
million
per
ton
of
semivolatile
metals
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
improved
particulate
matter
control
for
new
cement
kilns.
Feed
Control
of
Semivolatile
Metals
in
the
Hazardous
Waste.
We
also
believe
that
the
expense
for
further
reduction
in
semivolatile
metals
emissions
based
on
further
control
of
semivolatile
metals
concentrations
in
the
hazardous
waste
is
not
warranted.
We
also
evaluated
a
beyond­
the­
floor
level
of
5.0
x
10­
5
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste,
which
represents
a
20%
reduction
from
the
floor
level.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
evaluated
and
are
included
in
the
compliance
cost
estimates.
For
similar
reasons
discussed
above
for
existing
sources,
we
conclude
that
a
beyond­
the­
floor
standard
based
on
controlling
the
concentration
of
semivolatile
metals
levels
in
the
hazardous
waste
feed
would
not
be
justified
because
of
the
costs
coupled
with
estimated
emission
reductions.
Feed
Control
of
Semivolatile
Metals
in
the
Raw
Materials
and
Auxiliary
Fuels.
Cement
kilns
could
achieve
a
reduction
in
semivolatile
metals
emissions
by
substituting
a
raw
material
containing
lower
levels
of
cadmium
and
lead
for
a
primary
raw
material
with
a
higher
level.
For
a
new
source
at
an
existing
cement
plant,
we
believe
that
this
beyond­
the­
floor
option
would
not
be
cost­
effective
due
to
the
costs
of
transporting
large
quantities
of
an
alternate
source
of
raw
materials
to
the
cement
plant.
Given
that
the
plant
site
already
exists
and
sited
near
the
source
of
raw
material,
replacing
the
raw
materials
at
the
plant
site
with
lower
semivolatile
metalscontaining
materials
would
be
the
source's
only
option.
For
a
cement
kiln
constructed
at
a
new
greenfield
site,
we
are
not
aware
of
any
information
and
data
from
a
source
that
has
undertaken
or
is
currently
located
at
a
site
whose
raw
materials
are
inherently
lower
in
semivolatile
metals
that
would
consistently
achieve
reduced
semivolatile
metals
emissions.
Further,
we
are
uncertain
as
to
what
beyond­
the­
floor
standard
would
be
achievable
using
a
lower,
if
it
exists,
semivolatile
metals­
containing
raw
material.
Although
we
are
doubtful
that
selecting
a
new
plant
site
based
on
the
content
of
metals
in
the
raw
material
is
a
realistic
beyond­
the­
floor
option
considering
the
numerous
additional
factors
that
go
into
such
a
decision,
we
solicit
comment
on
whether
and
what
Redline­
strikeout
highlighting
changes
made
during
OMB
review
142
level
of
a
beyond­
the­
floor
standard
based
on
controlling
the
level
of
semivolatile
metals
in
the
raw
materials
is
appropriate.
We
also
considered
whether
fuel
switching
to
an
auxiliary
fuel
containing
a
lower
concentration
of
semivolatile
metals
would
be
an
appropriate
control
option
for
sources.
Given
that
most
cement
kilns
burning
hazardous
waste
also
burn
coal
as
a
fuel,
we
considered
switching
to
natural
gas
as
a
potential
beyond­
the­
floor
option.
For
the
same
reasons
discussed
for
mercury,
we
judge
a
beyond­
the­
floor
standard
based
on
fuel
switching
as
unwarranted.
For
the
reasons
discussed
above,
we
propose
to
establish
the
emission
standard
for
new
cement
kilns
at
6.2
x
10­
5
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste.
E.
What
Are
the
Proposed
Standards
for
Low
Volatile
Metals?
We
are
proposing
to
establish
standards
for
existing
and
new
cement
kilns
that
limit
emissions
of
low
volatile
metals
(
arsenic,
beryllium,
and
chromium,
combined)
to
1.4
x
10­
5
lbs
low
volatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste.
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
Low
volatile
metals
emissions
from
existing
cement
kilns
are
currently
limited
to
56
ug/
dscm
by
§
63.1204(
a)(
4).
This
standard
was
promulgated
in
the
Interim
Standards
Rule
(
See
67
FR
at
6796).
Cement
kilns
control
emissions
of
low
volatile
metals
with
baghouses
or
electrostatic
precipitators
and/
or
by
controlling
the
feed
concentration
of
low
volatile
metals
in
the
hazardous
waste.
We
have
compliance
test
emissions
data
representing
maximum
emissions
for
all
cement
kiln
sources.
For
most
sources,
we
have
compliance
test
emissions
data
from
more
than
one
compliance
test
campaign.
Low
volatile
metal
stack
emissions
range
from
approximately
1
to
100
ug/
dscm.
These
emissions
are
expressed
as
mass
of
low
volatile
metals
(
from
all
feedstocks)
per
unit
volume
of
stack
gas.
Hazardous
waste
thermal
emissions
range
from
9.2
x
10­
7
to
1.0
x
10­
5
lbs
per
million
Btu.
Hazardous
waste
thermal
emissions
represent
the
mass
of
low
volatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste.
For
nearly
every
cement
kiln,
chromium
was
the
most
significant
contributor
to
low
volatile
emissions.
To
identify
the
MACT
floor,
we
evaluated
the
compliance
test
emissions
data
associated
with
the
most
recent
test
campaign
using
the
SRE/
Feed
Approach.
The
calculated
floor
is
1.4
x
10­
5
lbs
low
volatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste,
which
considers
emissions
variability.
This
is
an
emission
level
that
the
average
of
the
best
performing
sources
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
We
estimate
that
this
emission
level
is
being
achieved
by
52%
of
sources
and
would
reduce
low
volatile
metals
emissions
by
0.10
tons
per
year.
To
put
the
proposed
floor
level
in
context
for
a
hypothetical
wet
process
cement
kiln
that
gets
50%
of
its
required
heat
input
from
hazardous
waste,
a
thermal
emissions
level
of
1.4
x
10­
5
lbs
low
volatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste
equates
approximately
to
a
stack
gas
concentration
of
7
ug/
dscm.
This
Redline­
strikeout
highlighting
changes
made
during
OMB
review
17
USEPA,
"
Draft
Technical
Support
Document
for
HWC
MACT
Replacement
Standards,
Volume
III:
Selection
of
MACT
Standards
and
Technologies,"
March
2004,
Chapter
23.

143
estimated
stack
gas
concentration
does
not
include
contributions
to
emission
from
other
low
volatile
metals­
containing
materials
such
as
raw
materials
and
fossil
fuels.
The
additional
contribution
to
stack
emissions
of
low
volatile
metals
in
an
average
raw
material
and
coal
is
estimated
to
range
from
less
than
1
to
15
ug/
dscm.
Thus,
for
the
hypothetical
wet
process
cement
kiln
the
thermal
emissions
floor
level
of
1.4
x
10­
5
lbs
low
volatile
metals
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste
is
estimated
to
be
less
than
22
ug/
dscm,
which
is
less
than
the
current
interim
standard
of
56
ug/
dscm.
Given
that
comparing
the
proposed
floor
level
to
the
interim
standard
requires
numerous
assumptions
(
as
just
illustrated)
including
hazardous
waste
fuel
replacement
rates,
heat
input
requirements
per
ton
of
clinker,
concentrations
of
low
volatile
metals
in
the
raw
material
and
coal,
and
system
removal
efficiency,
we
have
included
a
more
detailed
analysis
in
the
background
document.
17
Our
detailed
analysis
indicates
the
proposed
floor
level
is
as
least
as
stringent
as
the
interim
standard
(
so
as
to
avoid
any
backsliding
from
a
current
level
of
performance
achieved
by
all
cement
kilns,
and
hence,
the
level
of
minimal
stringency
at
which
EPA
could
calculate
the
MACT
floor).
Thus,
we
conclude
that
a
dual
standard
 
the
low
volatile
metals
standard
as
both
the
calculated
floor
level,
expressed
as
a
hazardous
waste
thermal
emissions
level,
and
the
current
interim
standard
 
is
not
needed
for
this
standard.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
identified
three
potential
beyond­
the­
floor
techniques
for
control
of
low
volatile
metals:
(
1)
improved
particulate
matter
control;
(
2)
control
of
low
volatile
metals
in
the
hazardous
waste
feed;
and
(
3)
control
of
the
low
volatile
metals
in
the
raw
materials.
For
reasons
discussed
below,
we
are
not
proposing
a
beyond­
the­
floor
standard
for
low
volatile
metals.
Improved
Particulate
Matter
Control.
Controlling
particulate
matter
also
controls
emissions
of
low
volatile
metals.
Our
data
show
that
all
cement
kilns
are
already
achieving
greater
than
99.9%
system
removal
efficiency
for
low
volatile
metals,
with
most
attaining
99.99%
removal.
Thus,
additional
control
of
particulate
matter
emissions
is
likely
to
result
in
only
a
small
increment
in
reduction
of
low
volatile
metals
emissions.
We
evaluated
a
beyond­
the­
floor
level
of
7.0
x
10­
6
lbs
low
volatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste,
which
represents
a
50%
reduction
in
emissions
from
MACT
floor
levels.
The
national
incremental
annualized
compliance
cost
for
cement
kilns
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
3.7
million
and
would
provide
an
incremental
reduction
in
low
volatile
metals
emissions
beyond
the
MACT
floor
controls
of
120
pounds
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
evaluated
to
estimate
the
impacts
between
further
improvements
to
control
particulate
matter
and
controls
likely
to
be
used
to
meet
the
floor
level.
We
estimate
that
this
beyond­
the­
floor
option
would
increase
the
amount
of
solid
waste
generated
by
72
tons
per
year
and
would
also
require
sources
to
use
an
additional
1.2
million
kW­
hours
per
year
beyond
Redline­
strikeout
highlighting
changes
made
during
OMB
review
144
the
requirements
to
achieve
the
floor
level.
The
costs
associated
with
these
impacts
are
accounted
for
in
the
national
annualized
compliance
cost
estimates.
Therefore,
based
on
these
factors
and
costs
of
approximately
$
63
million
per
additional
ton
of
low
volatile
metals
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
improved
particulate
matter
control.
Feed
Control
of
Low
Volatile
Metals
in
the
Hazardous
Waste.
We
also
evaluated
a
beyond­
the­
floor
level
of
1.1
x
10­
5
lbs
low
volatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste,
which
represents
a
20%
reduction
from
the
floor
level.
We
chose
a
20%
reduction
as
a
level
representing
the
practicable
extent
that
additional
feedrate
control
of
mercury
in
hazardous
waste
can
be
used
and
still
achieve
appreciable
emissions
reductions.
The
national
incremental
annualized
compliance
cost
for
cement
kilns
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
1.2
million
and
would
provide
an
incremental
reduction
in
low
volatile
metals
emissions
beyond
the
MACT
floor
controls
of
38
pounds
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
evaluated
and
are
included
in
the
cost
estimates.
Therefore,
based
on
these
factors
and
costs
of
approximately
$
64
million
per
additional
ton
of
low
volatile
metals
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
feed
control
of
low
volatile
metals
in
the
hazardous
waste.
Feed
Control
of
Low
Volatile
Metals
in
the
Raw
Materials
and
Auxiliary
Fuels.
Cement
kilns
could
achieve
a
reduction
in
low
volatile
metal
emissions
by
substituting
a
raw
material
containing
lower
levels
of
arsenic,
beryllium,
and/
or
chromium
for
a
primary
raw
material
with
higher
levels
of
these
metals.
We
believe
that
this
beyond­
the­
floor
option
would
even
be
less
cost­
effective
than
either
of
the
options
discussed
above,
however.
Given
that
cement
kilns
are
sited
near
the
primary
raw
material
supply,
acquiring
and
transporting
large
quantities
of
an
alternate
source
of
raw
materials
is
likely
to
be
cost­
prohibitive.
Therefore,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
limiting
low
volatile
metals
in
the
raw
material
feed.
We
also
considered
whether
fuel
switching
to
an
auxiliary
fuel
containing
a
lower
concentration
of
low
volatile
metals
would
be
an
appropriate
control
option
for
sources.
Given
that
most
cement
kilns
burning
hazardous
waste
also
burn
coal
as
a
fuel,
we
considered
switching
to
natural
gas
as
a
potential
beyond­
the­
floor
option.
For
the
same
reasons
discussed
for
mercury,
we
judge
a
beyond­
the­
floor
standard
based
on
fuel
switching
as
unwarranted.
For
the
reasons
discussed
above,
we
propose
to
establish
the
emission
standard
for
existing
cement
kilns
at
1.4
x
10­
5
lbs
low
volatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste.
3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
Low
volatile
metals
emissions
from
new
cement
kilns
are
currently
limited
to
54
ug/
dscm
by
§
63.1204(
b)(
4).
This
standard
was
promulgated
in
the
Interim
Standards
Rule
(
See
67
FR
at
6796,
February
13,
2002).
The
floor
level
for
new
sources
for
low
volatile
metals
would
be
1.4
x
10­
5
lbs
low
volatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste,
which
considers
emissions
variability.
This
is
an
emission
level
that
the
single
best
performing
source
identified
with
the
SRE/
Feed
Approach
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
conditions
identical
to
the
test
conditions
during
which
the
Redline­
strikeout
highlighting
changes
made
during
OMB
review
145
emissions
data
were
obtained.
To
put
the
proposed
floor
level
in
context
for
a
hypothetical
wet
process
cement
kiln
that
gets
50%
of
its
required
heat
input
from
hazardous
waste,
a
thermal
emissions
level
of
1.4
x
10­
5
lbs
low
volatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste
equates
approximately
to
a
stack
gas
concentration
of
22
ug/
dscm,
including
contributions
from
typical
raw
materials
and
coal.
Thus,
for
the
hypothetical
wet
process
cement
kiln
the
thermal
emissions
floor
level
of
6.2
x
10­
5
lbs
low
volatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste
is
estimated
to
be
more
stringent
than
the
current
interim
standard
for
new
sources
of
54
ug/
dscm.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
We
identified
the
same
three
potential
beyond­
the­
floor
techniques
for
control
of
low
volatile
metals:
(
1)
improved
control
of
particulate
matter;
(
2)
control
of
low
volatile
metals
in
the
hazardous
waste
feed;
and
(
3)
control
of
low
volatile
metals
in
the
raw
materials
and
fuels.
Improved
Particulate
Matter
Control.
Controlling
particulate
matter
also
controls
emissions
of
low
volatile
metals.
We
evaluated
improved
control
of
particulate
matter
based
on
a
state­
of­
the­
art
baghouse
using
a
high
quality
fabric
filter
bag
material
as
beyond­
the­
floor
control
for
further
reductions
in
low
volatile
metals
emissions.
We
evaluated
a
beyond­
the­
floor
level
of
6.0
x
10­
6
lbs
low
volatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste.
The
incremental
annualized
compliance
cost
for
a
new
cement
kiln
to
meet
this
beyond­
the­
floor
level,
rather
than
comply
with
the
floor
level,
would
be
approximately
$
0.38
million
and
would
provide
an
incremental
reduction
in
low
volatile
metals
emissions
of
approximately
33
pounds
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
evaluated
and
are
included
in
the
cost
estimates.
For
these
reasons
and
costs
of
$
23.5
million
per
ton
of
low
volatile
metals
removed,
we
are
not
proposing
a
beyond­
thefloor
standard
based
on
improved
particulate
matter
control
for
new
cement
kilns.
Feed
Control
of
Low
Volatile
Metals
in
the
Hazardous
Waste.
We
also
evaluated
a
beyond­
the­
floor
level
of
1.1
x
10­
5
lbs
low
volatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste,
which
represents
a
20%
reduction
from
the
floor
level.
We
believe
that
the
expense
for
further
reduction
in
low
volatile
metals
emissions
based
on
further
control
of
low
volatile
metals
concentrations
in
the
hazardous
waste
is
not
warranted
given
the
costs,
nonair
quality
health
and
environmental
impacts,
and
energy
effects.
Feed
Control
of
Low
Volatile
Metals
in
the
Raw
Materials
and
Auxiliary
Fuels.
Cement
kilns
could
achieve
a
reduction
in
low
volatile
metals
emissions
by
substituting
a
raw
material
containing
lower
levels
of
low
volatile
metals
for
a
primary
raw
material
with
a
higher
level.
For
a
new
source
at
an
existing
cement
plant,
we
believe
that
this
beyond­
the­
floor
option
would
not
be
cost­
effective
due
to
the
costs
of
transporting
large
quantities
of
an
alternate
source
of
raw
materials
to
the
cement
plant.
Given
that
the
plant
site
already
exists
and
sited
near
the
source
of
raw
material,
replacing
the
raw
materials
at
the
plant
site
with
lower
low
volatile
metalscontaining
materials
would
be
the
source's
only
option.
For
a
cement
kiln
constructed
at
a
new
greenfield
site,
we
are
not
aware
of
any
information
and
data
from
a
source
that
has
undertaken
or
is
currently
located
at
a
site
whose
raw
materials
are
inherently
lower
in
low
volatile
metals
that
would
consistently
achieve
reduced
low
volatile
metals
emissions.
Further,
we
are
uncertain
as
to
Redline­
strikeout
highlighting
changes
made
during
OMB
review
18
USEPA,
"
Draft
Technical
Support
Document
for
HWC
MACT
Replacement
Standards,
Volume
III:
Selection
of
MACT
Standards
and
Technologies"
Standards",
March
2004,
Chapter
2.

146
what
beyond­
the­
floor
standard
would
be
achievable
using
a
lower,
if
it
exists,
low
volatile
metals­
containing
raw
material.
Although
we
are
doubtful
that
selecting
a
new
plant
site
based
on
the
content
of
metals
in
the
raw
material
is
a
realistic
beyond­
the­
floor
option
considering
the
numerous
additional
factors
that
go
into
such
a
decision,
we
solicit
comment
on
whether
and
what
level
of
a
beyond­
the­
floor
standard
based
on
controlling
the
level
of
low
volatile
metals
in
the
raw
materials
is
appropriate.
We
also
considered
whether
fuel
switching
to
an
auxiliary
fuel
containing
a
lower
concentration
of
low
volatile
metals
would
be
an
appropriate
control
option
for
sources.
Given
that
most
cement
kilns
burning
hazardous
waste
also
burn
coal
as
a
fuel,
we
considered
switching
to
natural
gas
as
a
potential
beyond­
the­
floor
option.
For
the
same
reasons
discussed
for
mercury,
we
judge
a
beyond­
the­
floor
standard
based
on
fuel
switching
as
unwarranted.
Therefore,
we
are
proposing
a
low
volatile
metals
standard
of
1.4
x
10­
5
lbs
low
volatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste.
F.
What
Are
the
Proposed
Standards
for
Hydrogen
Chloride
and
Chlorine
Gas?
We
are
proposing
to
establish
standards
for
existing
and
new
cement
kilns
that
limit
total
chlorine
emissions
(
hydrogen
chloride
and
chlorine
gas,
combined,
reported
as
a
chloride
equivalent)
to
110
and
83
ppmv,
respectively.
However,
we
are
also
proposing
to
establish
alternative
risk­
based
standards,
pursuant
to
CAA
section
112(
d)(
4),
which
could
be
elected
by
the
source
in
lieu
of
the
MACT
emission
standards
for
total
chlorine.
The
emission
limits
would
be
based
on
national
exposure
standards
that
ensure
protection
of
public
health
with
an
ample
margin
of
safety.
See
Part
Two,
Section
XIII
for
additional
details.
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
Total
chlorine
emissions
from
existing
cement
kilns
are
limited
to
130
ppmv
by
§
63.1204(
a)(
6).
This
standard
was
promulgated
in
the
Interim
Standards
Rule
(
See
67
FR
at
6796).
None
of
the
cement
kilns
burning
hazardous
waste
use
a
dedicated
control
device,
such
as
a
wet
scrubber,
to
remove
total
chlorine
from
the
gas
stream.
However,
the
natural
alkalinity
in
some
of
the
raw
materials
is
highly
effective
at
removing
chlorine
from
the
gas
stream.
Our
data
base
shows
that
the
majority
of
the
system
removal
efficiency
(
SRE)
data
of
total
chlorine
 
over
80%
 
indicate
a
SRE
greater
than
95%.
This
scrubbing
effect,
though
quite
effective,
varies
across
different
sources
and
also
at
individual
sources
over
time
due
to
differences
in
raw
materials,
operating
conditions,
cement
kiln
dust
recycle
rates,
and
production
requirements.
Likewise,
our
data
show
that
total
chlorine
emissions
from
a
given
source
can
vary
over
a
considerable
range.
Based
on
these
data,
we
conclude
that
the
best
(
highest)
SRE
achieved
at
a
given
source
is
not
duplicable
or
replicable.
The
majority
of
the
chlorine
fed
to
the
cement
kiln
during
a
compliance
test
comes
from
the
hazardous
waste.
18
In
all
but
a
few
cases
the
hazardous
waste
contribution
to
the
total
Redline­
strikeout
highlighting
changes
made
during
OMB
review
19
We
are
also
requesting
comment
on
whether
the
hazardous
waste
feed
concentration
floor
level
should
be
the
standard
itself
(
i.
e.,
no
stack
emission
concentration
standard)
or
as
an
alternative
to
the
stack
emission
standard
(
e.
g.,
sources
have
the
option
to
comply
with
either
the
calculated
stack
emissions
concentration
or
the
hazardous
waste
feed
concentration
limit).

147
amount
of
chlorine
fed
to
the
kiln
represented
at
least
75%
of
the
total
chlorine
loading
to
the
kiln.
As
we
identified
in
the
September
1999
final
rule,
the
proposed
MACT
floor
control
for
total
chlorine
is
based
on
controlling
the
concentration
of
chlorine
in
the
hazardous
waste.
The
chlorine
concentration
in
the
hazardous
waste
will
affect
emissions
of
total
chlorine
at
a
given
SRE
because
emissions
increase
as
the
chlorine
loading
increases.
We
have
compliance
test
emissions
data
representing
maximum
emissions
for
all
cement
kiln
sources.
For
most
sources,
we
have
compliance
test
emissions
data
from
more
than
one
compliance
test
campaign.
Total
chlorine
emissions
range
from
less
than
1
ppmv
to
192
ppmv.
To
identify
the
MACT
floor,
we
evaluated
the
compliance
test
emissions
data
associated
with
the
most
recent
test
campaign
using
a
variant
of
the
SRE/
Feed
Approach
because
of
concerns
about
a
cement
kiln's
ability
to
replicate
a
given
SRE.
To
identify
the
floor
level
we
first
evaluated
the
chlorine
feed
level
in
the
hazardous
waste
for
all
sources.
The
best
performing
sources
had
the
lowest
maximum
theoretical
emissions
concentration
or
MTEC,
considering
variability.
We
then
applied
a
SRE
of
90%
to
the
best
performing
sources'
total
MTEC
(
i.
e.,
includes
chlorine
contributions
to
emissions
from
all
feedstreams
such
as
raw
material
and
fossil
fuels)
to
identify
the
floor
level.
Given
our
concerns
about
the
reproducibility
of
SREs
of
total
chlorine,
we
selected
a
SRE
of
90%
because
our
data
base
shows
that
all
sources
have
demonstrated
this
SRE
at
least
once
(
and
often
several
times)
during
a
compliance
test.
The
calculated
floor
is
110
ppmv,
which
considers
emissions
variability.
This
is
an
emission
level
that
the
best
performing
feed
control
sources
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
We
estimate
that
this
emission
level
is
being
achieved
by
93%
of
sources
and
would
reduce
total
chlorine
emissions
by
64
tons
per
year.
We
also
invite
comment
on
an
alternative
approach
to
establish
a
floor
level
expressed
as
a
hazardous
waste
thermal
feed
concentration.
19
A
hazardous
waste
thermal
feed
concentration
is
expressed
as
mass
of
chlorine
in
the
hazardous
waste
per
million
Btu
heat
input
contributed
by
the
hazardous
waste.
The
floor
would
be
based
on
the
best
five
performing
sources
with
the
lowest
thermal
feed
concentration
of
chlorine
in
the
hazardous
waste
considering
each
source's
most
recent
compliance
test
data.
One
advantage
of
this
approach
is
that
the
uncertainty
surrounding
the
capture
(
SRE)
of
chlorine
in
a
kiln
is
removed.
The
calculated
floor
level
would
be
2.4
lbs
chlorine
in
the
hazardous
waste
per
million
Btu
in
the
hazardous
waste,
which
considers
variability.
For
a
hypothetical
wet
process
cement
kiln
that
gets
50%
of
its
required
heat
input
from
hazardous
waste,
a
hazardous
waste
with
a
chlorine
concentration
of
2.4
lbs
chlorine
per
million
Btu
and
achieving
90%
SRE
equates
approximately
to
a
stack
gas
concentration
of
75
ppmv.
This
estimated
stack
gas
concentration
does
not
include
contributions
to
emission
from
Redline­
strikeout
highlighting
changes
made
during
OMB
review
148
other
chlorine­
containing
materials
such
as
raw
materials
and
fossil
fuels.
The
additional
contribution
to
stack
emissions
of
total
chlorine
in
an
average
raw
material
and
coal
is
estimated
to
range
from
less
than
1
to
35
ppmv.
Thus,
for
the
hypothetical
wet
process
cement
kiln
this
floor
level
is
estimated
to
be
less
than
110
ppmv,
which
is
less
than
the
current
interim
standard
of
130
ppmv.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
identified
three
potential
beyond­
the­
floor
techniques
for
control
of
total
chlorine:
(
1)
use
of
wet
scrubbers;
(
2)
control
of
chlorine
in
the
hazardous
waste
feed;
and
(
3)
control
of
the
chlorine
in
the
raw
materials.
For
reasons
discussed
below,
we
are
not
proposing
a
beyond­
thefloor
standard
for
total
chlorine.
Use
of
Wet
Scrubbers.
We
evaluated
the
use
of
wet
scrubbers
as
beyond­
the­
floor
control
for
further
reduction
of
mercury
emissions.
Wet
scrubbers
are
not
currently
being
used
at
any
hazardous
waste
burning
cement
kilns
to
capture
hydrogen
chloride.
We
evaluated
a
beyond­
thefloor
level
of
55
ppmv.
The
national
incremental
annualized
compliance
cost
for
cement
kilns
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
3.4
million
and
would
provide
an
incremental
reduction
in
total
chlorine
emissions
beyond
the
MACT
floor
controls
of
370
tons
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
evaluated
to
estimate
the
impacts
between
wet
scrubbing
and
controls
likely
to
be
used
to
meet
the
floor
level.
We
estimate
that
this
beyond­
the­
floor
option
would
increase
the
amount
of
water
usage
and
waste
water
generated
by
1.5
billion
gallon
per
year.
The
option
would
also
require
sources
to
use
an
additional
12
million
kW­
hours
per
year
beyond
the
requirements
to
achieve
the
floor
level.
The
costs
associated
with
these
impacts
are
accounted
for
in
the
national
annualized
compliance
cost
estimates.
Therefore,
based
on
these
factors
and
costs
of
approximately
$
9,300
per
additional
ton
of
total
chlorine
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
wet
scrubbing.
Feed
Control
of
Chlorine
in
the
Hazardous
Waste.
We
also
evaluated
a
beyond­
the­
floor
level
of
88
ppmv,
which
represents
a
20%
reduction
from
the
floor
level.
We
chose
a
20%
reduction
as
a
level
that
represents
the
practicable
extent
that
additional
feedrate
control
of
chlorine
in
the
hazardous
waste
can
be
used
and
still
achieve
modest
emissions
reductions.
The
national
incremental
annualized
compliance
cost
for
cement
kilns
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
1.1
million
and
would
provide
an
incremental
reduction
in
total
chlorine
emissions
beyond
the
MACT
floor
controls
of
100
tons
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
also
evaluated
and
are
included
in
the
compliance
cost
estimates.
Therefore,
based
on
these
factors
and
costs
of
approximately
$
11,000
per
additional
ton
of
total
chlorine,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
feed
control
of
chlorine
in
the
hazardous
waste.
Feed
Control
of
Chlorine
in
the
Raw
Materials
and
Auxiliary
Fuels.
Cement
kilns
could
achieve
a
reduction
in
total
chlorine
emissions
by
substituting
a
raw
material
containing
lower
levels
of
chlorine
for
a
primary
raw
material
with
higher
levels
of
chlorine.
We
believe
that
this
beyond­
the­
floor
option
would
even
be
less
cost­
effective
than
either
of
the
options
discussed
above
because
most
chlorine
feed
to
the
kiln
is
in
the
hazardous
waste.
In
addition,
given
that
cement
kilns
are
sited
near
the
primary
raw
material
supply,
acquiring
and
transporting
large
Redline­
strikeout
highlighting
changes
made
during
OMB
review
149
quantities
of
an
alternate
source
of
raw
materials
is
likely
to
be
cost­
prohibitive.
Therefore,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
limiting
chlorine
in
the
raw
material
feed.
We
also
considered
whether
fuel
switching
to
an
auxiliary
fuel
containing
a
lower
concentration
of
chlorine
would
be
an
appropriate
control
option
for
kilns.
Given
that
most
cement
kilns
burning
hazardous
waste
also
burn
coal
as
a
fuel,
we
considered
switching
to
natural
gas
as
a
potential
beyond­
the­
floor
option.
For
the
same
reasons
discussed
for
mercury,
we
judge
a
beyond­
thefloor
standard
based
on
fuel
switching
as
unwarranted.
For
the
reasons
discussed
above,
we
propose
not
to
adopt
a
beyond­
the­
floor
standard
for
total
chlorine
and
propose
to
establish
the
emission
standard
for
existing
cement
kilns
at
110
ppmv.
3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
Total
chlorine
emissions
from
new
cement
kilns
are
currently
limited
to
86
ppmv
by
§
63.1204(
b)(
6).
This
standard
was
promulgated
in
the
Interim
Standards
Rule
(
See
67
FR
at
6796).
The
MACT
floor
for
new
sources
for
total
chlorine
would
be
78
ppmv,
which
considers
emissions
variability.
This
is
an
emission
level
that
the
single
best
performing
source
identified
with
the
SRE/
Feed
Approach
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
conditions
identical
to
the
test
conditions
during
which
the
emissions
data
were
obtained.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
We
identified
similar
potential
beyond­
the­
floor
techniques
for
control
of
total
chlorine
for
new
sources:
(
1)
use
of
wet
scrubbing;
(
2)
control
of
chlorine
in
the
hazardous
waste
feed;
and
(
3)
control
of
chlorine
in
the
raw
materials
and
fuels.
Use
of
Wet
Scrubbers.
We
considered
wet
scrubbing
as
beyond­
the­
floor
control
for
further
reductions
in
total
chlorine
emissions
and
evaluated
a
beyond­
the­
floor
level
of
39
ppmv.
The
incremental
annualized
compliance
cost
for
a
new
cement
kiln
to
meet
this
beyond­
the­
floor
level,
rather
than
comply
with
the
floor
level,
would
be
approximately
$
1.2
million
and
would
provide
an
incremental
reduction
in
total
chlorine
emissions
of
approximately
22
tons
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
evaluated
and
are
included
in
the
cost
estimates.
For
these
reasons
and
costs
of
$
24,000
per
ton
of
total
chlorine
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
wet
scrubbing
for
new
cement
kilns.
Feed
Control
of
Low
Volatile
Metals
in
the
Hazardous
Waste.
We
also
evaluated
a
beyond­
the­
floor
level
of
62
ppmv,
which
represents
a
20%
reduction
from
the
floor
level.
We
believe
that
the
expense
for
further
reduction
in
total
chlorine
emissions
based
on
further
control
of
chlorine
concentrations
in
the
hazardous
waste
is
not
warranted
given
the
costs,
nonair
quality
health
and
environmental
impacts,
and
energy
effects.
Feed
Control
of
Chlorine
in
the
Raw
Materials
and
Auxiliary
Fuels.
Cement
kilns
could
achieve
a
reduction
in
total
chlorine
emissions
by
substituting
a
raw
material
containing
lower
levels
of
chlorine
for
a
primary
raw
material
with
a
higher
level.
For
a
new
source
at
an
existing
cement
plant,
we
believe
that
this
beyond­
the­
floor
option
would
not
be
cost­
effective
due
to
the
costs
of
transporting
large
quantities
of
an
alternate
source
of
raw
materials
to
the
cement
plant.
Given
that
the
plant
site
already
exists
and
sited
near
the
source
of
raw
material,
replacing
the
raw
Redline­
strikeout
highlighting
changes
made
during
OMB
review
20
A
greenfield
cement
kiln
is
a
kiln
that
commenced
construction
or
reconstruction
after
April
19,
1996
at
a
site
where
no
cement
kiln
previously
existed,
irrespective
of
the
class
of
kiln
(
i.
e.,
nonhazardous
waste
or
hazardous
waste
burning).
A
newly
constructed
or
reconstructed
cement
kiln
at
an
existing
site
is
not
classified
as
a
greenfield
cement
kiln,
and
is
subject
to
the
same
carbon
monoxide
and
hydrocarbon
standards
as
an
existing
cement
kiln.

150
materials
at
the
plant
site
with
lower
chlorine­
containing
materials
would
be
the
source's
only
option.
For
a
cement
kiln
constructed
at
a
new
greenfield
site,
we
are
not
aware
of
any
information
and
data
from
a
source
that
has
undertaken
or
is
currently
located
at
a
site
whose
raw
materials
are
inherently
lower
in
chlorine
that
would
consistently
achieve
reduced
total
chlorine
emissions.
Further,
we
are
uncertain
as
to
what
beyond­
the­
floor
standard
would
be
achievable
using
a
lower,
if
it
exists,
chlorine­
containing
raw
material.
Although
we
are
doubtful
that
selecting
a
new
plant
site
based
on
the
content
of
chlorine
in
the
raw
material
is
a
realistic
beyondthe
floor
option
considering
the
numerous
additional
factors
that
go
into
such
a
decision,
we
solicit
comment
on
whether
and
what
level
of
a
beyond­
the­
floor
standard
based
on
controlling
the
level
of
chlorine
in
the
raw
materials
is
appropriate.
We
also
considered
whether
fuel
switching
to
an
auxiliary
fuel
containing
a
lower
concentration
of
chlorine
would
be
an
appropriate
control
option
for
sources.
Given
that
most
cement
kilns
burning
hazardous
waste
also
burn
coal
as
a
fuel,
we
considered
switching
to
natural
gas
as
a
potential
beyond­
the­
floor
option.
For
the
same
reasons
discussed
for
mercury,
we
judge
a
beyond­
the­
floor
standard
based
on
fuel
switching
as
unwarranted.
Therefore,
we
are
proposing
a
total
chlorine
standard
of
78
ppmv
for
new
cement
kilns.
G.
What
Are
the
Standards
for
Hydrocarbons
and
Carbon
Monoxide?
Hydrocarbon
and
carbon
monoxide
standards
are
surrogates
to
control
emissions
of
organic
hazardous
air
pollutants
for
existing
and
new
cement
kilns.
For
cement
kilns
without
bypass
or
midkiln
sampling
systems,
the
standard
for
existing
sources
limit
hydrocarbon
or
carbon
monoxide
concentrations
to
20
ppmv
or
100
ppmv,
respectively.
The
standards
for
new
sources
limit
(
1)
hydrocarbons
to
20
ppmv;
or
(
2)
carbon
monoxide
to
100
ppmv.
New,
greenfield
kilns20,
that
elect
to
comply
with
the
100
ppmv
carbon
monoxide
standard,
however,
must
also
comply
with
a
50
ppmv
hydrocarbon
standard.
New
and
existing
sources
that
elect
to
comply
with
the
100
ppmv
carbon
monoxide
standard,
including
new
greenfield
kilns
that
elect
to
comply
with
the
carbon
monoxide
standard
and
50
ppmv
hydrocarbon
standard,
must
also
demonstrate
compliance
with
the
20
ppmv
hydrocarbon
standard
during
the
comprehensive
performance
test.
However,
continuous
hydrocarbon
monitoring
following
the
performance
test
is
not
required.
For
cement
kilns
with
bypass
or
midkiln
sampling
systems,
existing
cement
kilns
are
required
to
comply
with
either
a
carbon
monoxide
standard
of
100
ppmv
or
a
hydrocarbon
standard
of
10
ppmv.
Both
standards
apply
to
combustion
gas
sampled
in
the
bypass
or
a
midkiln
sampling
port
that
samples
representative
kiln
gas.
See
§
§
63.1204(
a)(
5)
and
(
b)(
5).
The
rationale
for
these
decisions
are
discussed
in
the
September
1999
final
rule
(
64
FR
at
52885).
We
view
the
standards
for
hydrocarbons
and
carbon
monoxide
as
unaffected
by
the
Court's
vacature
of
the
challenged
regulations
in
its
decision
of
July
24,
2001.
We
therefore
are
not
proposing
Redline­
strikeout
highlighting
changes
made
during
OMB
review
151
these
standards
for
cement
kilns,
but
rather
are
mentioning
them
here
for
the
reader's
convenience.
H.
What
Are
the
Standards
for
Destruction
and
Removal
Efficiency?
The
destruction
and
removal
efficiency
(
DRE)
standard
is
a
surrogate
to
control
emissions
of
organic
hazardous
air
pollutants
other
than
dioxin/
furans.
The
standard
for
existing
and
new
lightweight
aggregate
kilns
requires
99.99%
DRE
for
each
principal
organic
hazardous
constituent,
except
that
99.9999%
DRE
is
required
if
specified
dioxin­
listed
hazardous
wastes
are
burned.
See
§
§
63.1204(
c).
The
rationale
for
these
decisions
are
discussed
in
the
September
1999
final
rule
(
64
FR
at
52890).
We
view
the
standards
for
DRE
as
unaffected
by
the
Court's
vacature
of
the
challenged
regulations
in
its
decision
of
July
24,
2001.
We
therefore
are
not
proposing
these
standards
for
cement
kilns,
but
rather
are
mentioning
them
here
for
the
reader's
convenience.

IX.
How
Did
EPA
Determine
the
Proposed
Emission
Standards
for
Hazardous
Waste
Burning
Lightweight
Aggregate
Kilns?
In
this
section,
the
basis
for
the
proposed
emission
standards
is
discussed.
See
proposed
§
63.1205A1221.
The
proposed
emission
limits
apply
to
the
stack
gases
from
lightweight
aggregate
kilns
that
burn
hazardous
waste
and
are
summarized
in
the
table
below:

PROPOSED
STANDARDS
FOR
EXISTING
AND
NEW
LIGHTWEIGHT
AGGREGATE
KILNS
Hazardous
Air
Pollutant
or
Surrogate
Emission
Standard1
Existing
Sources
New
Sources
Dioxin
and
furan
0.40
ng
TEQ/
dscm
0.40
ng
TEQ/
dscm
Mercury2
67
ug/
dscm
67
ug/
dscm
Particulate
Matter
57
mg/
dscm
(
0.025
gr/
dscf)
23
mg/
dscm
(
0.0099
gr/
dscf)

Semivolatile
metals3
3.1
x
10­
4
lb/
MMBtu
and
250
ug/
dscm
2.4
x
10­
5
lb/
MMBtu
and
43
ug/
dscm
Low
volatile
metals3
9.5
x
10­
5
lb/
MMBtu
and
110
ug/
dscm
3.2
x
10­
5
lb/
MMBtu
and
110
ug/
dscm
Hydrogen
chloride
and
chlorine
gas4
Hydrocarbons5,
6
20
ppmv
(
or
100
ppmv
carbon
monoxide)
20
ppmv
(
or
100
ppmv
carbon
monoxide)
Redline­
strikeout
highlighting
changes
made
during
OMB
review
152
Destruction
and
removal
efficiency
For
existing
and
new
sources,
99.99%
for
each
principal
organic
hazardous
constituent
(
POHC).
For
sources
burning
hazardous
wastes
F020,
F021,
F022,
F023,
F026,
or
F027,
however,
99.9999%
for
each
POHC.

1
All
emission
standards
are
corrected
to
7%
oxygen,
dry
basis.
2
Mercury
standard
is
an
annual
limit.
3
Standards
are
expressed
as
mass
of
pollutant
emissions
contributed
by
hazardous
waste
per
million
British
thermal
unit
contributed
by
the
hazardous
waste.
4
Combined
standard,
reported
as
a
chloride
(
Cl(­))
equivalent.
5
Sources
that
elect
to
comply
with
the
carbon
monoxide
standard
must
demonstrate
compliance
with
the
hydrocarbon
standard
during
the
comprehensive
performance
test.
6
Hourly
rolling
average.
Hydrocarbons
reported
as
propane.

A.
What
Are
the
Proposed
Standards
for
Dioxin
and
Furan?
We
are
proposing
to
establish
standards
for
existing
and
new
lightweight
aggregate
kilns
that
limit
emissions
of
dioxin
and
furans
to
0.40
ng
TEQ/
dscm.
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
Dioxin
and
furan
emissions
for
existing
lightweight
aggregate
kilns
are
currently
limited
by
§
63.1205(
a)(
1)
to
0.20
ng
TEQ/
dscm
or
rapid
quench
of
the
flue
gas
at
the
exit
of
the
kiln
to
less
than
400
°
F.
This
standard
was
promulgated
in
the
Interim
Standards
Rule
(
See
67
FR
at
6797).
Since
promulgation
of
the
September
1999
final
rule,
we
have
obtained
additional
dioxin/
furan
emissions
data.
We
now
have
compliance
test
emissions
data
representing
maximum
emissions
for
all
lightweight
aggregate
kilns
that
burn
hazardous
waste.
The
compliance
test
dioxin/
furan
emissions
in
our
database
range
from
approximately
0.9
to
58
ng
TEQ/
dscm.
Quenching
kiln
gas
temperatures
at
the
exit
of
the
kiln
so
that
gas
temperatures
at
the
inlet
to
the
particulate
matter
control
device
are
below
the
temperature
range
of
optimum
dioxin/
furan
formation
(
400­
750
°
F)
may
be
problematic
for
some
of
these
sources.
Some
of
these
sources
have
extensive
(
long)
duct­
work
between
the
kiln
exit
and
the
inlet
to
the
control
device.
For
these
sources,
quenching
the
gases
at
the
kiln
exit
to
a
low
enough
temperature
to
limit
dioxin/
furan
formation
may
conflict
with
the
source's
ability
to
avoid
acid
gas
dew
point
related
problems
in
the
long
duct­
work
and
control
device.
As
a
result,
some
sources
quench
the
kiln
exit
gases
to
a
temperature
that
is
in
the
optimum
temperature
range
for
surface­
catalyzed
dioxin/
furan
formation.
Available
compliance
test
emissions
data
indicate
that
inlet
temperatures
to
the
control
device
range
from
435­
450
°
F.
This
means
that
temperatures
in
the
duct­
work
are
higher
and
well
within
the
range
of
optimum
dioxin/
furan
formation.
To
identify
the
MACT
floor,
we
evaluated
the
compliance
test
emissions
data
associated
with
the
most
recent
test
campaign
using
the
Emissions
Approach
described
in
Part
Two,
Section
VI
above.
The
calculated
floor
is
14
ng
TEQ/
dscm,
which
considers
emissions
variability.
However,
the
current
interim
emission
standard
 
0.20
ng
TEQ/
dscm
or
rapid
quench
of
the
flue
gas
at
the
exit
of
the
kiln
to
less
than
400
°
F
 
is
a
regulatory
limit
that
is
relevant
in
identifying
the
Redline­
strikeout
highlighting
changes
made
during
OMB
review
21
Even
though
all
sources
have
recently
demonstrated
compliance
with
the
interim
standards,
the
dioxin/
furan
data
in
our
data
base
preceded
the
compliance
demonstration.
This
explains
why
we
have
emissions
data
that
are
higher
than
the
interim
standard.

153
floor
level
because
it
fixes
a
level
of
performance
for
the
source
category.
We
estimate
that
sources
achieving
the
"
rapid
quench
of
the
flue
gas
at
the
exit
of
the
kiln
to
less
than
400
°
F"
part
of
the
current
standard
can
emit
up
to
6.1
ng
TEQ/
dscm.
Given
that
all
sources
are
achieving
the
interim
standard
and
that
the
interim
standard
is
judged
as
more
stringent
than
the
calculated
MACT
floor,
the
dioxin/
furan
floor
level
can
be
no
less
stringent
than
the
current
regulatory
limit.
21
We
are,
therefore,
proposing
the
dioxin/
furan
floor
level
as
the
current
emission
standard
of
0.20
ng
TEQ/
dscm
or
rapid
quench
of
the
flue
gas
at
the
exit
of
the
kiln
to
less
than
400
°
F.
This
emission
level
is
being
achieved
by
all
sources
because
it
is
the
interim
standard.
In
addition,
there
are
no
emissions
reductions
for
existing
lightweight
aggregate
kilns
to
comply
with
the
floor
level.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
evaluated
activated
carbon
injection
as
beyond­
the­
floor
control
for
further
reduction
of
dioxin/
furan
emissions.
Activated
carbon
has
been
demonstrated
for
controlling
dioxin/
furans
in
various
combustion
applications;
however,
no
lightweight
aggregate
kiln
that
burns
hazardous
waste
uses
activated
carbon
injection.
We
evaluated
a
beyond­
the­
floor
level
of
0.40
ng
TEQ/
dscm,
which
represents
a
level
that
is
considered
routinely
achievable
using
activated
carbon
injection.
In
addition,
we
assumed
for
costing
purposes
that
lightweight
aggregate
kilns
needing
activated
carbon
injection
to
achieve
the
beyond­
the­
floor
level
would
install
the
activated
carbon
injection
system
after
the
existing
particulate
matter
control
device
and
add
a
new,
smaller
baghouse
to
remove
the
injected
carbon
with
the
adsorbed
dioxin/
furans.
We
chose
this
costing
approach
to
address
potential
concerns
that
injected
carbon
may
interfere
with
lightweight
aggregate
dust
use
practices.
The
national
incremental
annualized
compliance
cost
for
lightweight
aggregate
kilns
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
21.28
million
and
would
provide
an
incremental
reduction
in
dioxin/
furan
emissions
beyond
the
MACT
floor
controls
of
1.9
grams
TEQ
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
evaluated
to
estimate
the
nonair
quality
health
and
environmental
impacts
between
activated
carbon
injection
and
controls
likely
to
be
used
to
meet
the
floor
level.
We
estimate
that
this
beyond­
the­
floor
option
would
increase
the
amount
of
solid
waste
generated
by
550
tons
per
year
and
would
require
sources
to
use
an
additional
1
million
kW­
hours
per
year
beyond
the
requirements
to
achieve
the
floor
level.
The
costs
associated
with
these
impacts
are
accounted
for
in
the
national
compliance
cost
estimates.
Therefore,
based
on
these
factors
and
costs
of
approximately
$
0.915
million
per
additional
gram
of
dioxin/
furan
TEQ
removed,
we
are
proposing
a
beyond­
the­
floor
standard
of
0.40
ng
TEQ/
dscm
for
existing
lightweight
aggregate
kilns.
We
judge
that
the
cost
to
achieve
this
beyond­
the­
floor
level
is
warranted
given
our
special
concern
about
dioxin/
furan.
Dioxin/
furan
are
some
of
the
most
toxic
compounds
known
due
to
their
bioaccumulation
potential
and
wide
Redline­
strikeout
highlighting
changes
made
during
OMB
review
154
range
of
health
effects,
including
carcinogenesis,
at
exceedingly
low
doses.
Exposure
via
indirect
pathways
is
a
chief
reason
that
Congress
singled
our
dioxin/
furan
for
priority
MACT
control
in
CAA
section
112(
c)(
6).
See
S.
Rep.
No.
128,
101st
Cong.
1st
Sess.
at
154­
155.
In
addition,
we
note
that
a
beyond­
the­
floor
standard
of
0.40
ng
TEQ/
dscm
is
consistent
with
historically
controlled
levels
under
MACT
for
hazardous
waste
incinerators
and
cement
kilns,
and
Portland
cement
plants.
See
§
§
63.1203(
a)(
1),
63.1204(
a)(
1),
and
63.1343(
d)(
3).
Also,
EPA
has
determined
previously
in
the
1999
Hazardous
Waste
Combustor
MACT
final
rule
that
dioxin/
furan
in
the
range
of
0.40
ng
TEQ/
dscm
or
less
are
necessary
for
the
MACT
standards
to
be
considered
generally
protective
of
human
health
under
RCRA
(
using
the
1985
cancer
slope
factor),
thereby
eliminating
the
need
for
separate
RCRA
standards
under
the
authority
of
RCRA
section
3005(
c)(
3)
and
40
CFR
270.10(
k).
Finally,
we
note
that
this
decision
is
not
inconsistent
with
EPA's
decision
not
to
promulgate
beyond­
the­
floor
standards
for
dioxin/
furan
for
hazardous
waste
burning
lightweight
aggregate
kilns,
cement
kilns,
and
incinerators
at
cost­
effectiveness
values
in
the
range
of
$
530,000
to
$
827,000
per
additional
gram
of
dioxin/
furan
TEQ
removed.
See
64
FR
at
52892,
52876,
and
52961.
In
those
cases,
EPA
determined
that
controlling
dioxin/
furan
emissions
from
a
level
of
0.40
ng
TEQ/
dscm
to
a
beyond­
the­
floor
level
of
0.20
ng
TEQ/
dscm
was
not
warranted
because
dioxin/
furan
levels
below
0.40
ng
TEQ/
dscm
are
generally
considered
to
be
below
the
level
of
health
risk
concern.

3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
Dioxin
and
furan
emissions
for
new
lightweight
aggregate
kilns
are
currently
limited
by
§
63.1205(
b)(
1)
to
0.20
ng
TEQ/
dscm
or
rapid
quench
of
the
flue
gas
at
the
exit
of
the
kiln
to
less
than
400
°
F.
This
standard
was
promulgated
in
the
Interim
Standards
Rule
(
See
67
FR
at
6797).
The
calculated
MACT
floor
for
new
sources
would
be
1.3
ng
TEQ/
dscm,
which
considers
emissions
variability,
or
rapid
quench
of
the
flue
gas
at
the
exit
of
the
kiln
to
less
than
400
°
F.
This
is
an
emission
level
that
the
single
best
performing
source
identified
by
the
Emissions
Approach.
However,
we
are
concerned
that
the
calculated
floor
level
of
1.3
ng
TEQ/
dscm
is
not
duplicable
by
all
sources
using
temperature
control
because
we
estimate
that
sources
rapidly
quenching
the
flue
gas
at
the
exit
of
the
kiln
to
less
than
400
°
F
can
emit
up
to
6.1
ng
TEQ/
dscm.
Therefore,
we
are
proposing
the
floor
as
the
current
emission
standard
of
0.20
ng
TEQ/
dscm
or
rapid
quench
of
the
flue
gas
at
the
exit
of
the
kiln
to
less
than
400
°
F.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
We
evaluated
activated
carbon
injection
as
beyond­
the­
floor
control
for
further
reduction
of
dioxin/
furan
emissions,
and
considered
a
beyond­
the­
floor
level
of
0.40
ng
TEQ/
dscm,
which
represents
a
level
that
is
considered
routinely
achievable
with
activated
carbon
injection.
In
addition,
we
assumed
for
costing
purposes
that
a
new
lightweight
aggregate
kiln
will
install
the
activated
carbon
injection
system
after
the
existing
particulate
matter
control
device
and
add
a
new,
smaller
baghouse
to
remove
the
injected
carbon
with
the
adsorbed
dioxin/
furan.
The
incremental
annualized
compliance
cost
for
a
new
source
to
meet
this
beyond­
the­
floor
level,
rather
than
comply
with
the
floor
level,
would
be
approximately
$
0.26
million
and
would
provide
an
incremental
reduction
in
dioxin/
furan
emissions
of
0.37
grams
per
year.
Nonair
quality
health,
environmental
impacts,
and
energy
effects
are
accounted
for
in
the
cost
estimates.
Therefore,
Redline­
strikeout
highlighting
changes
made
during
OMB
review
22
MTEC
is
a
term
to
compare
metals
and
chlorine
feedrates
across
sources
of
different
sizes.
MTEC
is
defined
as
the
metals
or
chlorine
feedrate
divided
by
the
gas
flow
rate
and
is
expressed
in
units
of
ug/
dscm.

23
Given
that
the
majority
of
feedrate
and
emissions
data
for
mercury
is
normal,
we
do
not
believe
it
is
appropriate
to
establish
a
hazardous
waste
thermal
emissions­
based
standard.

155
based
on
these
factors
and
cost
of
$
0.71
million
per
gram
TEQ
removed,
we
are
proposing
a
beyond­
the­
floor
standard
based
on
activated
carbon
injection.
We
believe
that
the
cost
to
achieve
this
beyond­
the­
floor
level
is
warranted
given
our
special
concern
about
dioxin/
furan.
Dioxin/
furan
are
some
of
the
most
toxic
compounds
known
due
to
their
bioaccumulation
potential
and
wide
range
of
health
effects,
including
carcinogenesis,
at
exceedingly
low
doses.
In
addition,
as
discussed
above,
we
note
that
the
beyond­
the­
floor
emission
level
of
0.40
ng
TEQ/
dscm
is
consistent
with
historically
controlled
levels
under
MACT
for
hazardous
waste
incinerators
and
cement
kilns,
and
Portland
cement
plants.
See
§
§
63.1203(
a)(
1),
63.1204(
a)(
1),
and
63.1343(
d)(
3).
EPA
has
determined
previously
in
the
1999
Hazardous
Waste
Combustor
MACT
final
rule
that
dioxin/
furan
in
the
range
of
0.40
ng
TEQ/
dscm
or
less
are
necessary
for
the
MACT
standards
to
be
considered
generally
protective
of
human
health
under
RCRA,
thereby
eliminating
the
need
for
separate
RCRA
standards.

B.
What
Are
the
Proposed
Standards
for
Mercury?
We
are
proposing
to
establish
standards
for
existing
and
new
lightweight
aggregate
kilns
that
limit
emissions
of
mercury
to
67
ug/
dscm.
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
Mercury
emissions
for
existing
lightweight
aggregate
kilns
are
currently
limited
to
120
ug/
dscm
by
§
63.1205(
a)(
2).
Existing
lightweight
aggregate
kilns
have
the
option
to
comply
with
an
alternative
mercury
standard
that
limits
the
hazardous
waste
maximum
theoretical
emissions
concentration
(
MTEC)
of
mercury
to
120
ug/
dscm.
22
This
standard
was
promulgated
in
the
Interim
Standards
Rule
(
See
67
FR
at
6797).
One
lightweight
aggregate
facility
with
two
kilns
uses
a
venturi
scrubber
to
remove
mercury
from
the
flue
gas
stream
and
the
remaining
sources
limit
the
feed
concentration
of
mercury
in
the
hazardous
waste
to
control
emissions.
We
have
compliance
test
emissions
data
representing
maximum
emissions
for
only
one
source;
however,
we
have
normal
emissions
data
for
all
sources.
For
most
sources,
we
have
normal
emissions
data
from
more
than
one
test
campaign.
We
used
these
emissions
data
to
represent
the
average
emissions
from
a
source
even
though
we
do
not
know
whether
the
emissions
represent
the
high
end,
low
end,
or
close
to
the
average
emissions.
The
normal
mercury
stack
emissions
range
from
less
than
1
to
47
ug/
dscm,
while
the
highest
compliance
test
emissions
data
is
1,050
ug/
dscm.
These
emissions
are
expressed
as
mass
of
mercury
(
from
all
feedstocks)
per
unit
volume
of
stack
gas.
To
identify
the
MACT
floor,
we
evaluated
all
normal
emissions
data
using
the
SRE/
Feed
Approach.
We
considered
normal
stack
emissions
data
from
all
test
campaigns.
23
For
example,
Redline­
strikeout
highlighting
changes
made
during
OMB
review
We
prefer
to
establish
emission
standards
under
the
hazardous
waste
thermal
emissions
format
using
compliance
test
data
because
the
metals
feedrate
information
from
compliance
tests
that
we
use
to
apportion
emissions
to
calculate
emissions
attributable
to
hazardous
waste
are
more
reliable
than
feedrate
data
measured
during
testing
under
normal,
typical
operations.

156
one
source
in
our
data
base
has
normal
emissions
data
for
three
different
testing
campaigns:
1992,
1995,
and
1999.
Under
this
approach
we
considered
the
emissions
data
from
the
three
separate
years
or
campaigns.
As
explained
earlier,
we
believe
this
approach
better
captures
the
range
of
average
emissions
for
a
source
than
only
considering
the
most
recent
normal
emissions.
In
addition,
for
sources
without
control
equipment
to
capture
mercury,
we
assumed
the
sources
achieved
a
SRE
of
zero.
The
effect
of
this
assumption
is
that
the
sources
(
without
control
equipment
for
mercury)
with
the
lower
mercury
concentrations
in
the
hazardous
waste
were
identified
as
the
better
performing
sources.
The
calculated
floor
is
67
ug/
dscm,
which
considers
emissions
variability,
based
on
a
hazardous
waste
maximum
theoretical
emissions
concentration
(
MTEC)
of
42
ug/
dscm.
This
is
an
emission
level
that
the
average
of
the
best
performing
sources
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
operating
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
We
estimate
that
this
emission
level
is
being
achieved
by
57%
of
sources
and
would
reduce
mercury
emissions
by
8
pounds
per
year.
If
we
were
to
adopt
such
a
floor
level,
we
are
proposing
that
sources
comply
with
the
limit
on
an
annual
basis
because
it
is
based
on
normal
emissions
data.
Under
this
approach,
compliance
would
not
be
based
on
the
use
of
a
total
mercury
continuous
emissions
monitoring
system
because
these
monitors
have
not
been
adequately
demonstrated
as
a
reliable
compliance
assurance
tool
at
all
types
of
incinerator
sources.
Instead,
a
source
would
maintain
compliance
with
the
mercury
standard
by
establishing
and
complying
with
short­
term
limits
on
operating
parameters
(
e.
g.,
for
pollution
control
equipment
and
annual
limits
on
maximum
total
mercury
feedrate
in
all
feedstreams)
on
an
annual
basis.
In
the
September
1999
final
rule,
we
acknowledged
that
a
lightweight
aggregate
kiln
using
properly
designed
and
operated
MACT
control
technologies,
including
controlling
the
levels
of
metals
in
the
hazardous
waste,
may
not
be
capable
of
achieving
a
given
emission
standard
because
of
process
raw
material
contributions
that
might
cause
an
exceedance
of
the
emission
standard.
To
address
this
concern,
we
promulgated
a
provision
that
allows
sources
to
petition
for
alternative
standards
provided
they
submit
site­
specific
information
that
shows
raw
material
hazardous
air
pollutant
contributions
to
the
emissions
prevent
the
source
from
complying
with
the
emission
standard
even
though
the
kiln
is
using
MACT
control.
See
§
63.1206(
b)(
9).
Today's
proposed
floor
of
67
ug/
dscm,
which
was
based
on
a
hazardous
waste
MTEC
of
42
ug/
dscm,
may
likewise
necessitate
such
an
alternative
because
contributions
of
mercury
in
the
raw
materials
and
fossil
fuels
at
some
sources
may
cause
an
exceedance
of
the
emission
standard.
Redline­
strikeout
highlighting
changes
made
during
OMB
review
24
Solite
Corporation
has
four
kilns
at
its
Cascade
facility
and
three
kilns
at
its
Arvonia
facility.
However,
only
three
kilns
and
two
kilns,
respectively,
can
be
fired
with
hazardous
waste
at
any
one
time.
For
purposes
of
today's
proposal,
Solite
Corporation
is
assumed
to
operate
a
total
of
five
kilns.

25
A
hazardous
waste
with
a
mercury
concentration
of
2
ppm
equates
approximately
to
a
mercury
emissions
level
of
200­
250
ug/
dscm,
and
a
source
firing
a
hazardous
waste
with
a
mercury
concentration
of
0.2
ppm
approximately
equates
to
20­
25
ug/
dscm.
The
existing
standard
of
120
ug/
dscm
allows
a
source
to
burn
a
hazardous
waste
with
a
mercury
concentration
of
approximately
1
ppm.

157
In
comments
submitted
to
EPA
in
1997,
Solite
Corporation
(
Solite),
owner
and
operator
of
five24
of
the
seven
lightweight
aggregate
kilns,
stated
that
the
normal
emissions
data
in
our
data
base
are
unrepresentative
of
average
emissions
of
mercury
because
the
normal
range
of
mercury
concentrations
in
the
hazardous
waste
burned
during
the
compliance
and
trial
burn
tests
was
not
captured
during
the
tests.
In
their
1997
comments,
Solite
provided
information
on
actual
mercury
concentrations
in
the
hazardous
waste
burn
tanks
over
a
year
and
a
quarter
period.
The
information
showed
that
87%
of
the
burn
tanks
contained
mercury
at
concentrations
below
the
facility's
detection
limit
of
2
ppm.
Additional
analyses
of
a
limited
number
of
these
samples
conducted
at
an
off­
site
lab
showed
that
the
majority
of
samples
were
actually
less
than
0.2
ppm.
25
We
examined
the
test
reports
of
the
five
best
performing
sources
that
are
the
basis
of
today's
proposed
floor
level
to
determine
the
concentration
level
of
mercury
in
the
hazardous
wastes.
The
hazardous
waste
burned
by
the
best
performing
sources
during
the
tests
that
generated
the
normal
emissions
data
had
mercury
concentrations
that
ranged
from
0.02
to
0.2
Redline­
strikeout
highlighting
changes
made
during
OMB
review
26
These
mercury
concentrations
were
analyzed
by
an
off­
site
lab
that
had
equipment
capable
of
detecting
mercury
at
lower
concentrations.
Sixteen
of
the
27
measurements
of
the
best
performers
were
reported
as
non­
detects.

158
ppm.
26
Even
though
the
concentrations
of
mercury
in
the
hazardous
waste
seem
low,
we
cannot
judge
how
these
snap
shot
concentrations
compare
to
long­
term
normal
concentrations
because
the
majority
of
the
burn
tank
concentration
data
submitted
by
Solite
are
nondetect
measurements
at
a
higher
detection
limit.
Solite
informed
us
in
July
2003
that
they
are
in
the
process
of
upgrading
the
analysis
equipment
at
their
on­
site
laboratory.
Once
completed,
Solite
expects
to
be
capable
of
detecting
mercury
in
the
hazardous
waste
at
concentrations
of
0.2
ppm.
Solite
also
indicated
that
they
intend
to
assemble
and
submit
to
EPA
several
months
of
burn
tank
concentration
data
analyzed
with
the
new
equipment.
We
will
add
these
data
to
the
docket
of
today's
proposal
once
available.
As
we
discussed
for
cement
kilns
for
mercury,
we
are
requesting
comment
on
approaches
to
establish
a
hazardous
waste
feed
concentration
standard
based
on
long­
term
feed
concentrations
of
mercury
in
the
hazardous
waste.
Likewise,
we
invite
comments
on
establishing
a
mercury
feed
concentration
standard
for
lightweight
aggregate
kilns.
We
also
invite
comment
on
whether
we
should
judge
an
annual
limit
of
67
ug/
dscm
as
less
stringent
than
either
the
current
emission
standard
of
120
ug/
dscm
or
the
hazardous
waste
MTEC
of
mercury
of
120
ug/
dscm
for
lightweight
aggregate
kilns
(
so
as
to
avoid
any
backsliding
from
a
current
level
of
performance
achieved
by
all
sources,
and
hence,
the
level
of
minimal
stringency
at
which
EPA
could
calculate
the
MACT
floor).
In
order
to
comply
with
the
current
emission
standard,
generally
a
source
must
conduct
manual
stack
sampling
to
demonstrate
compliance
with
the
mercury
emission
standard
and
then
establish
a
maximum
mercury
feedrate
limit
based
on
operations
during
the
performance
test.
Following
the
performance
test,
the
source
complies
with
a
limit
on
the
maximum
total
mercury
feedrate
in
all
feedstreams
on
a
12­
hour
rolling
average
(
not
an
annual
average).
Alternatively,
a
source
can
elect
to
comply
with
a
hazardous
waste
MTEC
of
mercury
of
120
ug/
dscm
that
would
require
the
source
to
limit
the
mercury
feedrate
in
the
hazardous
waste
on
a
12­
hour
rolling
average.
The
floor
level
of
67
ug/
dscm
proposed
today
would
allow
a
source
to
feed
more
variable
mercury­
containing
feedstreams
(
e.
g.,
a
hazardous
waste
with
an
mercury
MTEC
greater
than
120
ug/
dscm)
than
the
current
12­
hour
rolling
average
because
today's
proposed
floor
level
is
an
annual
limit.
For
example,
the
concentration
of
mercury
in
the
hazardous
waste
exceeded
a
hazardous
waste
MTEC
of
120
ug/
dscm
in
a
minimum
of
13%
of
the
burn
tanks
based
on
the
data
submitted
by
Solite
in
their
1997
comments
(
discussed
above).
As
mentioned
above,
Solite
intends
to
submit
several
months
of
burn
tank
concentration
data
using
upgraded
analysis
equipment
at
their
on­
site
laboratory
that
we
will
consider
when
comparing
the
relative
stringency
of
an
annual
limit
of
67
ug/
dscm
and
a
shortterm
limit
of
120
ug/
dscm.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
identified
three
potential
beyond­
the­
floor
techniques
for
control
of
mercury:
(
1)
activated
carbon
injection;
(
2)
control
of
mercury
in
the
hazardous
waste
feed;
and
(
3)
control
of
Redline­
strikeout
highlighting
changes
made
during
OMB
review
27
USEPA,
"
Draft
Technical
Support
Document
for
HWC
MACT
Replacement
Standards,
Volume
V:
Emission
Estimates
and
Engineering
Costs",
March
2004,
Chapter
4.

159
mercury
in
the
raw
materials
and
auxiliary
fuels.
For
reasons
discussed
below,
we
are
not
proposing
a
beyond­
the­
floor
standard
for
mercury.
Use
of
Activated
Carbon
Injection.
We
evaluated
activated
carbon
injection
as
beyondthe
floor
control
for
further
reduction
of
mercury
emissions.
Activated
carbon
has
been
demonstrated
for
controlling
mercury
in
several
combustion
applications;
however,
currently
no
lightweight
aggregate
kiln
that
burns
hazardous
waste
uses
activated
carbon
injection.
Given
this
lack
of
experience
using
activated
carbon
injection,
we
made
a
conservative
assumption
that
the
use
of
activated
carbon
injection
will
provide
70%
mercury
control
and
evaluated
a
beyond­
thefloor
level
of
20
ug/
dscm.
In
addition,
for
costing
purposes
we
assumed
that
sources
needing
activated
carbon
injection
to
achieve
the
beyond­
the­
floor
level
would
install
the
activated
carbon
injection
system
after
the
existing
baghouse
and
add
a
new,
smaller
baghouse
to
remove
the
injected
carbon
with
the
adsorbed
mercury.
We
chose
this
costing
approach
to
address
potential
concerns
that
injected
carbon
may
interfere
with
lightweight
aggregate
kiln
dust
use
practices.
The
national
incremental
annualized
compliance
cost
for
lightweight
aggregate
kilns
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
1.1
million
and
would
provide
an
incremental
reduction
in
mercury
emissions
beyond
the
MACT
floor
controls
of
11
pounds
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
evaluated
to
estimate
the
impacts
between
activated
carbon
injection
and
controls
likely
to
be
used
to
meet
the
floor
level.
We
estimate
that
this
beyond­
thefloor
option
would
increase
the
amount
of
solid
waste
generated
by
270
tons
per
year
and
would
require
sources
to
use
an
additional
1.2
million
kW­
hours
per
year
beyond
the
requirements
to
achieve
the
floor
level.
The
costs
associated
with
these
impacts
are
accounted
for
in
the
national
annualized
compliance
cost
estimates.
Therefore,
based
on
these
factors
and
costs
of
approximately
$
209
million
per
additional
ton
of
mercury
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
activated
carbon
injection.
Feed
Control
of
Mercury
in
the
Hazardous
Waste.
We
also
evaluated
a
beyond­
the­
floor
level
of
54
ug/
dscm,
which
represents
a
20%
reduction
from
the
floor
level.
We
chose
a
20%
reduction
as
a
level
representing
the
practicable
extent
that
additional
feedrate
control
of
mercury
in
hazardous
waste
(
beyond
feedrate
control
that
may
be
necessary
to
achieve
the
floor
level)
can
be
used
and
still
achieve
modest
emissions
reductions.
27
The
national
incremental
annualized
compliance
cost
for
lightweight
aggregate
kilns
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
0.3
million
and
would
provide
an
incremental
reduction
in
mercury
emissions
beyond
the
MACT
floor
controls
of
3
pounds
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
also
evaluated.
Therefore,
based
on
these
factors
and
costs
of
approximately
$
229
million
per
additional
ton
of
mercury
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
feed
control
of
mercury
in
the
hazardous
waste.
Feed
Control
of
Mercury
in
the
Raw
Materials
and
Auxiliary
Fuels.
Lightweight
Redline­
strikeout
highlighting
changes
made
during
OMB
review
160
aggregate
kilns
could
achieve
a
reduction
in
mercury
emissions
by
substituting
a
raw
material
containing
a
lower
level
of
mercury
for
a
primary
raw
material
with
a
higher
level.
We
believe
that
this
beyond­
the­
floor
option
would
be
even
less
cost­
effective
than
either
of
the
options
discussed
above,
however.
Given
that
sources
are
sited
near
the
supply
of
the
primary
raw
material,
transporting
large
quantities
of
an
alternate
source
of
raw
materials,
even
if
available,
is
likely
to
be
cost­
prohibitive,
especially
considering
the
small
expected
emissions
reductions
that
would
result.
We
also
considered
whether
fuel
switching
to
an
auxiliary
fuel
containing
a
lower
concentration
of
mercury
would
be
an
appropriate
control
option
for
sources.
Two
facilities
typically
burn
hazardous
waste
at
a
fuel
replacement
rate
of
100%,
while
one
facility
has
burned
a
combination
of
fuel
oil
and
natural
gas
in
addition
to
the
hazardous
waste.
We
considered
switching
only
to
natural
gas
as
the
auxiliary
fuel
as
a
potential
beyond­
the­
floor
option.
We
do
not
believe
that
switching
to
natural
gas
is
a
viable
control
option
for
the
same
reasons
discussed
above
for
cement
kilns.
For
the
reasons
discussed
above,
we
propose
to
establish
the
emission
standard
for
existing
lightweight
aggregate
kilns
at
67
ug/
dscm.
If
we
were
to
adopt
such
a
standard,
we
are
proposing
that
sources
comply
with
the
standard
on
an
annual
basis
because
it
is
based
on
normal
emissions
data.
3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
Mercury
emissions
from
new
lightweight
aggregate
kilns
are
currently
limited
to
120
ug/
dscm
by
§
63.1205(
b)(
2).
This
standard
was
promulgated
in
the
Interim
Standards
Rule
(
See
67
FR
at
6797).
The
MACT
floor
for
new
sources
for
mercury
would
be
67
ug/
dscm,
which
considers
emissions
variability.
This
is
an
emission
level
that
the
single
best
performing
source
identified
with
the
SRE/
Feed
Approach
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
operating
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
We
identified
the
same
three
potential
beyond­
the­
floor
techniques
for
control
of
mercury:
(
1)
use
of
activated
carbon;
(
2)
control
of
mercury
in
the
hazardous
waste
feed;
and
(
3)
control
of
the
mercury
in
the
raw
materials
and
auxiliary
fuels.
Use
of
Activated
Carbon
Injection.
We
evaluated
activated
carbon
injection
as
beyondthe
floor
control
for
further
reduction
of
mercury
emissions.
We
made
a
conservative
assumption
that
the
use
of
activated
carbon
injection
will
provide
70%
mercury
control
and
evaluated
a
beyond­
the­
floor
level
of
20
ug/
dscm.
The
incremental
annualized
compliance
cost
for
a
new
lightweight
aggregate
kiln
with
average
gas
flow
rate
to
meet
this
beyond­
the­
floor
level,
rather
than
comply
with
the
floor
level,
would
be
approximately
$
0.26
million
and
would
provide
an
incremental
reduction
in
mercury
emissions
of
approximately
42
pounds
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
are
accounted
for
in
the
national
annualized
compliance
cost
estimates.
Therefore,
based
on
these
factors
and
costs
of
$
12
million
per
ton
of
mercury
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
activated
carbon
injection
for
new
sources.
Redline­
strikeout
highlighting
changes
made
during
OMB
review
28
A
greenfield
source
is
a
kiln
constructed
at
a
site
where
no
lightweight
aggregate
kiln
previously
existed;
however,
a
newly
constructed
or
reconstructed
kiln
at
an
existing
site
would
not
be
considered
as
a
greenfield
kiln.

161
Feed
Control
of
Mercury
in
the
Hazardous
Waste.
We
also
believe
that
the
expense
for
further
reduction
in
mercury
emissions
based
on
further
control
of
mercury
concentrations
in
the
hazardous
waste
is
not
warranted.
A
beyond­
the­
floor
level
of
54
ug/
dscm,
which
represents
a
20%
reduction
from
the
floor
level,
would
result
in
little
additional
mercury
reductions.
For
similar
reasons
discussed
above
for
existing
sources,
we
conclude
that
a
beyond­
the­
floor
standard
based
on
controlling
the
mercury
in
the
hazardous
waste
feed
would
not
be
justified
because
of
the
costs
coupled
with
estimated
emission
reductions.
Feed
Control
of
Mercury
in
the
Raw
Materials
and
Auxiliary
Fuels.
Lightweight
aggregate
kilns
could
achieve
a
reduction
in
mercury
emissions
by
substituting
a
raw
material
containing
lower
levels
of
mercury
for
a
primary
raw
material
with
a
higher
level.
For
a
new
source
at
an
existing
lightweight
aggregate
plant,
we
believe
that
this
beyond­
the­
floor
option
would
not
be
cost­
effective
due
to
the
costs
of
transporting
large
quantities
of
an
alternate
source
of
raw
materials
to
the
facility.
Given
that
the
plant
site
already
exists
and
sited
near
the
source
of
raw
material,
replacing
the
raw
materials
at
the
plant
site
with
lower
mercury­
containing
materials
would
be
the
source's
only
option.
For
a
new
lightweight
aggregate
kiln
constructed
at
a
new
site
 
a
greenfield
site28
 
we
are
not
aware
of
any
information
and
data
from
a
source
that
has
undertaken
or
is
currently
located
at
a
site
whose
raw
materials
are
low
in
mercury
which
would
consistently
decrease
mercury
emissions.
Further,
we
are
uncertain
as
to
what
beyond­
the­
floor
standard
would
be
achievable
using
a
lower,
if
it
exists,
mercury­
containing
raw
material.
Although
we
are
doubtful
that
selecting
a
new
plant
site
based
on
the
content
of
metals
in
the
raw
material
is
a
realistic
beyond­
the­
floor
option
considering
the
numerous
additional
factors
that
go
into
such
a
decision,
we
solicit
comment
on
whether
and
what
level
of
a
beyond­
the­
floor
standard
based
on
controlling
the
level
of
mercury
in
the
raw
materials
is
appropriate.
We
also
considered
whether
fuel
switching
to
an
auxiliary
fuel
containing
a
lower
concentration
of
mercury
would
be
an
appropriate
control
option
for
sources.
We
considered
using
natural
gas
in
lieu
of
a
fuel
containing
higher
concentrations
of
mercury
as
a
potential
beyond­
the­
floor
option.
As
discussed
for
existing
sources,
we
are
concerned
about
the
availability
of
the
natural
gas
infrastructure
in
all
regions
of
the
United
States
and
believe
that
using
natural
gas
would
not
be
a
viable
control
option
for
all
new
sources.
Therefore,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
limiting
mercury
in
the
raw
material
feed
and
auxiliary
fuels.
Therefore,
we
propose
a
mercury
standard
of
67
ug/
dscm
for
new
sources.
If
we
were
to
adopt
such
a
standard,
we
are
proposing
that
sources
comply
with
the
standard
on
an
annual
basis
because
it
is
based
on
normal
emissions
data.
C.
What
Are
the
Proposed
Standards
for
Particulate
Matter?
We
are
proposing
to
establish
standards
for
existing
and
new
lightweight
aggregate
kilns
that
limit
emissions
of
particulate
matter
to
0.025
and
0.0099
gr/
dscf,
respectively.
This
standard
Redline­
strikeout
highlighting
changes
made
during
OMB
review
162
would
control
unenumerated
HAP
metals
in
hazardous
waste,
and
all
non­
Hg
HAP
metals
in
the
raw
material
and
fossil
fuel
inputs
to
the
kiln.
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
Particulate
matter
emissions
for
existing
lightweight
aggregate
kilns
are
currently
limited
to
0.025
gr/
dscf
(
57
mg/
dscm)
by
§
63.1205(
a)(
7).
This
standard
was
promulgated
in
the
Interim
Standards
Rule
(
See
67
FR
at
6797).
The
particulate
matter
standard
is
a
surrogate
control
for
the
non­
mercury
metal
HAP.
All
lightweight
aggregate
kilns
control
particulate
matter
with
baghouses.
We
have
compliance
test
emissions
data
representing
maximum
emissions
for
all
lightweight
aggregate
kiln
sources.
For
most
sources,
we
have
compliance
test
emissions
data
from
more
than
one
compliance
test
campaign.
Our
database
of
particulate
matter
stack
emissions
range
from
0.001
to
0.042
gr/
dscf.
To
identify
the
MACT
floor,
we
evaluated
the
compliance
test
emissions
data
associated
with
the
most
recent
test
campaign
using
the
APCD
Approach.
The
calculated
floor
is
0.029
gr/
dscf,
which
considers
emissions
variability.
This
is
an
emission
level
that
the
average
of
the
best
performing
sources
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
operating
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
The
calculated
floor
level
of
0.029
gr/
dscf
is
less
stringent
than
the
interim
standard
of
0.025
gr/
dscf,
which
is
a
regulatory
limit
relevant
in
identifying
the
floor
level
(
so
as
to
avoid
any
backsliding
from
a
current
level
of
performance
achieved
by
all
lightweight
aggregate
kilns,
and
hence,
the
level
of
minimal
stringency
at
which
EPA
could
calculate
the
MACT
floor).
Therefore,
we
are
proposing
the
floor
level
as
the
current
emission
standard
of
0.025
gr/
dscf.
This
emission
level
is
currently
being
achieved
by
all
sources.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
evaluated
improved
particulate
matter
control
to
achieve
a
beyond­
the­
floor
standard
of
29
mg/
dscm
(
0.013
gr/
dscf).
The
national
incremental
annualized
compliance
cost
for
lightweight
aggregate
kilns
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
0.32
million
and
would
provide
an
incremental
reduction
in
particulate
matter
emissions
beyond
the
MACT
floor
controls
of
8.6
tons
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
evaluated
to
estimate
the
impacts
between
further
improvements
to
control
particulate
matter
and
controls
likely
to
be
used
to
meet
the
floor
level.
We
estimate
that
this
beyond­
the­
floor
option
would
increase
the
amount
of
solid
waste
generated
by
9
tons
per
year
beyond
the
requirements
to
achieve
the
floor
level.
Therefore,
based
on
these
factors
and
costs
of
approximately
$
36,600
per
additional
ton
of
particulate
matter
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
improved
particulate
matter
control.
3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
Particulate
matter
emissions
from
new
lightweight
aggregate
kilns
are
currently
limited
to
0.025
gr/
dscf
by
§
63.1205(
b)(
7).
This
standard
was
promulgated
in
the
Interim
Standards
Rule
(
See
67
FR
at
6797,
February
13,
2002).
The
MACT
floor
for
new
sources
for
particulate
matter
would
be
23
mg/
dscm
(
0.0099
gr/
dscf),
which
considers
emissions
variability.
This
is
an
emission
level
that
the
single
best
Redline­
strikeout
highlighting
changes
made
during
OMB
review
163
performing
source
identified
with
the
APCD
Approach
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
operating
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
We
evaluated
improved
particulate
matter
control
to
achieve
a
beyond­
the­
floor
standard.
We
evaluated
a
beyond­
the­
floor
level
of
12
mg/
dscm
(
0.005
gr/
dscf).
The
incremental
annualized
compliance
cost
for
a
new
lightweight
aggregate
kiln
with
an
average
gas
flow
rate
to
meet
this
beyond­
the­
floor
level,
rather
than
comply
with
the
floor
level,
would
be
approximately
$
91,400
million
and
would
provide
an
incremental
reduction
in
particulate
matter
emissions
of
approximately
2
tons
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
also
evaluated
and
are
included
in
the
cost
estimates.
Therefore,
based
on
these
factors
and
costs
of
approximately
$
45,600
per
additional
ton
of
particulate
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
improved
particulate
matter
control
for
new
lightweight
aggregate
kilns.
Therefore,
we
propose
a
particulate
matter
standard
of
2.3
mg/
dscm
(
0.0099
gr/
dscf)
for
new
sources.
D.
What
Are
the
Proposed
Standards
for
Semivolatile
Metals?
We
are
proposing
to
establish
standards
for
existing
lightweight
aggregate
kilns
that
limit
emissions
of
semivolatile
metals
(
cadmium
and
lead,
combined)
to
3.1
x
10­
4
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste
and
250
ug/
dscm.
The
proposed
standard
for
new
sources
is
2.4
x
10­
5
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste
and
43
ug/
dscm.
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
Semivolatile
metals
emissions
from
existing
lightweight
aggregate
kilns
are
currently
limited
to
250
ug/
dscm
by
§
63.1205(
a)(
3).
This
standard
was
promulgated
in
the
Interim
Standards
Rule
(
See
67
FR
at
6797).
Lightweight
aggregate
kilns
control
emissions
of
semivolatile
metals
with
baghouses
and/
or
by
controlling
the
feed
concentration
of
semivolatile
metals
in
the
hazardous
waste.
We
have
compliance
test
emissions
data
representing
maximum
emissions
for
all
lightweight
aggregate
kiln
sources.
For
most
sources,
we
have
compliance
test
emissions
data
from
more
than
one
compliance
test
campaign.
Semivolatile
metal
stack
emissions
range
from
approximately
1
to
over
1,600
ug/
dscm.
These
emissions
are
expressed
as
mass
of
semivolatile
metals
(
from
all
feedstocks)
per
unit
volume
of
stack
gas.
Hazardous
waste
thermal
emissions
range
from
3.0
x
10­
6
to
1.1
x
10­
3
lbs
per
million
Btu.
Hazardous
waste
thermal
emissions
represent
the
mass
of
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste.
For
most
lightweight
aggregate
kilns,
lead
was
the
major
contributor
to
semivolatile
emissions.
To
identify
the
MACT
floor,
we
evaluated
the
compliance
test
emissions
data
associated
with
the
most
recent
test
campaign
using
the
SRE/
Feed
Approach.
The
calculated
floor
is
3.1
x
10­
4
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste,
which
considers
emissions
variability.
This
is
an
emission
level
that
the
average
of
the
best
performing
sources
could
be
expected
to
achieve
in
99
of
100
future
tests
Redline­
strikeout
highlighting
changes
made
during
OMB
review
29
USEPA,
"
Draft
Technical
Support
Document
for
HWC
MACT
Replacement
Standards,
Volume
III:
Selection
of
MACT
Standards
and
Technologies,"
March
2004,
Chapter
23.

164
when
operating
under
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
We
estimate
that
this
emission
level
is
being
achieved
by
71%
of
sources,
and
would
reduce
semivolatile
metals
emissions
by
30
pounds
per
year.
To
put
the
proposed
floor
level
in
context
for
a
hypothetical
lightweight
aggregate
kiln
that
gets
90%
of
its
required
heat
input
from
hazardous
waste,
a
thermal
emissions
level
of
3.1
x
10­
4
lbs
semivolatile
metals
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste
equates
approximately
to
a
stack
gas
concentration
of
300
ug/
dscm.
This
estimated
stack
gas
concentration
does
not
include
contributions
to
emission
from
other
semivolatile
metals­
containing
materials
such
as
raw
materials
and
fossil
fuels.
The
additional
contribution
to
stack
emissions
of
semivolatile
metals
in
an
average
raw
material
is
estimated
to
range
as
high
as
20
to
50
ug/
dscm.
Thus,
for
the
hypothetical
lightweight
aggregate
kiln
the
thermal
emissions
floor
level
of
3.1
x
10­
4
lbs
semivolatile
metals
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste
is
estimated
to
be
less
than
350
ug/
dscm,
which
is
higher
than
the
current
interim
standard
of
250
ug/
dscm.
Given
that
comparing
the
proposed
floor
level
to
the
interim
standard
requires
numerous
assumptions
(
as
just
illustrated)
including
hazardous
waste
fuel
replacement
rates,
heat
input
requirements
per
ton
of
clinker,
concentrations
of
semivolatile
metals
in
the
raw
material
and
fuels,
and
system
removal
efficiency,
we
have
included
a
more
detailed
analysis
in
the
background
document.
29
Our
detailed
analysis
indicates
the
proposed
floor
level
could
be
less
stringent
than
the
interim
standard
for
some
sources.
In
order
to
avoid
any
backsliding
from
the
current
level
of
performance
achieved
by
all
lightweight
aggregate
kilns,
we
propose
a
dual
standard:
the
semivolatile
metals
standard
as
both
the
calculated
floor
level,
expressed
as
a
hazardous
waste
thermal
emissions
level,
and
the
current
interim
standard.
This
would
ensure
that
all
sources
are
complying
with
a
limit
that
is
at
least
as
stringent
as
the
interim
standard.
In
the
September
1999
final
rule,
we
acknowledged
that
a
lightweight
aggregate
kiln
using
properly
designed
and
operated
MACT
control
technologies,
including
controlling
the
levels
of
metals
in
the
hazardous
waste,
may
not
be
capable
of
achieving
a
given
emission
standard
because
of
mineral
and
process
raw
material
contributions
that
might
cause
an
exceedance
of
the
emission
standard.
To
address
this
concern,
we
promulgated
a
provision
that
allows
kilns
to
petition
for
alternative
standards
provided
that
they
submit
site­
specific
information
that
shows
raw
material
hazardous
air
pollutant
contributions
to
the
emissions
prevent
the
source
from
complying
with
the
emission
standard
even
though
the
kiln
is
using
MACT
control.
See
§
63.1206(
b)(
9).
If
we
were
to
adopt
the
proposed
dual
semivolatile
(
and
low
volatile)
metals
standards
approach,
we
propose
to
retain
the
alternative
standard
provisions
under
§
63.1206(
b)(
9)
for
semivolatile
metals
(
and
low
volatile
metals).
We
invite
comment
on
this
approach.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
identified
three
potential
beyond­
the­
floor
techniques
for
control
of
semivolatile
Redline­
strikeout
highlighting
changes
made
during
OMB
review
165
metals:
(
1)
improved
particulate
matter
control;
(
2)
control
of
semivolatile
metals
in
the
hazardous
waste
feed;
and
(
3)
control
of
the
semivolatile
metals
in
the
raw
materials
and
fuels.
Improved
Particulate
Matter
Control.
Controlling
particulate
matter
also
controls
emissions
of
semivolatile
metals.
Our
data
show
that
all
lightweight
aggregate
kilns
are
already
achieving
greater
than
99.7%
system
removal
efficiency
for
semivolatile
metals,
with
many
attaining
99.9%
removal.
Thus,
additional
control
of
particulate
matter
are
likely
to
result
in
only
modest
additional
reductions
of
semivolatile
metals
emissions.
We
evaluated
a
beyond­
the­
floor
level
of
1.5
x
10­
4
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste,
which
represents
a
50%
reduction
in
emissions
from
MACT
floor
levels.
The
national
incremental
annualized
compliance
cost
for
lightweight
aggregate
kilns
to
meet
this
beyond­
the­
floor
level
rather
than
to
comply
with
the
floor
controls
would
be
approximately
$
84,200
and
would
provide
an
incremental
reduction
in
semivolatile
metals
emissions
beyond
the
MACT
floor
controls
of
20
pounds
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
evaluated
to
estimate
the
impacts
between
further
improvements
to
control
particulate
matter
and
controls
likely
to
be
used
to
meet
the
floor
level.
We
estimate
that
this
beyond­
the­
floor
option
would
increase
the
amount
of
solid
waste
generated
by
less
than
10
tons
per
year
and
would
also
require
sources
to
use
an
additional
2,000
kW­
hours
per
year
beyond
the
requirements
to
achieve
the
floor
level.
The
costs
associated
with
these
impacts
are
accounted
for
in
the
national
annualized
compliance
cost
estimates.
Therefore,
based
on
these
factors
and
costs
of
approximately
$
7.6
million
per
additional
ton
of
semivolatile
metals
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
improved
particulate
matter
control.
Feed
Control
of
Semivolatile
Metals
in
the
Hazardous
Waste.
We
also
evaluated
a
beyond­
the­
floor
level
of
2.5
x
10­
4
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste,
which
represents
a
20%
reduction
from
the
floor
level.
We
chose
a
20%
reduction
as
a
level
representing
the
practicable
extent
that
additional
feedrate
control
of
semivolatile
metals
in
hazardous
waste
can
be
used
and
still
achieve
appreciable
emissions
reductions.
The
national
incremental
annualized
compliance
cost
for
lightweight
aggregate
kilns
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
6,000
and
would
provide
an
incremental
reduction
in
semivolatile
metals
emissions
beyond
the
MACT
floor
controls
of
less
than
one
pound
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
evaluated
and
are
included
in
the
national
compliance
cost
estimates.
Therefore,
based
on
these
factors
and
costs
of
approximately
$
20
million
per
additional
ton
of
semivolatile
metals
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
feed
control
of
semivolatile
metals
in
the
hazardous
waste.
Feed
Control
of
Semivolatile
Metals
in
the
Raw
Materials
and
Auxiliary
Fuels.
Lightweight
aggregate
kilns
could
achieve
a
reduction
in
semivolatile
metal
emissions
by
substituting
a
raw
material
containing
lower
levels
of
cadmium
and/
or
lead
for
a
primary
raw
material
with
higher
levels
of
these
metals.
We
believe
that
this
beyond­
the­
floor
option
would
even
be
less
cost­
effective
than
either
of
the
options
discussed
above,
however.
Given
that
facilities
are
sited
near
the
primary
raw
material
supply,
acquiring
and
transporting
large
quantities
Redline­
strikeout
highlighting
changes
made
during
OMB
review
166
of
an
alternate
source
of
raw
materials
is
likely
to
be
cost­
prohibitive.
Therefore,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
limiting
semivolatile
metals
in
the
raw
material
feed.
We
also
considered
whether
fuel
switching
to
an
auxiliary
fuel
containing
a
lower
concentration
of
semivolatile
metals
would
be
an
appropriate
control
option
for
sources.
Two
facilities
typically
burn
hazardous
waste
at
a
fuel
replacement
rate
of
100%,
while
one
facility
has
burned
a
combination
of
fuel
oil
and
natural
gas
in
addition
to
the
hazardous
waste.
We
considered
switching
only
to
natural
gas
as
the
auxiliary
fuel
as
a
potential
beyond­
the­
floor
option.
We
do
not
believe
that
switching
to
natural
gas
is
a
viable
control
option
for
similar
reasons
discussed
above
for
cement
kilns.
For
the
reasons
discussed
above,
we
propose
to
establish
the
emission
standard
for
existing
lightweight
aggregate
kilns
at
3.1
x
10­
4
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste
and
250
ug/
dscm.
3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
Semivolatile
metals
emissions
from
new
lightweight
aggregate
kilns
are
currently
limited
to
43
ug/
dscm
by
§
63.1205(
b)(
3).
This
standard
was
promulgated
in
the
Interim
Standards
Rule
(
See
67
FR
at
6797).
The
MACT
floor
for
new
sources
for
semivolatile
metals
would
be
2.4
x
10­
5
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
in
the
hazardous
waste,
which
considers
emissions
variability.
This
is
an
emission
level
that
the
single
best
performing
source
identified
with
the
SRE/
Feed
Approach
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
operating
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
To
put
the
proposed
floor
level
in
context
for
a
hypothetical
lightweight
aggregate
kiln
that
gets
90%
of
its
required
heat
input
from
hazardous
waste,
a
thermal
emissions
level
of
2.4
x
10­
5
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste
can
equate
to
a
stack
gas
concentration
as
high
as
60
ug/
dscm,
including
contributions
from
typical
raw
materials.
Thus,
for
the
hypothetical
lightweight
aggregate
kiln
the
thermal
emissions
floor
level
of
2.4
x
10­
5
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste
is
estimated
to
be
as
high
as
60
ug/
dscm,
which
is
higher
than
the
current
interim
standard
of
43
ug/
dscm.
In
order
to
avoid
any
backsliding
from
the
current
level
of
performance
for
a
new
lightweight
aggregate
kiln
source,
we
propose
a
dual
standard:
the
semivolatile
metals
standard
as
both
the
calculated
floor
level,
expressed
as
a
hazardous
waste
thermal
emissions
level,
and
the
current
interim
standard.
This
would
ensure
that
all
sources
are
complying
with
a
limit
that
is
at
least
as
stringent
as
the
interim
standard.
Thus,
the
proposed
MACT
floor
for
new
lightweight
aggregate
kilns
is
2.4
x
10­
5
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste
and
43
ug/
dscm.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
We
identified
the
same
three
potential
beyond­
the­
floor
techniques
for
control
of
semivolatile
metals:
(
1)
improved
control
of
particulate
matter;
(
2)
control
of
semivolatile
metals
in
the
hazardous
waste
feed;
and
(
3)
control
of
semivolatile
metals
in
the
raw
materials
and
fuels.
Redline­
strikeout
highlighting
changes
made
during
OMB
review
167
Improved
Particulate
Matter
Control.
Controlling
particulate
matter
also
controls
emissions
of
semivolatile
metals.
We
evaluated
improved
control
of
particulate
matter
based
on
a
state­
of­
the­
art
baghouse
using
a
high
quality
fabric
filter
bag
material
as
beyond­
the­
floor
control
for
further
reductions
in
semivolatile
metals
emissions.
We
evaluated
a
beyond­
the­
floor
level
of
1.2
x
10­
5
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste.
The
incremental
annualized
compliance
cost
for
a
new
lightweight
aggregate
kiln
with
average
gas
flowrate
to
meet
this
beyond­
the­
floor
level,
rather
than
to
comply
with
the
floor
level,
would
be
approximately
$
0.11
million
and
would
provide
an
incremental
reduction
in
semivolatile
metals
emissions
of
approximately
13
pounds
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
evaluated
and
are
included
in
the
cost
estimates.
We
estimate
that
this
beyond­
the­
floor
option
would
increase
the
amount
of
solid
waste
generated
by
3
tons
per
year
and
would
also
require
sources
to
use
an
additional
0.3
million
kW­
hours
per
year
beyond
the
requirements
to
achieve
the
floor
level.
Therefore,
based
on
these
factors
and
costs
of
approximately
$
18
million
per
ton
of
semivolatile
metals
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
improved
particulate
matter
control
for
new
lightweight
aggregate
kilns.
Feed
Control
of
Semivolatile
Metals
in
the
Hazardous
Waste.
We
also
believe
that
the
expense
for
further
reduction
in
semivolatile
metals
emissions
based
on
further
control
of
semivolatile
metals
concentrations
in
the
hazardous
waste
is
not
warranted.
We
considered
a
beyond­
the­
floor
level
of
1.9
x
10­
5
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste,
which
represents
a
20%
reduction
from
the
floor
level.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
evaluated
and
are
included
in
the
compliance
cost
estimates.
For
similar
reasons
discussed
above
for
existing
sources,
we
conclude
that
a
beyond­
the­
floor
standard
based
on
controlling
the
concentration
of
semivolatile
metals
levels
in
the
hazardous
waste
feed
would
not
be
justified
because
of
the
costs
and
estimated
emission
reductions.
Feed
Control
of
Semivolatile
Metals
in
the
Raw
Materials
and
Auxiliary
Fuels.
Lightweight
aggregate
kilns
could
achieve
a
reduction
in
semivolatile
metals
emissions
by
substituting
a
raw
material
containing
lower
levels
of
cadmium
and
lead
for
a
primary
raw
material
with
a
higher
level.
For
a
new
source
at
an
existing
facility,
we
believe
that
this
beyondthe
floor
option
would
not
be
cost­
effective
due
to
the
costs
of
transporting
large
quantities
of
an
alternate
source
of
raw
material
to
the
facility.
Given
that
the
plant
site
already
exists
and
sited
near
the
source
of
raw
material,
replacing
the
raw
materials
at
the
plant
site
with
lower
semivolatile
metals­
containing
materials
would
be
the
source's
only
option.
For
a
kiln
constructed
at
a
new
greenfield
site,
we
are
not
aware
of
any
information
and
data
from
a
source
that
has
undertaken
or
is
currently
located
at
a
site
whose
raw
materials
are
inherently
lower
in
semivolatile
metals
that
would
consistently
achieve
reduced
semivolatile
metals
emissions.
Further,
we
are
uncertain
as
to
what
beyond­
the­
floor
standard
would
be
achievable
using,
if
it
exists,
a
lower
semivolatile
metals­
containing
raw
material.
Although
we
are
doubtful
that
selecting
a
new
plant
site
based
on
the
content
of
metals
in
the
raw
material
is
a
realistic
beyondthe
floor
option
considering
the
numerous
additional
factors
that
go
into
such
a
decision,
we
solicit
comment
on
whether
and
what
level
of
a
beyond­
the­
floor
standard
based
on
controlling
Redline­
strikeout
highlighting
changes
made
during
OMB
review
168
the
level
of
semivolatile
metals
in
the
raw
materials
is
appropriate.
We
also
considered
whether
fuel
switching
to
an
auxiliary
fuel
containing
a
lower
concentration
of
semivolatile
metals
would
be
an
appropriate
control
option
for
sources.
Two
facilities
typically
burn
hazardous
waste
at
a
fuel
replacement
rate
of
100%,
while
one
facility
has
burned
a
combination
of
fuel
oil
and
natural
gas
in
addition
to
the
hazardous
waste.
We
considered
switching
only
to
natural
gas
as
the
auxiliary
fuel
as
a
potential
beyond­
the­
floor
option.
We
do
not
believe
that
switching
to
natural
gas
is
a
viable
control
option
for
the
same
reasons
discussed
above
for
cement
kilns.
For
the
reasons
discussed
above,
we
propose
to
establish
the
emission
standard
for
new
lightweight
aggregate
kilns
at
2.4
x
10­
5
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
content
in
the
hazardous
waste
and
43
ug/
dscm.
E.
What
Are
the
Proposed
Standards
for
Low
Volatile
Metals?
We
are
proposing
to
establish
standards
for
existing
lightweight
aggregate
kilns
that
limit
emissions
of
low
volatile
metals
(
arsenic,
beryllium,
and
chromium)
to
9.5
x
10­
5
lbs
low
volatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste
and
110
ug/
dscm.
The
proposed
standard
for
new
sources
is
3.2
x
10­
5
lbs
low
volatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste
and
110
ug/
dscm.
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
Low
volatile
metals
emissions
from
existing
lightweight
aggregate
kilns
are
currently
limited
to
110
ug/
dscm
by
§
63.1205(
a)(
4).
This
standard
was
promulgated
in
the
Interim
Standards
Rule
(
See
67
FR
at
6797).
Lightweight
aggregate
kilns
control
emissions
of
low
volatile
metals
with
baghouses
and/
or
by
controlling
the
feed
concentration
of
low
volatile
metals
in
the
hazardous
waste.
We
have
compliance
test
emissions
data
representing
maximum
emissions
for
all
lightweight
aggregate
kiln
sources.
For
most
sources,
we
have
compliance
test
emissions
data
from
more
than
one
compliance
test
campaign.
Low
volatile
metal
stack
emissions
range
from
approximately
16
to
200
ug/
dscm.
These
emissions
are
expressed
as
mass
of
low
volatile
metals
(
from
all
feedstocks)
per
unit
volume
of
stack
gas.
Hazardous
waste
thermal
emissions
range
from
9.7
x
10­
6
to
1.8
x
10­
4
lbs
per
million
Btu.
Hazardous
waste
thermal
emissions
represent
the
mass
of
low
volatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste.
For
most
lightweight
aggregate
kilns,
chromium
was
the
major
contributor
to
low
volatile
emissions.
To
identify
the
MACT
floor,
we
evaluated
the
compliance
test
emissions
data
associated
with
the
most
recent
test
campaign
using
the
SRE/
Feed
Approach.
The
calculated
floor
is
9.5
x
10­
5
lbs
low
volatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste,
which
considers
emissions
variability.
This
is
an
emission
level
that
the
average
of
the
best
performing
sources
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
We
estimate
that
this
emission
level
is
being
achieved
by
57%
of
sources
and
would
reduce
low
volatile
metals
emissions
by
30
pounds
per
year.
To
put
the
proposed
floor
level
in
context
for
a
hypothetical
lightweight
aggregate
kiln
Redline­
strikeout
highlighting
changes
made
during
OMB
review
30
USEPA,
"
Draft
Technical
Support
Document
for
HWC
MACT
Replacement
Standards,
Volume
III:
Selection
of
MACT
Standards
and
Technologies,"
March
2004,
Chapter
23.

169
that
gets
90%
of
its
required
heat
input
from
hazardous
waste,
a
thermal
emissions
level
of
9.5
x
10­
5
lbs
low
volatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste
equates
approximately
to
a
stack
gas
concentration
of
90
ug/
dscm.
This
estimated
stack
gas
concentration
does
not
include
contributions
to
emission
from
other
low
volatile
metals­
containing
materials
such
as
raw
materials.
The
additional
contribution
to
stack
emissions
of
low
volatile
metals
in
an
average
raw
material
is
estimated
to
be
50
ug/
dscm.
Thus,
for
the
hypothetical
lightweight
aggregate
kiln
the
thermal
emissions
floor
level
of
9.5
x
10­
5
lbs
low
volatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste
is
estimated
to
be
150
ug/
dscm,
which
is
higher
than
the
current
interim
standard
of
110
ug/
dscm.
Given
that
comparing
the
proposed
floor
level
to
the
interim
standard
requires
numerous
assumptions
including
hazardous
waste
fuel
replacement
rates,
heat
input
requirements
per
ton
of
clinker,
concentrations
of
low
volatile
metals
in
the
raw
material
and
fuels,
and
system
removal
efficiency,
we
have
included
a
more
detailed
analysis
in
the
background
document.
30
Our
detailed
analysis
indicates
the
proposed
floor
level
could
be
less
stringent
than
the
interim
standard
for
some
sources.
In
order
to
avoid
any
backsliding
from
the
current
level
of
performance
achieved
by
all
lightweight
aggregate
kilns,
we
propose
a
dual
standard:
the
low
volatile
metals
standard
as
both
the
calculated
floor
level,
expressed
as
a
hazardous
waste
thermal
emissions
level,
and
the
current
interim
standard.
This
would
ensure
that
all
sources
are
complying
with
a
limit
that
is
at
least
as
stringent
as
the
interim
standard.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
identified
three
potential
beyond­
the­
floor
techniques
for
control
of
low
volatile
metals:
(
1)
improved
particulate
matter
control;
(
2)
control
of
low
volatile
metals
in
the
hazardous
waste
feed;
and
(
3)
control
of
the
low
volatile
metals
in
the
raw
materials
and
fuels.
Improved
Particulate
Matter
Control.
Controlling
particulate
matter
also
controls
emissions
of
low
volatile
metals.
Our
data
show
that
all
lightweight
aggregate
kilns
are
already
achieving
greater
than
99.8%
system
removal
efficiency
for
low
volatile
metals,
with
many
attaining
99.9%
or
greater
removal.
Thus,
additional
control
of
particulate
matter
emissions
is
likely
to
result
in
only
a
small
increment
in
reduction
of
low
volatile
metals
emissions.
We
evaluated
a
beyond­
the­
floor
level
of
4.7
x
10­
5
lbs
low
volatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste.
The
national
incremental
annualized
compliance
cost
for
lightweight
aggregate
kilns
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
0.24
million
and
would
provide
an
incremental
reduction
in
low
volatile
metals
emissions
beyond
the
MACT
floor
controls
of
28
pounds
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
evaluated
to
estimate
the
impacts
between
further
improvements
to
control
particulate
matter
and
controls
likely
to
be
used
to
meet
the
floor
level.
We
estimate
that
this
beyond­
the­
floor
option
would
increase
the
amount
of
solid
waste
generated
by
less
than
30
tons
Redline­
strikeout
highlighting
changes
made
during
OMB
review
170
per
year
and
would
also
require
sources
to
use
an
additional
46,000
kW­
hours
of
energy
per
year.
Therefore,
based
on
these
factors
and
costs
of
approximately
$
17
million
per
additional
ton
of
low
volatile
metals
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
improved
particulate
matter
control.
Feed
Control
of
Low
Volatile
Metals
in
the
Hazardous
Waste.
We
also
evaluated
a
beyond­
the­
floor
level
of
7.6
x
10­
5
lbs
low
volatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste,
which
represents
a
20%
reduction
from
the
floor
level.
We
chose
a
20%
reduction
as
a
level
representing
the
practicable
extent
that
additional
feedrate
control
of
low
volatile
metals
in
hazardous
waste
(
beyond
feedrate
control
that
may
be
necessary
to
achieve
the
floor
level)
can
be
used
and
still
achieve
modest
emissions
reductions.
The
national
incremental
annualized
compliance
cost
for
lightweight
aggregate
kilns
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
150,000
and
would
provide
an
incremental
reduction
in
low
volatile
metals
emissions
beyond
the
MACT
floor
controls
of
14
pounds
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
considered
and
are
included
in
the
cost
estimates.
Therefore,
based
on
these
factors
and
costs
of
approximately
$
22
million
per
additional
ton
of
low
volatile
metals
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
feed
control
of
low
volatile
metals
in
the
hazardous
waste.
Feed
Control
of
Low
Volatile
Metals
in
the
Raw
Materials
and
Auxiliary
Fuels.
Lightweight
aggregate
kilns
could
achieve
a
reduction
in
low
volatile
metal
emissions
by
substituting
a
raw
material
containing
lower
levels
of
arsenic,
beryllium,
and/
or
chromium
for
a
primary
raw
material
with
higher
levels
of
these
metals.
We
believe
that
this
beyond­
the­
floor
option
would
even
be
less
cost­
effective
than
either
of
the
options
discussed
above,
however.
Given
that
facilities
are
sited
near
the
primary
raw
material
supply,
acquiring
and
transporting
large
quantities
of
an
alternate
source
of
raw
materials
is
likely
to
be
cost­
prohibitive.
Therefore,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
limiting
low
volatile
metals
in
the
raw
material
feed.
We
also
considered
whether
fuel
switching
to
an
auxiliary
fuel
containing
a
lower
concentration
of
low
volatile
metals
would
be
an
appropriate
control
option
for
sources.
Two
facilities
typically
burn
hazardous
waste
at
a
fuel
replacement
rate
of
100%,
while
one
facility
has
burned
a
combination
of
fuel
oil
and
natural
gas
in
addition
to
the
hazardous
waste.
We
considered
switching
only
to
natural
gas
as
the
auxiliary
fuel
as
a
potential
beyond­
the­
floor
option.
We
do
not
believe
that
switching
to
natural
gas
is
a
viable
control
option
for
similar
reasons
discussed
above
for
cement
kilns.
For
the
reasons
discussed
above,
we
propose
to
establish
the
emission
standard
for
existing
lightweight
aggregate
kilns
at
9.5
x
10­
5
lbs
low
volatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste
and
110
ug/
dscm.
3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
Low
volatile
metals
emissions
from
new
lightweight
aggregate
kilns
are
currently
limited
to
110
ug/
dscm
by
§
63.1205(
b)(
4).
This
standard
was
promulgated
in
the
Interim
Standards
Rule
(
See
67
FR
at
6797).
The
MACT
floor
for
new
sources
for
low
volatile
metals
would
be
3.2
x
10­
5
lbs
low
Redline­
strikeout
highlighting
changes
made
during
OMB
review
171
volatile
metals
emissions
in
the
hazardous
waste
per
million
Btu
in
the
hazardous
waste,
which
considers
emissions
variability.
This
is
an
emission
level
that
the
single
best
performing
source
identified
with
the
SRE/
Feed
Approach
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
operating
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
As
discussed
for
existing
sources,
in
order
to
avoid
any
backsliding
from
the
current
level
of
performance
for
a
new
lightweight
aggregate
kiln
source,
we
propose
a
dual
standard:
the
low
volatile
metals
standard
as
both
the
calculated
floor
level,
expressed
as
a
hazardous
waste
thermal
emissions
level,
and
the
current
interim
standard.
This
would
ensure
that
all
sources
are
complying
with
a
limit
that
is
at
least
as
stringent
as
the
interim
standard.
Thus,
the
proposed
MACT
floor
for
new
lightweight
aggregate
kilns
is
3.2
x
10­
5
lbs
low
volatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste
and
110
ug/
dscm.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
We
considered
three
potential
beyond­
the­
floor
techniques
for
control
of
low
volatile
metals:
(
1)
improved
particulate
matter
control;
(
2)
control
of
low
volatile
metals
in
the
hazardous
waste
feed;
and
(
3)
control
of
the
low
volatile
metals
in
the
raw
materials
and
fuels.
Improved
Particulate
Matter
Control.
Controlling
particulate
matter
also
controls
emissions
of
low
volatile
metals.
We
evaluated
improved
control
of
particulate
matter
based
on
a
state­
of­
the­
art
baghouse
using
a
high
quality
fabric
filter
bag
material
as
beyond­
the­
floor
control
for
further
reductions
in
low
volatile
metals
emissions.
We
evaluated
a
beyond­
the­
floor
level
of
1.6
x
10­
5
lbs
low
volatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste.
The
incremental
annualized
compliance
cost
for
a
new
lightweight
aggregate
kiln
with
average
gas
flowrate
to
meet
this
beyond­
the­
floor
level,
rather
than
to
comply
with
the
floor
level,
would
be
approximately
$
0.11
million
and
would
provide
an
incremental
reduction
in
low
volatile
metals
emissions
of
approximately
16
pounds
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
evaluated
and
are
included
in
the
cost
estimates.
We
estimate
that
this
beyond­
the­
floor
option
would
increase
the
amount
of
solid
waste
generated
by
3
tons
per
year
and
would
also
require
sources
to
use
an
additional
0.3
million
kW­
hours
per
year
beyond
the
requirements
to
achieve
the
floor
level.
Therefore,
based
on
these
factors
and
costs
of
nearly
$
14
million
per
ton
of
low
volatile
metals
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
improved
particulate
matter
control
for
new
lightweight
aggregate
kilns.
Feed
Control
of
Low
Volatile
Metals
in
the
Hazardous
Waste.
We
also
believe
that
the
expense
for
further
reduction
in
low
volatile
metals
emissions
based
on
further
control
of
low
volatile
metals
concentrations
in
the
hazardous
waste
is
not
warranted.
We
considered
a
beyondthe
floor
level
of
2.6
x
10­
5
lbs
low
volatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste,
which
represents
a
20%
reduction
from
the
floor
level.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
evaluated
and
are
included
in
the
compliance
cost
estimates.
For
similar
reasons
discussed
above
for
existing
sources,
we
conclude
that
a
beyond­
the­
floor
standard
based
on
controlling
the
concentration
of
low
volatile
metals
levels
in
the
hazardous
waste
feed
would
not
be
justified
Redline­
strikeout
highlighting
changes
made
during
OMB
review
172
because
of
the
costs
and
estimated
emission
reductions.
Feed
Control
of
Low
Volatile
Metals
in
the
Raw
Materials
and
Auxiliary
Fuels.
Lightweight
aggregate
kilns
could
achieve
a
reduction
in
low
volatile
metals
emissions
by
substituting
a
raw
material
containing
lower
levels
of
arsenic,
beryllium,
and/
or
chromium
for
a
primary
raw
material
with
a
higher
level.
For
a
new
source
at
an
existing
facility,
we
believe
that
this
beyond­
the­
floor
option
would
not
be
cost­
effective
due
to
the
costs
of
transporting
large
quantities
of
an
alternate
source
of
raw
material
to
the
facility.
Given
that
the
plant
site
already
exists
and
sited
near
the
source
of
raw
material,
replacing
the
raw
materials
at
the
plant
site
with
lower
low
volatile
metals­
containing
materials
would
be
the
source's
only
option.
For
a
kiln
constructed
at
a
new
greenfield
site,
we
are
not
aware
of
any
information
and
data
from
a
source
that
has
undertaken
or
is
currently
located
at
a
site
whose
raw
materials
are
inherently
lower
in
low
volatile
metals
that
would
consistently
achieve
reduced
low
volatile
metals
emissions.
Further,
we
are
uncertain
as
to
what
beyond­
the­
floor
standard
would
be
achievable
using,
if
it
exists,
a
lower
low
volatile
metals­
containing
raw
material.
Although
we
are
doubtful
that
selecting
a
new
plant
site
based
on
the
content
of
metals
in
the
raw
material
is
a
realistic
beyondthe
floor
option
considering
the
numerous
additional
factors
that
go
into
such
a
decision,
we
solicit
comment
on
whether
and
what
level
of
a
beyond­
the­
floor
standard
based
on
controlling
the
level
of
low
volatile
metals
in
the
raw
materials
is
appropriate.
We
also
considered
whether
fuel
switching
to
an
auxiliary
fuel
containing
a
lower
concentration
of
low
volatile
metals
would
be
an
appropriate
control
option
for
sources.
Two
facilities
typically
burn
hazardous
waste
at
a
fuel
replacement
rate
of
100%,
while
one
facility
has
burned
a
combination
of
fuel
oil
and
natural
gas
in
addition
to
the
hazardous
waste.
We
considered
switching
only
to
natural
gas
as
the
auxiliary
fuel
as
a
potential
beyond­
the­
floor
option.
We
do
not
believe
that
switching
to
natural
gas
is
a
viable
control
option
for
the
same
reasons
discussed
above
for
cement
kilns.
For
the
reasons
discussed
above,
we
propose
to
establish
the
emission
standard
for
new
lightweight
aggregate
kilns
at
3.2
x
10­
5
lbs
low
volatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
content
in
the
hazardous
waste
and
110
ug/
dscm.
F.
What
Are
the
Proposed
Standards
for
Hydrogen
Chloride
and
Chlorine
Gas?
We
are
proposing
to
establish
standards
for
existing
and
new
lightweight
aggregate
kilns
that
limit
total
chlorine
emissions
(
hydrogen
chloride
and
chlorine
gas,
combined,
reported
as
a
chloride
equivalent)
to
.
However,
Although
we
are
also
proposing
to
invoke
CAA
section
112(
d)(
4)
to
establish
alternative
risk­
based
standards
in
lieu
of
the
MACT
emission
standards
for
total
chlorine.
The
emission
limits
would
be
based
on
national
exposure
standards
that
ensure
protection
of
public
health
with
an
ample
margin
of
safety,
the
risk­
based
standards
would
be
capped
at
the
interim
standards.
Given
that
we
are
proposing
MACT
standards
equivalent
to
the
interim
standards­­
600
ppmv,
an
emission
level
you
are
currently
achieving­­
you
would
not
be
eligible
for
the
section
112(
d)(
4)
risk­
based
standards.
See
Part
Two,
Section
XIII
for
additional
details.
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
Total
chlorine
emissions
from
existing
cement
kilns
are
limited
to
600
ppmv
by
§
63.1205(
a)(
6).
This
standard
was
promulgated
in
the
Interim
Standards
Rule
(
See
67
FR
at
Redline­
strikeout
highlighting
changes
made
during
OMB
review
31
We
also
considered
controlling
the
chlorine
levels
in
the
hazardous
waste
feed
and
controlling
the
chlorine
levels
in
the
raw
materials
as
potential
beyond­
the­
floor
techniques;
however,
it
is
our
judgment
that
they
are
not
likely
to
be
as
cost­
effective
as
dry
lime
scrubbing.

173
6797).
One
of
the
three
lightweight
aggregate
facilities
uses
a
venturi
scrubber
to
remove
total
chlorine
from
the
gas
stream.
The
system
removal
efficiency
(
SRE)
achieved
by
this
facility
during
compliance
testing
shows
removal
efficiencies
ranging
from
96
to
99%.
Sources
at
the
other
two
facilities
do
not
use
air
pollution
control
equipment
to
capture
emissions
of
total
chlorine,
and,
therefore,
SREs
are
negligible.
The
majority
of
the
chlorine
fed
to
the
lightweight
aggregate
kiln
during
a
compliance
test
comes
from
the
hazardous
waste.
In
all
but
a
few
cases
the
hazardous
waste
contribution
to
the
total
amount
of
chlorine
fed
to
the
kiln
represented
at
least
80%
of
the
total
loading
to
the
kiln.
The
proposed
MACT
floor
control
for
total
chlorine
is,
in
part,
based
on
controlling
the
concentration
of
chlorine
in
the
hazardous
waste.
The
chlorine
concentration
in
the
hazardous
waste
will
affect
emissions
of
total
chlorine
at
a
given
SRE
because
emissions
will
increase
as
the
chlorine
loading
increases.
We
have
compliance
test
emissions
data
representing
maximum
allowed
emissions
for
all
lightweight
aggregate
kiln
sources.
For
most
sources,
we
have
compliance
test
emissions
data
from
more
than
one
compliance
test
campaign.
Total
chlorine
emissions
range
from
14
to
116
ppmv
for
the
source
using
a
venturi
scrubber
and
range
from
500
to
2,400
ppmv
at
sources
without
scrubbing
control
equipment.
To
identify
the
MACT
floor,
we
evaluated
the
compliance
test
emissions
data
associated
with
the
most
recent
test
campaign
using
the
SRE/
Feed
Approach.
The
calculated
floor
is
3.0
lbs
total
chlorine
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste,
which
considers
emissions
variability.
This
is
an
emission
level
that
the
average
of
the
best
performing
sources
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
To
put
the
proposed
floor
level
in
context
for
a
hypothetical
lightweight
aggregate
kiln
that
gets
90%
of
its
required
heat
input
from
hazardous
waste,
a
thermal
emissions
level
of
3.0
lbs
total
chlorine
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste
equates
approximately
to
a
stack
gas
concentration
of
1,970
ppmv.
This
estimated
stack
gas
concentration
does
not
include
contributions
to
emission
from
other
chlorinecontaining
materials
such
as
raw
materials.
Given
that
the
calculated
floor
level
is
less
stringent
than
the
current
interim
emission
standard
of
600
ppmv.
In
order
to
avoid
any
backsliding
from
the
current
level
of
performance
achieved
by
all
lightweight
aggregate
kilns,
we
are
proposing
the
floor
standard
as
the
current
emission
standard
of
600
ppmv.
This
emission
level
is
currently
being
achieved
by
all
sources.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
considered
a
beyond­
the­
floor
standard
of
150
ppmv
based
on
the
assumption
that
dry
lime
scrubbing
will
provide
75%
control
of
hydrogen
chloride.
31
In
addition,
for
costing
purposes
Redline­
strikeout
highlighting
changes
made
during
OMB
review
174
we
assumed
that
lightweight
aggregate
kilns
needing
total
chlorine
reductions
to
achieve
the
beyond­
the­
floor
level
would
install
the
dry
scrubbing
system
after
the
existing
particulate
matter
control
device
and
add
a
new,
smaller
baghouse
to
remove
the
products
of
the
reaction
and
any
unreacted
lime.
We
chose
this
conservative
costing
approach
to
address
potential
concerns
that
unreacted
lime
and
collected
chloride
and
sulfur
salts
may
interfere
with
lightweight
aggregate
dust
use
practices.
The
national
incremental
annualized
compliance
cost
for
incinerators
lightweight
aggregate
kilns
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
1.9
million
and
would
provide
an
incremental
reduction
in
total
chlorine
emissions
beyond
the
MACT
floor
controls
of
275
tons
per
year.
Nonair280
tons
per
year,
for
a
cost­
effectiveness
of
$
6,800
per
additional
ton
of
total
chlorine
removed.
We
evaluated
nonair
quality
health
and
environmental
impacts
and
energy
effects
were
evaluated
to
estimate
the
nonair
quality
health
and
environmental
impacts
between
dry
scrubbing
and
controls
likely
to
be
used
to
meet
the
floor
level.
Weassociated
with
this
beyond­
the­
floor
standard
and
estimate
that
this
beyond­
the­
floor
option
would
increase
the
amount
of
solid
waste
generated
by
12,700
tons
per
year
and
would
also
require
sources
to
use
an
additional
175,000
kW­
hours
per
year
and
31
million
gallons
of
water
beyond
the
requirements
to
achieve
the
floor
level.

is
in
the
"
grey
area"
between
a
cost
the
Agency
has
concluded
is
cost­
effective
and
a
cost
the
Agency
has
concluded
is
not
cost­
effective
under
other
MACT
rules.
EPA
concluded
that
a
cost
of
$
1,100
per
ton
of
total
chlorine
removed
for
hazardous
waste
burning
lightweight
aggregate
kilns
was
cost­
effective
in
the
1999
MACT
final
rule.
See
68
FR
at
52900.
EPA
concluded,
however,
that
a
cost
of
$
45,000
per
ton
of
hydrogen
chloride
removed
was
not
cost­
effective
for
industrial
boilers.
See
68
FR
at
1677.
Consequently,
we
are
concerned
that
a
cost
of
$
6,800
per
additional
ton
of
total
chlorine
removed
is
not
warranted.

3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
Total
chlorine
emissions
from
new
lightweight
aggregate
kilns
are
currently
limited
to
600
ppmv
by
§
63.1205(
b)(
6).
This
standard
was
promulgated
in
the
Interim
Standards
Rule
(
See
67
FR
at
6797).
The
MACT
floor
for
new
sources
for
total
chlorine
would
be
0.93
lbs
chlorine
in
the
hazardous
waste
per
million
Btu
in
the
hazardous
waste,
which
considers
emissions
variability.
To
put
the
proposed
floor
level
in
context
for
a
hypothetical
lightweight
aggregate
kiln
that
gets
90%
of
its
required
heat
input
from
hazardous
waste,
a
thermal
emissions
level
of
0.93
lbs
total
chlorine
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste
equates
approximately
to
a
stack
gas
concentration
of
610
ppmv.
This
estimated
stack
gas
concentration
does
not
include
contributions
to
emission
from
other
chlorine­
containing
materials
such
as
raw
materials.
Given
that
the
calculated
floor
level
is
less
stringent
than
the
current
interim
emission
standard
of
600
ppmv.
In
order
to
avoid
any
backsliding
from
the
current
standard
for
a
new
lightweight
aggregate
kilns,
we
are
proposing
the
floor
standard
as
the
Redline­
strikeout
highlighting
changes
made
during
OMB
review
175
current
emission
standard
of
600
ppmv.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
Similar
to
existing
sources,
we
considered
a
beyond­
the­
floor
standard
of
150
ppmv
based
on
the
assumption
that
dry
lime
scrubbing
will
provide
75%
control
of
hydrogen
chloride.
The
incremental
annualized
compliance
cost
for
a
new
lightweight
aggregate
kiln
with
average
gas
flowrate
to
meet
this
beyond­
the­
floor
level,
rather
than
to
comply
with
the
floor
level,
would
be
approximately
$
0.42
million
and
would
provide
an
incremental
reduction
in
total
chlorine
emissions
of
approximately
118
pounds
per
year150
tons
per
year
for
a
cost­
effectiveness
of
approximately
$
2,800
per
additional
ton
of
total
chlorine
removed.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
were
evaluated
and
are
included
in
the
cost
estimates.
We
estimate
that
this
beyond­
the­
floor
option
would
increase
the
amount
of
solid
waste
generated
by
23
tons
per
year
and
would
also
require
sources
to
use
an
additional
0.3
million
kW­
hours
per
year
and
2
million
gallons
of
water
beyond
the
requirements
to
achieve
the
floor
level.

is
in
the
"
grey
area"
between
a
cost
the
Agency
has
concluded
is
cost­
effective
and
a
cost
the
Agency
has
concluded
is
not
cost­
effective
under
other
MACT
rules,
as
discussed
above.
Therefore,
we
are
concerned
that
a
cost­
effectiveness
of
$
2,800
per
additional
ton
of
total
chlorine
removed
may
not
be
warranted.

G.
What
Are
the
Standards
for
Hydrocarbons
and
Carbon
Monoxide?
Hydrocarbon
and
carbon
monoxide
standards
are
surrogates
to
control
emissions
of
organic
hazardous
air
pollutants
for
existing
and
new
lightweight
aggregate
kilns.
The
standards
limit
hydrocarbons
and
carbon
monoxide
concentrations
to
20
ppmv
or
100
ppmv.
See
§
§
63.1205(
a)(
5)
and
(
b)(
5).
Existing
and
new
lightweight
aggregate
kilns
can
elect
to
comply
with
either
the
hydrocarbon
limit
or
the
carbon
monoxide
limit
on
a
continuous
basis.
Sources
that
comply
with
the
carbon
monoxide
limit
on
a
continuous
basis
must
also
demonstrate
compliance
with
the
hydrocarbon
standard
during
the
comprehensive
performance
test.
However,
continuous
hydrocarbon
monitoring
following
the
performance
test
is
not
required.
The
rationale
for
these
decisions
are
discussed
in
the
September
1999
final
rule
(
64
FR
at
52900).
We
view
the
standards
for
hydrocarbons
and
carbon
monoxide
as
unaffected
by
the
Court's
vacature
of
the
challenged
regulations
in
its
decision
of
July
24,
2001.
We
therefore
are
not
proposing
these
standards
for
lightweight
aggregate
kilns,
but
rather
are
mentioning
them
here
for
the
reader's
convenience.
H.
What
Are
the
Standards
for
Destruction
and
Removal
Efficiency?
The
destruction
and
removal
efficiency
(
DRE)
standard
is
a
surrogate
to
control
emissions
of
organic
hazardous
air
pollutants
other
than
dioxin/
furans.
The
standard
for
existing
and
new
lightweight
aggregate
kilns
requires
99.99%
DRE
for
each
principal
organic
hazardous
constituent,
except
that
99.9999%
DRE
is
required
if
specified
dioxin­
listed
hazardous
wastes
are
Redline­
strikeout
highlighting
changes
made
during
OMB
review
176
burned.
See
§
§
63.1205(
c).
The
rationale
for
these
decisions
are
discussed
in
the
September
1999
final
rule
(
64
FR
at
52902).
We
view
the
standards
for
DRE
as
unaffected
by
the
Court's
vacature
of
the
challenged
regulations
in
its
decision
of
July
24,
2001.
We
therefore
are
not
proposing
these
standards
for
lightweight
aggregate
kilns,
but
rather
are
mentioning
them
here
for
the
reader's
convenience.

X.
How
Did
EPA
Determine
the
Proposed
Emission
Standards
for
Hazardous
Waste
Burning
Solid
Fuel­
Fired
Boilers?
The
proposed
standards
for
existing
and
new
solid
fuel­
fired
boilers
that
burn
hazardous
waste
are
summarized
in
the
table
below.
See
proposed
§
63.1216.

PROPOSED
STANDARDS
FOR
EXISTING
AND
NEW
SOLID
FUEL­
FIRED
BOILERS
Hazardous
Air
Pollutant
or
Surrogate
Emission
Standard1
Existing
Sources
New
Sources
Dioxin
and
furan
100
ppmv
carbon
monoxide
or
10
ppmv
hydrocarbons.
100
ppmv
carbon
monoxide
or
10
ppmv
hydrocarbons.

Mercury
10
ug/
dscm
10
ug/
dscm
Particulate
matter
69
mg/
dscm
(
0.030
gr/
dscf)
34
mg/
dscm
(
0.015
gr/
dscf)

Semivolatile
metals
170
ug/
dscm
170
ug/
dscm
Low
volatile
metals
210
ug/
dscm
190
ug/
dscm
Hydrogen
chloride
and
chlorine
gas2
73
ppmv
or
the
alternative
emission
limits
under
§
63.1215
Carbon
monoxide
or
hydrocarbons3
100
ppmv
carbon
monoxide
or
10
ppmv
hydrocarbons.
100
ppmv
carbon
monoxide
or
10
ppmv
hydrocarbons.

Destruction
and
Removal
Efficiency
For
existing
and
new
sources,
99.99%
for
each
principal
organic
hazardous
constituent
(
POHC).
For
sources
burning
hazardous
wastes
F020,
F021,
F022,
F023,
F026,
or
F027,
however,
99.9999%
for
each
POHC.

1
All
emission
standards
are
corrected
to
7%
oxygen,
dry
basis.
2
Combined
standard,
reported
as
a
chloride
(
Cl(­))
equivalent.
3
Hourly
rolling
average.
Hydrocarbons
reported
as
propane.

We
considered
whether
fuel
switching
could
be
considered
a
control
technology
to
Redline­
strikeout
highlighting
changes
made
during
OMB
review
32
C.
Leatherwood,
ERG,
to
J.
Eddinger,
OAQPS,
EPA,
Memorandum:
Development
of
Fuel
Switching
Costs
and
Emission
Reductions
for
Industrial/
Commercial/
Institutional
Boilers
and
Process
Heaters
National
Emission
Standards
for
Hazardous
Air
Pollutants,
October
2002.

177
achieve
MACT
floor
control.
We
investigated
whether
fuel
switching
would
achieve
lower
HAP
emissions
and
whether
it
could
be
technically
achieved
considering
the
existing
design
of
solid
fuel­
fired
boilers.
We
also
considered
the
availability
of
various
types
of
fuel.
After
considering
these
factors,
we
determined
that
fuel
switching
is
not
an
appropriate
control
technology
for
purposes
of
determining
the
MACT
floor
level
of
control.
This
decision
is
based
on
the
overall
effect
of
fuel
switching
on
HAP
emissions,
technical
and
design
considerations,
and
concerns
about
fuel
availability.
We
determined
that
while
fuel
switching
from
coal
to
natural
gas
or
oil
would
decrease
particulate
matter
and
some
metal
HAP
emissions,
emissions
of
some
organic
HAP
would
increase,
resulting
in
uncertain
benefits.
32
We
believe
that
it
is
inappropriate
in
a
MACT
rulemaking
to
consider
as
MACT
a
control
option
that
potentially
will
decrease
emissions
of
one
HAP
while
increasing
emissions
of
another
HAP.
In
order
to
adopt
such
a
strategy,
we
would
need
to
assess
the
relative
risk
associated
with
each
HAP
emitted,
and
determine
whether
requiring
the
control
in
question
would
result
in
overall
lower
risk.
Such
an
analysis
is
not
appropriate
at
this
stage
in
the
regulatory
process.
For
example,
the
term
  
clean
coal''
refers
to
coal
that
is
lower
in
sulfur
content
and
not
necessarily
lower
in
HAP
content.
Data
gathered
by
EPA
also
indicates
that
within
specific
coal
types
HAP
content
can
vary
significantly.
Switching
to
a
low
sulfur
coal
may
actually
increase
emissions
of
some
HAP.
Therefore,
it
is
not
appropriate
for
EPA
to
include
fuel
switching
to
a
low
sulfur
coal
as
part
of
the
MACT
standards
for
boilers
that
burn
hazardous
waste.
We
also
considered
the
availability
of
alternative
fuel
types.
Natural
gas
pipelines
are
not
available
in
all
regions
of
the
U.
S.,
and
natural
gas
is
simply
not
available
as
a
fuel
for
many
solid
fuel­
fired
boilers.
Moreover,
even
where
pipelines
provide
access
to
natural
gas,
supplies
of
natural
gas
may
not
be
adequate.
For
example,
it
is
common
practice
in
cities
during
winter
months
(
or
periods
of
peak
demand)
to
prioritize
natural
gas
usage
for
residential
areas
before
industrial
usage.
Requiring
EPA
regulated
combustion
units
to
switch
to
natural
gas
would
place
an
even
greater
strain
on
natural
gas
resources.
Consequently,
even
where
pipelines
exist,
some
units
would
not
be
able
to
run
at
normal
or
full
capacity
during
these
times
if
shortages
were
to
occur.
Therefore,
under
any
circumstances,
there
would
be
some
units
that
could
not
comply
with
a
requirement
to
switch
to
natural
gas.
In
addition,
we
have
significant
concern
that
switching
fuels
would
be
infeasible
for
sources
designed
and
operated
to
burn
specific
fuel
types.
Changes
in
the
type
of
fuel
burned
by
a
boiler
may
require
extensive
changes
to
the
fuel
handling
and
feeding
system
(
e.
g.,
a
stoker­
fired
boiler
using
coal
as
primary
fuel
would
need
to
be
redesigned
to
handle
fuel
oil
or
gaseous
fuel
as
the
primary
fuel).
Additionally,
burners
and
combustion
chamber
designs
are
generally
not
Redline­
strikeout
highlighting
changes
made
during
OMB
review
33
Uncontrolled
hydrogen
chloride
in
combustion
gas
was
approximately
700
ppmv.

34
USEPA,
"
Draft
Technical
Support
Document
for
HWC
MACT
Replacement
Standards,
Volume
III:
Selection
of
MACT
Standards
and
Technologies,"
March
2004,
Chapter
2.

178
capable
of
handling
different
fuel
types,
and
generally
cannot
accommodate
increases
or
decreases
in
the
fuel
volume
and
shape.
Design
changes
to
allow
different
fuel
use,
in
some
cases,
may
reduce
the
capacity
and
efficiency
of
the
boiler.
Reduced
efficiency
may
result
in
less
complete
combustion
and,
thus,
an
increase
in
organic
HAP
emissions.
For
the
reasons
discussed
above,
we
conclude
that
fuel
switching
to
cleaner
solid
fuels
or
to
liquid
or
gaseous
fuels
is
not
an
appropriate
criteria
for
identifying
the
MACT
floor
level
of
control
for
solid
fuel­
fired
boilers.
A.
What
Is
the
Rationale
for
the
Proposed
Standards
for
Dioxin
and
Furan?
The
proposed
standard
for
dioxin/
furan
for
existing
and
new
sources
is
compliance
with
the
proposed
carbon
monoxide
or
hydrocarbon
(
CO/
HC)
emission
standard
and
compliance
with
the
proposed
destruction
and
removal
efficiency
(
DRE)
standard.
The
CO/
HC
and
DRE
standards
control
emissions
of
organic
HAPs
in
general,
and
are
discussed
in
Sections
G
and
H
below.
This
standard
ensures
that
boilers
operate
under
good
combustion
practices
as
a
surrogate
for
dioxin/
furan
control.
Operating
under
good
combustion
practices
minimizes
levels
of
products
of
incomplete
combustion,
including
potentially
dioxin/
furan,
and
organic
compounds
that
could
be
precursors
for
post­
combustion
formation
of
dioxin/
furan.
The
rationale
for
the
dioxin/
furan
standard
is
discussed
below.
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
The
proposed
MACT
floor
control
for
existing
sources
is
compliance
with
the
proposed
CO/
HC
emission
standard
and
compliance
with
the
proposed
DRE
standard.
Solid
fuel­
fired
boilers
that
burn
hazardous
waste
cofire
the
hazardous
waste
with
coal
at
firing
rates
of
6­
33%
of
total
heat
input.
We
have
dioxin/
furan
emission
data
for
one
source,
and
those
emissions
are
0.07
ng
TEQ/
dscm.
Although
dioxin/
furan
can
be
formed
post­
combustion
in
an
electrostatic
precipitator
or
baghouse
that
is
operated
at
temperatures
within
the
range
of
400

to
750

F,
the
boiler
for
which
we
have
dioxin/
furan
emissions
data
is
equipped
with
an
electrostatic
precipitator
that
operated
at
500

F
during
the
emissions
test.
Although
this
is
well
within
the
optimum
temperature
range
for
formation
of
dioxin/
furan,
dioxin/
furan
emissions
were
low.
In
addition,
this
boiler
fed
chlorine
at
levels
four
times
greater
than
any
other
solid
fuel
boiler.
33
We
also
have
emissions
data
from
16
nonhazardous
waste
coal­
fired
boilers
equipped
with
electrostatic
precipitators
and
baghouses
operated
at
temperatures
up
to
480

F,
all
of
which
have
dioxin/
furan
emissions
below
0.3
ng
TEQ/
dscm.
34
We
conclude
from
these
data
and
the
information
discussed
below
that
rapid
quench
of
post­
combustion
gas
temperatures
to
below
400

F­­
the
control
technique
that
is
the
basis
for
the
MACT
standards
for
hazardous
waste
burning
incinerators,
and
cement
and
lightweight
aggregate
kilns­­
is
not
the
dominant
dioxin/
furan
control
mechanism
for
coal­
fired
boilers.
Redline­
strikeout
highlighting
changes
made
during
OMB
review
35
Section
266.104
requires
compliance
with
a
CO
limit
of
100
ppmv
or
a
HC
limit
of
20
ppmv,
while
we
are
proposing
today
a
CO
limit
of
100
ppmv
or
a
HC
limit
of
10
ppmv
(
see
Section
X.
H
in
the
text).
Although
today's
proposed
HC
limit
is
more
stringent
than
the
current
limit
for
boilers,
all
solid
fuel
boilers
chose
to
comply
with
the
100
ppmv
CO
limit.
Moreover,
for
179
We
believe
that
sulfur
contributed
by
the
coal
fuel
is
a
dominant
control
mechanism
by
inhibiting
formation
of
dioxin/
furan.
Coal
generally
contributes
from
65%
to
95%
percent
of
the
boiler's
heat
input
with
the
remainder
provided
by
hazardous
waste
fuel.
The
presence
of
sulfur
in
combustor
feedstocks
has
been
shown
to
dramatically
inhibit
the
catalytic
formation
of
dioxin/
furan
in
downstream
temperature
zones
from
400

F
to
750

F.
High
sulfur
coals
tend
to
inhibit
dioxin/
furan
formation
better
than
low
sulfur
coals.
Id.
Adsorption
of
any
dioxin/
furan
that
may
be
formed
on
coal
fly
ash,
and
subsequent
capture
in
the
electrostatic
precipitator
or
baghouse,
also
may
contribute
to
the
low
dioxin/
furan
emissions
despite
some
boilers
operating
at
relatively
high
back­
end
gas
temperatures.
This
effect
is
similar
to
that
of
using
activated
carbon
injection
to
control
dioxin/
furan
emissions.
Adsorption
of
dioxin/
furan
on
fly
ash
is
related
to
the
carbon
content
of
the
fly
ash,
and,
thus,
the
type
of
coal
burned.
Id.
Operating
under
good
combustion
conditions
to
minimize
emissions
of
organic
compounds
such
as
polychlorinated
biphenols,
benzene,
and
phenol
that
can
be
precursors
to
dioxin/
furan
formation
is
an
important
requisite
to
control
dioxin/
furan
emissions.
Although
sulfur­
induced
inhibition
may
be
the
dominant
mechanism
to
control
dioxin/
furan
emissions
from
coal­
fired
boilers,
minimizing
dioxin/
furan
precursors
by
operating
under
good
combustion
practices
certainly
plays
a
part
in
controlling
dioxin/
furan
emissions.
We
propose
to
use
the
CO/
HC
and
DRE
standards
as
surrogates
to
ensure
that
boilers
operate
under
good
combustion
conditions
because
quantified
levels
of
control
provided
by
sulfur
in
the
coal
and
adsorption
onto
collected
fly
ash
may
not
be
replicable
by
the
best
performing
sources
nor
duplicable
by
other
sources.
Although
coal
sulfur
content
may
be
a
dominant
factor
affecting
dioxin/
furan
emissions,
we
do
not
know
what
minimum
level
of
sulfur
provides
significant
control.
Moreover,
sulfur
in
coal
causes
emissions
of
sulfur
oxides,
a
major
criteria
pollutant,
and
particulate
sulfates.
Similarly,
we
cannot
quantify
a
minimum
carbon
content
of
coal
that
would
form
carbonaceous
fly
ash
with
superior
dioxin/
furan
adsorptive
properties.
In
addition,
restricting
coal
types
that
may
be
burned
based
on
carbon
content
may
have
an
adverse
impact
on
energy
production
at
sources
burning
hazardous
waste
as
fuel.
(
These
considerations
raise
the
question
of
whether
boilers
operating
under
these
conditions
would
still
be
"
best"
performers
when
these
adverse
impacts
are
taken
into
account.)
For
these
reasons,
and
because
we
have
emissions
data
from
only
one
source,
we
cannot
establish
a
numerical
dioxin/
furan
emission
standard.
Operating
under
good
combustion
practices
is
floor
control
because
all
hazardous
waste
burning
boilers
are
required
by
existing
RCRA
regulations
to
operate
under
good
combustion
conditions
to
minimize
emissions
of
toxic
organic
compounds.
See
§
266.104
requiring
compliance
with
DRE
and
CO/
HC
emission
standards.
35
We
also
find,
as
required
by
CAA
Redline­
strikeout
highlighting
changes
made
during
OMB
review
those
liquid­
fuel
fired
boilers
that
chose
to
comply
with
the
20
ppmv
HC
limit,
their
HC
emissions
are
below
10
ppmv.

36
We
considered
a
beyond­
the­
floor
standard
of
0.20
ng
TEQ/
dscm
but
determined
that
it
may
not
result
in
emissions
reductions
because
the
majority
of
sources
(
the
hazardous
waste
coal­
fired
boiler
and
the
nonhazardous
waste
coal­
fired
boilers)
appear
to
emit
dioxin/
furan
at
levels
below
0.20
ng
TEQ/
dscm.

180
Section
112(
h)(
1),
that
these
proposed
standards
are
consistent
with
Section
112(
d)'
s
objective
of
reducing
emissions
of
these
HAPs
to
the
extent
achievable.
We
request
comment
on
an
alternative
floor
that
would
be
established
as
the
highest
dioxin/
furan
emission
level
in
our
data
base.
Because
we
have
dioxin/
furan
emission
data
from
only
one
coal­
fired
boiler
that
burns
hazardous
waste,
we
would
combine
that
data
point
with
emissions
data
from
coal­
fired
boilers
that
do
not
burn
hazardous
waste
since
the
factors
that
affect
dioxin/
furan
emissions
from
these
boilers
are
not
significantly
influenced
by
hazardous
waste.
These
additional
data
would
better
represent
the
range
of
emissions
from
coal­
fired
boilers.
Under
this
approach,
the
dioxin/
furan
floor
would
be
an
emission
level
of
0.30
ng
TEQ/
dscm.
We
would
also
use
this
approach
to
establish
the
same
floor
for
new
sources.
Finally,
we
note
that
we
propose
to
require
a
one­
time
dioxin/
furan
emission
test
for
sources
that
would
not
be
subject
to
a
numerical
dioxin/
furan
emission
standard,
such
as
solid
fuel­
fired
boilers.
As
discussed
in
Part
Two,
Section
XIV.
B
below,
the
testing
would
assist
in
developing
both
Section
112(
d)(
6)
standards
and
Section
112(
f)
residual
risk
standards.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
As
discussed
above,
we
propose
to
use
the
CO/
HC
and
DRE
standards
as
surrogates
to
ensure
good
combustion
conditions,
and
thus,
control
of
dioxin/
furan
emissions.
We
are
not
proposing
beyond­
the­
floor
standards
for
CO/
HC
and
DRE,
as
discussion
in
Sections
G
and
H
below..
We
investigated
use
of
activated
carbon
injection
or,
for
sources
equipped
with
baghouses,
catalytically
impregnated
fabric
felt/
membrane
filter
materials
to
achieve
a
beyond­
thefloor
standard
of
0.10
ng
TEQ/
dscm.
36
To
estimate
the
cost­
effectiveness
of
these
beyond­
thefloor
control
techniques,
we
imputed
dioxin/
furan
emissions
levels
for
the
six
sources
for
which
we
don't
have
measured
emissions
data.
To
impute
the
missing
emissions
levels,
we
used
the
emissions
data
from
the
hazardous
waste
burning
boiler
as
well
as
the
emissions
data
from
nonhazardous
waste
coal­
fired
boilers.
It
may
be
appropriate
to
meld
these
emissions
data
because
hazardous
waste
burning
should
not
affect
dioxin/
furan
emissions
from
coal­
fired
boilers.
In
fact,
the
nonhazardous
waste
coal­
fired
boilers
had
somewhat
higher
emissions
than
the
hazardous
waste
coal­
fired
boiler.
(
The
emissions
from
the
nonhazardous
waste
coal­
fired
boilers
may
simply
represent
the
range
of
emissions
that
could
be
expected
from
hazardous
waste
coalfired
boilers,
as
well,
given
that
we
have
emissions
data
from
only
one
hazardous
waste
boiler.)
The
national
incremental
annualized
compliance
cost
for
solid
fuel­
fired
boilers
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
Redline­
strikeout
highlighting
changes
made
during
OMB
review
37
To
estimate
the
cost
of
a
beyond­
the­
floor
standard
conservatively,
we
assumed
the
solid
waste
generated
would
be
subject
to
regulation
as
hazardous
waste.
These
costs
are
likely
over­
estimated,
however,
because
these
residues
are
not
likely
to
fail
the
criteria
for
retaining
the
Bevill
exclusion
under
40
CFR
266.112.

38
We
note
that
we
propose
to
require
solid
fuel­
fired
boilers
(
and
liquid
fuel­
fired
boilers
that
are
not
subject
to
a
numerical
dioxin/
furan
standard)
to
conduct
a
one­
time
dioxin/
furan
emission
test
to
provide
data
to
assist
in
developing
both
Section
112(
d)(
6)
standards
and
Section
112(
f)
residual
risk
standards.
See
discussion
in
Section
XIV.
B
of
the
preamble.

181
$
3.4
million
and
would
provide
an
incremental
reduction
in
dioxin/
furan
emissions
beyond
the
MACT
floor
controls
of
0.26
grams
TEQ
tons
per
year.
We
also
evaluated
the
nonair
quality
health
and
environmental
impacts
and
energy
effects
between
activated
carbon
injection
and
controls
likely
to
be
used
to
meet
the
floor
level.
We
estimate
that
this
beyond­
the­
floor
option
would
increase
the
amount
of
hazardous
waste37
generated
by
3,300
tons
per
year
and
would
also
require
sources
to
use
an
additional
1.2
million
kW­
hours
per
year.
Based
on
these
impacts
and
costs
of
approximately
$
13
million
per
additional
grams
of
dioxin/
furan
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
activated
carbon
injection.
For
these
reasons,
we
propose
a
floor
standard
for
dioxin/
furan
for
existing
sources
of
compliance
with
the
proposed
CO/
HC
emission
standard
and
compliance
with
the
proposed
DRE
standard.
38
3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
As
discussed
above,
we
propose
to
use
the
CO/
HC
and
DRE
standards
as
surrogates
to
ensure
good
combustion
conditions,
and
thus,
control
of
dioxin/
furan
emissions.
Because
we
are
proposing
the
same
DRE
and
CO/
HC
standards
for
existing
sources
and
new
sources
as
discussion
in
Sections
G
and
H
below,
we
are
proposing
the
same
dioxin/
furan
floor
for
new
and
existing
sources.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
We
are
not
proposing
beyond­
the­
floor
standards
for
CO/
HC
for
dioxin/
furan
for
new
solid
fuel­
fired
boilers
because
we
are
not
proposing
beyond­
the­
floor
standards
for
CO/
HC
and
DRE
for
new
sources.
See
discussion
in
Sections
G
and
H
below.
In
addition,
we
evaluated
activated
carbon
injection
or,
for
sources
equipped
with
baghouses,
use
of
catalytically
impregnated
fabric
felt/
membrane
filter
materials
as
beyond­
thefloor
control
for
further
reduction
of
dioxin/
furan
emissions
to
achieve
a
beyond­
the­
floor
level
of
0.15
ng
TEQ/
dscm.
The
incremental
annualized
compliance
cost
for
a
new
solid
fuel­
fired
boiler
with
average
gas
flowrate
to
meet
this
beyond­
the­
floor
level,
rather
than
comply
with
the
floor
level,
would
be
approximately
$
0.28
million
and
would
provide
an
incremental
reduction
in
dioxin/
furan
emissions
of
approximately
0.21
grams
TEQ
per
year,
for
a
cost­
effectiveness
of
$
1.3
million
per
gram
of
dioxin/
furan
removed.
We
estimate
that
this
beyond­
the­
floor
option
would
increase
the
amount
of
hazardous
waste
(
or
solid
waste
if
the
source
retains
the
Bevill
exclusion
under
40
CFR
266.112)
generated
for
a
new
solid
fuel­
fired
boiler
with
average
gas
flowrate
by
Redline­
strikeout
highlighting
changes
made
during
OMB
review
39
As
information,
EPA
proposed
MACT
standards
for
mercury
for
solid
fuel­
fired
industrial,
commercial,
and
institutional
boilers
that
do
not
burn
hazardous
waste
of
5.3
ug/
dscm
for
existing
sources
and
3.4
ug/
dscm
for
new
sources.
See
68
FR
1660
(
Jan.
13,
2003).
These
standards
are
based
on
use
of
fabric
filters
to
control
mercury
emissions.

40
Owners
and
operators
have
used
the
emissions
data
from
the
three
boilers
as
"
data
in
lieu
of
testing"
emissions
from
other,
identical
boilers
at
the
same
facility.
One
of
the
three
boilers
as
two
such
sister
identical
boilers,
and
the
other
two
boilers
each
have
a
sister
identical
boiler.
Thus,
emissions
from
these
three
boilers
represent
emissions
from
seven
of
the
12
solid
fuel­
fired
boilers.

41
Memo
from
Frank
Princiotta,
USEPA,
to
John
Seitz,
USEPA,
entitled
"
Control
of
Mercury
Emissions
from
Coal­
fired
Utility
Boilers,"
dated
October
25,
2000.

182
270
tons
per
year
and
would
require
a
source
to
use
an
additional
0.1
million
kW­
hours
per
year
beyond
the
requirements
to
achieve
the
floor
level.
After
considering
these
impacts
and
a
cost
of
$
1.3
million
per
gram
of
dioxin/
furan
removed,
we
conclude
that
a
beyond­
the­
floor
standard
based
on
activated
carbon
injection
or
catalytically
impregnated
fabric
felt/
membrane
filter
is
not
warranted
for
new
sources.
Consequently,
we
propose
a
floor
standard
for
dioxin/
furan
for
new
sources:
compliance
with
the
proposed
CO/
HC
and
DRE
emissions
standards.
B.
What
Is
the
Rationale
for
the
Proposed
Standards
for
Mercury?
The
proposed
standard
for
mercury
for
solid
fuel­
fired
boilers
is
10
ug/
dscm
for
both
existing
sources
and
new
sources.
39
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
The
MACT
floor
for
existing
sources
is
10
ug/
dscm
based
on
adsorption
of
mercury
onto
coal
fly
ash
and
removal
of
fly
ash
by
the
electrostatic
precipitator
or
baghouse.
All
solid
fuel­
fired
boilers
are
equipped
with
electrostatic
precipitators
or
baghouses.
We
have
compliance
test
emissions
data
for
three
sources
equipped
with
electrostatic
precipitators
which
document
maximum
mercury
emissions
ranging
from
3
ug/
dscm
to
11
ug/
dscm
and
system
removal
efficiencies
of
83%
to
96%.
These
three
sources
represent
seven
of
the
12
solid
fuelfired
boilers.
40
The
Agency
has
also
determined
that
coal­
fired
utility
boilers
can
achieve
significant
control
of
mercury
by
adsorption
on
fly
ash
and
particulate
matter
control.
41
To
identify
the
MACT
floor,
we
evaluated
the
compliance
test
emissions
data
using
the
SRE/
Feed
Approach.
The
calculated
floor
is
10
ug/
dscm,
which
considers
emissions
variability.
This
is
an
emission
level
that
the
average
of
the
best
performing
sources
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
operating
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
We
estimate
that
this
emission
level
is
being
achieved
by
67%
of
sources
and
would
provide
a
reduction
in
mercury
emissions
of
0.015
tons
per
year.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
Redline­
strikeout
highlighting
changes
made
during
OMB
review
183
We
identified
two
potential
beyond­
the­
floor
techniques
for
control
of
mercury:
(
1)
activated
carbon
injection;
and
(
2)
control
of
mercury
in
the
hazardous
waste
feed.
For
reasons
discussed
below,
we
are
not
proposing
a
beyond­
the­
floor
standard
for
mercury.
a.
Use
of
Activated
Carbon
Injection.
We
evaluated
activated
carbon
injection
as
beyond­
the­
floor
control
for
further
reduction
of
mercury
emissions.
Activated
carbon
has
been
demonstrated
for
controlling
mercury
from
waste
combustion
systems
and
has
achieved
efficiencies
ranging
from
80%
to
greater
than
90%
depending
on
factors
such
as:
activated
carbon
type/
impregnation;
injection
rate;
mercury
speciation
in
the
flue
gas;
and
flue
gas
temperature.
We
made
a
conservative
assumption
that
the
use
of
activated
carbon
will
provide
70%
mercury
control
for
coal­
fired
boilers
given
the
low
mercury
levels
at
the
floor.
Applying
this
activated
carbon
removal
efficiency
to
the
mercury
floor
level
of
10
ug/
dscm
would
provide
a
beyond­
the­
floor
level
of
3.0
ug/
dscm.
The
national
incremental
annualized
compliance
cost
for
solid
fuel
boilers
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
1.1
million
and
would
provide
an
incremental
reduction
in
mercury
emissions
beyond
the
MACT
floor
controls
of
0.03
tons
per
year.
We
evaluated
nonair
quality
health
and
environmental
impacts
and
energy
effects
and
estimate
that
this
beyond­
the­
floor
option
would
increase
the
amount
of
hazardous
waste
(
or
solid
waste
if
the
source
retains
the
Bevill
exclusion
under
40
CFR
266.112)
generated
by
1,000
tons
per
year
and
would
require
sources
to
use
an
additional
0.35
million
kWhours
per
year
beyond
the
requirements
to
achieve
the
floor
level.
Based
on
these
factors
and
costs
of
approximately
$
35
million
per
additional
ton
of
mercury
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
activated
carbon
injection.
b.
Feed
Control
of
Mercury
in
the
Hazardous
Waste.
We
also
evaluated
a
beyond­
thefloor
level
of
8
ug/
dscm,
which
represents
a
20%
reduction
from
the
floor
level.
The
national
incremental
annualized
compliance
cost
for
solid
fuel
boilers
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
0.11
million
and
would
provide
an
incremental
reduction
in
mercury
emissions
beyond
the
MACT
floor
controls
of
0.005
tons
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
are
not
significant
factors
for
feedrate
control.
We
are
not
proposing
a
beyond­
the­
floor
standard
based
on
feed
control
of
mercury
in
the
hazardous
waste
because
it
would
not
be
cost­
effective
at
approximately
$
23
million
per
additional
ton
of
mercury
removed.
Consequently,
we
propose
a
floor
standard
for
mercury
for
existing
sources
of
10
ug/
dscm.
3.
What
Is
the
Rationale
for
MACT
Floor
for
New
Sources?
MACT
floor
for
new
sources
would
be
10
ug/
dscm,
the
same
as
the
floor
for
existing
sources.
This
is
an
emission
level
that
the
single
best
performing
source
identified
by
the
SRE/
Feed
Approach
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
operating
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
We
identified
the
same
two
potential
beyond­
the­
floor
techniques
for
control
of
mercury:
(
1)
use
of
activated
carbon
injection;
and
(
2)
control
of
mercury
in
the
hazardous
waste
feed.
Redline­
strikeout
highlighting
changes
made
during
OMB
review
42
As
information,
EPA
proposed
MACT
standards
for
particulate
matter
for
solid
fuel­
fired
industrial,
commercial,
and
institutional
boilers
that
do
not
burn
hazardous
waste
of
0.035
gr/
dscf
for
existing
sources
and
0.013
gr/
dscf
for
new
sources.
See
68
FR
1660
(
Jan.
13,
2003).
These
standards
are
based
on
control
of
particulate
matter
emissions
using
a
fabric
filter.

43
Owners
and
operators
have
determined
that
emissions
from
these
seven
boilers
represent
emissions
from
five
other
identical,
sister
boilers.
Owners
and
operators
have
used
the
emissions
from
these
seven
boilers
as
"
data
in
lieu
of
testing"
emissions
from
the
other
five
identical
boilers.

44
Although
particulate
matter
emissions
are
predominantly
attributable
to
coal
ash
rather
than
ash
from
hazardous
waste
fuel,
we
did
not
combine
emissions
data
for
coal­
fired
boilers
that
do
not
burn
hazardous
waste
with
the
data
for
boilers
that
burn
hazardous
waste
because:
(
1)
we
have
particulate
emissions
data
for
all
boilers
that
burn
hazardous
waste;
and
(
2)
the
best
performing
sources
for
the
two
categories
of
boilers
may
be
achieving
different
emission
levels
given
that
they
are
subject
to
different
emission
standards
(
e.
g.,
hazardous
waste
burning
184
We
evaluated
use
of
carbon
injection
for
new
sources
to
achieve
a
beyond­
the­
floor
emission
level
of
5.0
ug/
dscm.
The
incremental
annualized
compliance
cost
for
a
new
solid
fuel
boiler
with
average
gas
flowrate
to
meet
this
beyond­
the­
floor
level,
rather
than
comply
with
the
floor
level,
would
be
approximately
$
0.28
million
and
would
provide
an
incremental
reduction
in
mercury
emissions
of
approximately
0.008
tons
per
year,
for
a
cost­
effectiveness
of
$
37
million
per
ton
of
mercury
removed.
We
estimate
that
this
beyond­
the­
floor
option
would
increase
the
amount
of
hazardous
waste
(
or
solid
waste
if
the
source
retains
the
Bevill
exclusion
under
40
CFR
266.112)
generated
for
a
new
solid
fuel­
fired
boiler
with
average
gas
flowrate
by
270
tons
per
year
and
would
require
a
source
to
use
an
additional
0.1
million
kW­
hours
per
year
beyond
the
requirements
to
achieve
the
floor
level.
After
considering
these
impacts
and,
primarily,
costeffectiveness
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
activated
carbon
injection
for
new
sources.
Consequently,
we
propose
a
floor
standard
for
mercury
of
10
ug/
dscm
for
new
sources.
C.
What
Is
the
Rationale
for
the
Proposed
Standards
for
Particulate
Matter?
The
proposed
standards
for
particulate
matter
for
solid
fuel­
fired
boilers
are
69
mg/
dscm
(
0.030
gr/
dscf)
for
existing
sources
and
34
mg/
dscm
(
0.015
gr/
dscf)
for
new
sources.
42
The
particulate
matter
standard
serves
as
a
surrogate
for
nonmercury
HAP
metals
in
emissions
from
the
coal
burned
in
the
boiler,
and
for
nonenumerated
HAP
metal
emissions
attributable
to
the
hazardous
waste
fuel
burned
in
the
boiler.
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
All
solid
fuel­
fired
boilers
are
equipped
with
electrostatic
precipitators
or
baghouses.
We
have
compliance
test
emissions
data
representing
maximum
allowable
emissions
for
seven
boilers.
Emissions
from
these
seven
boilers
represent
emissions
from
all
12
solid
fuel­
fired
boilers.
43
Particulate
emissions
range
from
0.021
gr/
dscf
to
0.037
gr/
dscf.
44
Redline­
strikeout
highlighting
changes
made
during
OMB
review
coal­
fired
boilers
are
subject
to
the
RCRA
particulate
matter
emission
standard
of
0.08
gr/
dscf).

45
Note
that
we
are
not
proposing
beyond­
the­
floor
particulate
matter
standards
for
incinerators,
cement
kilns,
lightweight
aggregate
kilns,
and
liquid
fuel­
fired
boilers
because
those
standards
would
have
a
cost­
effectiveness
of
$
12,000
to
$
80,000
per
ton
of
particulate
matter
removed,
substantially
higher
than
the
$
3,200
per
ton
cost­
effectiveness
of
a
beyond­
the­
floor
standard
for
solid
fuel­
fired
boilers.

185
To
identify
the
floor
level,
we
evaluated
the
compliance
test
emissions
data
associated
with
the
most
recent
test
campaign
using
the
air
pollution
control
device
approach.
See
discussion
in
Part
Two,
Section
VI.
A.
2.
a.
The
calculated
floor
is
140
mg/
dscm
(
0.063
gr/
dscf),
which
considers
emissions
variability.
This
is
an
emission
level
that
the
average
of
the
best
performing
sources
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
We
estimate
that
this
emission
level
is
being
achieved
by
75%
of
sources.
Compliance
with
the
floor
level
would
reduce
particulate
matter
emissions
by
33
tons
per
year.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
evaluated
improved
design,
operation,
and
maintenance
of
the
existing
electrostatic
precipitators
(
e.
g.,
humidification
to
improve
gas
conditioning)
and
baghouses
(
e.
g.,
improved
bags)
for
these
boilers
to
achieve
a
beyond­
the­
floor
emission
level
of
69
mg/
dscm
(
0.030
gr/
dscf).
We
also
evaluated
a
more
stringent
standard
based
on
adding
a
polishing
fabric
filter
to
achieve
a
beyond­
the­
floor
emission
level
of
0.015
gr/
dscf.
The
national
incremental
annualized
compliance
cost
for
solid
fuel
boilers
to
meet
a
beyond­
the­
floor
level
of
69
mg/
dscm
rather
than
comply
with
the
floor
controls
would
be
approximately
$
1.53
million
and
would
provide
an
incremental
reduction
in
particulate
matter
emissions
beyond
the
MACT
floor
controls
of
400
tons
per
year
and
an
incremental
reduction
in
metal
HAP
of
6.8
tons
per
year.
We
evaluated
nonair
quality
health
and
environmental
impacts
and
energy
effects
and
estimate
that
this
beyondthe
floor
option
would
increase
the
amount
of
hazardous
waste
(
or
solid
waste
is
the
source
retains
its
Bevill
exclusion
under
40
CFR
266.112)
generated
by
380
tons
per
year
and
would
require
sources
to
use
an
additional
3.3
million
kW­
hours
per
year
and
to
use
an
additional
160
million
gallons
of
water
beyond
the
requirements
to
achieve
the
floor
level.
Notwithstanding
these
nonair
quality
health
and
environmental
impacts
and
energy
effects,
a
beyond­
the­
floor
standard
of
69
mg/
dscm
(
0.030
gr/
dscf)
based
on
improved
particulate
matter
control
is
warranted
because
it
is
cost­
effective
at
a
cost
of
approximately
$
3,600200
per
additional
ton
of
particulate
matter
removed
and
a
cost
of
approximately
$
190,000
per
additional
ton
of
metal
HAP
removed.
45
In
addition,
the
average
incremental
annualized
cost
would
be
only
$
120,000
per
facility,
and
it
would
remove
additional
nonenumerated
metal
HAP.
131
We
also
note
that,
although
Section
112(
d)
only
authorizes
control
of
HAPs,
and
particulate
matter
is
not
itself
a
HAP
but
a
surrogate
for
HAP
metals,
Congress
expected
the
MACT
program
to
result
in
significant
emissions
reductions
of
criteria
air
pollutants
(
of
which
particulate
matter
is
one),
and
viewed
this
as
an
important
benefit
of
the
MACT
(
and
residual
risk)
provisions.
See
5
Legislative
Redline­
strikeout
highlighting
changes
made
during
OMB
review
186
History
at
8512
(
Senate
Committee
Report).
Finally,
we
note
that
this
beyond­
the­
floor
standard
of
0.030
gr/
dscf
would
be
comparable
to
the
floor­
based
standard
the
Agency
recently
proposedpromulgated
for
solid
fuel­
fired
boilers
that
do
not
burn
hazardous
waste:
0.07
lb/
MM
Btu
(
approximately
0.034
gr/
dscf).
See
68
FR
at
1660
(
January
13,
2003)
NESHAP
for
Industrial/
Commercial/
Institutional
Boilers
and
Process
Heaters,
signed
Feb.
26,
2004.
Because
hazardous
waste
does
not
contribute
substantially
to
particulate
matter
emissions
from
coal­
fired
boilers,
MACT
standards
for
solid
fuel
boilers
should
be
similar
irrespective
of
whether
they
burn
hazardous
waste.
A
34
mg/
dscm
beyond­
the­
floor
standard
for
existing
sources
based
on
use
of
a
polishing
fabric
filter
would
remove
an
additional
570
tons
per
year
of
particulate
matter
beyond
the
floor
level
at
a
cost­
effectiveness
of
$
9,800
per
ton
removed.
We
conclude
that
this
standard
would
not
be
as
cost­
effective
as
a
69
mg/
dscm
standard
and
would
result
in
greater
nonair
quality
health
and
environmental
impacts
and
energy
effects.
For
these
reasons,
we
propose
a
beyond­
the­
floor
particulate
matter
standard
of
0.030
gr/
dscf
(
69
mg/
dscm)
for
existing
sources.

3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
MACT
floor
for
new
sources
would
be
90
mg/
dscm
(
0.040
gr/
dscf),
considering
emissions
variability.
This
is
an
emission
level
that
the
single
best
performing
source
identified
by
the
APCD
Approach
(
i.
e.,
the
source
using
a
fabric
filter
with
the
lowest
emissions)
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
operating
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
We
evaluated
use
of
a
fabric
filter
to
achieve
a
beyond­
the­
floor
emission
level
of
34
mg/
dscm
(
0.015
gr/
dscf).
The
incremental
annualized
cost
for
a
new
solid
fuel­
fired
boiler
with
average
gas
flowrate
to
meet
this
beyond­
the­
floor
level,
rather
than
comply
with
the
floor
level,
would
be
approximately
$
280,000
and
would
provide
an
incremental
reduction
in
particulate
emissions
of
approximately
44
tons
per
year,
for
a
cost­
effectiveness
of
$
6,400
per
ton
of
particulate
matter
removed.
We
estimate
that
this
beyond­
the­
floor
option
would
increase
the
amount
of
hazardous
waste
(
or
solid
waste
if
the
source
retains
the
Bevill
exclusion
under
40
CFR
266.112)
generated
for
a
new
solid
fuel­
fired
boiler
with
average
gas
flowrate
by
44
tons
per
year
and
would
require
a
source
to
use
an
additional
1.1
million
kW­
hours
per
year
beyond
the
requirements
to
achieve
the
floor
level.
Notwithstanding
these
impacts,
a
standard
of
34
mg/
dscm
(
0.015
gr/
dscf)
is
warranted
because
it
would
be
cost­
effective
and
it
would
remove
additional
nonenumerated
metal
HAP.
We
also
note
that
this
beyond­
the­
floor
standard
of
0.015
gr/
dscf
for
new
sources
would
be
comparable
to
the
floor­
based
standard
the
Agency
recently
proposedpromulgated
for
new
solid
fuel­
fired
boilers
that
do
not
burn
hazardous
waste:
0.0265
lb/
MM
Btu
(
approximately
0.0132
gr/
dscf).
See
68
FR
at
1660
(
January
13,
2003)
NESHAP
for
Industrial/
Commercial/
Institutional
Boilers
and
Process
Heaters,
signed
Feb.
26,
2004..
For
these
reasons,
we
propose
a
beyond­
the­
floor
particulate
matter
standard
of
34
mg/
dscm
(
0.015
gr/
dscf)
for
new
sources.

D.
What
Is
the
Rationale
for
the
Proposed
Standards
for
Semivolatile
Metals?
Redline­
strikeout
highlighting
changes
made
during
OMB
review
46
As
information,
EPA
proposed
to
control
nonmercury
metal
HAP
emissions
for
industrial,
commercial,
and
institutional
boilers
that
do
not
burn
hazardous
waste
with
a
particulate
matter
emission
standard
only.
See
68
FR
1660
(
Jan.
13,
2003).
For
hazardous
waste
combustors,
we
propose
to
control
specific,
enumerated
semivolatile
and
low
volatile
metals
with
separate
emission
standards
because
hazardous
waste
can
have
a
wide
range
of
concentrations
of
these
metals,
and,
thus,
particulate
matter
may
contain
a
wide
range
of
metal
concentrations.
Thus,
particulate
matter
may
not
be
an
effective
surrogate
for
particular
metal
HAP.
Nonetheless,
for
practical
reasons,
we
rely
on
particulate
matter
to
control
nonenumerated
metal
HAP.

47
Owners
and
operators
have
determined
that
emissions
from
these
four
boilers
represent
emissions
from
five
other
identical,
sister
boilers.
Owners
and
operators
have
used
the
emissions
from
these
four
boilers
as
"
data
in
lieu
of
testing"
emissions
from
the
other
five
identical
boilers.

187
The
proposed
standard
for
semivolatile
metals
(
lead
and
cadmium,
combined)
for
solid
fuel­
fired
boilers
is
170
ug/
dscm
for
both
existing
and
new
sources.
46
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
We
have
compliance
test
emissions
data
representing
maximum
allowed
emissions
for
four
boilers.
Emissions
from
these
four
boilers
represent
emissions
from
nine
of
the
12
solid
fuel­
fired
boilers.
47
Semivolatile
metal
emissions
range
from
62
ug/
dscm
to
170
ug/
dscm.
These
emissions
are
expressed
as
mass
of
semivolatile
metals
(
from
all
feedstocks)
per
unit
of
stack
gas.
To
identify
the
MACT
floor,
we
evaluated
the
compliance
test
emissions
data
associated
with
the
most
recent
test
campaign
using
the
SRE/
Feed
Approach.
The
calculated
floor
is
170
ug/
dscm,
which
considers
emissions
variability.
This
is
an
emission
level
that
the
average
of
the
best
performing
sources
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
We
estimate
that
this
floor
level
is
being
achieved
by
42%
of
sources
and
would
reduce
semivolatile
metals
emissions
by
0.22
tons
per
year.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
evaluated
three
beyond­
the­
floor
approaches
for
semivolatile
metals
for
existing
sources:
(
1)
improved
control
of
particulate
matter;
(
2)
control
of
semivolatile
metals
in
the
hazardous
waste
feed;
and
(
3)
a
no­
cost
standard
derived
from
the
beyond­
the­
floor
particulate
matter
standard.
For
reasons
discussed
below,
we
are
not
proposing
a
beyond­
the­
floor
standard
for
semivolatile
metals.
a.
Improved
Particulate
Matter
Control.
Controlling
particulate
matter
also
controls
emissions
of
semivolatile
metals.
Consequently,
we
evaluated
a
beyond­
the­
floor
level
of
85
ug/
dscm
,
a
50
percent
reduction
in
semivolatile
metal
emissions,
that
would
be
achieved
by
reducing
particulate
matter
emissions.
The
national
incremental
annualized
compliance
cost
for
solid
fuel
boilers
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
0.29
million
and
would
provide
an
incremental
reduction
in
semivolatile
metals
emissions
beyond
the
MACT
floor
controls
of
0.29
tons
per
year.
We
evaluated
the
nonair
Redline­
strikeout
highlighting
changes
made
during
OMB
review
188
quality
health
and
environmental
impacts
and
energy
effects
of
this
beyond­
the­
floor
standard
and
estimate
that
the
amount
of
hazardous
waste
generated
would
increase
by
approximately
133
tons
per
year,
an
additional
61
million
gallons
per
year
of
water
would
be
used,
and
an
additional
1.3
million
kW­
hours
per
year
of
electricity
would
be
used.
Therefore,
based
on
these
factors
and
costs
of
approximately
$
1
million
per
additional
ton
of
semivolatile
metals
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
improved
particulate
matter
control.
b.
Feed
Control
of
Semivolatile
Metals
in
the
Hazardous
Waste.
We
also
evaluated
a
beyond­
the­
floor
level
of
140
ug/
dscm
based
on
additional
control
of
semivolatile
metals
in
the
hazardous
waste
feed.
This
represents
a
20%
reduction
from
the
floor
level.
The
national
incremental
annualized
compliance
cost
for
solid
fuel
boilers
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
36,000
and
would
provide
an
incremental
reduction
in
semivolatile
metals
emissions
beyond
the
MACT
floor
controls
of
0.046
tons
per
year.
Although
nonair
quality
health
and
environmental
impacts
and
energy
effects
are
not
significant
factors,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
feed
control
of
semivolatile
metals
in
the
hazardous
waste
because
it
is
not
cost­
effective
at
approximately
$
0.78
million
per
additional
ton
of
semivolatile
metals
removed.
c.
No­
cost
Standard
Derived
from
the
Beyond­
the­
Floor
Particulate
Matter
Standard.
The
beyond­
the­
floor
standard
for
particulate
matter
would
also
provide
beyond­
the­
floor
control
for
semivolatile
metals
if
sources
were
to
comply
with
the
beyond­
the­
floor
particulate
matter
standard
using
improved
particulate
matter
control
rather
than
by
reducing
the
feedrate
of
ash.
To
identify
a
beyond­
the­
floor
emission
level
for
semivolatile
metals
that
would
derive
from
the
beyond­
the­
floor
particulate
matter
standard,
we
assumed
that
emissions
of
semivolatile
metals
would
be
reduced
by
the
same
percentage
that
sources
would
need
to
reduce
particulate
matter
emissions.
We
then
developed
a
revised
semivolatile
metal
emission
data
base
considering
these
particulate
matter
standard­
derived
reductions
and
reductions
needed
to
meet
the
semivolatile
metal
floor
level.
We
analyzed
these
revised
emissions
to
identify
the
best
performing
sources
and
an
emission
level
that
the
average
of
the
best
performers
could
achieve
99
out
of
100
future
tests.
This
emission
level­­
82
ug/
dscm­­
is
a
beyond­
the­
floor
semivolatile
metal
standard
that
can
be
achieved
at
no
cost
because
the
costs
have
been
allocated
to
the
particulate
matter
beyond­
thefloor
standard.
We
are
concerned,
however,
that
sources
may
choose
to
comply
with
the
beyond­
thefloor
particulate
matter
standard
by
controlling
the
feedrate
of
ash
in
the
hazardous
waste
feed,
which
may
or
may
not
reduce
the
feedrate
and
emissions
of
metal
HAP.
If
so,
it
would
be
inappropriate
to
consider
the
beyond­
the­
floor
standard
for
semivolatile
metals
discussed
above
as
a
no­
cost
standard.
We
specifically
request
comment
on
whether
sources
may
comply
with
beyond­
the­
floor
particulate
matter
standard
by
controlling
the
feedrate
of
ash..
For
these
reasons,
we
propose
a
floor
standard
for
semivolatile
metals
of
170
ug/
dscm
for
existing
sources.
3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
MACT
floor
for
new
sources
would
be
170
ug/
dscm,
considering
emissions
variability.
This
is
the
same
as
the
floor
for
existing
sources.
This
is
an
emission
level
that
the
single
best
performing
source
identified
by
the
SRE/
Feed
Approach
could
be
expected
to
achieve
in
99
of
Redline­
strikeout
highlighting
changes
made
during
OMB
review
189
100
future
tests
when
operating
under
operating
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
We
evaluated
three
beyond­
the­
floor
approaches
for
semivolatile
metals
for
new
sources:
(
1)
improved
particulate
matter
controls;
(
2)
control
of
semivolatile
metals
in
the
hazardous
waste
feed;
and
(
3)
a
no­
cost
standard
derived
from
the
beyond­
the­
floor
particulate
matter
standard.
a.
Improved
Particulate
Matter
Controls.
We
evaluated
improved
control
of
particulate
matter
using
a
fabric
filter
as
beyond­
the­
floor
control
for
further
reductions
in
semivolatile
metals
emissions.
We
evaluated
a
beyond­
the­
floor
level
of
71
ug/
dscm.
The
incremental
annualized
compliance
cost
for
a
new
solid
fuel
boiler
with
average
gas
flowrate
to
meet
this
beyond­
thefloor
level,
rather
than
comply
with
the
floor
level,
would
be
approximately
$
0.28
million
and
would
provide
an
incremental
reduction
in
semivolatile
metals
emissions
of
approximately
0.15
tons
per
year,
for
a
cost­
effectiveness
of
$
1.8
million
per
ton
of
semivolatile
metals
removed.
We
estimate
that
this
beyond­
the­
floor
option
would
increase
the
amount
of
hazardous
waste
(
or
solid
waste
if
the
source
retains
the
Bevill
exclusion
under
40
CFR
266.112)
generated
for
a
new
solid
fuel­
fired
boiler
with
average
gas
flowrate
by
44
tons
per
year
and
would
require
the
source
to
use
an
additional
1.2
million
kW­
hours
per
year
beyond
the
requirements
to
achieve
the
floor
level.
After
considering
these
impacts
and
cost­
effectiveness,
we
conclude
that
a
beyond­
thefloor
standard
for
new
sources
based
on
use
of
a
fabric
filter
to
improve
control
of
particulate
matter
is
not
warranted.
b.
Feedrate
Control.
For
similar
reasons
discussed
above
for
existing
sources,
we
conclude
that
a
beyond­
the­
floor
standard
based
on
controlling
the
semivolatile
metals
in
the
hazardous
waste
feed
would
not
be
cost­
effective.
c.
No­
cost
Standard
Derived
from
the
Beyond­
the­
Floor
Particulate
Matter
Standard.
As
discussed
above
in
the
context
of
existing
sources,
the
beyond­
the­
floor
standard
for
particulate
matter
would
also
provide
beyond­
the­
floor
control
for
semivolatile
metals
if
sources
were
to
comply
with
the
beyond­
the­
floor
particulate
matter
standard
using
improved
particulate
matter
control
rather
than
by
reducing
the
feedrate
of
ash.
Under
this
approach,
the
no­
cost
beyond­
thefloor
standard
for
semivolatile
metals
for
new
sources
would
be
44
ug/
dscm.
As
discussed
above,
however,
we
are
concerned
that
sources
may
choose
to
comply
with
the
beyond­
the­
floor
particulate
matter
standard
by
controlling
the
feedrate
of
ash
in
the
hazardous
waste
feed,
which
may
or
may
not
reduce
the
feedrate
and
emissions
of
metal
HAP.
If
so,
it
would
be
inappropriate
to
consider
this
beyond­
the­
floor
standard
as
a
no­
cost
standard.
We
specifically
request
comment
on
whether
sources
may
comply
with
beyond­
the­
floor
particulate
matter
standard
by
controlling
the
feedrate
of
ash.
For
these
reasons,
we
propose
a
semivolatile
metals
standard
of
170
ug/
dscm
for
new
sources.
E.
What
Is
the
Rationale
for
the
Proposed
Standards
for
Low
Volatile
Metals?
The
proposed
standards
for
low
volatile
metals
(
arsenic,
beryllium,
and
chromium)
for
solid
fuel­
fired
boilers
is
210
ug/
dscm
for
existing
sources
and
190
ug/
dscm
for
new
sources.
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
We
have
compliance
test
emissions
data
representing
maximum
allowed
emissions
for
four
Redline­
strikeout
highlighting
changes
made
during
OMB
review
48
Owners
and
operators
have
determined
that
emissions
from
these
four
boilers
represent
emissions
from
five
other
identical,
sister
boilers.
Owners
and
operators
have
used
the
emissions
from
these
four
boilers
as
"
data
in
lieu
of
testing"
emissions
from
the
other
five
identical
boilers.

190
boilers.
Emissions
from
these
four
boilers
represent
emissions
from
10
of
the
12
solid
fuel­
fired
boilers.
48
Low
volatile
metal
emissions
range
from
41
ug/
dscm
to
230
ug/
dscm.
These
emissions
are
expressed
as
mass
of
low
volatile
metals
(
from
all
feedstocks)
per
unit
of
stack
gas.
To
identify
the
MACT
floor,
we
evaluated
the
compliance
test
emissions
data
associated
with
the
most
recent
test
campaign
using
the
SRE/
Feed
Approach.
The
calculated
floor
is
210
ug/
dscm,
which
considers
emissions
variability.
This
is
an
emission
level
that
the
average
of
the
best
performing
sources
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
We
estimate
that
this
emission
level
is
being
achieved
by
67%
of
sources
and
that
it
would
reduce
low
volatile
metals
emissions
by
0.45
tons
per
year.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
evaluated
three
beyond­
the­
floor
approaches
for
low
volatile
metals
for
existing
sources:
(
1)
improved
control
of
particulate
matter;
(
2)
control
of
low
volatile
metals
in
the
hazardous
waste
feed;
and
(
3)
a
no­
cost
standard
derived
from
the
beyond­
the­
floor
particulate
matter
standard.
For
reasons
discussed
below,
we
are
not
proposing
a
beyond­
the­
floor
standard
for
low
volatile
metals.
a.
Improved
Particulate
Matter
Control.
Controlling
particulate
matter
also
controls
emissions
of
low
volatile
metals.
We
evaluated
a
beyond­
the­
floor
level
of
105
ug/
dscm.
The
national
incremental
annualized
compliance
cost
for
solid
fuel
boilers
to
meet
this
beyond­
thefloor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
0.32
million
and
would
provide
an
incremental
reduction
in
low
volatile
metals
emissions
beyond
the
MACT
floor
controls
of
0.37
tons
per
year.
We
evaluated
the
nonair
quality
health
and
environmental
impacts
and
energy
effects
of
this
beyond­
the­
floor
standard
and
estimate
that
the
amount
of
hazardous
waste
generated
would
increase
by
approximately
83
tons
per
year,
an
additional
54
million
gallons
of
water
per
year
would
be
used,
and
electricity
consumption
would
increase
by
1.2
million
kW­
hours
per
year.
Considering
these
impacts
and
a
cost
of
approximately
$
0.87
million
per
additional
ton
of
low
volatile
metals
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
improved
particulate
matter
control.
b.
Feed
Control
of
Low
Volatile
Metals
in
the
Hazardous
Waste.
We
also
evaluated
a
beyond­
the­
floor
level
of
170
ug/
dscm,
which
represents
a
20%
reduction
from
the
floor
level.
The
national
incremental
annualized
compliance
cost
for
solid
fuel
boilers
to
meet
this
beyondthe
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
98,000
and
would
provide
an
incremental
reduction
in
low
volatile
metals
emissions
beyond
the
MACT
floor
controls
of
0.13
tons
per
year.
Although
nonair
quality
health
and
environmental
impacts
and
energy
effects
are
not
significant
factors,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
feedrate
control
of
low
volatile
metals
in
the
hazardous
waste
because
it
would
not
be
cost­
Redline­
strikeout
highlighting
changes
made
during
OMB
review
191
effective
at
approximately
$
0.78
million
per
additional
ton
of
low
volatile
metals
removed.
c.
No­
cost
Standard
Derived
from
the
Beyond­
the­
Floor
Particulate
Matter
Standard.
As
discussed
above
in
the
context
of
semivolatile
metals,
the
beyond­
the­
floor
standard
for
particulate
matter
would
also
provide
beyond­
the­
floor
control
for
low
volatile
metals
if
sources
were
to
comply
with
the
beyond­
the­
floor
particulate
matter
standard
using
improved
particulate
matter
control
rather
than
by
reducing
the
feedrate
of
ash.
To
identify
a
beyond­
the­
floor
emission
level
for
low
volatile
metals
that
would
derive
from
the
beyond­
the­
floor
particulate
matter
standard,
we
assumed
that
emissions
of
low
volatile
metals
would
be
reduced
by
the
same
percentage
that
sources
would
need
to
reduce
particulate
matter
emissions.
We
then
developed
a
revised
low
volatile
metal
emission
data
base
considering
these
particulate
matter
standardderived
reductions
and
reductions
needed
to
meet
the
low
volatile
metal
floor
level.
We
analyzed
these
revised
emissions
to
identify
the
best
performing
sources
and
an
emission
level
that
the
average
of
the
best
performers
could
achieve
99
out
of
100
future
tests.
This
emission
level­­
110
ug/
dscm­­
is
a
beyond­
the­
floor
low
volatile
metal
standard
that
can
be
achieved
at
no
cost
because
the
costs
have
been
allocated
to
the
particulate
matter
beyond­
the­
floor
standard.
We
are
concerned,
however,
that
sources
may
choose
to
comply
with
the
beyond­
thefloor
particulate
matter
standard
by
controlling
the
feedrate
of
ash
in
the
hazardous
waste
feed,
which
may
or
may
not
reduce
the
feedrate
and
emissions
of
metal
HAP.
If
so,
it
would
be
inappropriate
to
consider
the
beyond­
the­
floor
standard
for
low
volatile
metals
discussed
above
as
a
no­
cost
standard.
We
specifically
request
comment
on
whether
sources
may
comply
with
beyond­
the­
floor
particulate
matter
standard
by
controlling
the
feedrate
of
ash.
For
these
reasons,
we
propose
a
floor
standard
for
low
volatile
metals
of
210
ug/
dscm
for
existing
sources.
3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
MACT
floor
for
low
volatile
metals
for
new
sources
would
be
190
ug/
dscm,
considering
emissions
variability.
This
is
an
emission
level
that
the
single
best
performing
source
identified
by
the
SRE/
Feed
Approach
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
operating
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
We
evaluated
three
beyond­
the­
floor
approaches
for
low
volatile
metals
for
new
sources:
(
1)
improved
particulate
matter
control;
(
2)
control
of
low
volatile
metals
in
the
hazardous
waste
feed;
and
(
3)
a
no­
cost
standard
derived
from
the
beyond­
the­
floor
particulate
matter
standard.
a.
Improved
Particulate
Matter
Control.
We
evaluated
improved
control
of
particulate
matter
using
a
fabric
filter
to
achieve
an
emission
level
of
79
ug/
dscm
as
beyond­
the­
floor
control
for
low
volatile
metals
emissions.
The
incremental
annualized
compliance
cost
for
a
new
solid
fuel
boiler
to
meet
this
beyond­
the­
floor
level,
rather
than
comply
with
the
floor
level,
would
be
approximately
$
0.28
million
and
would
provide
an
incremental
reduction
in
low
volatile
metals
emissions
of
approximately
0.17
tons
per
year,
for
a
cost­
effectiveness
of
$
1.7
million
per
ton
of
low
volatile
metals
removed.
We
estimate
that
this
beyond­
the­
floor
option
would
increase
the
amount
of
hazardous
waste
(
or
solid
waste
if
the
source
retains
the
Bevill
exclusion
under
40
CFR
266.112)
generated
for
a
new
solid
fuel­
fired
boiler
with
average
gas
flowrate
by
44
tons
per
year
Redline­
strikeout
highlighting
changes
made
during
OMB
review
49
As
information,
EPA
proposed
MACT
standards
for
hydrogen
chloride
for
solid
fuel­
fired
industrial,
commercial,
and
institutional
boilers
that
do
not
burn
hazardous
waste
of
68
ppmv
for
existing
sources
and
15
ppmv
for
new
sources.
See
68
FR
1660
(
Jan.
13,
2003).
These
standards
are
based
on
use
of
wet
scrubbers
to
control
hydrogen
chloride.

50
Owners
and
operators
have
determined
that
emissions
from
these
five
boilers
represent
emissions
from
five
other
identical,
sister
boilers.
Owners
and
operators
have
used
the
emissions
from
these
five
boilers
as
"
data
in
lieu
of
testing"
emissions
from
the
other
five
identical
boilers.

192
and
would
require
the
source
to
use
an
additional
1.2
million
kW­
hours
per
year
beyond
the
requirements
to
achieve
the
floor
level.
After
considering
these
impacts
and
cost­
effectiveness,
we
conclude
that
a
beyond­
the­
floor
standard
based
on
improved
particulate
matter
control
using
a
fabric
filter
for
new
sources
is
not
warranted.
b.
Feedrate
Control.
For
similar
reasons
discussed
above
for
existing
sources,
we
conclude
that
a
beyond­
the­
floor
standard
based
on
controlling
the
low
volatile
metals
in
the
hazardous
waste
feed
would
not
be
cost­
effective.
c.
No­
cost
Standard
Derived
from
the
Beyond­
the­
Floor
Particulate
Matter
Standard.
As
discussed
above
in
the
context
of
existing
sources,
the
beyond­
the­
floor
standard
for
particulate
matter
would
also
provide
beyond­
the­
floor
control
for
low
volatile
metals
if
sources
were
to
comply
with
the
beyond­
the­
floor
particulate
matter
standard
using
improved
particulate
matter
control
rather
than
by
reducing
the
feedrate
of
ash.
Under
this
approach,
the
no­
cost
beyond­
thefloor
standard
for
low
volatile
metals
for
new
sources
would
be
34
ug/
dscm.
As
discussed
above,
however,
we
are
concerned
that
sources
may
choose
to
comply
with
the
beyond­
the­
floor
particulate
matter
standard
by
controlling
the
feedrate
of
ash
in
the
hazardous
waste
feed,
which
may
or
may
not
reduce
the
feedrate
and
emissions
of
metal
HAP.
If
so,
it
would
be
inappropriate
to
consider
this
beyond­
the­
floor
standard
as
a
no­
cost
standard.
We
specifically
request
comment
on
whether
sources
may
comply
with
beyond­
the­
floor
particulate
matter
standard
by
controlling
the
feedrate
of
ash.
For
these
reasons,
we
propose
a
low
volatile
metals
standard
of
190
ug/
dscm
for
new
sources.
F.
What
Is
the
Rationale
for
the
Proposed
Standards
for
Total
Chlorine?
The
proposed
standards
for
hydrogen
chloride
and
chlorine
gas
(
i.
e.,
total
chlorine,
reported
as
a
hydrogen
chloride
equivalents)
for
solid
fuel­
fired
boilers
are
ppmv
for
existing
sources
and
73
ppmv
for
new
sources.
49
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
Solid
fuel­
fired
boilers
that
burn
hazardous
waste
are
equipped
with
electrostatic
precipitators
or
baghouses
and
do
not
have
back­
end
controls
for
total
chlorine.
Total
chlorine
emissions
are
controlled
by
controlling
the
feedrate
of
chlorine
in
the
hazardous
waste
feed.
We
have
compliance
test
emissions
data
representing
maximum
allowable
emissions
for
five
boilers.
Emissions
from
these
five
boilers
represent
emissions
from
10
of
the
12
solid
fuel­
fired
boilers.
50
Redline­
strikeout
highlighting
changes
made
during
OMB
review
51
See
discussion
in
Part
Four,
Section
VIII.
D.
1
of
the
preamble
for
more
information.

193
Total
chlorine
emissions
range
from
60
ppmv
to
700
ppmv.
To
identify
the
MACT
floor,
we
evaluated
the
compliance
test
emissions
data
associated
with
the
most
recent
test
campaign
using
the
SRE/
Feed
Approach.
The
calculated
floor
is
440
ppmv,
which
considers
emissions
variability.
This
is
an
emission
level
that
the
best
performing
feed
control
sources
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
We
estimate
that
this
emission
level
is
being
achieved
by
83%
of
sources
and
that
it
would
reduce
total
chlorine
emissions
by
420
tons
per
year.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
evaluated
dry
scrubbing
to
achieve
a
beyond­
the­
floor
emission
level
of
110
ppmv
for
total
chlorine
for
existing
sources,
assuming
conservatively
a
75%
removal
efficiency.
The
national
annualized
incremental
compliance
cost
for
solid
fuel­
fired
boilers
to
comply
with
this
beyond­
the­
floor
standardlevel
rather
than
the
floor
level
would
be
$
3.7
million,
and
emissions
of
total
chlorine
would
be
reduced
by
an
additional
790
tons
per
year,
for
a
cost­
effectiveness
of
$
4,700
per
ton
of
total
chlorine
removed.
We
evaluated
the
nonair
quality
health
and
environmental
impacts
and
energy
effects
of
this
beyond­
the­
floor
standardlevel
and
estimate
that
the
amount
of
hazardous
waste
generated
would
increase
by
18,000
tons
per
year,
an
additional
27
million
gallons
of
water
per
year
would
be
used,
and
electricity
consumption
would
increase
by
0.11
million
kW­
hours
per
year.

We
note
that
a
cost
of
is
in
the
"
grey
area"
between
a
cost
the
Agency
has
concluded
is
cost­
effective
and
a
cost
the
Agency
has
concluded
is
not
cost­
effective
under
other
MACT
rules.
EPA
concluded
that
a
cost
of
$
1,100
per
ton
of
total
chlorine
removed
for
hazardous
waste
burning
lightweight
aggregate
kilns
was
cost­
effective
in
the
1999
MACT
final
rule.
See
68
FR
at
52900.
EPA
concluded,
however,
that
a
cost
of
$
45,000
per
ton
of
hydrogen
chloride
removed
was
not
cost­
effective
for
industrial
boilers.
See
68
FR
at
1677.
Although
a
beyond­
the­
floor
standard
of
110
ppmv
for
solid
fuel
boilers
under
today's
rule
would
provide
health
benefits
from
collateral
reductions
in
SO
2
emissions,
51
we
are
concerned
that
a
cost
of
$
4,700
per
additional
ton
of
total
chlorine
removed
is
not
warranted.

We
also
evaluated
use
of
feedrate
control
of
chlorine
in
the
hazardous
waste
to
achieve
a
beyond­
the­
floor
level
of
350
ppmv,
which
represents
a
20%
reduction
from
the
floor
level.
Redline­
strikeout
highlighting
changes
made
during
OMB
review
52
Although
we
assumed
dry
scrubbing
can
readily
achieve
75%
removal
of
total
chlorine
for
beyond­
the­
floor
control
for
existing
sources,
assuming
50%
removal
for
beyond­
thefloor
control
for
new
sources
is
appropriate.
This
is
because
the
floor
for
new
sources­­
73
ppmv­
­
is
substantially
lower
than
the
floor
for
existing
sources­­
440
ppmv­­
and
dry
scrubbing
is
less
efficient
at
lower
uncontrolled
emission
levels.

194
Because
this
standard
would
achieve
emission
reductions
of
only
44
tonsThe
national
annualized
incremental
compliance
cost
for
solid
fuel­
fired
boilers
to
comply
with
this
beyond­
the­
floor
level
rather
than
the
floor
level
would
be
$
0.08
million,
and
emissions
of
total
chlorine
per
year,
we
proposewould
be
reduced
by
an
additional
40
tons
per
year,
for
a
cost­
effectiveness
of
$
2,000
per
ton
of
total
chlorine
removed.
Although
nonair
quality
health
and
environmental
impacts
and
energy
effects
are
not
significant
factors
for
feedrate
control,
we
are
not
proposing
a
beyond­
thefloor
standard
of
110
ppmv
based
on
dry
scrubbing
for
existing
sourcesbased
on
hazardous
waste
feedrate
control
because
we
are
concerned
about
the
practicability
of
achieving
these
emissions
reductions,
and
our
estimate
of
the
associated
cost,
using
feedrate
control.

3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
MACT
floor
for
new
sources
would
be
73
ppmv.
This
is
an
emission
level
that
the
single
best
performing
source
identified
by
the
Emissions
Approach
(
i.
e.,
the
source
with
the
lowest
emissions)
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
operating
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
We
evaluated
dry
lime
scrubbing
to
achieve
a
beyond­
the­
floor
emission
level
of
37
ppmv
for
total
chlorine
for
new
sources,
assuming
conservatively
a
50%
removal
efficiency.
52
The
incremental
annualized
compliance
cost
for
a
new
solid
fuel
boiler
with
average
gas
flowrate
to
meet
this
beyond­
the­
floor
level,
rather
than
comply
with
the
floor
level,
would
be
approximately
$
610,000
and
would
provide
an
incremental
reduction
in
total
chlorine
emissions
of
approximately
42
tons
per
year.
Although
nonair
quality
health
and
environmental
impacts
and
energy
effects
are
not
significant
factors,
we
conclude
that
a
beyond­
the­
floor
standard
of
37
ppmv
is
not
warranted
because
it
would
not
be
cost­
effective
at
approximately
$
14,000
per
additional
ton
of
total
chlorine
removed.
For
these
reasons,
we
propose
a
floor
standard
for
total
chlorine
of
73
ppmv
for
new
sources.
G.
What
Is
the
Rationale
for
the
Proposed
Standards
for
Carbon
Monoxide
or
Hydrocarbons?
To
control
emissions
of
organic
HAP,
existing
and
new
sources
would
be
required
to
comply
with
either
a
carbon
monoxide
standard
of
100
ppmv
or
a
hydrocarbon
standard
of
10
Redline­
strikeout
highlighting
changes
made
during
OMB
review
53
As
information,
EPA
proposed
MACT
standards
for
carbon
monoxide
for
new
solid
fuel­
fired
industrial,
commercial,
and
institutional
boilers
that
do
not
burn
hazardous
waste
of
400
ppmv
corrected
to
3%
oxygen.
See
68
FR
1660
(
Jan.
13,
2003).

195
ppmv.
53
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
Solid
fuel­
fired
boilers
that
burn
hazardous
waste
are
currently
subject
to
RCRA
standards
that
require
compliance
with
either
a
carbon
monoxide
standard
of
100
ppmv,
or
a
hydrocarbon
standard
of
20
ppmv.
Compliance
is
based
on
an
hourly
rolling
average
as
measured
with
a
CEMS.
See
§
266.104(
a).
We
are
proposing
today
floor
standards
of
100
ppmv
for
carbon
monoxide
or
10
ppmv
for
hydrocarbons.
Floor
control
for
existing
sources
is
operating
under
good
combustion
practices
including:
(
1)
providing
adequate
excess
air
with
use
of
oxygen
CEMS
and
feedback
air
input
control;
(
2)
providing
adequate
fuel/
air
mixing;
(
3)
homogenizing
hazardous
waste
fuels
(
such
as
by
blending
or
size
reduction)
to
control
combustion
upsets
due
to
very
high
or
very
low
volatile
content
wastes;
(
4)
regulating
waste
and
air
feedrates
to
ensure
proper
combustion
temperature
and
residence
time;
(
5)
characterizing
waste
prior
to
burning
for
combustion­
related
composition
(
including
parameters
such
as
heating
value,
volatile
content,
liquid
waste
viscosity,
etc.);
(
6)
ensuring
the
source
is
operated
by
qualified,
experienced
operators;
and
(
7)
periodic
inspection
and
maintenance
of
combustion
system
components
such
as
burners,
fuel
and
air
supply
lines,
injection
nozzles,
etc.
Given
that
there
are
many
interdependent
parameters
that
affect
combustion
efficiency
and
thus
carbon
monoxide
and
hydrocarbon
emissions,
we
are
not
able
to
quantify
"
good
combustion
practices."
Ten
of
12
solid
fuel­
fired
boilers
are
currently
complying
with
the
RCRA
carbon
monoxide
limit
of
100
ppmv
on
an
hourly
rolling
average.
The
remaining
two
boilers
are
complying
with
the
RCRA
hydrocarbon
limit
of
20
ppmv
on
an
hourly
rolling
average.
Those
boilers
have
hydrocarbon
levels
below
5
ppmv,
however,
indicative
of
operating
under
good
combustion
practices.
We
propose
a
floor
level
for
carbon
monoxide
level
of
100
ppmv
because
it
is
a
currently
enforceable
Federal
standard.
Although
the
best
performing
sources
are
achieving
carbon
monoxide
levels
below
100
ppmv,
it
is
not
appropriate
to
establish
a
lower
floor
level
because
carbon
monoxide
is
a
surrogate
for
nondioxin/
furan
organic
HAP.
As
such,
lowering
the
carbon
monoxide
floor
may
not
significantly
reduce
organic
HAP
emissions.
In
addition,
it
would
be
inappropriate
to
apply
a
MACT
methodology
to
the
carbon
monoxide
emissions
from
the
best
performing
sources
because
those
sources
may
not
be
able
to
replicate
their
emission
levels.
This
is
because
there
are
myriad
factors
that
affect
combustion
efficiency
and,
subsequently,
carbon
monoxide
emissions.
Extremely
low
carbon
monoxide
emissions
cannot
be
assured
by
controlling
only
one
or
two
operating
parameters
We
note
also
that
we
used
this
rationale
to
establish
a
carbon
monoxide
standard
of
100
ppmv
for
Phase
I
sources
in
the
September
1999
Final
Rule.
We
propose
a
floor
level
for
hydrocarbons
of
10
ppmv
even
though
the
currently
enforceable
standard
is
20
ppmv
because:
(
1)
the
two
sources
that
comply
with
the
RCRA
Redline­
strikeout
highlighting
changes
made
during
OMB
review
196
hydrocarbon
standard
can
readily
achieve
10
ppmv;
and
(
2)
reducing
hydrocarbon
emissions
within
the
range
of
20
ppmv
to
10
ppmv
should
reduce
emissions
of
nondioxin/
furan
organic
HAP.
We
do
not
apply
a
prescriptive
MACT
methodology
to
establish
a
hydrocarbon
floor
below
10
ppmv,
however,
because
we
have
data
from
only
two
sources.
In
addition,
we
note
that
the
hydrocarbon
emission
standard
for
Phase
I
sources
established
in
the
September
1999
Final
Rule
is
10
ppmv
also.
There
would
be
no
incremental
emission
reductions
associated
with
these
floors
because
all
sources
are
currently
achieving
the
floor
levels.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
considered
beyond­
the­
floor
levels
for
carbon
monoxide
and
hydrocarbons
based
on
use
of
better
combustion
practices
but
conclude
that
they
may
not
be
replicable
by
the
best
performing
sources
nor
duplicable
by
other
sources
given
that
we
cannot
quantify
good
combustion
practices.
Moreover,
we
cannot
ensure
that
carbon
monoxide
or
hydrocarbon
levels
lower
than
the
floors
would
significantly
reduce
emissions
of
nondioxin/
furan
organic
HAP.
This
is
because
the
portion
of
hydrocarbons
that
is
comprised
of
nondioxin/
furan
organic
HAP
is
likely
to
become
lower
as
combustion
efficiency
improves
and
hydrocarbon
levels
decrease.
Thus,
at
beyond­
the­
floor
hydrocarbon
levels,
we
would
expect
a
larger
portion
of
residual
hydrocarbons
to
be
compounds
that
are
not
organic
HAP.
Nonair
quality
health
and
environmental
impacts
and
energy
requirements
are
not
significant
factors
for
use
of
better
combustion
practices
as
beyond­
the­
floor
control.
For
these
reasons,
we
conclude
that
beyond­
the­
floor
standards
for
carbon
monoxide
and
hydrocarbons
are
not
warranted
for
existing
sources.
3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
MACT
floor
for
new
sources
would
be
the
same
as
the
floor
for
existing
sources­­
100
ppmv
for
carbon
monoxide
and
10
ppmv
for
hydrocarbons­­
and
based
on
the
same
rationale.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
As
discussed
in
the
context
of
beyond­
the­
floor
considerations
for
existing
sources,
we
considered
beyond­
the­
floor
standards
for
carbon
monoxide
and
hydrocarbons
for
new
sources
based
on
use
of
better
combustion
practices.
But,
we
conclude
that
beyond
the
floor
standards
may
not
be
replicable
by
the
best
performing
sources
nor
duplicable
by
other
sources
given
that
we
cannot
quantify
good
combustion
practices.
Moreover,
we
cannot
ensure
that
carbon
monoxide
or
hydrocarbon
levels
lower
than
the
floors
would
significantly
reduce
emissions
of
nondioxin/
furan
organic
HAP.
Nonair
quality
health
and
environmental
impacts
and
energy
requirements
are
not
significant
factors
for
use
of
better
combustion
practices
as
beyond­
the­
floor
control.
For
these
reasons,
we
conclude
that
beyond­
the­
floor
standards
for
carbon
monoxide
and
hydrocarbons
are
not
warranted
for
new
sources.
H.
What
Is
the
Rationale
for
the
Proposed
Standard
for
Destruction
and
Removal
Efficiency?
To
control
emissions
of
organic
HAP,
existing
and
new
sources
would
be
required
to
comply
with
a
destruction
and
removal
efficiency
(
DRE)
of
99.99%
for
organic
HAP.
For
sources
burning
hazardous
wastes
F020,
F021,
F022,
F023,
F026,
or
F027,
however,
the
DRE
standard
is
99.9999%
for
organic
HAP.
Redline­
strikeout
highlighting
changes
made
during
OMB
review
54
The
carbon
monoxide/
hydrocarbon
emission
standard
would
control
organic
HAP
that
are
products
of
incomplete
combustion
by
also
ensuring
use
of
good
combustion
practices.

197
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
Solid
fuel­
fired
boilers
that
burn
hazardous
waste
are
currently
subject
to
RCRA
DRE
standards
that
require
99.99%
destruction
of
designated
principal
organic
hazardous
constituents
(
POHCs).
For
sources
that
burn
hazardous
wastes
F020,
F021,
F022,
F023,
F026,
or
F027,
however,
the
DRE
standard
is
99.9999%
destruction
of
designated
POHCs.
See
§
266.104(
a).
The
DRE
standard
helps
ensure
that
a
combustor
is
operating
under
good
combustion
practices
and
thus
minimizing
emissions
of
organic
HAP.
Under
the
MACT
compliance
regime,
sources
would
designate
POHCs
that
are
organic
HAP
or
that
are
surrogates
for
organic
HAP.
We
propose
to
establish
the
RCRA
DRE
standard
as
the
floor
for
existing
sources
because
it
is
a
currently
enforceable
Federal
standard.
There
would
be
no
incremental
emission
reductions
associated
with
this
floor
because
sources
are
currently
complying
with
the
standard.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
considered
a
beyond­
the­
floor
level
for
DRE
based
on
use
of
better
combustion
practices
but
conclude
that
it
may
not
be
replicable
by
the
best
performing
sources
nor
duplicable
by
other
sources
given
that
we
cannot
quantify
better
combustion
practices.
Moreover,
we
cannot
ensure
that
a
higher
DRE
standard
would
significantly
reduce
emissions
of
organic
HAP
given
that
DRE
measures
the
destruction
of
organic
HAP
present
in
the
boiler
feed
rather
than
gross
emissions
of
organic
HAP.
Although
a
source's
combustion
practices
may
be
adequate
to
destroy
particular
organic
HAP
in
the
feed,
other
organic
HAP
that
may
be
emitted
as
products
of
incomplete
combustion
may
not
be
controlled
by
the
DRE
standard.
54
For
these
reasons,
and
after
considering
non­
air
quality
health
and
environmental
impacts
and
energy
requirements,
we
are
not
proposing
a
beyond­
the­
floor
DRE
standard
for
existing
sources.
3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
We
propose
to
establish
the
RCRA
DRE
standard
as
the
floor
for
new
sources
because
it
is
a
currently
enforceable
Federal
standard.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
Using
the
same
rationale
as
we
used
to
consider
a
beyond­
the­
floor
DRE
standard
for
existing
sources,
we
conclude
that
a
beyond­
the­
floor
DRE
standard
for
new
sources
is
not
warranted.
Consequently,
after
considering
non­
air
quality
health
and
environmental
impacts
and
energy
requirements,
we
are
proposing
the
floor
DRE
standard
for
new
sources.

XI.
How
Did
EPA
Determine
the
Proposed
Emission
Standards
for
Hazardous
Waste
Burning
Liquid
Fuel­
Fired
Boilers?

The
proposed
standards
for
existing
and
new
liquid
fuel­
fired
boilers
that
burn
hazardous
waste
are
summarized
in
the
table
below.
See
proposed
§
63.1217.
Redline­
strikeout
highlighting
changes
made
during
OMB
review
198
PROPOSED
STANDARDS
FOR
EXISTING
AND
NEW
LIQUID
FUEL­
FIRED
BOILERS
Hazardous
Air
Pollutant
or
Surrogate
Emission
Standard1
Existing
Sources
New
Sources
Dioxin
and
furan:
sources
equipped
with
dry
air
pollution
control
system2
0.40
ng
TEQ/
dscm
0.015
ng
TEQ/
dscm
or
control
of
flue
gas
temperature
not
to
exceed
400
°
F
at
the
inlet
to
the
particulate
matter
control
device.

Dioxin
and
furan:
sources
equipped
with
wet
or
with
no
air
pollution
control
systems2
100
ppmv
carbon
monoxide
or
10
ppmv
hydrocarbons
100
ppmv
carbon
monoxide
or
10
ppmv
hydrocarbons
Mercury3
3.7E­
6
lbs/
MM
Btu
3.8E­
7
lbs/
MM
BTU
Particulate
matter
72
mg/
dscm
(
0.032
gr/
dscf)
17
mg/
dscm
(
0.0076
gr/
dscf)

Semivolatile
metals3
1.1E­
5
lbs/
MM
BTU
4.3E­
6
lbs/
MM
BTU
Low
volatile
metals:
chromium
only
3,
4
1.1E­
4
lbs/
MM
BTU
3.6E­
5
lbs/
MM
BTU
Hydrogen
chloride
and
chlorine
gas3,
5
2.5E­
2
lbs/
MM
BTU
or
the
alternative
emission
limits
under
§
63.1215
7.2E­
4
lbs/
MM
BTU
or
the
alternative
emission
limits
under
§
63.1215
Carbon
monoxide
or
hydrocarbons6
100
ppmv
carbon
monoxide
or
10
ppmv
hydrocarbons.
100
ppmv
carbon
monoxide
or
10
ppmv
hydrocarbons.

Destruction
and
Removal
Efficiency
For
existing
and
new
sources,
99.99%
for
each
principal
organic
hazardous
constituent
(
POHC).
For
sources
burning
hazardous
wastes
F020,
F021,
F022,
F023,
F026,
or
F027,
however,
99.9999%
for
each
POHC.

1
All
emission
standards
are
corrected
to
7%
oxygen,
dry
basis.
2
A
wet
air
pollution
system
followed
by
a
dry
air
pollution
control
system
is
not
considered
to
be
a
dry
air
pollution
control
system
for
purposes
of
this
standard.
A
dry
air
pollution
systems
followed
a
wet
air
pollution
control
system
is
considered
to
be
a
dry
air
pollution
control
system
for
purposes
of
this
standard.
3
Standards
are
expressed
as
mass
of
pollutant
emissions
contributed
by
hazardous
waste
per
million
Btu
contributed
by
the
hazardous
waste.
Redline­
strikeout
highlighting
changes
made
during
OMB
review
55
Sources
with
a
wet
air
pollution
system
followed
by
a
dry
air
pollution
control
system
is
not
considered
to
be
a
dry
air
pollution
control
system
for
purposes
of
this
standard.
Sources
with
a
dry
air
pollution
systems
followed
a
wet
air
pollution
control
system
is
considered
to
be
a
dry
air
pollution
control
system
for
purposes
of
this
standard.

199
4
Standard
is
for
chromium
only
and
does
not
include
arsenic
and
beryllium.
5
Combined
standard,
reported
as
a
chloride
(
Cl(­))
equivalent.
6
Hourly
rolling
average.
Hydrocarbons
reported
as
propane.

We
considered
whether
fuel
switching
could
be
considered
a
MACT
floor
control
technology
for
liquid
fuel­
fired
boilers
to
achieve
lower
HAP
emissions.
We
conclude
that
HAP
emissions
from
liquid
fuel­
fired
boilers
are
attributable
primarily
to
the
hazardous
waste
fuels
rather
than
the
natural
gas
or
fuel
oil
that
these
boilers
burn.
Consequently,
we
conclude
that
fuel
switching
is
not
an
effective
MACT
floor
control
technology
to
reduce
HAP
emissions
for
liquid
fuel­
fired
boilers.
A.
What
Are
the
Proposed
Standards
for
Dioxin
and
Furan?
We
propose
to
establish
a
dioxin/
furan
standards
for
existing
liquid
fuel­
fired
boilers
equipped
with
dry
air
pollution
control
devices
that
limit
emissions
of
dioxin/
furan
to
0.40
ng
TEQ/
dscm
(.
The
standard
for
existingnew
sources)
and
would
be
0.015
ng
TEQ/
dscm
(
for
new
sources).
We
also
propose
to
establish
standards
foror
control
of
flue
gas
temperature
not
to
exceed
400
°
F
at
the
inlet
to
the
particulate
matter
control
device.
For
liquid
fuel­
fired
boilers
equipped
either
with
wet
air
pollution
control
systems
or
with
no
air
pollution
systems,
we
propose
a
standard
for
both
existing
and
new
sources
as
compliance
with
the
proposed
standards
for
carbon
monoxide/
hydrocarbon
emission
standard
and
compliance
with
the
proposedand
destruction
and
removal
efficiency.
In
addition,
we
note
that
we
propose
to
require
a
one­
time
dioxin/
furan
emission
test
for
sources
that
would
not
be
subject
to
a
numerical
dioxin/
furan
emission
standard,
including
liquid
fuel­
fired
boilers
with
wet
or
no
emission
control
device,
and
new
liquid
fuel­
fired
boilers
equipped
with
a
dry
air
pollution
control
device.
As
discussed
in
Part
Two,
Section
XIV.
B
below,
the
testing
would
assist
in
developing
both
section
112(
d)(
6)
standards
and
section
112(
f)
residual
risk
standards.
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
As
discussed
in
Part
Two,
Section
I.
B.
5,
we
used
a
statistical
analysis
to
conclude
that
liquid
boilers
equipped
with
dry
air
pollution
control
devices
have
different
dioxin/
furan
emission
characteristics
compared
to
sources
with
either
wet
air
pollution
control
or
no
air
pollution
control
devices.
55
Note
that
we
consider
the
type
of
emission
control
device
as
a
basis
for
subcategorization
because
the
type
of
control
device
affects
formation
of
dioxin/
furan:
dioxin/
furan
can
form
in
dry
particulate
matter
control
devices
while
it
cannot
form
in
wet
(
or
no)
control
devices.
We
therefore
believe
subcategorization
is
warranted
and
we
propose
to
identify
separate
floor
levels
for
sources
equipped
with
dry
particulate
matter
control
devices
versus
sources
with
wet
or
no
emission
control
device.
a.
MACT
Floor
for
Boilers
Equipped
with
Dry
Control
Systems.
To
identify
the
floor
Redline­
strikeout
highlighting
changes
made
during
OMB
review
200
level
for
liquid
fuel
boilers
equipped
with
dry
air
pollution
control
systems,
we
evaluated
the
compliance
test
emissions
data
using
the
Emissions
Approach.
A
numerical
dioxin/
furan
floor
level
is
appropriate
for
these
sources
because
dioxin/
furan
emissionsconsidered
whether
dioxin/
furan
can
be
controlled
by
controlling
the
temperature
at
the
inlet
to
the
particulate
matter
control
device.
Thus,
the
floor
level
can
be
replicated
by
the
best
performing
sources
and
duplicated
by
other
sources.
In
contrast,
We
conclude
that
this
control
mechanism
may
not
be
the
predominant
factor
that
affects
dioxin/
furan
emissions
for
solid
fuel­
fired
boilers
could
not
be
replicated
by
the
best
performing
sources
nor
duplicated
by
other
sources,
even
though
all
sources
arefrom
these
sources.
We
have
emissions
data
for
three
boilers
equipped
with
electrostatic
precipitators
or
fabric
filters.
Emissions
from
two
of
the
boilers
are
below
0.03
ng
TEQ/
dscm.
We
do
not
have
data
on
the
gas
temperature
at
the
inlet
to
the
emission
control
device
for
these
sources.
The
third
boiler,
however,
has
dioxin/
furan
emissions
of
2.4
ng
TEQ/
dscm
when
the
flue
gas
temperature
at
the
inlet
to
the
fabric
filter
is
410

F.
We
conclude
from
this
information
that
this
boiler
is
not
likely
to
be
able
to
achieve
dioxin/
furan
emissions
below
0.40
ng
TEQ/
dscm
if
the
gas
temperature
is
reduced
to
below
400

F.
This
is
contrary
to
the
finding
we
made
for
cement
kilns
and
incinerators
without
heat
recovery
boilers
and
equipped
with
dry
particulate
matter
control
devices,
because.
In
those
cases,
we
conclude
that
gas
temperature
control
at
the
dry
particulate
matter
control
device
is
the
predominant
factor
affecting
dioxin/
furan
emissions.
See
discussions
in
Sections
VII
and
VIII
above.
Consequently,
other
factors
are
likely
contributing
to
high
dioxin/
furan
emissions
are
governed
by
other
factors,
principally
the
inhibiting
effect
onfrom
the
liquid
fuel­
fired
boiler
equipped
with
a
fabric
filter
operated
at
a
gas
temperature
of
410

F,
such
as
metals
in
the
waste
feed
or
soot
on
boiler
tubes
that
may
catalyze
dioxin/
furan
formation
of
sulfur
in
the
coal,
but
also
including
operating
under
good
combustion
practices.
The
calculated
floor
for
sources
equipped
with
dry
air
pollution
control
systems
would
bereactions.
We
evaluated
the
compliance
test
emissions
data
using
the
Emissions
Approach
and
calculated
a
numerical
dioxin/
furan
floor
level
of
3.0
ng
TEQ/
dscm,
which
considers
emissions
variability.
As
discussed
above,
however,
one
of
the
three
sources
for
which
we
have
emissions
data
is
not
likely
to
be
able
to
achieve
this
emission
level
using
gas
temperature
control
at
the
inlet
to
the
dry
particulate
matter
control
device.
Consequently,
we
propose
to
identify
the
floor
level
as
3.0
ng
TEQ/
dscm
or
control
of
flue
gas
temperature
not
to
exceed
400
°
F
at
the
inlet
to
the
particulate
matter
control
device.
This
floor
level
is
duplicable
by
all
sources,
and
would
minimize
dioxin/
furan
emissions
for
sources
where
flue
gas
temperature
at
the
control
device
substantially
affects
dioxin/
furan
emissions.
We
estimate
that
this
emission
level
is
being
achieved
by
99%
ofall
sources
and,
thus,
would
not
reduce
dioxin/
furan
emissions
by
0.14
grams/
year.
.
b.
MACT
Floor
for
Boilers
Equipped
with
Wet
or
No
Control
Systems.
We
have
dioxin/
furan
emissions
data
for
33
liquid
fuel­
fired
boilers
equipped
with
a
wet
or
no
particulate
matter
control
device.
Emissions
levels
are
below
0.1
ng
TEQ/
dscm
for
30
of
the
sources.
Emission
levels
for
the
other
three
sources
are
0.19,
0.36,
and
0.44
ng
TEQ/
dscm.
As
previously
discussed
in
Part
Two,
Section
VII.
A,
we
believe
that
it
would
be
inappropriate
to
establish
a
numerical
dioxin/
furan
emission
floor
level
for
sources
using
wet
or
no
Redline­
strikeout
highlighting
changes
made
during
OMB
review
56
The
fact
that
we
determined
floor
control
for
existing
sources
as
good
combustion
practices
does
not
mean
that
all
sources
using
floor
control
will
have
low
dioxin/
furan
emissions.
As
discussed
in
Part
Two,
Section
XIV.
B.,
we
are
proposing
to
require
liquid
fuel­
fired
boilers
that
would
not
be
subject
to
a
numerical
dioxin/
furan
emission
standard
to
perform
a
one­
time
dioxin/
furan
emissions
test
to
quantify
the
effectiveness
of
today's
proposed
surrogate
for
dioxin/
furan
emission
control.

57
Although
the
floor
for
liquid
fuel
boilers
equipped
with
a
dry
emission
control
device
would
not
be
a
numerical
standard
(
i.
e.,
3.0
ng
TEQ/
dscm
or
control
of
temperature
of
flue
gas
at
the
inlet
to
the
control
device
to
400

F),
we
propose
a
numerical
beyond­
the­
floor
standard
for
those
boilers,
as
discussed
below
in
the
text.

201
air
pollution
control
systems
based
on
the
emissions
achieved
by
the
best
performing
sources
because
a
numerical
floor
level
would
not
be
replicable
by
the
best
performing
sources
nor
duplicable
by
other
sources.
As
a
result,
we
propose
to
define
the
MACT
floor
control
for
sources
with
wet
or
no
emission
control
devices
as
operating
under
good
combustion
practices
by
complying
with
the
destruction
and
removal
efficiency
and
carbon
monoxide/
hydrocarbon
standards.
56
There
would
be
no
emissions
reductions
for
these
existing
boilers
to
comply
with
the
floor
level
because
they
are
currently
complying
with
the
carbon
monoxide/
hydrocarbon
standard
and
destruction
and
removal
efficiency
standard
pursuant
to
RCRA
requirements.
We
also
request
comment
on
an
alternative
MACT
floor
expressed
as
a
dioxin/
furan
emission
concentration
for
sourcesliquid
fuel
boilers
with
wet
or
no
emission
control
devices.
57
Although
it
would
be
inappropriate
to
identify
a
floor
concentration
based
on
the
average
emissions
of
the
best
performing
sources
as
discussed
above,
we
possibly
could
identify
the
floor
as
the
highest
emission
concentration
from
any
source
in
our
data
base,
after
considering
emissions
variability.
Dioxin/
furan
emissions
from
those
liquid
fuel
boilers
in
our
data
base
range
from
less
than
0.001
to
1.6
ng
TEQ/
dscm
with
all
except
one
source
emitting
at
levels
below
0.44
ng
TEQ/
dscm.
Finally,
we
note
that
we
propose
to
require
a
one­
time
dioxin/
furan
emission
test
for
sources
that
would
not
be
subject
to
a
numerical
dioxin/
furan
emission
standard,
including
liquid
fuel­
fired
boilers
with
wet
or
no
emission
control
device.
As
discussed
in
Part
Two,
Section
XIV.
B
below,
the
testing
would
assist
in
developing
both
Section
112(
d)(
6)
standards
and
Section
112(
f)
residual
risk
standards.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
evaluated
use
of
activated
carbon
injection
systems
or
carbon
beds
as
beyond­
thefloor
control
for
further
reduction
of
dioxin/
furan
emissions.
Activated
carbon
has
been
demonstrated
for
controlling
dioxin/
furans
in
various
combustion
applications.
a.
Beyond­
the­
Floor
Considerations
for
Boilers
Equipped
with
Dry
Control
Systems.
For
liquid
fuel­
fired
boilers
using
dry
air
pollution
control
equipment,
we
evaluated
a
beyond­
the­
floor
level
of
0.40
ng
TEQ/
dscm
based
on
activated
carbon
injection
or
control
of
flue
gas
temperature
not
to
exceed
400
°
F
at
the
inlet
to
the
particulate
matter
control
device.
The
national
incremental
Redline­
strikeout
highlighting
changes
made
during
OMB
review
202
annualized
compliance
cost
for
sources
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
80,000
and
would
provide
an
incremental
reduction
in
dioxin/
furan
emissions
beyond
the
MACT
floor
controls
of
0.06
grams
TEQ
per
year
for
a
cost­
effectiveness
of
$
1.3
million
per
additional
gram
of
dioxin/
furan
removed.
We
evaluated
the
nonair
quality
health
and
environmental
impacts
and
energy
effects
of
this
beyondthe
floor
standard
and
estimate
that
the
amount
of
hazardous
waste
generated
would
increase
by
100
tons
per
year,
an
additional
25
trillion
Btu
per
year
of
natural
gas
would
be
consumed,
and
electricity
consumption
would
increase
by
0.50
million
kW­
hours
per
year.
Notwithstanding
these
impacts
and
although
We
judge
that
the
cost
to
achieve
this
beyond­
the­
floor
level
is
warranted
given
our
special
concern
about
dioxin/
furan.
Dioxin/
furan
are
some
of
the
most
toxic
compounds
known
due
to
their
bioaccumulation
potential
and
wide
range
of
health
effects,
including
carcinogenesis,
at
exceedingly
low
doses.
Exposure
via
indirect
pathways
is
a
chief
reason
that
Congress
singled
our
dioxin/
furan
for
priority
MACT
control
in
CAA
section
112(
c)(
6).
See
S.
Rep.
No.
128,
101st
Cong.
1st
Sess.
at
154­
155.
In
addition,
we
note
that
the
beyond­
the­
floor
emission
level
of
0.40
ng
TEQ/
dscm
is
consistent
with
historically
controlled
levels
under
MACT
for
hazardous
waste
incinerators
and
cement
kilns,
and
Portland
cement
plants.
See
§
§
63.1203(
a)(
1),
63.1204(
a)(
1),
and
63.1343(
d)(
3).
Also,
EPA
has
determined
previously
in
the
1999
Hazardous
Waste
Combustor
MACT
final
rule
that
dioxin/
furan
in
the
range
of
0.40
ng
TEQ/
dscm
or
less
are
necessary
for
the
MACT
standards
to
be
considered
generally
protective
of
human
health
under
RCRA
(
using
the
1985
cancer
slope
factor),
thereby
eliminating
the
need
for
separate
RCRA
standards
under
the
authority
of
RCRA
section
3005(
c)(
3)
and
40
CFR
270.10(
k).
Finally,
we
note
that
this
decision
is
not
inconsistent
with
EPA's
decision
not
to
promulgate
beyond­
the­
floor
standards
for
dioxin/
furan
for
hazardous
waste
burning
lightweight
aggregate
kilns,
cement
kilns,
and
incinerators
at
cost­
effectiveness
values
in
the
range
of
$
530,000
to
$
827,000
per
additional
gram
of
dioxin/
furan
TEQ
removed.
See
64
FR
at
52892,
52876,
and
52961.
In
those
cases,
EPA
determined
that
controlling
dioxin/
furan
emissions
from
a
level
of
0.40
ng
TEQ/
dscm
to
a
beyond­
the­
floor
level
of
0.20
ng
TEQ/
dscm
was
not
warranted
because
dioxin/
furan
levels
below
0.40
ng
TEQ/
dscm
are
generally
considered
to
be
below
the
level
of
health
risk
concern.
For
these
reasons,
we
believe
that
proposing
a
beyond­
the­
floor
standard
would
reduce
dioxin/
furan
emissions
by
only
0.06
grams
TEQ
per
year,
it
would
be
cost­
effective
atof
0.40
ng
TEQ/
dscm
is
warranted
notwithstanding
the
nonair
quality
health
and
environmental
impacts
and
energy
effects
identified
above
and
costs
of
approximately
$
1.3
million
per
additional
gram
of
dioxin/
furansdioxin/
furan
TEQ
removed.
he
particular
health
hazard
that
dioxin/
furan
pose,
0.40
ng
TEQ/
dscm
is
warranted
for
liquid
fuel­
fired
boilers
using
dry
air
pollution
control
systems.
.
b.
Beyond­
the­
Floor
Considerations
for
Boilers
Equipped
with
Wet
or
No
Control
Systems.
For
liquid
fuel­
fired
boilers
equipped
with
wet
or
with
no
air
pollution
control
systems,
we
also
evaluated
a
beyond­
the­
floor
level
of
0.20
ng
TEQ/
dscm
based
on
activated
carbon
injection.
The
national
incremental
annualized
compliance
cost
for
these
sources
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
Redline­
strikeout
highlighting
changes
made
during
OMB
review
58
Note
that
we
are
not
proposing
beyond­
the­
floor
particulate
matter
standards
for
incinerators,
cement
kilns,
lightweight
aggregate
kilns,
and
liquid
fuel­
fired
boilers
because
those
standards
would
have
a
cost­
effectiveness
of
$
12,000
to
$
80,000
per
ton
of
particulate
matter
removed,
substantially
higher
than
the
$
3,600
per
ton
cost­
effectiveness
of
a
beyond­
the­
floor
standard
for
solid
fuel­
fired
boilers.
fThese
data
were
recently
obtained
and
are
not
in
the
MACT
data
base.
See
"
Region
4
Boiler
Dioxin
Data,"
Excel
spreadsheet,
March
10,
2004.

203
$
550,000
and
would
provide
an
incremental
reduction
in
dioxin/
furan
emissions
beyond
the
MACT
floor
controls
of
0.12
grams
TEQ
per
year.
We
evaluated
the
nonair
quality
health
and
environmental
impacts
and
energy
effects
of
this
beyond­
the­
floor
standard
and
estimate
that
the
amount
of
hazardous
waste
generated
would
increase
by
100
tons
per
year,
an
additional
25
trillion
Btu
per
year
of
natural
gas
would
be
consumed,
an
additional
4
million
gallons
per
year
of
water
would
be
used,
and
electricity
consumption
would
increase
by
0.50
million
kW­
hours
per
year.
We
are
not
proposing
a
beyond­
the­
floor
standard
of
0.20
ng
TEQ/
dscm
for
liquid
boilers
that
use
a
wet
or
no
air
pollution
control
system
because
it
would
not
be
cost­
effective
at
$
4.6
million
per
gram
of
TEQ
removed.
We
are
also
considering
an
alternative
beyond­
the­
floor
standard
for
existing
liquid
fuel
boilers
with
wet
or
no
particulate
matter
control
devices
of
0.40
ng
TEQ/
dscm.
Although
all
but
one
source
for
which
we
have
data
are
currently
achieving
this
emission
level,
boilers
for
which
we
do
not
have
dioxin/
furan
emissions
data
may
have
emissions
higher
than
0.40
ng
TEQ/
dscm.
In
addition,
dioxin/
furan
emissions
from
a
given
boiler
may
vary
over
time.
Other
factors
that
may
contribute
substantially
to
dioxin/
furan
formation,
such
as
the
level
and
type
of
soot
on
boiler
tubes,
or
feeding
metals
that
catalyze
dioxin/
furan
formation
reactions,
differ
across
boilers
and
may
change
over
time
at
a
given
boiler.
Thus,
dioxin/
furan
levels
for
these
sources
may
be
higher
than
0.40
ng
TEQ/
dscm.
For
example,
we
recently
obtained
dioxin/
furan
emissions
data
for
a
liquid
fuel­
fired
boiler
equipped
with
a
wet
emission
control
system
documenting
emissions
of
1.4
ng
TEQ/
dscm.
58
To
control
dioxin/
furan
emissions
to
a
beyond­
the­
floor
standard
of
0.40
ng
TEQ/
dscm,
you
would
use
activated
carbon.

3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
The
calculated
floor
level
for
new
liquid
fuel
boilers
equipped
with
dry
air
pollution
control
systems
is
0.015
ng
TEQ/
dscm,
which
we
identified
using
the
Emissions
Approach.
This
is
anIf
dioxin/
furan
emissions
could
be
controlled
predominantly
by
controlling
the
gas
temperature
at
the
inlet
to
the
dry
particulate
matter
control
device,
this
would
be
the
emission
level
that
the
single
best
performing
source
could
be
expected
to
achieve
in
99
out
of
100
future
tests
when
operating
under
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
This
emission
level
may
not
be
replicable
by
this
source
and
duplicable
by
other
(
new)
sources,
however,
because
factors
other
than
flue
gas
temperature
control
at
the
control
device
may
affect
dioxin/
furan
emissions.
See
discussion
of
this
issue
in
the
context
of
the
floor
level
for
existing
sources.
Therefore,
we
propose
to
establish
the
floor
level
as
0.015
ng
TEQ/
dscm
or
control
of
flue
gas
temperature
not
to
exceed
400
°
F
at
the
inlet
to
the
Redline­
strikeout
highlighting
changes
made
during
OMB
review
204
particulate
matter
control
device.
As
previously
discussed,
we
believe
that
it
would
be
inappropriate
to
establish
a
numerical
dioxin/
furan
emission
floorsfloor
level
for
liquid
boilers
with
wet
or
with
no
air
pollution
control
systems.
Therefore,
we
propose
floor
control
for
these
units
as
good
combustion
practices
provided
by
complying
with
the
proposed
destruction
and
removal
efficiency
and
carbon
monoxide/
hydrocarbon
standards.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
We
evaluated
use
of
activated
carbon
injection
systems
as
beyond­
the­
floor
control
for
further
reduction
of
dioxin/
furan
emissions.
Activated
carbon
has
been
demonstrated
for
controlling
dioxin/
furan
in
various
combustion
applications.
Wa.
Beyond­
the­
Floor
Considerations
for
Boilers
Equipped
with
Dry
Control
Systems.
For
liquid
fuel­
fired
boilers
using
dry
air
pollution
control
equipment,
we
evaluated
a
beyond­
thefloor
level
of
0.01
ng
TEQ/
dscm
using
activated
carbon
for
liquid
fuel­
fired
boilers
equipped
with
dry
air
pollution
control
systemsinjection.
The
national
incremental
annualized
compliance
cost
for
a
source
with
an
average
gas
flowrate
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
0.15
million
and
would
provide
an
incremental
reduction
in
dioxin/
furan
emissions
beyond
the
MACT
floor
controls
of
0.005
grams
TEQ
per
year.
We
evaluated
the
nonair
quality
health
and
environmental
impacts
and
energy
effects
of
this
beyond­
the­
floor
standard
and
estimate
that,
for
a
new
liquid
fuel­
fired
boiler
with
average
gas
flowrate,
the
amount
of
hazardous
waste
generated
would
increase
by
120
tons
per
year
and
electricity
consumption
would
increase
by
0.1
million
kW­
hours
per
year.
After
considering
these
impacts
and
costs
of
approximately
$
32
million
per
additional
gram
of
dioxin/
furan
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
of
0.01
ng
TEQ/
dscm
for
liquid
fuel­
fired
boilers
using
dry
air
pollution
control
systems.
We
are
also
considering
an
alternative
beyond­
the­
floor
standard
of
0.40
ng
TEQ/
dscm
for
new
liquid
fuel
boilers
equipped
with
a
dry
particulate
matter
control
device.
A
new
source
that
achieves
the
floor
level
by
controlling
the
gas
temperature
at
the
inlet
to
the
dry
particulate
matter
control
device
to
400

F
may
have
dioxin/
furan
emissions
at
levels
far
exceeding
0.40
ng
TEQ/
dscm.
See
discussion
above
regarding
factors
other
than
gas
temperature
at
the
control
device
that
can
affect
dioxin/
furan
emissions
from
liquid
fuel­
fired
boilers
(
and
discussion
of
emissions
of
2.4
ng
TEQ/
dscm
for
a
boiler
operating
a
fabric
filter
at
410

F).
Therefore,
it
may
be
appropriate
to
establish
a
beyond­
the­
floor
standard
to
limit
emissions
to
0.40
ng
TEQ/
dscm
based
on
use
of
activated
carbon
injection.
We
also
note
that
this
beyond­
the­
floor
standard
may
be
appropriate
to
ensure
that
emission
levels
from
new
sources
do
not
exceed
the
proposed
0.40
ng
TEQ/
dscm
beyond­
the­
floor
standard
for
existing
sources.
Because
standards
for
new
sources
are
based
on
the
single
best
performing
source
while
standards
for
existing
sources
are
based
on
the
average
of
the
best
12%
(
or
best
5)
performing
sources,
standards
for
new
sources
should
not
be
less
stringent
than
standards
for
existing
sources.

b.
Beyond­
the­
Floor
Considerations
for
Boilers
Equipped
with
Wet
or
No
Control
Systems.
We
evaluated
a
beyond­
the­
floor
level
of
0.20
ng
TEQ/
dscm
for
liquid
fuel­
fired
boilers
Redline­
strikeout
highlighting
changes
made
during
OMB
review
59
As
information,
EPA
did
not
propose
MACT
emission
standards
for
mercury
for
liquid
fuel­
fired
boilers
that
do
not
burn
hazardous
waste.
See
68
FR
1660
(
Jan.
13,2003).
Note
that,
in
today's
rule,
we
propose
to
control
mercury
only
in
hazardous
waste
fuels,
an
option
obviously
not
available
to
boilers
that
do
not
burn
hazardous
waste.

205
equipped
with
wet
or
with
no
air
pollution
control
systems
based
on
use
of
activated
carbon
injection.
The
national
incremental
annualized
compliance
cost
for
a
source
with
average
gas
flowrate
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
0.15
million
and
would
provide
an
incremental
reduction
in
dioxin/
furan
emissions
beyond
the
MACT
floor
controls
of
0.06
grams
TEQ
per
year.
We
evaluated
the
nonair
quality
health
and
environmental
impacts
and
energy
effects
of
this
beyond­
the­
floor
standard
and
estimate
that,
for
a
source
with
average
gas
flowrate,
the
amount
of
hazardous
waste
generated
would
increase
by
120
tons
per
year
and
electricity
consumption
would
increase
by
0.1
million
kW­
hours
per
year.
After
considering
these
impacts
and
costs
of
approximately
$
2.4
million
per
additional
gram
of
dioxin/
furan
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
for
liquid
fuel­
fired
boilers
using
a
wet
or
no
air
pollution
control
system.
We
are
also
considering
an
alternative
beyond­
the­
floor
standard
of
0.40
ng
TEQ/
dscm
for
new
liquid
fuel
boilers
equipped
with
wet
or
with
no
air
pollution
control
systems.
A
new
source
that
achieves
the
floor
level­­
compliance
with
the
standards
for
carbon
monoxide/
hydrocarbon
and
destruction
and
removal
efficiency­­
may
have
high
dioxin/
furan
emissions
at
levels
far
exceeding
0.40
ng
TEQ/
dscm.
See
discussion
above
regarding
factors
other
than
gas
temperature
at
the
control
device
that
can
affect
dioxin/
furan
emissions
from
liquid
fuel­
fired
boilers.
Therefore,
it
may
be
appropriate
to
establish
a
beyond­
the­
floor
standard
to
limit
emissions
to
0.40
ng
TEQ/
dscm
based
on
use
of
activated
carbon.

B.
What
Is
the
Rationale
for
the
Proposed
Standards
for
Mercury?
We
propose
to
establish
standards
for
existing
liquid
fuel­
fired
boilers
that
limit
emissions
of
mercury
to
3.7E­
6
lbs
mercury
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste.
The
proposed
standards
for
new
sources
would
be
3.8E­
7
lbs
mercury
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste.
59
These
standards
are
expressed
as
hazardous
waste
thermal
emission
concentrations
because
liquid
fuel­
fired
boilers
burn
hazardous
waste
for
energy
recovery.
See
discussion
in
Part
Two,
Section
IV.
B
of
the
preamble.
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
MACT
floor
for
existing
sources
is
3.7E­
6
lbs
mercury
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste,
which
is
based
primarily
by
controlling
the
feed
concentration
of
mercury
in
the
hazardous
waste.
Approximately
11%
of
liquid
boilers
also
use
wet
scrubbers
that
can
control
emissions
of
mercury.
We
have
normal
emissions
data
within
the
range
of
normal
emissions
for
32%
of
the
Redline­
strikeout
highlighting
changes
made
during
OMB
review
60
Several
owners
and
operators
have
used
the
emissions
data
as
"
data
in
lieu
of
testing"
emissions
from
other,
identical
boilers
at
the
same
facility.
For
purposes
of
identifying
the
number
of
boilers
represented
in
this
paragraph,
the
percentage
includes
the
data­
in­
lieu
sources.

206
sources.
60
The
normal
mercury
stack
emissions
in
our
data
base
are
all
less
than
7
ug/
dscm.
These
emissions
are
expressed
as
mass
of
mercury
(
from
all
feedstocks)
per
unit
of
stack
gas.
Hazardous
waste
thermal
emissions,
available
for
12%
of
sources,
range
from
1.0E­
7
to
1.0E­
5
lbs
mercury
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste.
Hazardous
waste
thermal
emissions
represent
the
mass
of
mercury
contributed
by
the
hazardous
waste
per
million
Btu
contributed
by
the
hazardous
waste.
To
identify
the
MACT
floor,
we
evaluated
all
normal
emissions
data
using
the
Emissions
Approach.
The
calculated
floor
is
3.7E­
6
lbs
mercury
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste.
This
is
an
emission
level
that
the
average
of
the
best
performing
sources
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
We
estimate
that
this
floor
level
is
being
achieved
by
40%
of
sources
and
would
reduce
mercury
emissions
by
0.68
tons
per
year.
Because
the
floor
level
is
based
on
normal
emissions
data,
compliance
would
be
documented
by
complying
with
a
hazardous
waste
mercury
thermal
feed
concentration
on
an
annual
rolling
average.
See
discussion
in
Part
Two,
Section
XIV.
F
below.
We
did
not
use
the
SRE/
Feed
Approach
to
identify
the
floor
level
because
the
vast
majority
of
mercury
feed
levels
in
the
hazardous
waste
and
the
emissions
measurements
did
not
have
detectable
concentrations
of
mercury.
Given
that
a
system
removal
efficiency,
or
SRE,
is
the
percentage
of
mercury
emitted
compared
to
the
amount
fed,
we
concluded
that
it
would
be
inappropriate
to
base
this
analysis
on
SREs
that
were
derived
from
measurements
below
detectable
levels.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
identified
two
potential
beyond­
the­
floor
techniques
for
control
of
mercury:
(
1)
activated
carbon
injection;
and
(
2)
control
of
mercury
in
the
hazardous
waste
feed.
For
reasons
discussed
below,
we
are
not
proposing
a
beyond­
the­
floor
standard
for
mercury.
a.
Use
of
Activated
Carbon
Injection.
We
evaluated
activated
carbon
injection
as
beyond­
the­
floor
control
for
further
reduction
of
mercury
emissions.
Activated
carbon
has
been
demonstrated
for
controlling
mercury
in
several
combustion
applications;
however,
currently
no
liquid
fuel
boilers
burning
hazardous
waste
uses
activated
carbon
injection.
We
evaluated
a
beyond­
the­
floor
level
of
1.1E­
6
lbs
mercury
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste.
The
national
incremental
annualized
compliance
cost
for
liquid
fuel­
fired
boilers
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
12
million
and
would
provide
an
incremental
reduction
in
mercury
emissions
beyond
the
MACT
floor
controls
of
0.097
tons
per
year.
We
evaluated
nonair
quality
health
and
environmental
impacts
and
energy
effects
of
using
activated
carbon
injection
to
Redline­
strikeout
highlighting
changes
made
during
OMB
review
61
We
note
that
the
beyond­
the­
floor
dioxin/
furan
standard
we
propose
for
liquid
fuel­
fired
boilers
equipped
with
dry
particulate
matter
control
devices
would
also
provide
no­
cost
beyond­
the­
floor
mercury
control
for
sources
that
use
activated
carbon
injection
to
control
dioxin/
furan.
If
such
sources
achieve
the
beyond­
the­
floor
dioxin/
furan
standard
by
other
means
(
control
of
temperature
at
the
inlet
to
the
control
device;
control
of
feedrate
of
metals
that
may
catalyze
formation
of
dioxin/
furan),
however,
collateral
reductions
in
mercury
emissions
would
not
be
realized.

207
meet
this
beyond­
the­
floor
emission
level
and
estimate
that
the
amount
of
hazardous
waste
generated
would
increase
by
4,800
tons
per
year
and
that
sources
would
consume
an
additional
44
trillion
Btu
per
year
of
natural
gas
and
use
an
additional
9.6
million
kW­
hours
per
year
beyond
the
requirements
to
achieve
the
floor
level.
Therefore,
based
on
these
factors
and
costs
of
approximately
$
124
million
per
additional
ton
of
mercury
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
activated
carbon
injection.
61
b.
Feed
Control
of
Mercury
in
the
Hazardous
Waste.
We
also
evaluated
a
beyond­
thefloor
level
of
3.0E­
6
lbs
mercury
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste,
which
represents
a
20%
reduction
from
the
floor
level.
The
national
incremental
annualized
compliance
cost
for
liquid
fuel­
fired
boilers
to
meet
this
beyondthe
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
4.2
million
and
would
provide
an
incremental
reduction
in
mercury
emissions
beyond
the
MACT
floor
controls
of
0.036
tons
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
are
not
significant
factors
for
feedrate
control.
Therefore,
based
on
these
factors
and
costs
of
approximately
$
115
million
per
additional
ton
of
mercury
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
feed
control
of
mercury
in
the
hazardous
waste.
For
the
reasons
discussed
above,
we
do
not
propose
a
beyond­
the­
floor
standard
for
mercury
for
existing
sources.
We
propose
a
standard
based
on
the
floor
level:
3.7E­
6
lbs
mercury
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste.
3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
The
MACT
floor
for
new
sources
for
mercury
would
be
3.8E­
7
lbs
mercury
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste
and
would
be
implemented
as
an
annual
average
because
it
is
based
on
normal
emissions
data.
This
is
an
emission
level
that
the
single
best
performing
source
identified
with
the
Emissions
Approach
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
operating
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
We
evaluated
activated
carbon
injection
as
beyond­
the­
floor
control
to
achieve
an
emission
level
of
2.0E­
7
lbs
mercury
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste.
The
incremental
annualized
compliance
cost
for
a
new
liquid
fuel­
fired
boiler
with
average
gas
flowrate
to
meet
this
beyond­
the­
floor
level,
rather
than
comply
with
the
floor
level,
would
be
approximately
$
0.15
million
and
would
provide
an
Redline­
strikeout
highlighting
changes
made
during
OMB
review
62
As
information,
EPA
proposed
MACT
standards
for
particulate
matter
for
solid
fuel­
fired
industrial,
commercial,
and
institutional
boilers
that
do
not
burn
hazardous
waste
of
0.035
gr/
dscf
for
existing
sources
and
0.013
gr/
dscf
for
new
sources.

208
incremental
reduction
in
mercury
emissions
of
less
than
0.0002
tons
per
year,
for
a
costeffectiveness
of
$
1
billion
per
ton
of
mercury
removed.
We
evaluated
the
nonair
quality
health
and
environmental
impacts
and
energy
effects
of
this
beyond­
the­
floor
standard
and
estimate
that,
for
a
new
liquid
fuel­
fired
boiler
with
average
gas
flowrate,
the
amount
of
hazardous
waste
generated
would
increase
by
120
tons
per
year
and
electricity
consumption
would
increase
by
0.1
million
kW­
hours
per
year.
Although
nonair
quality
health
and
environmental
impacts
and
energy
effects
are
not
significant
factors,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
activated
carbon
injection
for
new
sources
because
it
would
not
be
cost­
effective.
Therefore,
we
propose
a
mercury
standard
based
on
the
floor
level:
3.8E­
7
lbs
mercury
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste.
C.
What
Is
the
Rationale
for
the
Proposed
Standards
for
Particulate
Matter?
The
proposed
standards
for
particulate
matter
for
liquid
fuel­
fired
boilers
are
59
mg/
dscm
(
0.026
gr/
dscf)
for
existing
sources
and
17
mg/
dscm
(
0.0076
gr/
dscf)
for
new
sources.
62
The
particulate
matter
standard
serves
as
a
surrogate
for
nonenumerated
HAP
metal
emissions
attributable
to
the
hazardous
waste
fuel
burned
in
the
boiler.
Although
the
particulate
matter
standard
would
also
control
nonmercury
HAP
metal
from
nonhazardous
waste
fuels,
the
natural
gas
or
fuel
oil
these
boilers
burn
as
primary
or
auxiliary
fuel
do
not
contain
significant
levels
of
metal
HAP.
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
Few
liquid
fuel­
fired
boilers
are
equipped
particulate
matter
control
equipment
such
as
electrostatic
precipitators
and
baghouses,
and,
therefore,
many
sources
control
particulate
matter
emissions
by
limiting
the
ash
content
of
the
hazardous
waste.
We
have
compliance
test
emissions
data
from
nearly
all
liquid
boilers
representing
maximum
allowable
emissions.
Particulate
emissions
range
from
0.0008
to
0.078
gr/
dscf.
To
identify
the
floor
level,
we
evaluated
the
compliance
test
emissions
data
associated
with
the
most
recent
test
campaign
using
the
APCD
Approach.
The
calculated
floor
is
72
mg/
dscm
(
0.032
gr/
dscf),
which
considers
emissions
variability.
This
is
an
emission
level
that
the
average
of
the
performing
sources
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
operating
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
We
estimate
that
this
floor
level
is
being
achieved
by
44%
of
sources
and
would
reduce
particulate
matter
emissions
by
1,200
tons
per
year.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
evaluated
use
of
fabric
filters
to
improve
particulate
matter
control
to
achieve
a
beyond­
the­
floor
standard
of
36
mg/
dscm
(
0.016
gr/
dscf).
The
national
incremental
annualized
compliance
cost
for
liquid
fuel­
fired
boilers
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
16
million
and
would
provide
an
incremental
reduction
in
particulate
matter
emissions
beyond
the
MACT
floor
controls
of
520
tons
per
year.
Redline­
strikeout
highlighting
changes
made
during
OMB
review
63
The
source
also
is
equipped
with
a
high
efficiency
particulate
air
(
HEPA)
filter.

209
We
evaluated
the
nonair
quality
health
and
environmental
impacts
and
energy
effects
of
this
beyond­
the­
floor
standard
and
estimate
that
the
amount
of
hazardous
waste
generated
would
increase
by
520
tons
per
year
and
electricity
consumption
would
increase
by
13
million
kW­
hours
per
year.
After
considering
these
factors
and
costs
of
approximately
$
30,000
per
additional
ton
of
particulate
matter
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard.
For
the
reasons
discussed
above,
we
propose
a
standard
for
particulate
matter
for
existing
liquid
fuel­
fired
boilers
based
on
the
floor
level:
72
mg/
dscm
(
0.032
gr/
dscf).
3.
What
Is
the
Rational
for
the
MACT
Floor
for
New
Sources?
MACT
floor
for
new
sources
would
be
17
mg/
dscm
(
0.0076
gr/
dscf),
considering
emissions
variability.
This
is
an
emission
level
that
the
single
best
performing
source
identified
by
the
APCD
Approach
(
i.
e.,
the
source
using
a
fabric
filter63
with
the
lowest
emissions)
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
operating
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
We
evaluated
use
of
an
advanced
fabric
filter
using
high
efficiency
membrane
bag
material
and
a
low
air
to
cloth
ratio
to
achieve
a
beyond­
the­
floor
emission
level
of
9
mg/
dscm
(
0.0040
gr/
dscf).
The
incremental
annualized
cost
for
a
new
liquid
fuel­
fired
boiler
with
average
gas
flowrate
to
meet
this
beyond­
the­
floor
level,
rather
than
comply
with
the
floor
level,
would
be
approximately
$
0.15
million
and
would
provide
an
incremental
reduction
in
particulate
emissions
of
approximately
2.9
tons
per
year,
for
a
cost­
effectiveness
of
$
53,000
per
ton
of
particulate
matter
removed.
We
evaluated
the
nonair
quality
health
and
environmental
impacts
and
energy
effects
of
this
beyond­
the­
floor
standard
and
estimate
that,
for
a
new
liquid
fuel­
fired
boiler
with
average
gas
flowrate,
the
amount
of
hazardous
waste
generated
would
increase
by
3
tons
per
year
and
electricity
consumption
would
increase
by
0.54
million
kW­
hours
per
year.
Considering
these
factors
and
cost­
effectiveness,
we
conclude
that
a
beyond­
the­
floor
standard
of
9
mg/
dscm
is
not
warranted.
For
the
reasons
discussed
above,
we
propose
a
floor­
based
standard
for
particulate
matter
for
new
liquid
fuel­
fired
boilers:
9.8
mg/
dscm
(
0.0043
gr/
dscf)
D.
What
Is
the
Rationale
for
the
Proposed
Standards
for
Semivolatile
Metals?
We
propose
a
standard
for
existing
liquid
fuel­
fired
boilers
that
limits
emissions
of
semivolatile
metals
(
cadmium
and
lead,
combined)
to
1.1E­
5
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste.
The
proposed
standard
for
new
sources
is
4.3E­
6
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste.
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
MACT
floor
for
existing
sources
is
1.1E­
5
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste,
which
is
based
on
particulate
matter
control
(
for
those
few
sources
using
a
control
device)
and
controlling
the
feedrate
of
semivolatile
metals
in
the
hazardous
waste.
Redline­
strikeout
highlighting
changes
made
during
OMB
review
64
Several
owners
and
operators
have
used
the
emissions
data
as
"
data
in
lieu
of
testing"
emissions
from
other,
identical
boilers
at
the
same
facility.
For
purposes
of
identifying
the
number
of
boilers
represented
in
this
paragraph,
the
percentages
include
the
data­
in­
lieu
sources.

65
We
propose
to
use
the
Emissions
Approach
rather
than
the
SRE/
Feed
approach
because
our
data
base
is
comprised
of
emissions
obtained
during
normal
rather
than
compliance
test
operations.
Because
of
the
relatively
low
semivolatile
metal
feedrates
during
normal
operations,
we
are
concerned
that
the
system
removal
efficiencies
that
we
would
calculate
may
be
inaccurate
(
e.
g.,
sampling
and
analysis
imprecision
at
low
feed
rates
can
have
a
substantial
impact
on
calculated
system
removal
efficiencies).

210
We
have
emissions
data
within
the
range
of
normal
emissions
for
nearly
40%
of
the
sources.
64
The
normal
semivolatile
stack
emissions
in
our
database
range
from
less
than
1
to
46
ug/
dscm.
These
emissions
are
expressed
conventionally
as
mass
of
semivolatile
metals
(
from
all
feedstocks)
per
unit
of
stack
gas.
Hazardous
waste
thermal
emissions,
available
for
25%
of
sources,
range
from
1.2E­
6
to
4.8E­
5
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
of
the
hazardous
waste.
We
identified
a
MACT
floor
of
1.1E­
5
expressed
as
a
hazardous
waste
thermal
emission
by
applying
the
Emissions
Approach
to
the
normal
hazardous
waste
thermal
emissions
data.
65
This
is
an
emission
level
that
the
average
of
the
best
performing
sources
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
We
estimate
that
this
floor
level
is
being
achieved
by
33%
of
sources
and
would
reduce
semivolatile
metals
emissions
by
1.7
tons
per
year.
Because
the
floor
level
is
based
on
normal
emissions
data,
compliance
would
be
documented
by
complying
with
a
hazardous
waste
mercury
thermal
feed
concentration
on
an
annual
rolling
average.
See
discussion
in
Part
Two,
Section
XIV.
F
below.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
identified
two
potential
beyond­
the­
floor
techniques
for
control
of
semivolatile
metals:
(
1)
improved
particulate
matter
control;
and
(
2)
control
of
mercury
in
the
hazardous
waste
feed.
For
reasons
discussed
below,
we
are
not
proposing
a
beyond­
the­
floor
standard
for
semivolatile
metals.
a.
Improved
Particulate
Matter
Control.
We
evaluated
installation
of
a
new
fabric
filter
or
improved
design,
operation,
and
maintenance
of
the
existing
electrostatic
precipitator
and
fabric
filter
as
beyond­
the­
floor
control
for
further
reduction
of
semivolatile
metals
emissions.
We
evaluated
a
beyond­
the­
floor
level
of
5.5E­
6
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste.
The
national
incremental
annualized
compliance
cost
for
liquid
fuel­
fired
boilers
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
6.5
million
and
would
provide
an
incremental
reduction
in
semivolatile
metals
emissions
beyond
the
MACT
floor
controls
of
0.06
Redline­
strikeout
highlighting
changes
made
during
OMB
review
66
We
use
the
Emissions
Approach
rather
than
the
SRE/
Feed
Approach
when
we
use
normal
rather
than
compliance
test
data
to
establish
the
standard,
as
discussed
previously.

211
tons
per
year.
We
evaluated
nonair
quality
health
and
environmental
impacts
and
energy
effects
and
determined
that
this
beyond­
the­
floor
option
would
increase
the
amount
of
hazardous
waste
generated
by
approximately
45
tons
per
year
and
would
increase
electricity
usage
by
0.8
million
kW­
hours
per
year.
After
considering
these
factors
and
costs
of
approximately
$
100
million
per
additional
ton
of
semivolatile
metals
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
improved
particulate
matter
control.
b.
Feed
Control
of
Semivolatile
Metals
in
the
Hazardous
Waste.
We
also
evaluated
a
beyond­
the­
floor
level
of
8.8E­
6
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste,
which
represents
a
20%
reduction
from
the
floor
level.
The
national
incremental
annualized
compliance
cost
for
liquid
fuel­
fired
boilers
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
4.8
million
and
would
provide
an
incremental
reduction
in
semivolatile
metals
emissions
beyond
the
MACT
floor
controls
of
0.06
tons
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
are
not
significant
factors
for
feedrate
control.
Therefore,
considering
these
factors
and
costs
of
approximately
$
81
million
per
additional
ton
of
semivolatile
metals
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
feed
control
of
semivolatile
metals
in
the
hazardous
waste.
For
the
reasons
discussed
above,
we
propose
a
floor
standard
for
semivolatile
metals
for
existing
liquid
fuel­
fired
boilers
of
1.1E­
5
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste.
3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
The
MACT
floor
for
new
sources
for
semivolatile
metals
would
be
4.3E­
6
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste.
This
is
an
emission
level
that
the
single
best
performing
source
identified
with
the
Emissions
Approach66
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
operating
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
Because
the
floor
level
is
based
on
normal
emissions
data,
compliance
would
be
documented
by
complying
with
a
hazardous
waste
mercury
thermal
feed
concentration
on
an
annual
rolling
average.
See
discussion
in
Part
Two,
Section
XIV.
F
below.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
We
evaluated
a
beyond­
the­
floor
level
of
2.1E­
6
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste
based
on
an
advanced
fabric
filter
using
high
efficiency
membrane
bag
material
and
a
low
air
to
cloth
ratio.
The
incremental
annualized
compliance
cost
for
a
new
liquid
fuel­
fired
boiler
with
average
gas
flowrate
to
meet
this
beyond­
the­
floor
level,
rather
than
comply
with
the
floor
level,
would
be
approximately
$
0.15
million
and
would
provide
an
incremental
reduction
in
semivolatile
metals
emissions
of
less
than
0.002
tons
per
year,
for
a
cost­
effectiveness
of
$
87
million
per
ton
of
Redline­
strikeout
highlighting
changes
made
during
OMB
review
212
semivolatile
metals
removed.
We
evaluated
the
nonair
quality
health
and
environmental
impacts
and
energy
effects
of
this
beyond­
the­
floor
standard
and
estimate
that,
for
a
new
liquid
fuel­
fired
boiler
with
average
gas
flowrate,
the
amount
of
hazardous
waste
generated
would
increase
by
2
tons
per
year
and
electricity
consumption
would
increase
by
0.54
million
kW­
hours
per
year.
Considering
these
factors
and
cost­
effectiveness,
we
conclude
that
a
beyond­
the­
floor
standard
is
not
warranted.
Therefore,
we
propose
a
semivolatile
metals
standard
based
on
the
floor
level:
4.3E­
6
lbs
semivolatile
metals
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste
for
new
sources.
E.
What
Is
the
Rationale
for
the
Proposed
Standards
for
Chromium?
We
propose
to
establish
standards
for
existing
and
new
liquid
fuel­
fired
boilers
that
limit
emissions
of
chromium
to
1.1E­
4
lbs
and
3.6E­
5
lbs
chromium
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste,
respectively.
We
propose
to
establish
emission
standards
on
chromium­
only
because
our
data
base
has
very
limited
compliance
test
data
on
emissions
of
total
low
volatile
metals:
arsenic,
beryllium,
and
chromium.
We
have
compliance
test
data
on
only
two
sources
for
total
low
volatile
metals
emissions
while
we
have
compliance
test
data
for
12
sources
for
chromium­
only.
Although
we
have
total
low
volatile
metals
emissions
for
12
sources
when
operating
under
normal
operations,
we
prefer
to
use
compliance
test
data
to
establish
the
floor
because
they
better
address
emissions
variability.
By
establishing
a
low
volatile
metal
floor
based
on
chromium
emissions
only
we
are
relying
on
the
particulate
matter
standard
to
control
the
other
enumerated
low
volatile
metals­­
arsenic
and
beryllium­­
as
well
as
nonenumerated
metal
HAP.
We
request
comment
on
this
approach
and
note
that,
as
discussed
below,
an
alternative
approach
would
be
to
establish
a
MACT
floor
based
on
normal
emissions
data
for
all
three
enumerated
low
volatile
metals.
We
request
comment
on
whether
the
compliance
test
data
for
chromium­
only
are
appropriate
for
establishing
a
MACT
floor
for
chromium.
We
are
concerned
that
some
sources
in
our
data
base
may
have
used
chromium
as
a
surrogate
for
arsenic
and
beryllium
during
RCRA
compliance
testing
such
that
their
chromium
emissions
may
be
more
representative
of
their
total
low
volatile
metals
emissions
than
only
chromium.
If
we
determine
this
to
be
the
case,
we
could
apply
the
floor
we
calculate
using
chromium
emissions
to
total
low
volatile
metal
emissions.
Alternatively,
we
could
use
the
normal
emissions
data
we
have
on
12
sources
and
our
MACT
methodology
to
establish
a
total
low
volatile
metals
floor.
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
MACT
floor
for
existing
sources
is
1.1E­
4
lbs
chromium
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste,
which
is
based
on
particulate
matter
control
(
for
those
few
sources
using
a
control
device)
and
controlling
the
feed
concentration
of
chromium
in
the
hazardous
waste.
We
have
compliance
test
emissions
data
representing
maximum
emissions
for
Redline­
strikeout
highlighting
changes
made
during
OMB
review
67
Several
owners
and
operators
have
used
the
emissions
data
as
"
data
in
lieu
of
testing"
emissions
from
other,
identical
boilers
at
the
same
facility.
For
purposes
of
identifying
the
number
of
boilers
represented
in
this
paragraph,
the
percentages
include
the
data­
in­
lieu
sources.

213
approximately
17%
of
the
sources.
67
The
compliance
test
chromium
stack
emissions
in
our
database
range
from
2
to
900
ug/
dscm.
These
emissions
are
expressed
as
mass
of
chromium
(
from
all
feedstocks)
per
unit
of
stack
gas.
Hazardous
waste
thermal
emissions,
available
for
13%
of
sources,
range
from
3.2E­
6
to
8.8E­
4
lbs
chromium
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste.
To
identify
the
floor
level,
we
evaluated
all
compliance
test
thermal
emissions
data
using
the
SRE/
Feed
Approach
(
see
discussion
in
Section
VI.
C
above).
The
calculated
floor
is
1.1E­
4
lbs
chromium
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste
feed,
which
considers
emissions
variability.
This
is
an
emission
level
that
the
average
of
the
best
performing
sources
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
We
estimate
that
this
floor
level
is
being
achieved
by
36%
of
sources
and
would
reduce
chromium
emissions
by
9.4
tons
per
year.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
identified
two
potential
beyond­
the­
floor
techniques
for
control
of
chromium
emissions:
(
1)
use
of
a
fabric
filter
to
improve
particulate
matter
control;
and
(
2)
control
of
chromium
in
the
hazardous
waste
feed.
For
reasons
discussed
below,
we
are
not
proposing
a
beyond­
the­
floor
standard
for
chromium.
a.
Use
of
a
Fabric
Filter
to
Improve
Particulate
Matter
Control.
We
evaluated
use
of
a
fabric
filter
as
beyond­
the­
floor
control
for
further
reduction
of
chromium
emissions.
We
evaluated
a
beyond­
the­
floor
level
of
5.5E­
5
lbs
chromium
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste.
The
national
incremental
annualized
compliance
cost
for
liquid
fuel­
fired
boilers
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
5.9
million
and
would
provide
an
incremental
reduction
in
chromium
emissions
beyond
the
MACT
floor
controls
of
0.50
tons
per
year.
We
evaluated
nonair
quality
health
and
environmental
impacts
and
energy
effects
and
determined
that
this
beyond­
the­
floor
option
would
increase
the
amount
of
hazardous
waste
generated
by
approximately
160
tons
per
year
and
would
increase
electricity
usage
by
3.0
million
kW­
hours
per
year.
Based
on
these
impacts
and
a
cost
of
approximately
$
12
million
per
additional
ton
of
chromium
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
improved
particulate
matter
control.
b.
Feed
Control
of
Chromium
in
the
Hazardous
Waste.
We
evaluated
additional
feed
control
of
chromium
in
the
hazardous
waste
as
a
beyond­
the­
floor
control
technique
to
reduce
floor
emission
levels
by
25%
to
achieve
a
standard
of
8.8E­
5
lbs
chromium
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste.
This
beyond­
thefloor
level
of
control
would
reduce
chromium
by
an
additional
0.20
tons
per
year
at
a
cost­
Redline­
strikeout
highlighting
changes
made
during
OMB
review
214
effectiveness
of
$
22
million
per
ton
of
chromium
removed.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
are
not
significant
factors
for
feedrate
control.
We
conclude
that
use
of
additional
hazardous
waste
chromium
feedrate
control
would
not
be
costeffective
and
are
not
proposing
a
beyond­
the­
floor
standard
based
on
this
control
technique.
For
the
reasons
discussed
above,
we
do
not
propose
a
beyond­
the­
floor
standard
for
chromium.
Consequently,
we
propose
to
establish
the
emission
standard
for
existing
liquid
fuelfired
boilers
at
the
floor
level:
a
hazardous
waste
thermal
emission
standard
of
1.1E­
4
lbs
chromium
emissions
attributable
to
hazardous
waste
per
million
Btu
of
hazardous
waste
feed.
3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
The
MACT
floor
for
new
sources
for
chromium
would
be
3.6E­
5
lbs
chromium
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste
feed.
This
is
an
emission
level
that
the
single
best
performing
source
identified
with
the
SRE/
Feed
Approach
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
operating
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
We
evaluated
use
of
an
advanced
fabric
filter
using
high
efficiency
membrane
bag
material
and
a
low
air
to
cloth
ratio
as
beyond­
the­
floor
control
to
reduce
chromium
emissions
to
a
beyond­
the­
floor
level
of
1.8E­
5
lbs
chromium
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste.
The
incremental
annualized
compliance
cost
for
a
new
liquid
fuel­
fired
boiler
with
average
gas
flowrate
to
meet
this
beyond­
the­
floor
level,
rather
than
comply
with
the
floor
level,
would
be
approximately
$
0.15
million
and
would
provide
an
incremental
reduction
in
chromium
emissions
of
0.014
tons
per
year,
for
a
cost­
effectiveness
of
$
11
million
per
ton
of
chromium
removed.
We
evaluated
the
nonair
quality
health
and
environmental
impacts
and
energy
effects
of
this
beyond­
the­
floor
standard
and
estimate
that,
for
a
new
liquid
fuel­
fired
boiler
with
average
gas
flowrate,
the
amount
of
hazardous
waste
generated
would
increase
by
2
tons
per
year
and
electricity
consumption
would
increase
by
0.54
million
kWhours
per
year.
Considering
these
factors
and
cost­
effectiveness,
we
conclude
that
a
beyond­
thefloor
standard
is
not
warranted.
Therefore,
we
propose
a
chromium
emission
standard
for
new
sources
based
on
the
floor
level:
3.6E­
5
lbs
chromium
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste
feed.
F.
What
Is
the
Rationale
for
the
Proposed
Standards
for
Total
Chlorine?
We
are
proposing
to
establish
a
standard
for
existing
liquid
fuel­
fired
boilers
that
limit
emissions
of
hydrogen
chloride
and
chlorine
gas
(
i.
e.,
total
chlorine)
to
2.5E­
2
lbs
total
chlorine
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste.
The
proposed
standard
for
new
sources
would
be
7.2E­
4
lbs
total
chlorine
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste.
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
Most
liquid
fuel­
fired
boilers
that
burn
hazardous
waste
do
not
have
back­
end
controls
such
as
wet
scrubbers
for
total
chlorine
control.
For
these
sources,
total
chlorine
emissions
are
controlled
by
most
sources
by
controlling
the
feedrate
of
chlorine
in
the
hazardous
waste
feed.
Approximately
15%
of
sources
use
wet
scrubbing
systems
to
control
total
chlorine
emissions.
Redline­
strikeout
highlighting
changes
made
during
OMB
review
215
We
have
compliance
test
data
representing
maximum
emissions
for
40%
of
the
boilers.
Total
chlorine
emissions
range
from
less
than
1
to
900
ppmv.
Hazardous
waste
thermal
emissions,
available
for
27%
of
boilers,
range
from
1.00E­
4
to
1.4
lbs
total
chlorine
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste.
The
calculated
floor
is
2.5E­
2
lbs
total
chlorine
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste
using
the
SRE/
Feed
Approach
to
identify
the
best
performing
sources
(
see
discussion
in
Section
VI.
C
above).
This
is
an
emission
level
that
the
average
of
the
performing
sources
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
operating
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
We
estimate
that
this
floor
level
is
being
achieved
by
70%
of
sources
and
would
reduce
total
chlorine
emissions
by
660
tons
per
year.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
identified
two
potential
beyond­
the­
floor
techniques
for
control
of
total
chlorine
emissions:
(
1)
use
of
a
wet
scrubber;
and
(
2)
control
of
chlorine
in
the
hazardous
waste
feed.
For
reasons
discussed
below,
we
are
not
proposing
a
beyond­
the­
floor
standard
for
total
chlorine.
a.
Use
of
Wet
Scrubbing.
We
considered
a
beyond­
the­
floor
standard
of
1.3E­
2
lbs
total
chlorine
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste
based
on
wet
scrubbing
to
reduce
emissions
beyond
the
floor
level
by
50
percent.
The
national
incremental
annualized
compliance
cost
for
liquid
fuel­
fired
boilers
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
7.8
million
and
would
provide
an
incremental
reduction
in
total
chlorine
emissions
beyond
the
MACT
floor
controls
of
430
tons
per
year.
We
evaluated
nonair
quality
health
and
environmental
impacts
and
energy
effects
and
determined
that
this
beyond­
the­
floor
option
would
increase
both
the
amount
of
hazardous
wastewater
generated
and
water
usage
by
approximately
3.2
billion
gallons
per
year
and
would
increase
electricity
usage
by
30
million
kW­
hours
per
year.
Considering
these
impacts
and
a
cost­
effectiveness
of
approximately
$
18,000
per
additional
ton
of
total
chlorine
removed,
we
are
not
proposing
a
beyond­
the­
floor
standard
based
on
wet
scrubbing.
b.
Feed
Control
of
Chromium
in
the
Hazardous
Waste.
We
evaluated
additional
feed
control
of
chlorine
in
the
hazardous
waste
as
a
beyond­
the­
floor
control
technique
to
reduce
floor
emission
levels
by
20%
to
achieve
a
standard
of
2.0E­
2
lbs
total
chlorine
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste.
The
national
incremental
annualized
compliance
cost
for
liquid
fuel­
fired
boilers
to
meet
this
beyond­
the­
floor
level
rather
than
comply
with
the
floor
controls
would
be
approximately
$
3.9
million
and
would
provide
an
incremental
reduction
in
total
chlorine
emissions
beyond
the
MACT
floor
controls
of
170
tons
per
year.
Nonair
quality
health
and
environmental
impacts
and
energy
effects
are
not
significant
factors
for
feedrate
control.
We
conclude
that
use
of
additional
hazardous
waste
chlorine
feedrate
control
would
not
be
cost­
effective
at
$
23,000
per
ton
of
total
chlorine
removed
and
are
not
proposing
a
beyond­
the­
floor
standard
based
on
this
control
technique.
For
the
reasons
discussed
above,
we
propose
a
total
chlorine
standard
for
existing
liquid
fuel­
fired
boilers
based
on
the
floor
level:
2.5E­
2
lbs
total
chlorine
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste.
Redline­
strikeout
highlighting
changes
made
during
OMB
review
216
3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
The
MACT
floor
for
new
sources
for
total
chlorine
would
be
7.2E­
4
lbs
total
chlorine
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste.
This
is
an
emission
level
that
the
single
best
performing
source
identified
with
the
SRE/
Feed
Approach
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
operating
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
We
evaluated
wet
scrubbing
as
beyond­
the­
floor
control
for
further
reductions
in
total
chlorine
emissions
to
achieve
a
beyond­
the­
floor
level
of
3.6E­
4
lbs
total
chlorine
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste.
The
incremental
annualized
compliance
cost
for
a
new
liquid
fuel­
fired
boiler
with
an
average
gas
flowrate
to
meet
this
beyond­
the­
floor
level,
rather
than
comply
with
the
floor
level,
would
be
approximately
$
0.44
million
and
would
provide
an
incremental
reduction
in
total
chlorine
emissions
of
approximately
0.13
tons
per
year,
for
a
cost­
effectiveness
of
$
3.3
million
per
ton
of
total
chlorine
removed.
We
evaluated
nonair
quality
health
and
environmental
impacts
and
energy
effects
and
determined
that,
for
a
new
source
with
average
an
average
gas
flowrate,
this
beyondthe
floor
option
would
increase
both
the
amount
of
hazardous
wastewater
generated
and
water
usage
by
approximately
140
million
gallons
per
year
and
would
increase
electricity
usage
by
1.3
million
kW­
hours
per
year.
After
considering
these
impacts
and
cost­
effectiveness,
we
conclude
that
a
beyond­
the­
floor
standard
based
on
wet
scrubbing
for
new
liquid
fuel­
fired
boilers
is
not
warranted.
For
the
reasons
discussed
above,
we
propose
a
total
chlorine
standard
for
new
sources
based
on
the
floor
level:
7.2E­
4
lbs
total
chlorine
emissions
attributable
to
the
hazardous
waste
per
million
Btu
heat
input
from
the
hazardous
waste.
G.
What
Is
the
Rationale
for
the
Proposed
Standards
for
Carbon
Monoxide
or
Hydrocarbons?
To
control
emissions
of
organic
HAP,
existing
and
new
sources
would
be
required
to
comply
with
either
a
carbon
monoxide
standard
of
100
ppmv
or
a
hydrocarbon
standard
of
10
ppmv.
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
Liquid
fuel­
fired
boilers
that
burn
hazardous
waste
are
currently
subject
to
RCRA
standards
that
require
compliance
with
either
a
carbon
monoxide
standard
of
100
ppmv,
or
a
hydrocarbon
standard
of
20
ppmv.
Compliance
is
based
on
an
hourly
rolling
average
as
measured
with
a
CEMS.
See
§
266.104(
a).
We
are
proposing
today
floor
standards
of
100
ppmv
for
carbon
monoxide
or
10
ppmv
for
hydrocarbons.
Floor
control
for
existing
sources
is
operating
under
good
combustion
practices
including:
(
1)
providing
adequate
excess
air
with
use
of
oxygen
CEMS
and
feedback
air
input
control;
(
2)
providing
adequate
fuel/
air
mixing;
(
3)
homogenizing
hazardous
waste
fuels
(
such
as
by
blending
or
size
reduction)
to
control
combustion
upsets
due
to
very
high
or
very
low
volatile
content
wastes;
(
4)
regulating
waste
and
air
feedrates
to
ensure
proper
combustion
temperature
and
residence
time;
(
5)
characterizing
waste
prior
to
burning
for
combustion­
related
composition
Redline­
strikeout
highlighting
changes
made
during
OMB
review
217
(
including
parameters
such
as
heating
value,
volatile
content,
liquid
waste
viscosity,
etc.);
(
6)
ensuring
the
source
is
operated
by
qualified,
experienced
operators;
and
(
7)
periodic
inspection
and
maintenance
of
combustion
system
components
such
as
burners,
fuel
and
air
supply
lines,
injection
nozzles,
etc.
Given
that
there
are
many
interdependent
parameters
that
affect
combustion
efficiency
and
thus
carbon
monoxide
and
hydrocarbon
emissions,
we
are
not
able
to
quantify
"
good
combustion
practices."
All
liquid
fuel­
fired
boilers
are
currently
complying
with
the
RCRA
carbon
monoxide
limit
of
100
ppmv
on
an
hourly
rolling
average.
No
boilers
are
complying
with
the
RCRA
hydrocarbon
limit
of
20
ppmv
on
an
hourly
rolling
average.
We
propose
a
floor
level
for
carbon
monoxide
level
of
100
ppmv
because
it
is
a
currently
enforceable
Federal
standard.
Although
the
best
performing
sources
are
achieving
carbon
monoxide
levels
below
100
ppmv,
it
is
not
appropriate
to
establish
a
lower
floor
level
because
carbon
monoxide
is
a
surrogate
for
nondioxin/
furan
organic
HAP.
As
such,
lowering
the
carbon
monoxide
floor
may
not
significantly
reduce
organic
HAP
emissions.
In
addition,
it
would
be
inappropriate
to
apply
a
MACT
methodology
to
the
carbon
monoxide
emissions
from
the
best
performing
sources
because
those
sources
may
not
be
able
to
replicate
their
emission
levels.
This
is
because
there
are
myriad
factors
that
affect
combustion
efficiency
and,
subsequently,
carbon
monoxide
emissions.
Extremely
low
carbon
monoxide
emissions
cannot
be
assured
by
controlling
only
one
or
two
operating
parameters
We
note
also
that
we
used
this
rationale
to
establish
a
carbon
monoxide
standard
of
100
ppmv
for
Phase
I
sources
in
the
September
1999
Final
Rule.
We
propose
a
floor
level
for
hydrocarbons
of
10
ppmv
even
though
the
currently
enforceable
standard
is
20
ppmv
because:
(
1)
the
two
sources
that
comply
with
the
RCRA
hydrocarbon
standard
can
readily
achieve
10
ppmv;
and
(
2)
reducing
hydrocarbon
emissions
within
the
range
of
20
ppmv
to
10
ppmv
should
reduce
emissions
of
nondioxin/
furan
organic
HAP.
We
do
not
apply
a
prescriptive
MACT
methodology
to
establish
a
hydrocarbon
floor
below
10
ppmv,
however,
because
we
have
data
from
only
two
sources.
In
addition,
we
note
that
the
hydrocarbon
emission
standard
for
Phase
I
sources
established
in
the
September
1999
Final
Rule
is
10
ppmv
also.
There
would
be
no
incremental
emission
reductions
associated
with
these
floors
because
all
sources
are
currently
achieving
the
floor
levels.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
considered
beyond­
the­
floor
levels
for
carbon
monoxide
and
hydrocarbons
based
on
use
of
better
combustion
practices
but
conclude
that
they
may
not
be
replicable
by
the
best
performing
sources
nor
duplicable
by
other
sources
given
that
we
cannot
quantify
good
combustion
practices.
Moreover,
as
discussed
above,
we
cannot
ensure
that
lower
carbon
monoxide
or
hydrocarbon
levels
would
significantly
reduce
emissions
of
nondioxin/
furan
organic
HAP.
Nonair
quality
health
and
environmental
impacts
and
energy
requirements
are
not
significant
factors
for
use
of
better
combustion
practices
as
beyond­
the­
floor
control.
For
these
reasons,
we
conclude
that
beyond­
the­
floor
standards
for
carbon
monoxide
and
hydrocarbons
are
not
warranted
for
existing
sources.
3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
Redline­
strikeout
highlighting
changes
made
during
OMB
review
218
MACT
floor
for
new
sources
would
be
the
same
as
the
floor
for
existing
sources­­
100
ppmv
for
carbon
monoxide
and
10
ppmv
for
hydrocarbons­­
and
based
on
the
same
rationale.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
As
discussed
in
the
context
of
beyond­
the­
floor
considerations
for
existing
sources,
we
considered
beyond­
the­
floor
standards
for
carbon
monoxide
and
hydrocarbons
for
new
sources
based
on
use
of
better
combustion
practices.
But
we
conclude
that
beyond
the
floor
standards
may
not
be
replicable
by
the
best
performing
sources
nor
duplicable
by
other
sources
given
that
we
cannot
quantify
good
combustion
practices.
Moreover,
we
cannot
ensure
that
lower
carbon
monoxide
or
hydrocarbon
levels
would
significantly
reduce
emissions
of
nondioxin/
furan
organic
HAP.
Nonair
quality
health
and
environmental
impacts
and
energy
requirements
are
not
significant
factors
for
use
of
better
combustion
practices
as
beyond­
the­
floor
control.
For
these
reasons,
we
are
not
proposing
a
beyond­
the­
floor
standard
for
carbon
monoxide
and
hydrocarbons.
H.
What
Is
the
Rationale
for
the
Proposed
Standard
for
Destruction
and
Removal
Efficiency?
To
control
emissions
of
organic
HAP,
existing
and
new
sources
would
be
required
to
comply
with
a
destruction
and
removal
efficiency
(
DRE)
of
99.99%
for
organic
HAP.
For
sources
burning
hazardous
wastes
F020,
F021,
F022,
F023,
F026,
or
F027,
however,
the
DRE
standard
is
99.9999%
for
organic
HAP.
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
Liquid
fuel­
fired
boilers
that
burn
hazardous
waste
are
currently
subject
to
RCRA
DRE
standards
that
require
99.99%
destruction
of
designated
principal
organic
hazardous
constituents
(
POHCs).
For
sources
that
burn
hazardous
wastes
F020,
F021,
F022,
F023,
F026,
or
F027,
however,
the
DRE
standard
is
99.9999%
destruction
of
designated
POHCs.
See
§
266.104(
a).
The
DRE
standard
helps
ensure
that
a
combustor
is
operating
under
good
combustion
practices
and
thus
minimizing
emissions
of
organic
HAP.
Under
the
MACT
compliance
regime,
sources
would
designate
POHCs
that
are
organic
HAP
or
that
are
surrogates
for
organic
HAP.
We
propose
to
establish
the
RCRA
DRE
standard
as
the
floor
for
existing
sources
because
it
is
a
currently
enforceable
Federal
standard.
There
would
be
no
incremental
costs
or
emission
reductions
associated
with
this
floor
because
sources
are
currently
complying
with
the
standard.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
considered
a
beyond­
the­
floor
level
for
DRE
based
on
use
of
better
combustion
practices
but
conclude
that
it
may
not
be
replicable
by
the
best
performing
sources
nor
duplicable
by
other
sources
given
that
we
cannot
quantify
better
combustion
practices.
Moreover,
we
cannot
ensure
that
a
higher
DRE
standard
would
significantly
reduce
emissions
of
organic
HAP
given
that
DRE
measures
the
destruction
of
organic
HAP
present
in
the
boiler
feed
rather
than
gross
emissions
of
organic
HAP.
Although
a
source's
combustion
practices
may
be
adequate
to
destroy
particular
organic
HAP
in
the
feed,
other
organic
HAP
that
may
be
emitted
as
products
of
Redline­
strikeout
highlighting
changes
made
during
OMB
review
68
The
carbon
monoxide/
hydrocarbon
emission
standard
would
control
organic
HAP
that
are
products
of
incomplete
combustion
by
also
ensuring
use
of
good
combustion
practices.

219
incomplete
combustion
may
not
be
controlled
by
the
DRE
standard.
68
For
these
reasons,
and
after
considering
nonair
quality
health
and
environmental
impacts
and
energy
requirements,
we
are
not
proposing
a
beyond­
the­
floor
DRE
standard
for
existing
sources.
3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
We
propose
to
establish
the
RCRA
DRE
standard
as
the
floor
for
new
sources
because
it
is
a
currently
enforceable
Federal
standard.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
Using
the
same
rationale
as
we
used
to
consider
a
beyond­
the­
floor
DRE
standard
for
existing
sources,
we
conclude
that
a
beyond­
the­
floor
DRE
standard
for
new
sources
is
not
warranted.
Consequently,
after
considering
nonair
quality
health
and
environmental
impacts
and
energy
requirements,
we
are
proposing
the
floor
DRE
standard
for
new
sources.

XII.
How
Did
EPA
Determine
the
Proposed
Emission
Standards
for
Hazardous
Waste
Burning
Hydrochloric
Acid
Production
Furnaces?
The
proposed
standards
for
existing
and
new
hydrochloric
acid
production
furnaces
that
burn
hazardous
waste
are
summarized
in
the
table
below.
See
proposed
§
63.1218.
Redline­
strikeout
highlighting
changes
made
during
OMB
review
220
PROPOSED
STANDARDS
FOR
EXISTING
AND
NEW
HYDROCHLORIC
ACID
PRODUCTION
FURNACES
Hazardous
Air
Pollutant
or
Surrogate
Emission
Standard1
Existing
Sources
New
Sources
Dioxin
and
furan
0.40
ng
TEQ/
dscm
0.40
ng
TEQ/
dscm
Hydrochloric
acid
and
chlorine
gas2
14
ppmv
or
99.9927%
System
Removal
Efficiency
or
the
alternative
emission
limits
under
§
63.1215
1.2
ppmv
or
99.99937%
System
Removal
Efficiency
or
the
alternative
emission
limits
under
§
63.1215
Carbon
monoxide
or
hydrocarbons3
100
ppmv
carbon
monoxide
or
10
ppmv
hydrocarbons
100
ppmv
carbon
monoxide
or
10
ppmv
hydrocarbons
Destruction
and
Removal
Efficiency
For
existing
and
new
sources,
99.99%
for
each
principal
organic
hazardous
constituent
(
POHC).
For
sources
burning
hazardous
wastes
F020,
F021,
F022,
F023,
F026,
or
F027,
however,
99.9999%
for
each
POHC.

1
All
emission
standards
are
corrected
to
7%
oxygen,
dry
basis.
2
Combined
standard,
reported
as
a
hydrogen
chloride
equivalent.
3
Hourly
rolling
average.
Hydrocarbons
reported
as
propane.

A.
What
Is
the
Rationale
for
the
Proposed
Standards
for
Dioxin
and
Furan?
The
proposed
standard
for
dioxin/
furan
for
existing
and
new
sources
is
0.40
ng
TEQ/
dscm.
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
The
proposed
MACT
floor
for
existing
sources
is
compliance
with
the
proposed
CO/
HC
emission
standard
and
compliance
with
the
proposed
DRE
standard.
Hydrochloric
acid
production
furnaces
use
wet
scrubbers
to
remove
hydrochloric
acid
from
combustion
gases
to
produce
the
hydrochloric
acid
product
and
to
minimize
residual
emissions
of
hydrochloric
acid
and
chlorine
gas.
Thus,
dioxin/
furan
cannot
be
formed
on
particulate
surfaces
in
the
emission
control
device
as
can
happen
with
electrostatic
precipitators
and
fabric
filters.
Nonetheless,
dioxin/
furan
emissions
from
hydrochloric
acid
production
furnaces
can
be
very
high.
We
have
dioxin/
furan
emissions
data
for
18
test
conditions
representing
14
of
the
17
sources.
Dioxin/
furan
emissions
range
from
0.02
ng
TEQ/
dscm
to
6.8
ng
TEQ/
dscm.
We
investigated
whether
it
would
be
appropriate
to
establish
separate
dioxin/
furan
standards
for
furnaces
equipped
with
waste
heat
recovery
boilers
versus
those
without
boilers.
Ten
of
the
17
hydrochloric
acid
production
furnaces
are
equipped
with
boilers.
We
considered
whether
waste
heat
recovery
boilers
may
be
causing
the
elevated
dioxin/
furan
emissions,
as
appeared
to
be
the
case
for
incinerators
equipped
with
boilers.
See
62
FR
at
24220
(
May
2,
Redline­
strikeout
highlighting
changes
made
during
OMB
review
69
Section
266.104
requires
compliance
with
a
carbon
monoxide
limit
of
100
ppmv
or
a
hydrocarbon
limit
of
20
ppmv,
while
we
are
proposing
today
a
carbon
monoxide
limit
of
100
ppmv
or
a
hydrocarbon
limit
of
10
ppmv
(
see
Section
XII.
H
in
the
text).
Although
today's
proposed
hydrocarbon
limit
is
more
stringent
than
the
current
limit
for
hydrochloric
acid
production
furnaces,
all
sources
chose
to
comply
with
the
100
ppmv
carbon
monoxide
limit.

221
1997)
where
we
explain
that
heat
recovery
boilers
preclude
rapid
temperature
quench
of
combustion
gases,
thus
allowing
particle­
catalyzed
formation
of
dioxin/
furan.
The
dioxin/
furan
data
for
hydrochloric
acid
production
furnaces
indicate,
however,
that
furnaces
with
boilers
have
dioxin/
furan
emissions
ranging
from
0.05
to
6.8
ng
TEQ/
dscm,
while
furnaces
without
boilers
have
dioxin/
furan
emissions
ranging
from
0.02
to
1.7
ng
TEQ/
dscm.
Based
on
a
statistical
analysis
of
the
data
sets
(
see
discussion
in
Part
Two,
Section
II.
E),
we
conclude
that
the
dioxin/
furan
emissions
for
furnaces
equipped
with
boilers
are
not
significantly
different
from
dioxin/
furan
emissions
for
furnaces
without
boilers.
Thus,
we
conclude
that
separate
dioxin/
furan
emission
standards
are
not
warranted.
We
cannot
or
identify
or
quantify
a
dioxin/
furan
control
mechanism
for
these
furnaces.
Consequently,
we
conclude
that
establishing
a
floor
emission
level
based
on
emissions
from
the
best
performing
sources
would
not
be
appropriate
because
the
best
performing
sources
may
not
be
able
to
replicate
their
emission
levels,
and
other
sources
may
not
be
able
to
duplicate
those
emission
levels.
We
note,
however,
that
dioxin/
furan
emissions
can
be
affected
by
the
furnace's
combustion
efficiency.
Operating
under
poor
combustion
conditions
can
generate
dioxin/
furan
and
organic
precursors
that
may
contribute
to
post­
combustion
dioxin/
furan
formation.
Because
we
cannot
quantify
a
dioxin/
furan
floor
level
and
because
hydrochloric
acid
production
furnaces
are
currently
required
to
operate
under
good
combustion
practices
by
RCRA
standards
for
carbon
monoxide/
hydrocarbons
and
destruction
and
removal
efficiency,
we
identify
those
RCRA
standards
as
the
proposed
MACT
floor.
See
§
266.104
requiring
compliance
with
destruction
and
removal
efficiency
and
carbon
monoxide/
hydrocarbon
emission
standards.
69
We
also
find,
as
required
by
CAA
Section
112(
h)(
1),
that
these
proposed
standards
are
consistent
with
Section
112(
d)'
s
objective
of
reducing
emissions
of
these
HAP
to
the
extent
achievable.
We
also
request
comment
on
an
alternative
MACT
floor
expressed
as
a
dioxin/
furan
emission
concentration.
Although
it
would
be
inappropriate
to
identify
a
floor
concentration
based
on
the
average
emissions
of
the
best
performing
sources
as
discussed
above,
we
could
identify
the
floor
as
the
highest
emission
concentration
from
any
source
in
our
data
base,
after
considering
emissions
variability.
Under
this
approach,
the
highest
emitting
source
could
be
expected
to
achieve
the
floor
99
out
of
100
future
tests
when
operating
under
the
same
conditions
as
it
did
when
the
emissions
data
were
obtained.
A
floor
that
is
expressed
as
a
dioxin/
furan
emission
level
would
prevent
sources
from
emitting
at
levels
higher
than
the
(
currently)
worstcase
source
(
actually,
the
worst­
case
performance
test
result)
currently
emits.
We
specifically
request
comment
on
this
alternative
MACT
floor.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
Redline­
strikeout
highlighting
changes
made
during
OMB
review
70
USEPA,
"
Draft
Technical
Support
Document
for
HWC
MACT
Replacement
Standards,
Volume
IIIV:
Selection
of
MACT
StandardsEmissions
Estimates
and
TechnologiesEngineering
Costs,"
March
2004,
Chapter
4.

71
Please
note
that,
under
the
proposed
floor
level,
sources
would
not
incur
retrofit
costs
or
achieve
dioxin/
furan
emissions
reductions
because
they
currently
comply
with
the
floor
controls
under
current
RCRA
regulations
at
40
CFR
266.104.

222
We
evaluated
use
of
an
activated
carbon
bed
(
preceded
by
gas
reheating
to
above
the
dewpoint)
as
beyond­
the­
floor
control
for
dioxin/
furan.
Carbon
beds
can
achieve
greater
than
99%
reduction
in
dioxin/
furan
emissions.
70
We
considered
alternative
beyond­
the­
floor
levels
of
0.40
ng
TEQ/
dscm
and
0.20
ng
TEQ/
dscm.
The
incremental
annualized
cost
of
a
beyond­
the­
floor
emission
level
of
0.40
ng
TEQ/
dscm
would
be
$
1.9
million
and
would
provide
an
incremental
reduction
in
dioxin/
furan
emissions
of
2.3
grams
TEQ
per
year,
for
a
cost­
effectiveness
of
$
0.823
million
per
gram
TEQ
removed.
71
A
beyond­
the­
floor
emission
level
of
0.20
ng
TEQ/
dscm
would
provide
very
little
incremental
emissions
reduction­­
0.1
grams
TEQ
per
year­­
at
additional
costs.
We
evaluated
nonair
quality
health
and
environmental
impacts
and
energy
effects
and
determined
that
this
beyond­
the­
floor
option
would
increase
the
amount
of
hazardous
wastewater
generated
by
210
tons
per
year,
and
would
increase
electricity
usage
by
1.8
million
kW­
hours
per
year
and
natural
gas
consumption
by
96
trillion
Btu
per
year.
Notwithstanding
these
impacts,
we
conclude
that
We
judge
that
the
cost
to
achieve
a
beyond­
the­
floor
standard
of
0.40
ng
TEQ/
dscm
is
warranted
given
our
special
concern
about
dioxin/
furan.
Dioxin/
furan
are
some
of
the
most
toxic
compounds
known
due
to
their
bioaccumulation
potential
and
wide
range
of
health
effects,
including
carcinogenesis,
at
exceedingly
low
doses.
Exposure
via
indirect
pathways
is
a
chief
reason
that
Congress
singled
our
dioxin/
furan
for
priority
MACT
control
in
CAA
section
112(
c)(
6).
See
S.
Rep.
No.
128,
101st
Cong.
1st
Sess.
at
154­
155.
In
addition,
we
note
that
the
beyond­
the­
floor
emission
level
of
0.40
ng
TEQ/
dscm
is
consistent
with
historically
controlled
levels
under
MACT
for
hazardous
waste
incinerators
and
cement
kilns,
and
Portland
cement
plants.
See
§
§
63.1203(
a)(
1),
63.1204(
a)(
1),
and
63.1343(
d)(
3).
Also,
EPA
has
determined
previously
in
the
1999
Hazardous
Waste
Combustor
MACT
final
rule
that
dioxin/
furan
in
the
range
of
0.40
ng
TEQ/
dscm
or
less
are
necessary
for
the
MACT
standards
to
be
considered
generally
protective
of
human
health
under
RCRA
(
using
the
1985
cancer
slope
factor),
thereby
eliminating
the
need
for
separate
RCRA
standards
under
the
authority
of
RCRA
section
3005(
c)(
3)
and
40
CFR
270.10(
k).
Finally,
we
note
that
this
decision
is
not
inconsistent
with
EPA's
decision
not
to
promulgate
beyond­
the­
floor
standards
for
dioxin/
furan
for
hazardous
waste
burning
lightweight
aggregate
kilns,
cement
kilns,
and
incinerators
at
cost­
effectiveness
values
in
the
range
of
$
530,000
to
$
827,000
per
additional
gram
of
dioxin/
furan
TEQ
removed.
See
64
FR
at
52892,
52876,
and
52961.
In
those
cases,
EPA
determined
that
controlling
dioxin/
furan
emissions
from
a
level
of
0.40
ng
TEQ/
dscm
to
a
beyond­
the­
floor
level
of
0.20
ng
TEQ/
dscm
was
not
warranted
because
dioxin/
furan
levels
below
0.40
ng
TEQ/
dscm
would
be
Redline­
strikeout
highlighting
changes
made
during
OMB
review
72
We
estimate
beyond­
the­
floor
control
costs
assuming
a
new
source
emits
the
highest
levels
likely
under
floor
control
based
on
compliance
with
the
carbon
monoxide
and
destruction
and
removal
efficiency
standards.

223
warranted
considering
the
special
hazard
dioxin/
furan
emissions
pose.
For
the
reasons
discussed
above,
we
areare
generally
considered
to
be
below
the
level
of
health
risk
concern.
For
these
reasons,
we
believe
that
proposing
a
beyond­
the­
floor
standard
for
dioxin/
furan
of
0.40
ng
TEQ/
dscm
for
existing
sources.
Thirty­
five
percent
of
sources
are
currently
meeting
this
standard.
is
warranted
notwithstanding
the
nonair
quality
health
and
environmental
impacts
and
energy
effects
identified
above
and
costs
of
approximately
$
0.83
million
per
additional
gram
of
dioxin/
furan
TEQ
removed.

3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
MACT
floor
for
new
sources
is
the
same
as
for
existing
sources
under
the
same
rationale:
compliance
with
the
carbon
monoxide/
hydrocarbon
emission
standard
and
compliance
with
the
destruction
and
removal
efficiency
standard.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
As
for
existing
sources,
we
evaluated
use
of
an
activated
carbon
bed
as
beyond­
the­
floor
control
for
new
sources
to
achieve
an
emission
level
of
0.40
ng
TEQ/
dscm.
We
estimate
that
the
incremental
annualized
cost
for
a
new
hydrochloric
acid
production
furnace
with
average
gas
flowrate
to
reduce
dioxin/
furan
emissions
at
the
floor
of
0.68
ng
TEQ/
dscm72
to
achieve
a
beyond­
the­
floor
emission
level
of
0.40
ng
TEQ/
dscm
would
be
$
0.15
million.
These
controls
would
provide
an
incremental
reduction
in
dioxin/
furan
emissions
of
0.66
grams
TEQ
per
year,
for
a
cost­
effectiveness
of
$
230,000
per
gram
TEQ
removed.
We
evaluated
nonair
quality
health
and
environmental
impacts
and
energy
effects
and
determined
that,
for
a
new
source
with
average
an
average
gas
flowrate,
this
beyond­
the­
floor
option
would
increase
the
amount
of
hazardous
wastewater
generated
by
9
tons
per
year,
and
would
increase
electricity
usage
by
0.14
million
kW­
hours
per
year
and
natural
gas
consumption
by
9.2
trillion
Btu
per
year.
Notwithstanding
these
impacts,
we
conclude
that
a
beyond­
the­
floor
level
of
0.40
ng
TEQ/
dscm
would
be
costeffective
considering
the
special
hazard
dioxin/
furan
emissions
pose.
For
these
reasons,
we
propose
We
judge
that
the
cost
to
achieve
a
beyond­
the­
floor
standard
forof
0.40
ng
TEQ/
dscm
is
warranted
given
our
special
concern
about
dioxin/
furan.
Dioxin/
furan
are
some
of
the
most
toxic
compounds
known
due
to
their
bioaccumulation
potential
and
wide
range
of
health
effects,
including
carcinogenesis,
at
exceedingly
low
doses.
Exposure
via
indirect
pathways
is
a
chief
reason
that
Congress
singled
our
dioxin/
furan
for
new
sourcespriority
MACT
control
in
CAA
section
112(
c)(
6).
See
S.
Rep.
No.
128,
101st
Cong.
1st
Sess.
at
154­
155.
In
addition,
we
note
that
the
beyond­
the­
floor
standard
of
0.40
ng
TEQ/
dscm
is
consistent
with
historically
controlled
levels
under
MACT
for
hazardous
waste
incinerators
and
cement
kilns,
and
Portland
cement
plants.
See
§
§
63.1203(
a)(
1),
63.1204(
a)(
1),
and
63.1343(
d)(
3).
Also,
EPA
has
determined
Redline­
strikeout
highlighting
changes
made
during
OMB
review
73
USEPA,
"
Draft
Technical
Support
Document
for
HWC
MACT
Replacement
Standards,
Volume
III:
Selection
of
MACT
Standards
and
Technologies,"
March
2004,
Chapter
2.

74
Except
that
one
source
emitted
330
ug/
dscm
low
volatile
metals
and
0.043
gr/
dscf
particulate
matter
during
compliance
testing.
This
source
apparently
detuned
the
acid
gas
absorber
and
other
acid
gas
control
equipment
given
that
it
achieved
less
than
99%
system
removal
efficiency
for
total
chlorine
and
had
total
chlorine
emissions
of
500
ppmv.
This
source
would
be
not
be
allowed
to
operate
under
these
conditions
under
today's
proposed
rule:
14
ppmv
total
chlorine
emission
limit,
or
99.9927
system
removal
efficiency.
Thus,
under
the
proposed
rule,
emissions
of
low
volatile
metals
and
particulate
matter
would
be
substantially
lower.

224
previously
in
the
1999
Hazardous
Waste
Combustor
MACT
final
rule
that
dioxin/
furan
in
the
range
of
0.40
ng
TEQ/
dscm
or
less
are
necessary
for
the
MACT
standards
to
be
considered
generally
protective
of
human
health
under
RCRA
(
using
the
1985
cancer
slope
factor),
thereby
eliminating
the
need
for
separate
RCRA
standards
under
the
authority
of
RCRA
section
3005(
c)(
3)
and
40
CFR
270.10(
k).
For
these
reasons,
we
believe
that
proposing
a
beyond­
the­
floor
standard
of
0.40
ng
TEQ/
dscm
is
warranted
notwithstanding
the
nonair
quality
health
and
environmental
impacts
and
energy
effects
identified
above
and
costs
of
approximately
$
0.23
million
per
additional
gram
of
dioxin/
furan
TEQ
removed.

B.
What
Is
the
Rationale
for
the
Proposed
Standards
for
Mercury,
Semivolatile
Metals,
and
Low
Volatile
Metals?
We
propose
to
require
compliance
with
the
total
chlorine
standard
as
a
surrogate
for
the
mercury,
semivolatile
metals,
and
low
volatile
metals
standards.
As
discussed
above,
hydrochloric
acid
production
furnaces
use
wet
scrubbers
to
remove
hydrochloric
acid
from
combustion
gases
to
produce
the
hydrochloric
acid
product
and
to
minimize
residual
emissions
of
hydrochloric
acid
and
chlorine
gas.
Wet
scrubbers
also
remove
metal
HAP,
including
mercury,
from
combustion
gases.
To
minimize
contamination
of
hydrochloric
acid
product
with
metals,
hydrochloric
acid
production
furnaces
generally
feed
hazardous
waste
with
low
levels
of
metal
HAP.
Moreover,
the
wet
scrubbers
used
to
recover
the
hydrochloric
acid
product
and
minimize
residual
emissions
of
hydrochloric
acid
and
chlorine
gas
also
control
emissions
of
metal
HAP
to
very
low
levels.
Based
on
emissions
testing
within
the
range
of
normal
emissions
(
i.
e.,
not
compliance
test,
maximum
allowed
emissions),
hydrochloric
acid
production
furnaces
emit
mercury
at
levels
from
0.1
to
0.4
ug/
dscm,
semivolatile
metals
at
levels
from
0.1
to
4.1
ug/
dscm,
and
low
volatile
metals
at
levels
from
0.1
to
43
ug/
dscm.
73,74
We
also
note
that
these
sources
emit
low
levels
of
particulate
matter.
Compliance
test,
maximum
allowable
emissions
of
particulate
matter
range
from
0.001
to
0.013
gr/
dscf.
Because
wet
scrubbers
designed
to
recover
the
hydrochloric
acid
product
and
control
residual
emissions
of
hydrogen
chloride
and
chlorine
gas
also
control
emissions
of
mercury,
and
Redline­
strikeout
highlighting
changes
made
during
OMB
review
225
semivolatile
and
low
volatile
metals
(
including
nonenumerated
metals),
use
of
MACT
wet
scrubbers
to
comply
with
the
proposed
total
chlorine
standard
discussed
below
will
also
ensure
MACT
control
of
metal
HAP.
Accordingly,
we
propose
to
use
the
total
chlorine
standard
as
a
surrogate
for
the
mercury,
semivolatile
metals,
and
low
volatile
metals
standards.
C.
What
Is
the
Rationale
for
the
Proposed
Standards
for
Total
Chlorine?
The
proposed
standards
for
total
chlorine
are
14
ppmv
or
99.9927
percent
total
chlorine
system
removal
efficiency
(
SRE)
for
existing
sources
and
1.2
ppmv
or
99.99937
percent
total
chlorine
SRE
for
new
sources.
A
source
may
elect
to
comply
with
either
standard.
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
The
proposed
MACT
floor
for
existing
sources
is
compliance
with
either
a
total
chlorine
emission
level
of
14
ppmv
or
a
total
chlorine
SRE
of
99.9927
percent.
Hydrochloric
acid
production
furnaces
use
wet
scrubbers
to
remove
hydrochloric
acid
from
combustion
gases
to
produce
the
hydrochloric
acid
product
and
to
minimize
residual
emissions
of
hydrochloric
acid
and
chlorine
gas.
We
have
compliance
test,
maximum
allowable
total
chlorine
emissions
data
for
all
17
hydrochloric
acid
production
furnaces.
Total
chlorine
emissions
range
from
0.4
to
500
ppmv,
and
total
chlorine
system
removal
efficiencies
(
SRE)
range
from
98.967
to
99.9995
percent.
As
discussed
in
Section
VI.
C
above,
control
of
the
feedrate
of
chlorine
in
hazardous
waste
fed
to
the
furnace
is
not
an
appropriate
MACT
emission
control
technique
because
hydrochloric
acid
production
furnaces
are
designed
to
produce
hydrochloric
acid
from
chlorinated
feedstocks.
Consequently,
the
approaches
we
normally
use
to
identify
the
best
performing
sources­­
SRE/
Feed
Approach
or
Emissions
Approach­­
are
not
appropriate
because
they
directly
or
indirectly
consider
chlorine
feedrate.
More
simply,
limiting
feedrate
means
not
producing
the
intended
product,
a
result
inconsistent
with
MACT.
See
2
Legislative
History
at
3352
(
House
Report)
("
MACT
is
not
intended
to
...
drive
sources
to
the
brink
of
shutdown").
To
avoid
this
concern,
we
identify
a
floor
SRE,
and
provide
an
alternative
floor
as
a
total
chlorine
emission
limit
based
on
floor
SRE
and
the
highest
chlorine
feedrate
for
any
source
in
the
data
base.
By
using
the
highest
chlorine
feedrate
to
calculate
the
alternative
total
chlorine
emission
limit,
we
ensure
that
feedrate
control
(
i.
e.
nonproduction
of
product)
is
not
a
factor
in
identifying
the
proposed
MACT
floor.
The
alternative
total
chlorine
emission
limit
would
require
a
source
that
may
not
be
achieving
floor
SRE
to
achieve
total
chlorine
emission
levels
no
greater
than
the
level
that
would
be
emitted
by
any
source
achieving
floor
SRE.
The
floor
SRE
is
99.9927
percent.
It
is
calculated
from
the
five
best
SREs,
and
considers
emissions
variability.
Floor
SRE
is
an
SRE
that
the
average
of
the
performing
sources
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
We
estimate
that
this
SRE
is
being
achieved
by
29%
of
sources.
The
alternative
floor
emission
limit
is
14
ppmv,
and
is
the
emission
level
that
the
source
with
the
highest
chlorine
feedrate­­
2.9E+
8
ug/
dscm
 
would
achieve
when
achieving
99.9927
percent
SRE.
Approximately
24%
of
sources
are
achieving
the
alternative
floor
levels,
and
these
floor
levels
would
reduce
total
chlorine
emissions
by
145
tons
per
year.
Redline­
strikeout
highlighting
changes
made
during
OMB
review
226
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
evaluated
improved
design,
operation,
and
maintenance
of
existing
scrubbers
to
achieve
a
beyond­
the­
floor
emission
level
of
7
ppmv
for
total
chlorine
for
existing
sources,
assuming
a
50%
reduction
in
emissions
from
the
floor
level.
The
national
annualized
compliance
cost
for
hydrochloric
acid
production
furnaces
to
comply
with
this
beyond­
the­
floor
standard
would
be
$
0.25
million,
and
emissions
of
total
chlorine
would
be
reduced
by
3
tons
per
year.
The
cost­
effectiveness
of
this
beyond­
the­
floor
standard
would
be
$
76,000
per
ton
of
total
chlorine
removed.
We
evaluated
nonair
quality
health
and
environmental
impacts
and
energy
effects
and
determined
that
this
beyond­
the­
floor
option
would
increase
both
the
amount
of
hazardous
wastewater
generated
and
water
usage
by
approximately
82
million
gallons
per
year
and
would
increase
electricity
usage
by
0.34
million
kW­
hours
per
year.
Generation
of
nonwastewater
hazardous
waste
would
decrease
by
7
tons
per
year.
Considering
these
impacts
and
costeffectiveness
as
well,
we
conclude
that
a
beyond­
the­
floor
standard
for
existing
sources
would
not
be
warranted.
For
these
reasons,
we
propose
a
floor
total
chlorine
standard
of
14
ppmv
or
99.9927%
SRE
for
existing
sources.
3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
The
proposed
MACT
floor
for
new
sources
is
compliance
with
either
a
total
chlorine
emission
level
of
1.2
ppmv
or
a
total
chlorine
SRE
of
99.99937
percent.
We
use
the
same
rationale
for
identifying
alternative
floors
for
new
sources
as
discussed
above
in
the
context
of
existing
sources.
The
new
source
floor
SRE
is
the
SRE
that
the
single
best
performing
source
(
i.
e,
source
with
the
best
SRE)
could
be
expected
to
achieve
in
99
of
100
future
tests
when
operating
under
conditions
identical
to
the
compliance
test
conditions
during
which
the
emissions
data
were
obtained.
The
new
source
floor
alternative
emission
limit
is
an
emission
level
that
the
source
with
the
highest
chlorine
feedrate­­
2.9E+
8
ug/
dscm
 
would
achieve
when
achieving
99.99937
percent
SRE.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
We
evaluated
a
beyond­
the­
floor
standard
for
new
sources
of
0.60
ppmv
based
on
achieving
a
50
percent
reduction
in
emissions
by
improving
the
design/
operation/
maintenance
of
the
wet
scrubber.
The
incremental
annualized
cost
for
a
new
solid
fuel­
fired
boiler
with
average
gas
flowrate
to
meet
a
beyond­
the­
floor
level
of
0.60
ppmv
would
be
approximately
$
0.15
million
and
would
provide
an
incremental
reduction
in
total
chlorine
emissions
of
0.07
tons
per
year,
for
a
cost­
effectiveness
of
$
2.1
million
per
ton
of
total
chlorine
removed.
We
evaluated
nonair
quality
health
and
environmental
impacts
and
energy
effects
and
determined
that,
for
a
new
source
with
average
gas
flowrate,
this
beyond­
the­
floor
option
would
increase
both
the
amount
of
hazardous
wastewater
generated
and
water
usage
by
approximately
26
million
gallons
per
year
and
would
increase
electricity
usage
by
0.25
million
kW­
hours
per
year.
Considering
these
impacts
and
cost­
effectiveness
as
well,
we
conclude
that
a
beyond­
thefloor
standard
for
new
sources
would
not
be
warranted.
For
the
reasons
discussed
above,
we
propose
a
total
chlorine
standard
of
1.2
ppmv
or
a
Redline­
strikeout
highlighting
changes
made
during
OMB
review
227
total
chlorine
SRE
of
99.99937
percent
for
new
sources.
D.
What
Is
the
Rationale
for
the
Proposed
Standards
for
Carbon
Monoxide
or
Hydrocarbons?
To
control
emissions
of
organic
HAP,
existing
and
new
sources
would
be
required
to
comply
with
either
a
carbon
monoxide
standard
of
100
ppmv
or
a
hydrocarbon
standard
of
10
ppmv.
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
Hydrochloric
acid
production
furnaces
that
burn
hazardous
waste
are
currently
subject
to
RCRA
standards
that
require
compliance
with
either
a
carbon
monoxide
standard
of
100
ppmv,
or
a
hydrocarbon
standard
of
20
ppmv.
Compliance
is
based
on
an
hourly
rolling
average
as
measured
with
a
CEMS.
See
§
266.104(
a).
All
hydrochloric
acid
production
furnaces
have
elected
to
comply
with
the
100
ppmv
carbon
monoxide
standard.
We
propose
floor
standards
of
100
ppmv
for
carbon
monoxide
or
10
ppmv
for
hydrocarbons
for
the
same
reasons
discussed
above
in
the
context
of
liquid
fuel­
fired
boilers.
There
would
be
no
incremental
emission
reductions
associated
with
these
floors
because
sources
are
currently
achieving
the
carbon
monoxide
standard..
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
Our
considerations
for
beyond­
the­
floor
standards
for
existing
hydrochloric
acid
production
furnaces
are
identical
to
those
discussed
above
for
existing
liquid
fuel­
fired
boilers.
For
the
reasons
discussed
above
in
the
context
of
liquid
fuel­
fired
boilers,
we
conclude
that
beyond­
the­
floor
standards
for
carbon
monoxide
and
hydrocarbons
for
existing
hydrochloric
acid
production
furnaces
are
not
warranted.
3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
MACT
floor
for
new
sources
would
be
the
same
as
the
floor
for
existing
sources­­
100
ppmv
for
carbon
monoxide
and
10
ppmv
for
hydrocarbons­­
and
based
on
the
same
rationale.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
Our
considerations
for
beyond­
the­
floor
standards
for
new
hydrochloric
acid
production
furnaces
are
identical
to
those
discussed
above
for
new
liquid
fuel­
fired
boilers.
For
the
reasons
discussed
above
in
the
context
of
liquid
fuel­
fired
boilers,
we
conclude
that
beyond­
the­
floor
standards
for
carbon
monoxide
and
hydrocarbons
for
new
hydrochloric
acid
production
furnaces
are
not
warranted.
E.
What
Is
the
Rationale
for
the
Proposed
Standard
for
Destruction
and
Removal
Efficiency?
To
control
emissions
of
organic
HAP,
existing
and
new
sources
would
be
required
to
comply
with
a
destruction
and
removal
efficiency
(
DRE)
of
99.99%
for
organic
HAP.
For
sources
burning
hazardous
wastes
F020,
F021,
F022,
F023,
F026,
or
F027,
however,
the
DRE
standard
is
99.9999%
for
organic
HAP.
1.
What
Is
the
Rationale
for
the
MACT
Floor
for
Existing
Sources?
Hydrochloric
acid
production
furnaces
that
burn
hazardous
waste
are
currently
subject
to
RCRA
DRE
standards
that
require
99.99%
destruction
of
designated
principal
organic
hazardous
constituents
(
POHCs).
For
sources
that
burn
hazardous
wastes
F020,
F021,
F022,
F023,
F026,
or
F027,
however,
the
DRE
standard
is
99.9999%
destruction
of
designated
POHCs.
See
§
266.104(
a).
Redline­
strikeout
highlighting
changes
made
during
OMB
review
228
The
DRE
standard
helps
ensure
that
a
combustor
is
operating
under
good
combustion
practices
and
thus
minimizing
emissions
of
organic
HAP.
Under
the
MACT
compliance
regime,
sources
would
designate
POHCs
that
are
organic
HAPs
or
that
are
surrogates
for
organic
HAPs.
We
propose
to
establish
the
RCRA
DRE
standard
as
the
floor
for
existing
sources
because
it
is
a
currently
enforceable
Federal
standard.
There
would
be
no
incremental
emission
reductions
associated
with
this
floor
because
sources
are
currently
complying
with
the
standard.
2.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
Existing
Sources
We
considered
a
beyond­
the­
floor
level
for
DRE
based
on
use
of
better
combustion
practices
but
conclude
that
it
may
not
be
replicable
by
the
best
performing
sources
nor
duplicable
by
other
sources
given
that
we
cannot
quantify
better
combustion
practices.
Moreover,
we
cannot
ensure
that
a
higher
DRE
standard
would
significantly
reduce
emissions
of
organic
HAP
given
that
DRE
measures
the
destruction
of
organic
HAP
present
in
the
boiler
feed
rather
than
gross
emissions
of
organic
HAP.
Although
a
source's
combustion
practices
may
be
adequate
to
destroy
particular
organic
HAP
in
the
feed,
other
organic
HAP
may
be
emitted
as
products
of
incomplete
combustion.
For
these
reasons,
and
after
considering
nonair
quality
health
and
environmental
impacts
and
energy
requirements,
we
are
not
proposing
a
beyond­
the­
floor
DRE
standard
for
existing
sources.
3.
What
Is
the
Rationale
for
the
MACT
Floor
for
New
Sources?
We
propose
to
establish
the
RCRA
DRE
standard
as
the
floor
for
new
sources
because
it
is
a
currently
enforceable
Federal
standard.
4.
EPA's
Evaluation
of
Beyond­
the­
Floor
Standards
for
New
Sources
Using
the
same
rationale
as
we
used
to
consider
a
beyond­
the­
floor
DRE
standard
for
existing
sources,
we
conclude
that
a
beyond­
the­
floor
DRE
standard
for
new
sources
is
not
warranted.
Consequently,
after
considering
nonair
quality
health
and
environmental
impacts
and
energy
requirements,
we
are
proposing
the
floor
DRE
standard
for
new
sources.
