1
MEMORANDUM
DATE:
January
7,
2004
SUBJECT:
CO
Removal
Efficiency
as
a
Surrogate
for
HAP
Removal
Efficiency
FROM:
Melanie
Taylor
and
Jennifer
Snyder
Alpha­
Gamma
Technologies,
Inc.

TO:
Sims
Roy,
EPA
OAQPS
ESD
Combustion
Group
The
purpose
of
this
memorandum
is
to
establish
whether
carbon
monoxide
(
CO)
removal
efficiency
can
be
used
as
a
surrogate
for
hazardous
air
pollutants
(
HAP)
removal
efficiency
from
engines
equipped
with
HAP
control.
The
purpose
of
this
memorandum
is
also
to
determine
if
CO
concentration
can
be
used
as
a
surrogate
for
HAP
concentration.
It
has
been
suggested
that
the
oxidation
catalyst
removal
efficiencies
of
CO
and
HAP
exhibit
the
same
behavior
for
spark
ignition
lean
burn
engines
and
compression
ignition
engines,
making
CO
removal
efficiency
a
possible
surrogate
for
HAP
removal
efficiency.
It
is
beneficial
for
both
industry
and
EPA
to
monitor
HAP
removal
through
the
use
of
CO
as
a
surrogate,
if
that
is
appropriate.
It
is
much
easier
to
test
for
one
pollutant
instead
of
all
HAP.
Testing
for
CO
is
easier
and
more
economical
than
testing
for
several
HAP,
and
can
be
done
either
through
source
testing
or
continuous
emission
monitoring
systems,
making
it
more
flexible
than
testing
for
several
HAP.

A
previous
version
of
this
memorandum
based
on
initial
data
from
the
testing
at
Colorado
State
University
was
developed
at
the
time
of
proposal
and
entered
into
the
docket
(
Document
ID
Number
OAR­
2002­
0059­
0208
or
A­
95­
35
II­
F­
4).
This
memorandum
has
been
updated
to
this
version
and
reflects
emissions
data
from
final
test
reports.

In
order
to
determine
if
CO
is
an
appropriate
surrogate
for
HAP,
emissions
data
from
EPA's
emissions
tests
on
lean
burn
natural
gas
and
diesel
engines
and
an
industry
test
on
rich
burn
natural
gas
engines
were
analyzed.
To
simplify
this
analysis,
CO
and
formaldehyde
emissions
were
compared.
Formaldehyde
is
representative
of
HAP
emissions
because
it
is
the
hazardous
air
pollutant
emitted
in
the
largest
concentrations.
2
In
summary,
the
emissions
data
suggest
that
CO
removal
efficiency
is
a
reasonable
surrogate
for
HAP
removal
efficiency
for
an
oxidation
catalyst.
The
emissions
data
do
not
indicate
that
CO
removal
efficiency
is
an
appropriate
surrogate
for
HAP
removal
efficiency
for
non­
selective
catalytic
reduction
(
NSCR).
Emissions
data
do
not
indicate
that
CO
concentration
is
a
surrogate
for
HAP
concentration.
A
more
detailed
discussion
is
provided
below
for
each
engine
type.

2­
Stroke
Lean
Burn
Natural
Gas
Engine
Table
1
presents
the
CO
and
formaldehyde
emissions
data
(
corrected
to
15%
O
2,
dry
basis)
for
the
2­
stroke
lean
burn
engine
tested.
Figure
1
displays
the
percent
reduction
of
CO
and
formaldehyde
versus
catalyst
temperature
for
the
2­
stroke
lean
burn
engine.
The
catalyst
control
efficiency
increases
as
the
catalyst
temperature
increases,
as
demonstrated
by
the
least
squares
best­
fit
line.
Higher
CO
removal
results
in
higher
formaldehyde
removal,
as
shown
in
Figure
1.
Because
the
control
efficiencies
of
CO
and
formaldehyde
behave
similarly
with
respect
to
the
catalyst
temperature,
CO
reduction
is
considered
a
surrogate
for
formaldehyde
reduction
and
other
HAP
reduction
for
2­
stroke
lean
burn
natural
gas
and
other
spark
ignition
lean
burn
engines.

4­
Stroke
Lean
Burn
Natural
Gas
Engine
Table
2
presents
the
CO
and
formaldehyde
emissions
data
(
corrected
to
15%
O
2,
dry
basis)
for
the
4­
stroke
lean
burn
engine
tested.
Figure
2
shows
the
percent
reduction
versus
catalyst
temperature.
Catalyst
temperature
is
known
to
have
an
effect
on
the
catalyst
removal
efficiency.
Using
least
squares
analysis,
the
best­
fit
lines
for
CO
and
formaldehyde
show
that
the
control
efficiency
for
both
CO
and
formaldehyde
decreases
slightly
with
increasing
temperature.
Both
lines
have
nearly
identical
slopes.
The
figure
shows
that
higher
CO
removal
levels
correspond
to
higher
HAP
removal
levels.
From
this
it
can
be
concluded
that
the
control
efficiencies
of
CO
and
formaldehyde
exhibit
similar
behavior,
and
thus
CO
reduction
is
a
reasonable
surrogate
for
HAP
reduction
for
4­
stroke
lean
burn
natural
gas
engines.
Natural
gas
is
a
clean
burn
fuel
and
represents
the
majority
of
engines
in
the
spark
ignition
engine
subcategory.
In
addition,
natural
gas
engines
share
the
same
design
as
other
clean
gaseous
engines;
therefore,
CO
reduction
can
be
considered
a
surrogate
for
other
4­
stroke
spark
ignition
lean
burn
engines
burning
fuels
such
as
propane,
liquid
petroleum
gas,
and
process
gas.

4­
Stroke
Diesel
Engine
The
CO
and
formaldehyde
emissions
data
(
corrected
to
15%
O
2,
dry
basis)
for
the
4­
stroke
diesel
engine
are
shown
in
Table
3.
Figure
3
presents
the
percent
reduction
of
CO
and
formaldehyde
versus
catalyst
temperature
for
the
4­
stroke
diesel
engine.
The
least
squares
best­
fit
lines
in
the
figure
show
that
the
control
efficiencies
of
CO
and
formaldehyde
behave
similarly.
The
removal
efficiency
of
formaldehyde
is
actually
higher
than
the
removal
efficiency
of
CO
for
this
engine.
The
figure
demonstrates
that
high
CO
removal
corresponds
to
high
HAP
removal
for
the
4­
stroke
diesel
engine.
3
Because
of
similarities
in
engine
operating
parameters,
emissions
from
2­
stroke
diesel
engines
are
expected
to
behave
similarly
to
emissions
from
4­
stroke
diesel
engines.
Thus,
CO
removal
efficiency
is
a
surrogate
for
HAP
removal
efficiency
for
diesel
engines,
which
are
compression
ignited.
The
compression
ignition
engine
subcategory
also
contains
dual
fuel
engines,
which
use
natural
gas
and
diesel
fuel.
Carbon
monoxide
reduction
is
found
to
be
a
surrogate
for
diesel
engines
and
for
natural
gas
engines,
hence
it
is
considered
a
surrogate
for
dual
fuel
engines.

4­
Stroke
Rich
Burn
Natural
Gas
Engine
Table
4
presents
the
CO
and
formaldehyde
emissions
data
(
corrected
to
15%
O
2,
dry
basis)
for
one
of
the
4­
stroke
rich
burn
engines
tested
(
engine
7A).
Figure
4
shows
the
percent
reduction
versus
catalyst
temperature
for
engine
7A.
Engine
exhaust
temperatures
for
the
other
engine
tested
(
engine
9A)
were
not
recorded
during
the
field
test.
Using
least
squares
analysis,
the
best­
fit
lines
for
CO
and
formaldehyde
show
that
the
control
efficiency
for
CO
decreases
while
the
formaldehyde
control
efficiency
increases
with
increasing
temperature.
The
figure
shows
that
with
increasing
temperature,
the
control
efficiencies
for
CO
and
formaldehyde
behave
differently.
From
this
it
can
be
concluded
that
the
control
efficiencies
of
CO
and
formaldehyde
do
not
exhibit
similar
behavior,
and
thus
CO
reduction
is
not
a
reasonable
surrogate
for
HAP
reduction
for
4­
stroke
rich
burn
natural
gas
engines.
As
previously
mentioned,
natural
gas
engines
are
similar
to
other
clean
gaseous
fueled
engines.
For
that
reason,
CO
reduction
cannot
be
considered
a
surrogate
for
other
4­
stroke
rich
burn
engines
burning
fuels
such
as
propane,
liquid
petroleum
gas,
and
process
gas.

In
addition
to
investigating
whether
CO
reduction
can
be
used
as
surrogate
for
HAP
reduction
for
4­
stroke
rich
burn
engines,
the
relationship
between
NOx
reduction
and
HAP
reduction
was
also
examined.
It
has
been
suggested
that
NOx
monitoring
should
be
allowed
as
a
surrogate
for
NSCR
function.
The
data
in
Table
4
show
the
reduction
levels
of
NOx
and
formaldehyde
with
varying
temperatures
for
the
4­
stroke
rich
burn
engine
tested
where
temperature
data
was
obtained.
As
the
data
show,
there
is
not
a
consistent
trend.
For
example,
in
one
case,
as
the
temperature
was
increased
from
887

F
to
911

F,
the
NOx
reduction
increased
slightly,
while
the
formaldehyde
reduction
decreased.
The
opposite
occurred
when
the
temperature
was
increased
from
911

F
to
945

F.
In
this
case,
the
NOx
reduction
decreased
while
the
formaldehyde
reduction
increased.
Based
on
this
it
can
be
concluded
that
NOx
reduction
is
not
an
appropriate
surrogate
for
HAP
reduction
for
4­
stroke
rich
burn
engines
with
NSCR.

Conclusion
Figures
1­
3
show
that
CO
and
formaldehyde
reductions
achieved
with
oxidation
catalysts
exhibit
the
same
behavior
for
spark
ignition
lean
burn
engines
and
compression
ignition
engines.
Therefore,
CO
removal
efficiency
is
a
reasonable
surrogate
for
HAP
removal
efficiency
for
spark
ignition
lean
burn
engines
and
compression
ignition
engines.
High
CO
reduction
efficiency
is
a
good
indicator
of
high
HAP
reduction
for
oxidation
catalysts.
This
conclusion
is
advantageous
for
both
industry
and
EPA
because
testing
for
CO
is
easier
than
testing
for
several
HAP.
Figure
4
4
shows
that
CO
and
formaldehyde
reductions
achieved
with
NSCR
do
not
exhibit
the
same
behavior
for
spark
ignition
4­
stroke
rich
burn
engines.
Therefore,
CO
removal
efficiency
is
not
a
reasonable
surrogate
for
HAP
removal
efficiency
for
spark
ignition
4­
stroke
rich
burn
engines.
It
was
also
concluded
that
NOx
removal
efficiency
is
not
an
appropriate
surrogate
for
HAP
removal
efficiency
for
4­
stroke
rich
burn
engines
with
NSCR.

In
addition,
the
emissions
data
presented
in
Tables
1­
3
were
analyzed
to
determine
if
there
is
a
relationship
between
formaldehyde
and
CO
concentration.
As
Figures
5­
7
indicate,
the
inlet
concentration
of
formaldehyde
and
inlet
concentration
of
CO
do
not
behave
similarly
with
varying
engine
exhaust
temperatures.
As
Figure
5
shows,
the
CO
inlet
concentration
reduced
drastically
as
the
formaldehyde
inlet
concentration
remained
fairly
consistent,
with
increasing
engine
exhaust
temperatures.
Figures
6
and
7
indicate
a
different
trend;
the
CO
inlet
concentration
increased
slightly
as
the
formaldehyde
concentration
remained
relatively
consistent,
with
increasing
engine
exhaust
temperatures.
For
these
reasons,
a
relationship
between
formaldehyde
and
CO
concentration
could
not
be
established
and
CO
concentration
is
not
an
appropriate
surrogate
for
HAP
concentration.
5
Table
1:
Emissions
Data
for
2­
Stroke
Lean
Burn
Natural
Gas
Engine
(
corrected
to
15%
O2,
dry
basis)
Catalyst
CH2O
CH
2
O
CH
2
O
CO
CO
CO
Temp.
inlet
outlet
Percent
inlet
outlet
Percent
(
deg
F)
(
ppm)
(
ppm)
Reduction
(
ppm)
(
ppm)
Reduction
556
16.7
9.0
46.1
83.0
27.1
67.3
483
24.1
15.9
34.0
258.9
80.1
69.1
451
22.4
17.3
22.8
244.7
91.3
62.7
518
15.3
8.2
46.4
74.9
28.8
61.5
537
17.2
10.5
39.0
115.6
40.1
65.3
567
16.1
7.5
53.4
75.3
26.9
64.3
500
19.0
10.0
47.4
134.5
44.0
67.3
548
16.7
8.7
47.9
77.6
28.1
63.8
558
17.4
8.7
50.0
78.9
27.4
65.3
503
18.0
12.2
32.2
122.6
50.9
58.5
508
18.4
11.9
35.3
119.6
47.9
59.9
569
18.2
9.4
48.4
80.6
29.2
63.8
538
15.8
8.3
47.5
95.0
31.8
66.5
559
18.3
9.5
48.1
91.8
29.9
67.4
557
18.0
9.2
48.9
87.6
29.6
66.2
Table
2:
Emissions
Data
for
4­
Stroke
Lean
Burn
Natural
Gas
Engine
(
corrected
to
15%
O2,
dry
basis)
Catalyst
CH2O
CH
2
O
CH
2
O
CO
CO
CO
Temp.
inlet
outlet
Percent
inlet
outlet
Percent
(
deg
F)
(
ppm)
(
ppm)
Reduction
(
ppm)
(
ppm)
Reduction
704
34.1
10.9
68.0
329.7
22.0
93.3
676
36.6
9.3
74.6
314.7
14.3
95.5
640
35.0
6.4
81.7
305.0
11.6
96.2
657
33.4
9.8
70.7
314.0
16.0
94.9
685
41.8
14.0
66.5
420.1
30.2
92.8
723
33.8
8.2
75.7
334.9
19.6
94.1
694
37.9
7.5
80.2
322.7
13.2
95.9
641
37.2
12.8
65.6
363.8
20.3
94.4
704
33.8
10.5
68.9
326.1
21.4
93.4
705
34.4
11.0
68.0
332.5
22.2
93.3
701
34.0
10.9
67.9
330.1
22.2
93.3
704
34.9
10.7
69.3
328.6
21.9
93.3
734
36.6
9.4
74.3
348.3
23.7
93.2
685
33.9
11.4
66.4
348.8
22.7
93.5
707
35.1
10.6
69.8
335.3
22.7
93.2
698
34.8
10.6
69.5
333.9
22.4
93.3
6
Table
3:
Emissions
Data
for
4­
Stroke
Diesel
Engine
(
corrected
to
15%
O2,
dry
basis)
Catalyst
CH2O
CH
2
O
CH
2
O
CO
CO
CO
Temp.
inlet
outlet
Percent
inlet
outlet
Percent
(
deg
F)
(
ppm)
(
ppm)
Reduction
(
ppm)
(
ppm)
Reduction
809
1.5
0.58
61.3
44.1
12.1
72.6
830
0.93
0.16
82.8
44.1
12.4
71.9
837
1.4
0.14
90.0
76.8
13.2
82.8
804
1.6
0.13
91.9
76.2
12.0
84.3
878
1.3
0.16
87.7
39.8
11.4
71.4
792
1.6
0.17
89.4
43.5
12.1
72.2
809
1.4
0.15
89.3
41.6
12.0
71.2
775
1.6
0.15
90.6
42.7
11.8
72.4
813
1.6
0.16
90.0
45.2
11.8
73.9
811
1.4
0.15
89.3
41.6
12.5
70.0
Table
4:
Emissions
Data
for
4­
Stroke
Rich
Burn
Engine
(
7A)
(
corrected
to
15%
O2,
dry
basis)
Catalyst
CH
2
O
CH
2
O
CH
2
O
CO
CO
CO
NOx
NOx
NOx
Temp.
inlet
outlet
Percent
inlet
outlet
Percent
inlet
outlet
Percent
(
deg
F)
(
ppm)
(
ppm)
Reduction
(
ppm)
(
ppm)
Reduction
(
ppm)
(
ppm)
Reduction
955
1.72
0.34
80.47
915.73
369.25
59.68
645.48
18.23
97.18
945
1.51
0.35
76.47
1059.9
745.10
29.70
615.67
20.02
96.75
887
1.47
0.37
74.52
1215.0
664.90
45.28
588.75
16.34
97.23
911
1.35
0.36
73.11
1535.3
597.42
61.09
524.38
13.41
97.44
7
Figure
1:
Percent
Reduction
vs.
Catalyst
Temperature
2­
Stroke
Lean
Burn
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
80.0
90.0
100.0
440
460
480
500
520
540
560
580
Catalyst
Temperature
(
deg
F)

Percent
Reduction
(

corrected
to
15%

O2)
CH2O
Percent
Reduction
CO
Percent
Reduction
8
Figure
2:
Percent
Reduction
vs.
Catalyst
Temperature
4­
Stroke
Lean
Burn
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
80.0
90.0
100.0
620
640
660
680
700
720
740
Catalyst
Temperature
(
deg
F)

Percent
Reduction
(

corrected
to
15%

O2)
CH2O
Percent
Reduction
CO
Percent
Reduction
9
Figure
3:
Percent
Reduction
vs.
Catalyst
Temperature
4­
Stroke
Diesel
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
80.0
90.0
100.0
760
780
800
820
840
860
880
900
Catalyst
Temperature
(
deg
F)

Percent
Reduction
(

corrected
to
15%

O2)
CH2O
Percent
Reduction
CO
Percent
Reduction
10
Figure
4:
Percent
Reduction
vs.
Catalyst
Temperature
4­
Stroke
Rich
Burn
0.00
10.00
20.00
30.00
40.00
50.00
60.00
70.00
80.00
90.00
880.00
890.00
900.00
910.00
920.00
930.00
940.00
950.00
960.00
Catalyst
Temperature
(
deg
F)

Percent
Reduction
(

corrected
to
15%

O2)
CH2O
Percent
Reduction
CO
Percent
Reduction
Linear
(
CO
Percent
Reduction)

Linear
(
CH2O
Percent
Reduction)
11
Figure
5:
Inlet
Concentration
vs.
Temperature
2­
Stroke
Lean
Burn
0.0
50.0
100.0
150.0
200.0
250.0
300.0
440
460
480
500
520
540
560
580
Catalyst
Temperature
(
deg
F)

Concentration
(

ppm)

(

corrected
to
15%

O2)
CH2O
Concentration
CO
Concentration
Linear
(
CO
Concentration)

Linear
(
CH2O
Concentration)
12
Figure
6:
Inlet
Concentration
vs.
Temperature
4­
Stroke
Lean
Burn
0.0
50.0
100.0
150.0
200.0
250.0
300.0
350.0
400.0
450.0
620
640
660
680
700
720
740
Catalyst
Temperature
(
deg
F)

Concentration
(

ppm)

(

corrected
to
15%

O2)
CH2O
Concentration
CO
Concentration
Linear
(
CO
Concentration)

Linear
(
CH2O
Concentration)
13
Figure
7:
Inlet
Concentration
vs.
Temperature
4­
Stroke
Diesel
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
80.0
90.0
760
780
800
820
840
860
880
900
Catalyst
Temperature
(
deg
F)

Concentration
(

ppm)

(

corrected
to
15%

O2)
CH2O
Concentration
CO
Concentration
Linear
(
CO
Concentration)

Linear
(
CH2O
Concentration)
