COST
ANALYSIS
OF
MERCURY
MONITORING
TECHNIQUES
TASK
1:
COST
OF
MERCURY
MEASUREMENT
TECHNIQUES
DRAFT
REPORT,
Revision
1
3
NOVEMBER
2003
Infrastructure,
buildings,

environment,
communications
Cost
Analysis
of
Mercury
Monitoring
Techniques
Task
1:
Cost
of
Mercury
Monitoring
Techniques
Draft
Report,
Revision
1
Brent
Hall
Project
Manager
Carl
Singer
Project
Engineer
Michiel
Doorn
Project
Engineer
Prepared
for:

U.
S.
Environmental
Protection
Agency
Clean
Air
Markets
Division
Prepared
by:

ARCADIS
G&
M,
Inc.

4915
Prospectus
Drive
Suite
F
Durham
North
Carolina
27713
Tel
919
544
4535
Fax
919
544
5690
Our
Ref.:

RN096402.0001
Date:

3
November
2003
This
document
is
intended
only
for
the
use
of
the
individual
or
entity
for
which
it
was
prepared
and
may
contain
information
that
is
privileged,
confidential,
and
exempt
from
disclosure
under
applicable
law.
Any
dissemination,
distribution,
or
copying
of
this
document
is
strictly
prohibited.
i
Table
of
Contents
List
of
Tables
iii
List
of
Acronyms
iv
1.
Introduction
1
1.1
Scope
1
1.2
Clear
Skies
Act
2
1.2.1
Common
Provisions
3
1.2.2
Mercury
Emissions
Reductions
3
1.2.3
Performance
Standards
for
New
Sources
4
1.3
General
Assumptions
4
2.
Ontario
Hydro
(
OH)
Analysis
7
2.1
Background
7
2.1.1
Scope
and
Applicability
7
2.1.2
Technique
7
2.1.3
Accuracy
and
Precision
8
2.2
Assumptions
8
2.3
Capital
Costs
8
2.4
Annual
Costs
9
2.4.1
Labor
Costs
9
2.4.2
Subcontract
and
Other
Direct
Costs
(
ODC)
9
2.4.3
Annualized
Capital
Costs
9
2.4.4
Total
Annualized
Cost
10
2.5
Uncertainties
10
3.
Carbon
Trap
Analysis
11
3.1
Background
11
3.1.1
Preliminary
Methods
11
3.1.2
Scope
and
Applicability
13
3.1.3
Technique
13
3.1.4
Accuracy
and
Precision
15
3.2
General
Assumptions
15
3.3
Capital
Costs
17
3.4
Annual
Costs
18
ii
Table
of
Contents
3.4.1
Labor
Costs
18
3.4.2
Subcontract
and
ODC
19
3.4.3
Annualized
Capital
costs
19
3.4.4
Total
Annualized
Costs
19
3.5
Uncertainties
20
4.
CEMS
Analysis
22
4.1
Background
22
4.1.1
Scope
and
Applicability
23
4.1.2
Technique
29
4.1.3
Accuracy
and
Precision
29
4.2
General
Assumptions
31
4.3
Capital
Costs
31
4.4
Annual
Costs
33
4.4.1
Annual
Labor
Costs
33
4.4.2
Subcontract
and
ODC
34
4.4.3
Annualized
Capital
Costs
35
4.4.4
Total
Annualized
Costs
35
4.5
Uncertainties
35
5.
Fuel
Analysis
36
5.1
Background
and
Methods
36
5.1.1
Scope
and
Applicability
36
5.1.2
Accuracy
and
Precision
37
5.2
Assumptions
38
5.3
Annual
Costs
39
5.3.1
Labor
Costs
39
5.3.2
Subcontract
and
ODC
40
5.3.3
Total
Annualized
Costs
40
5.4
Uncertainties
40
6.
Discussion
and
Summary
42
6.1
Comparison
Carl:
please
update
this
section
with
revised
costs.
42
6.2
Barriers
to
Cost
Analysis
43
6.3
Suggestions
for
Improvements
and
Additional
Research
44
iii
Table
of
Contents
List
of
Tables
Table
1­
1.
Projected
Emissions
Reductions
under
Clear
Skies
Act
(
EPA,
2002i.)
3
Table
3­
1.
Draft
MQO
for
Sorbent
Total
Mercury
by
internal
Frontier
Geosciences
Method
12
Table
3.2:
Assumed
Carbon
Trap
Analytical
Summary
17
Table
4­
1.
Overview
of
Mercury
CEMs
that
are
(
expected
to
be)
Commercially
Available
in
the
United
States
24
Table
4­
2.
Operation
&
Maintenance
27
Table
4­
3.
CEMS
Capital
Cost
(
March
2003)
32
Table
5­
1.
ASTM
D­
3684­
01
Bias
37
Table
6­
1.
Summary
of
Mercury
Annual
Analysis
Costs
42
iv
Table
of
Contents
List
of
Acronyms
ADAES
ADA
Environmental
Solutions
ASTM
American
Society
for
Testing
and
Materials
AWMA
Air
and
Waste
Management
Association
BACT
Best
Available
Control
Technology
BART
Best
Available
Retrofit
Technology
CAA
Clean
Air
Act
CEM
Continuous
Emission
Monitor
CEMS
Continuous
Emission
Monitoring
System
CRM
Certified
Reference
Material
CVAAS
Cold
Vapor
Atomic
Absorption
Spectroscopy
CVAFS
Cold
Vapor
Atomic
Fluorescence
Spectroscopy
DOE
Department
of
Energy
EPA
United
States
Environmental
Protection
Agency
EPRI
Electric
Power
Research
Institute
ETV
Environmental
Test
Verification
ICR
Information
Collection
Request
IGCC
Integrated
Gasification
Combined
Cycle
LAER
Lowest
Achievable
Emission
Rate
MESA
Mercury
Speciation
Adsorption
MQO
Method
Quality
Objectives
v
Table
of
Contents
NAAQS
National
Ambient
Air
Quality
Standards
NIST
National
Institute
of
Standards
&
Technology
NSPS
New
Source
Performance
Standard
NSR
New
Source
Review
OAQPS
Office
of
Air
Quality
Planning
and
Standards
ODC
Other
Direct
Costs
OH
Ontario
Hydro
PM
Particulate
Matter
QA
Quality
Assurance
QSEM
Quick
Silver
Emissions
Monitor
RATA
Relative
Accuracy
Test
Audit
RDF
Refuse
Derived
Fuel
RSD
Relative
Standard
Deviation
SIP
State
Implementation
Plan
SOP
Standard
Operating
Procedures
TSCA
Toxic
Substances
Control
Act
WRAP
Western
Regional
Air
Partnership
1
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
1.
Introduction
1.1
Scope
Anthropogenic
mercury
emissions
are
considered
a
potential
serious
health
hazard.
As
a
by­
product
of
combustion
processes,
mercury
emitted
to
the
air
can
be
transformed
into
the
highly
toxic
methyl­
mercury
form
by
means
of
a
natural
process
in
the
atmosphere.
The
Mercury
Study
and
the
Utility
Hazardous
Air
Pollutant
Reports1
to
congress
have
identified
coal­
fired
power
plants
as
the
largest
anthropogenic
source
of
mercury
emissions.
In
December
2000,
the
United
States
Environmental
Protection
Agency
(
EPA)
made
the
determination
to
regulate
mercury
emissions
from
coal­
fired
power
plants
under
section
112
of
the
Clean
Air
Act
(
CAA)
Amendments
of
1990.
EPA
is
required
to
promulgate
mercury
emissions
standards
to
electricity
generating
utility
boilers
by
December
2004,
to
be
implemented
by
December
2007.
According
to
the
explicit
language
of
the
Clean
Air
Act,
these
standards
must
reflect
the
utilization
of
"
Maximum
Achievable
Control
Technology"
(
MACT)
for
mercury.

As
a
possible
alternative
the
Bush
Administration
proposed
legislation
in
2002
under
the
Clear
Skies
Initiative
to
establish
a
flexible,
market­
based
program
to
reduce
and
cap
emissions
of
sulfur
dioxide,
nitrogen
oxides,
and
mercury
from
coal­
fired
power
plants
and
other
electric
power
generators.
Such
a
program
is
already
in
place
for
sulfur
dioxide
and
nitrogen
oxides
under
the
Acid
Rain
Program
and
the
NOX
State
Implementation
Plan
(
SIP)
Call.

As
part
of
a
mercury
control
or
cap
and
trade
program,
EPA
will
need
to
implement
monitoring
requirements
that
are
cost­
effective,
technically
viable,
contain
sufficient
quality
assurance
(
QA),
and
are
based
on
technologies
and
equipment
that
are
compliance­
capable.
Mercury
emission
reductions
under
CAA­
MACT
or
under
Clear
Skies
may
have
different
monitoring
requirements.

EPA
has
conducted
several
studies
to
describe
and
compare
various
mercury
monitoring
technologies,
as
well
as
a
cost­
effectiveness
study
based
on
preliminary
1
U.
S.
EPA,
1998.
Study
of
Hazardous
Air
Pollutant
Emissions
from
Electric
Utility
Steam
Generating
Units
 
Final
Report
to
Congress.
U.
S.
EPA,
Office
of
Air
Quality
Planning
and
Standards
Research
Triangle
Park,
NC
27711.
February
1998.
EPA­
453/
R­
98­
004a
2
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
information.
2
This
report
is
a
continuation
of
this
previous
study
and
provides
a
detailed
cost
analysis
of
four
mercury­
monitoring
technologies
applicable
at
coal­
fired
power
plants.
The
technologies
are
fuel
analysis,
the
American
Society
for
Testing
and
Materials
(
ASTM)
Ontario
Hydro
(
OH)
method,
iodated
carbon
traps,
as
well
as
continuous
emissions
monitoring
systems
(
CEMS).

1.2
Clear
Skies
Act
The
Clear
Skies
Act
would
amend
Title
IV
of
the
CAA
to
establish
new
cap­
and­
trade
programs
requiring
reductions
of
SO2,
NOX,
and
mercury
emissions
from
electric
generating
facilities
and
amends
Title
I
of
the
CAA
to
provide
an
alternative
regulatory
classification
for
units
subject
to
the
cap­
and­
trade
programs.
The
market­
based
approach
should
allow
for
the
implementation
of
multi­
pollutant
control
systems.
Also,
the
ability
to
carry
over
unused
allowances
is
believed
to
create
incentives
for
early
reductions,
delivering
environmental
and
human
health
benefits
sooner,
in
exchange
for
a
longer
compliance
window
for
sources.
Ultimately,
the
cap
ensures
that
the
emission
reduction
goals
are
met
and
maintained.
The
EPA
projects
emission
reductions
for
SO2,
NOX,
and
mercury
of
73,
67,
and
69
percent
respectively,
compared
to
2000
emissions.
Table
1­
1
includes
projected
emissions
reductions.
The
Act
does
not
require
reductions
in
CO2
emissions,
although
there
is
a
provision
to
maintain
CO2
monitoring
and
reporting
requirements
that
currently
exist
under
the
CAA.

As
revised
by
the
Clear
Skies
Act,
Title
IV
of
the
CAA
has
five
Parts.
Part
A
contains
provisions
common
to
the
control
of
all
three
pollutants.
Part
B
contains
provisions
specifically
for
sulfur
dioxide
emission
reductions.
Part
C
contains
provisions
specifically
for
nitrogen
oxides
emission
reductions.
Part
D
contains
provisions
specifically
for
mercury
emission
reductions.
Part
E
contains
performance
standards
for
affected
units
and
provisions
for
research,
environmental
monitoring,
and
assessment.
Part
E
of
the
Clear
Skies
Act
includes
substantial
changes
in
the
New
Source
Review
(
NSR)
and
New
Source
Performance
Standard
(
NSPS)
sections
of
the
existing
CAA
(
EPA,
2002i).
Below,
summaries
are
provided
for
Parts
A,
D,
and
E.

2
Cadmus
Group,
2003.
Mercury
Monitoring
Cost­
Effectiveness
Analysis.
Draft
report
prepared
for
Ruben
Deza,
U.
S.
EPA.
Clean
Air
Markets
Division,
Washington
DC.
Prepared
by
The
Cadmus
Group,
6330
Quadrangle
Drive,
Chapel
Hill.
3
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
1.2.1
Common
Provisions
The
Clear
Skies
Act
establishes
a
new
Part
A
of
Title
IV
of
the
CAA,
which
contains
the
program
elements
shared
by
the
SO2,
NOX,
and
mercury
programs.
A
cap­
and­
trade
program
will
be
implemented
for
each
pollutant.
Common
definitions,
allowance
system
procedures,
monitoring,
permitting
and
compliance
requirements,
penalties
for
non­
compliance,
opt­
ins
and
auction
procedures
apply
to
the
new
trading
programs
and
are
modeled
largely
after
the
existing
Acid
Rain
Program.

Table
1­
1.
Projected
Emissions
Reductions
under
Clear
Skies
Act
(
EPA,
2002i.)

Clear
Skies
Act
emission
caps
Actual
emissions
in
2000
First
phase
of
reductions
Second
phase
of
reductions
Total
reduction
at
full
implementation
SO2
(
million
tons)
11.2
4.5
(
in
20101)
3
(
in
20181)
73%

NOX
2
(
million
tons)
5.1
2.1
(
in
20081)
1.7
(
in
20181)
67%

Mercury
(
tons)
48
26
(
in
2010)
15
(
in
20181)
69%

Because
sources
can
reduce
emissions
early,
earn
allowances
for
those
actions,
and
use
these
allowances
later,
actual
emission
levels
will
be
higher
that
the
cap
in
the
first
years
of
these
phases.

The
NOX
cap
is
divided
between
two
zones
with
separate
trading
programs
under
each
zone.
Zone
1
generally
includes
Eastern
states
and
Zone
2
includes
Western
states
and
territories.

The
Administrator
must
establish
an
allowance
system
for
SO2,
NOX,
and
mercury
that
is
essentially
the
same
as
the
current
Acid
Rain
Program
but
provides
for
a
safety
valve,
i.
e.,
a
direct
sale
of
allowances
by
the
Administrator
at
a
fixed
price
for
use
in
meeting
the
requirement
to
hold
allowances
at
least
equal
to
annual
emissions.
Criteria
and
the
process
will
be
established
by
which
the
Administrator
may
recommend
to
Congress
adjustment
of
the
total
amounts
of
allowances
available
(
whether
through
allocation
or
auction)
starting
in
2018
under
the
new
SO2,
NOX,
and
mercury
trading
programs.

1.2.2
Mercury
Emissions
Reductions
The
Clear
Skies
Act
establishes
Part
D,
which
contains
the
new,
annual
caps
on
total
mercury
allowances
and
new,
allocation
procedures
starting
January
1,
2010.
The
new
mercury
trading
program
is
analogous
to
the
new
SO2
and
NOX
trading
programs
and
cover
the
same
coal­
fired
units.
Annual
mercury
emissions
are
capped
at
26
tons
starting
in
2010
and
15
tons
starting
in
2018.
Each
year,
the
percentages
of
allowances
4
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
allocated
and
auctioned
are
the
same
as
under
the
new
sulfur
and
NOX
trading
programs.

1.2.3
Performance
Standards
for
New
Sources
To
ensure
that
all
new
affected
units
have
appropriate
controls,
Part
E
establishes
performance
standards
for
all
new
boilers,
combustion
turbines,
and
IGCC
plants
covered
under
the
Act.
"
New"
units
are
those
that
commence
construction
or
reconstruction
after
the
date
of
enactment.

These
statutory
performance
standards
include
emission
limits
for
four
air
pollutants:
NOX,
SO2,
mercury,
and
particulate
matter
(
PM).
The
mercury
emission
limit
applies
only
to
coal.
In
addition,
a
PM
emission
limit
is
established
for
existing
oil­
fired
boilers
to
ensure
reductions
of
nickel
from
such
units.

All
units
subject
to
a
performance
standard
must
monitor
emissions
using
continuous
emission
monitoring
systems
(
CEMS)
and
use
averaging
times
similar
to
current
NSPS.

1.3
General
Assumptions
The
sampling
program
is
assumed
to
be
based
on
a
single
stack
or
single
unit
testing.
No
discount
has
been
allowed
for
multiple
site
testing
at
a
plant
or
multiple
plant
testing.
Multiple
units
being
served
by
a
common
stack
would
be
tested
under
a
single
sampling
program.

For
cost
estimation,
it
is
assumed
that
the
plant
will
contract
for
OH
sampling
and
analysis.
Costs
for
the
plant
to
administer
the
OH
analysis
program
for
Hg
are
based
on
the
Information
Collection
Request
(
ICR)
supporting
statement,
3
which
assumes
the
following
hours
on
a
yearly
basis:


For
obtaining
an
OH
sampling
contract,
40
technical
hours.

3"
U.
S.
EPA,
"
Standard
Form
83­
1
Supporting
Statement
for
OMB
Review
of
ICR
No._____:[
sic]
Information
Collection
Request
for
Electric
Utility
Steam
Generating
Unit
Mercury
Emissions
Information
Collection
Effort,"
November
16,
1998.
available
at
http://
www.
epa.
gov/
ttn/
atw/
combust/
utiltox/
114_
ss19.
pdf.
5
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1

For
oversight
of
OH
sampling
events,
8
technical
hours
per
test.


To
compile
and
review
analytical
data,
8
technical
hours
per
sampling
event.


To
submit
test
results,
1
technical
hour
per
event.


For
all
technical
work,
associated
management
hours
have
been
assumed
to
total
5
percent
of
technical
hours
and
clerical
hours
have
been
assumed
to
total
10
percent
of
technical
hours.

Cost
for
plant
labor
hours
have
been
assumed
to
be
the
same
as
used
in
the
ICR
Supporting
Statement:

Technical
hours
$
47.50,
Management
hours
$
57.27,
and
Clerical
hours
$
26.96.

To
estimate
the
cost
associated
with
OH
sampling,
ARCADIS
obtained
three
estimates
for
a
sampling
event
of
three
OH
runs
at
one
location
at
a
utility
stack
within
500
miles
traveling
distance
from
the
sampling
firm:
$
8,500,4
15,000,5
and
between
12,000
and
15,0006.
The
ICR
supporting
statement
contained
an
estimated
cost
of
$
40,020
for
concurrent
OH
tests
at
two
locations7.
Consequently,
costs
for
contracting
for
OH
sampling
and
analysis
are
estimated
at
$
15,000
per
sampling
event
of
three
tests.
It
is
assumed
that
this
number
can
be
broken
out
as
follows:
$
4,000
for
travel
and
lodging
4
Personal
communication
from
Brian
Jacobs,
METCO
Environmental,
to
Carl
Singer,
ARCADIS,
on
March
26,
2003.

5
Personal
communication
from
Gene
Stephenson,
ARCADIS,
to
Carl
Singer,
ARCADIS,
on
March
11,
2003.

6
Personal
communication
from
Ralph
Roberson,
RMB
Consulting,
to
Carl
Singer,
ARCADIS,
on
March
11,
2003.

7"
U.
S.
EPA,
"
Standard
Form
83­
1
Supporting
Statement
for
OMB
Review
of
ICR
No._____:[
sic]
Information
Collection
Request
for
Electric
Utility
Steam
Generating
Unit
Mercury
Emissions
Information
Collection
Effort,"
November
16,
1998.
available
at
http://
www.
epa.
gov/
ttn/
atw/
combust/
utiltox/
114_
ss19.
pdf
6
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
of
the
OH
contractor
and
$
11,000
for
the
three­
test
event.
This
can
be
used
to
estimate
the
costs
of
different
sampling
events.
For
example,
a
nine­
test
event
would
cost
(
3
×
$
11,000)
+
$
4,000
=
$
37,000.
7
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
2.
Ontario
Hydro
(
OH)
Analysis
2.1
Background
The
ASTM
recently
issued
the
OH
method
under
designation
D6784­
02
approved
April
10,
2002.
The
OH
method
is
a
gas
sampling
method
that
differentiates
mercury
into
particle­
bound,
oxidized,
and
elemental
fractions.
The
OH
method
was
widely
used
as
the
basis
for
sampling
plans
under
the
recent
mercury
ICR
phase
III.

2.1.1
Scope
and
Applicability
The
OH
method
is
applicable
to
the
determination
of
elemental,
oxidized,
particlebound
and
total
mercury
emissions
from
coal­
fired
stationary
sources.
Resulting
emissions
are
expressed
in
concentration
terms,
µ
g/
Nm3.
Conversion
of
OH
results
into
an
annual
mass
emission
will
require
application
of
additional
methods.

Though
the
method
explicitly
claims
applicability
to
coal­
fired
stationary
sources,
the
method
does
not
appear
to
exclude
the
use
of
other
fuels
such
as
coke,
orimulsion,
oil,
tires,
refuse
derived
fuel
(
RDF),
or
biomass.
Use
of
alternate
fuels
may
produce
ash
that
may
further
confound
speciation
bias
in
the
OH
method
but
is
not
expected
to
effect
total
mercury.

2.1.2
Technique
The
OH
method
involves
extracting
a
flue
gas
sample
through
either
a
modified
method
5
or
method
17
sampling
train.
Sample
location
and
traverse
points
are
determined
using
Methods
1
or
2.
Mercury
is
absorbed
from
the
sample
into
various
solutions
in
the
train.
The
gas
volume
sampled
is
metered
as
the
sample
exits
the
sampling
train.
At
the
conclusion
of
an
integrated
sampling
event,
the
train
is
recovered
generating
analytical
samples
specific
to
particular
mercury
forms.
The
analytical
samples
are
digested
and
analyzed
in
a
laboratory.
8
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
2.1.3
Accuracy
and
Precision
There
are
no
known
biases
to
the
OH
method
for
total
Hg,
however
several
conditions
can
affect
the
speciation
of
the
sample.
8
The
relative
standard
deviations
for
gas
phase
elemental
and
oxidized
Hg
were
less
than
11
percent
for
concentrations
greater
than
3
µ
g/
m3
and
less
than
34
percent
for
concentrations
less
than
3
µ
g/
m3.9
A
number
of
revisions
to
the
OH
method
have
been
made
since
the
Method
301
evaluation
cited
in
the
OH
method.
The
precision
of
the
OH
method
is
likely
to
have
improved
since
the
initial
Method
301
evaluation.

2.2
Assumptions
For
the
purposes
of
calculating
total
mercury
emissions,
12
OH
sampling
events
are
assumed
with
3
OH
tests
required
for
each
sampling
event.
The
resulting
36
tests
are
expected
to
provide
an
accuracy
of
±
8.4%,
similar
to
that
expected
from
a
nine­
point
Relative
Accuracy
Test
Audit
(
RATA)
for
a
CEMS.

For
cost
estimation,
it
is
assumed
that
the
plant
will
contract
for
OH
sampling
and
analysis.
Further,
the
sampling
program
is
assumed
to
be
based
on
each
stack
such
that
multiple
units
being
served
by
a
common
stack
would
be
tested
under
a
single
sampling
program.

2.3
Capital
Costs
An
OH
analysis
program
is
likely
to
be
based
on
sampling
after
the
particulate
control
device
and
is
not
expected
to
result
in
any
significant
capital
costs.
Current
stack
sampling
locations
are
expected
to
be
adequate
for
OH
sampling
and
no
system
modifications
will
be
required.

For
purposes
of
estimation,
it
is
assumed
that
plants
will
contract
with
sampling
firms
for
OH
sampling.
Sampling
firms
are
expected
to
supply
their
own
sampling
equipment
as
part
of
their
cost
of
doing
business.

8
D6784­
02
Standard
Test
Method
for
Elemental,
Oxidized,
Particle­
Bound,
and
Total
Mercury
in
Flue
Gas
Generated
from
Coal­
Fired
Stationary
Sources
(
Ontario
Hydro
Method),
ASTM
International,
West
Conshohoken
Pennsylvania,
June
1,
2002.

9
ibid.
9
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
2.4
Annual
Costs
2.4.1
Labor
Costs
Costs
for
administering
the
OH
analysis
program
are
estimated
using
values
from
the
ICR
supporting
statement.
10
The
following
technical
hours
have
been
assumed
per
year:


For
obtaining
an
OH
sampling
contract,
40
technical
hours.


For
oversight
of
12
OH
sampling
events,
288
technical
hours.


To
compile
and
review
analytical
data,
96
technical
hours.


To
submit
test
results,
12
technical
hours
have
been
assumed
annually.

Total
Technical
hours
for
administration
of
the
program
is
therefore
expected
to
be
436
hours
per
year.
Accordingly,
labor
costs
for
the
Hg
analysis
program
at
the
stack
level
are
estimated
at
$
23,134
annually,
including
management
and
clerical
costs.

2.4.2
Subcontract
and
Other
Direct
Costs
(
ODC)

As
indicated
in
Section
1.1.1
costs
for
contracting
for
OH
sampling
and
analysis
are
estimated
a
$
15,000
per
sampling
event
of
three
tests.
Subcontracting
and
ODC
costs
for
an
OH
sampling
program
incorporating
12
sampling
events
is,
therefore,
estimated
at
$
180,000
per
year
for
each
stack
in
the
program.

2.4.3
Annualized
Capital
Costs
No
significant
capital
costs
are
expected
for
an
OH
based
monitoring
plan.
Annualized
capital
costs
are
therefore
estimated
at
$
0/
year.

10
U.
S.
EPA,
"
Standard
Form
83­
1
Supporting
Statement
for
OMB
Review
of
ICR
No._____:[
sic]
Information
Collection
Request
for
Electric
Utility
Steam
Generating
Unit
Mercury
Emissions
Information
Collection
Effort,"
November
16,
1998.
available
at
http://
www.
epa.
gov/
ttn/
atw/
combust/
utiltox/
114_
ss19.
pdf.
10
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
2.4.4
Total
Annualized
Cost
Total
Annualized
costs
for
the
OH
emissions
monitoring
plan
is
estimated
at
$
203,134.

2.5
Uncertainties
OH
is
presumed
to
be
the
reference
method
for
potential
mercury
air
emissions
regulation.
The
draft
performance
specification
for
mercury
continuous
monitoring
systems
cites
40CFR266,
Appendix
IX,
Section
3.1.
Use
of
different
reference
methods
such
as
Method
29,
Method
101A
or
Method
0060
are
not
expected
to
drastically
affect
the
cost
of
each
test,
however,
the
monitoring
program
may
require
adjusting
to
achieve
acceptable
accuracy
and
precision.

The
cost
of
an
OH
sampling
program
for
determining
emissions
is
largely
driven
by
the
frequency
of
sampling
and
the
number
of
samples.
For
this
analysis,
we
have
assumed
triplicate
runs
of
OH
on
a
monthly
basis.
Sites
may
require
more
or
fewer
samples
to
achieve
the
same
confidence
interval
depending
on
the
actual
variability
encountered.
Variability
is
expected
to
arise
from
unit
operation
and
method
chosen
for
analysis.
36
OH
runs
were
selected
as
a
basis
assuming
a
25%
relative
standard
deviation
(
RSD)
and
a
desired
confidence
interval
of
±
8.4%.

An
OH
sampling
program
for
determining
emissions
relies
on
a
representative
sample
being
collected.
Such
a
program
must
be
designed
to
collect
a
representative
sample.
Confounding
factors
may
include
changes
in
fuel,
daily,
weekly,
and
seasonal
variations
in
operations,
and
catastrophic
events.
If
there
is
significant
serial
correlation
in
mercury
sampling,
the
number
of
tests
may
need
to
be
increased
to
achieve
the
same
confidence
interval.

While
it
appears
probable
that
an
OH
sampling
program
could
be
designed
to
provide
an
accurate
estimate
of
the
amount
of
mercury
emitted
from
a
stack,
OH
sampling
may
be
inappropriate
for
an
emission
trading
program.
Specifically,
it
is
doubtful
that
an
OH
sampling
program
could
provide
information
with
the
same
timeliness
as
a
Hg
CEMS.
With
an
OH
sampling
program,
it
would
be
difficult
to
identify
changes
in
emissions
resulting
from
either
operational
or
mechanical
changes
in
the
system;
failure
in
a
control
system,
such
as
a
carbon
injection
system,
would
not
likely
be
discernable
using
OH
sampling.
The
applicability
of
an
OH
sample
program
is
therefore
likely
dependent
on
other
records
and
a
workable
data
substitution
mechanism.
Long
intervals
between
sampling
events
risk
compromising
the
credibility
of
OH­
based
emissions
estimates.
11
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
3.
Carbon
Trap
Analysis
3.1
Background
The
carbon
sampling
trap
method
is
currently
under
further
development
in
the
U.
S.
It
is
in
the
laboratory/
pilot
phase
and
is
being
tested
on
a
limited
scale.
No
references
were
found
that
support
that
the
method
has
been
applied
elsewhere.

Carbon
trap
sampling
involves
drawing
a
gas
sample
through
traps
filled
with
a
sorbent,
generally
activated
carbon.
Gaseous
mercury
present
in
the
gas
is
adsorbed
onto
the
carbon.
The
traps
are
subsequently
removed
from
the
sampling
apparatus
and
shipped
to
a
laboratory
for
analysis.
For
mercury
emissions
monitoring,
exposure
of
nominally
one
week
is
envisioned.

3.1.1
Preliminary
Methods
Through
literature
research
and
industry
contacts,
two
general
carbon
trap
methods
under
development
in
the
U.
S.
were
identified.
One
method,
using
intermittent
integrating
technology
coupled
with
Method
101A,
is
being
designed
by
a
team
headed
by
Walter
S.
Smith
and
Associates,
Inc.,
of
Apex,
North
Carolina.
The
method
is
described
by
the
developer
as
a
prototype;
technical
specifics
are
not
available,
and
no
tests
have
been
performed.
11
The
second
method
is
an
adaptation
by
the
Electric
Power
Research
Institute
(
EPRI,
"
the
manufacturer")
of
a
method
that
was
originally
developed
by
scientists
at
Brooks
Rand
Limited
and
Frontier
Geosciences
of
Seattle,
Washington,
in
1991.12
The
earlier
version
of
the
method
was
identified
as
Mercury
Speciation
Adsorption
(
MESA).
The
most
recent
adaptation
combines
a
larger
version
of
the
iodated
carbon
trap
used
with
the
MESA
method
and
a
specially
designed
gas
meter
and
probe
assembly
to
provide
a
cumulative
average
mercury
concentration
for
an
extended
period
(
1
week
to
1
month).
The
EPRI
method,
the
Quick
Silver
Emissions
Monitor
(
QSEM),
differs
significantly
from
MESA,
and
no
peer­
reviewed
publications
are
yet
available.
The
QSEM
is
currently
undergoing
testing
in
North
Carolina
and
Missouri.
Several
groups
are
assisting
with
the
development
of
this
11
Personal
communication
from
Walter
Smith,
Walter
Smith
Associates,
Inc.
to
Michiel
Doorn,

ARCADIS,
by
email,
June
26
and
email
June
24,
2003.

12
Personal
communication
from
Sharon
Sjostrom,
ADAES,
to
Michiel
Doorn,
ARCADIS,
by
email
June
24,
2003.
12
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
method
under
contract
to
EPRI:
ADA
Environmental
Solutions
(
ADAES)
designed
the
sampling
consoles
and
is
currently
coordinating
the
testing
efforts
and
subcontractor
activities,
Frontier
Geosciences
provides
the
iodated
carbon
traps
and
conducts
the
trap
analysis,
and
Apex
Instruments
is
currently
providing
the
sampling
hardware.
Results
from
the
Missouri/
North
Carolina
field
tests
are
expected
in
the
near
future.
Additional
testing
is
planned
for
late
summer
2003.
Draft
Method
Quality
Objectives
(
MQO)
for
laboratory
analysis
of
the
iodated
carbon
traps
for
sorbent
total
mercury
are
included
in
Table
3­
1
below.

Further
discussion
will
only
cover
the
QSEM
method.
A
recent
modification
(
Model
2)
includes
an
adapter
to
allow
insertion
of
a
smaller­
diameter
trap
appropriate
for
a
three­
hour
test
to
allow
for
better
comparison
to
an
OH
test.
A
draft
operating
manual
for
Model
1
is
available
from
the
manufacturer.
Also
available
are
two
proprietary
standard
operating
procedures
(
SOP).
The
first
SOP
addresses
total
mercury
analysis
by
Cold
Vapor
Atomic
Fluorescence
Spectroscopy
(
CVAFS)
and
is
a
modified
version
of
EPA
Method
1631E.
The
second
SOP
is
entitled
"
Digestion
for
Gas/
Air
Samples
Collected
on
Iodated
Carbon
Traps
for
Total
Mercury
Analysis"
and
is
a
peerreviewed
published
method.
A
recent
change
to
this
SOP
consists
of
shipping
the
traps
with
gloves
to
be
worn
by
personnel
that
work
with
the
traps,
to
minimize
contamination.
13
Table
3­
1.
Draft
MQO
for
Sorbent
Total
Mercury
by
internal
Frontier
Geosciences
Method
QA
Measurement
MQO
Value
Laboratory
Replicate
±
25%
RPD
Laboratory
SRM
Spike
Recovery
100
±
20%

Lab
Analysis
Matrix
Spike
Recovery
100
±
20%

Field
Sample
Spike
Recovery
100
±
30%

Field
Sample
Duplicate
±
30%
RPD
Method
Detection
Limit
0.05
µ
g/
m3
@
30
liters
0.002
µ
g/
m3
@
1500
liters
Advantages
of
the
carbon
trap
method
include
ease
of
operation
and
a
simple
technology.
The
method
can
be
cost
effective,
depending
on
verification
requirements.

13
Personal
communication
from
Sharon
Sjostrom,
ADAES,
to
Michiel
Doorn,
Carl
Singer,
and
Brent
Hall,
ARCADIS,
July
7,
2003
13
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
The
main
disadvantage
is
that
the
method
provides
a
delayed,
"
blanket"
cumulative
average
without
the
possibility
to
identify
peaks
or
periods
with
minimal
emissions.
Because
continuous
information
may
prove
necessary
for
timely
response
to
control
device
operations,
this
could
be
a
very
important
disadvantage.
The
method
is
neither
real­
time,
nor
continuous
and,
furthermore,
is
still
in
the
development
stages.
In
addition,
only
one
manufacturer
in
the
U.
S.
can
currently
offer
a
pilot­
test­
ready
device
or
the
treated
traps.

3.1.2
Scope
and
Applicability
Once
the
carbon
trap
method
has
reached
commercial
status,
it
should
be
applicable
to
capturing
total
mercury
with
100
percent
efficiency
from
coal­
fired
stationary
sources.
The
method
takes
a
continuous,
cumulative
sample
over
a
given
time
period
that
can
range
from
one
hour
to
one
week
and
possibly
up
to
a
month.
After
this
time
period
the
trap
is
exchanged
for
a
fresh
one,
and
the
used
trap
is
sent
to
an
off­
site
laboratory
to
be
analyzed
according
to
standard
methods.
14
For
use
in
wet,
saturated
environments,
the
manufacturer
has
designed
a
modified
tip
for
the
trap
to
minimize
wetting
by
large
droplets,
as
well
as
a
heated
carbon
trap
to
avoid
operational
difficulties
due
to
the
potential
high
pressure
drop
across
the
trap.
This
modified
design
has
not
yet
been
tested.
15
3.1.3
Technique
The
carbon
trap
technique
uses
6
mm
or
10
mm
diameter
cylindrical
traps
containing
activated
carbon
impregnated
with
potassium
iodide
to
adsorb
mercury
from
flue
gas.
The
sampling
system
consists
of
a
sample
probe,
Teflon
®
tubing,
a
glass
wool
plug
to
collect
PM,
a
two­
part
sorbent
trap,
a
sample
pump,
and
a
gas
meter
mounted
in
a
console.
The
front
section
of
the
sorbent
trap
is
used
to
chemically
adsorb
mercury.
The
back
section,
separated
from
the
front
by
glass
wool,
collects
any
mercury
that
breaks
through
the
front
section.
The
flue
gas
is
extracted
through
the
carbon­
packed
trap
where
the
Hg
is
adsorbed.
The
sample
pump
mounted
in
the
console,
along
with
a
14
Personal
communication
from
Sharon
Sjostrom,
ADAES,
to
Michiel
Doorn,
ARCADIS,
by
email
June
24,
2003.

15
Personal
communication
from
Sharon
Sjostrom,
ADAES,
to
Michiel
Doorn,
ARCADIS,
by
email
June
24,
2003.
14
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
flow
meter
and
temperature
gauges,
creates
the
vacuum
on
the
carbon
trap.
The
flow
meter
can
be
adjusted
to
achieve
any
desired
flow
rate
through
the
sampling
system
according
to
the
time
period
of
sampling
and
the
flow
rate
of
the
stack
gas.
Typically,
the
carbon
trap
and
the
probe
that
secures
it
are
the
only
things
extending
into
the
flue
gas
path.
Once
the
sample
has
passed
the
probe
and
carbon
trap,
it
travels
through
a
heated
Teflon
line
to
prevent
condensation
of
any
moisture
that
may
build
up
and
cause
operational
difficulties.
The
sample
then
travels
through
a
water
knockout
and
drying
column
where
the
moisture
is
removed
before
passing
through
the
sample
pump
and
gas
meter.

Experiments
by
the
manufacturer
have
shown
that
the
carbon
captures
both
elemental
and
oxidized
mercury
with
100
percent
efficiency
over
at
least
one
week
for
flue
gas
concentrations
typical
of
coal­
fired
power
plants.
16
The
sampling
system
can
be
modified
to
sample
isokinetically
for
particulate
mercury
as
well
as
gas
phase
mercury.
In
the
current
application,
the
sampling
system
has
been
designed
to
exclude
particulate
Hg
collection.
However,
some
incidental
PM
may
collect
on
the
glass
wool
plug
prior
to
the
carbon
bed.
A
separate
analysis
has
indicated
that
particulate
Hg
is
less
than
1
percent
of
total
Hg
in
gas
streams
after
a
particulate
control
device,
which
is
similar
to
results
reported
by
the
OH
Method.
17
The
system
currently
operates
by
drawing
gas
continuously
through
the
trap
at
a
constant
rate.
The
samples
are
typically
collected
for
a
week,
at
which
time
the
carbon
trap
is
removed,
sealed,
and
shipped
to
an
analytical
laboratory,
where
the
mercury
is
leached
from
the
carbon
using
a
strong
acid
and
subsequently
analyzed
with
Cold
Vapor
Atomic
Absorption
Spectroscopy
(
CVAAS).

Research
is
underway
to
develop
a
proportional
sampling
approach.
The
new
consoles
being
used
incorporate
a
barometric
pressure
sensor,
a
digital
manometer,
and
a
flow
control
valve.
The
sampling
consoles
are
designed
to
monitor
the
duct
flow
using
a
standard
pitot
connected
to
the
digital
manometer.
Upon
installation,
the
operator
can
configure
the
sampling
console
to
use
either
the
duct
flow
measured
by
the
pitot,
or
an
external
0­
5
volt
signal
(
such
as
stack
flow
from
the
CEMS,
or
boiler
load)
to
control
the
sample
rate
set
point.
The
micro
controller
in
the
sampling
console
automatically
16
Personal
communication
from
Eric
Prestbo,
Frontier
Geosciences,
Inc.
to
Michiel
Doorn,
ARCADIS,
email
June
25,
2003
17
Ibid.
15
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
adjusts
the
sample
flow
rate
based
upon
the
set
point
signal.
The
association
of
the
control
signal
(
i.
e.
digital
manometer,
stack
flow,
load,
etc.)
to
the
sample
flow
set
point
can
be
defined
by
the
operator.
18
3.1.4
Accuracy
and
Precision
The
carbon
trap
is
analyzed
in
an
off­
site
laboratory.
Currently,
the
only
laboratory
that
analyzes
these
traps
is
Frontier
Geosciences.
The
quantity
of
mercury
is
measured
using
a
method
with
a
detection
limit
of
0.5
 g/
L
(
5
ppt)
in
the
leachate.
Frontier
Geosciences
estimates
that
this
detection
limit
is
equivalent
to
about
1
 g/
m3
for
a
oneweek
sampling
period.
19
In
addition,
Frontier
Geosciences
maintains
that
they
can
easily
match
the
OH
Method
for
total
mercury.
20
There
is
data
available
from
the
MESA
method
indicating
very
good
accuracy
and
precision.
This
method
employed
a
solid
trap
upstream
of
the
activated
carbon
to
capture
oxidized
mercury.
Frontier
Geosciences
suggests
that
the
new
Iodated
Carbon
Traps
should
perform
similarly
to
the
carbon
portion
of
the
MESA
traps
and
that
data
should
be
comparable.
Recent
tests
with
Iodated
Carbon
Traps
suggest
very
good
precision
in
the
laboratory
analysis.
The
manufacturer
is
using
components
from
standard
sampling
equipment
(
i.
e.
calibrated
dry
gas
meters)
and
expects
the
total
flow
recorded
during
each
sampling
run
to
be
within
acceptable
accuracy.
In
addition,
the
manufacturer
is
currently
conducting
tests
to
assess
the
precision
and
accuracy
of
the
current
equipment
and
carbon
traps
by
running
duplicate
simultaneous
samples
in
conjunction
with
OH
testing.

3.2
General
Assumptions
For
the
purposes
of
cost
estimation,
it
is
assumed
that
the
method
will
be
developed
in
adequate
detail
for
EPA
to
include
carbon
traps
as
an
acceptable
method
of
18
Personal
communication
from
Sharon
Sjostrom,
ADAES,
to
Michiel
Doorn,
ARCADIS,
by
email
June
24,
2003.

19
Personal
communication
from
Eric
Prestbo,
Frontier
Geosciences,
Inc.
to
Michiel
Doorn
and
Carl
Singer,
ARCADIS,
May
30,
2003
20
Personal
communication
from
Eric
Prestbo,
Frontier
Geosciences,
Inc.
to
Michiel
Doorn,

ARCADIS,
email,
June
24,
2003
16
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
determining
annual
total
mercury
emissions.
We
further
assume
carbon
traps
will
typically
sample
for
one
week
consistent
with
the
current
design
strategies.
A
range
of
sampling
strategies
may
prove
viable;
each
with
QA/
QC
adequate
to
demonstrate
annual
emissions
accuracy.

Since
the
currently
proposed
carbon
trap
techniques
utilized
fixed
point
sampling,
similar
to
other
CEMS
measurements
covered
in
Part
75,
it
is
assumed
that
a
QA/
QC
program
will
be
required
to
ensure
representativeness
and
accuracy
for
carbon
trap
measurements.
The
reference
method
for
mercury
is
assumed
to
be
the
OH
method.
While
a
RATA
of
carbon
traps
sampling
for
one
week
does
not
appear
practical
due
to
widely
differing
sampling
times
with
OH,
a
RATA
of
carbon
traps
sampling
for
3
hours,
using
an
adapter
and
smaller
traps,
is
assumed
to
be
adequate
for
determining
representativeness
and
accuracy.
For
costing
purposes,
an
annual
9­
sample
RATA
against
OH
is
assumed.

In
addition
to
a
RATA,
it
is
assumed
that
some
QA/
QC
program
will
be
implemented
to
ensure
accurate
emission
estimates
are
made
throughout
the
year.
For
cost
estimating
purposes,
the
QA/
QC
program
will
consist
of
4
quarterly
checks
of
each
sample
gas
metering
system
and
3
quarterly
matrix
spike
checks.
The
fully
developed
carbon
trap
technique
is
assumed
to
provide
proportional
sampling
of
the
stack
gas;
auditing
of
proportionality
is
assumed
to
be
integral
to
the
monitoring
system.
Accuracy
of
the
total
sample
volume
measurements
is
expected
to
be
confirmed
with
quarterly
calibration
of
the
sample
gas
metering
system.
It
is
assumed
that
Hg
concentration
accuracy
will
be
confirmed
with
some
form
of
spiked
carbon
trap
analysis.
For
costing
purposes,
it
is
assumed
that
mercury
spiked
carbon
traps
will
be
available
and
will
sample
flue
gas
in
parallel
with
an
unspiked
carbon
trap
over
the
typical
one­
week
sampling
period.
It
is
further
assumed
that
an
unexposed
carbon
trap
will
be
analyzed
as
a
blank
on
a
quarterly
basis.

It
is
assumed
that
carbon
trap
sampling
will
be
performed
in
duplicate.
If
the
QA/
QC
validating
carbon
trap
based
emission
measurements
is
performed
on
a
quarterly
basis,
failure
to
meet
QA/
QC
objectives
could
invalidate
a
large
portion
of
the
annual
emissions
measurement.
Replication
is
one
method
of
ensuring
against
the
need
to
use
currently
undefined
data
substitution.
Duplication
is
therefore
expected
to
represent
the
minimum
application
of
replication
adequate
for
QA/
QC
requirements
and
reasonable
data
invalidation
protection.
Assumed
annual
requirements
for
carbon
trap
emission
measurements
are
summarized
in
Table
3.2
17
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
Table
3.2:
Assumed
Carbon
Trap
Analytical
Summary
Carbon
Trap
Requirements
Other
QA/
QC
Requirements
RATA
9
3­
h
traps
+
1
weekly
tube
9
OH
runs
+
2
gas
meter
cals
12
weeks
24
weekly
traps
QA/
QC
check
1
3
weekly
traps:
matrix,
matrix
spike,
blank
2
gas
meter
cals
12
weeks
24
weekly
traps
QA/
QC
check
2
3
weekly
traps:
matrix,
matrix
spike,
blank
2
gas
meter
cals
12
weeks
24
weekly
traps
QA/
QC
check
3
3
weekly
traps:
matrix,
matrix
spike,
blank
2
gas
meter
cals
12
weeks
24
weekly
traps
Annual
Total
9
3­
h
traps
+
106
weekly
traps
9
OH
runs
+
8
gas
meter
cals
Travel
expenses
associated
with
installation,
training,
and
maintenance
by
the
manufacturer
are
not
included.
Travel
expenses
associated
with
the
RATA
are
included
in
the
cost
of
subcontracting.

3.3
Capital
Costs
Capital
costs
for
a
carbon
trap
system
(
sampling
console,
probe,
water
knockout
assembly)
is
approximately
$
8,500.21
Installation
costs
(
for
equipment
location,
setup,
electrical
hook­
up
and
probe
insertion)
are
estimated
to
require
two
hours
of
technical
labor
for
a
total
labor
cost
of
$
106
per
carbon
trap
system.
As
two
carbon
trap
systems
are
assumed,
total
capital
costs
are
estimated
at
$
17,212.

21
Personal
communication
from
Sharon
Sjostrom,
ADAES,
to
Michiel
Doorn,
ARCADIS,
by
email
June
24,
2003.
18
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
3.4
Annual
Costs
3.4.1
Labor
Costs
This
section
addresses
annual
labor
costs
for
the
plant.
The
following
technical
hours
are
estimated
to
be
associated
with
this
method
on
an
annual
basis:


For
operations,
4
hours
per
week
per
carbon
trap
system,
to
remove
the
used
trap,
install
a
new
trap,
mail
the
used
trap
to
the
manufacturer
(
or
other
future
laboratories/
providers),
and
complete
associated
paperwork.


For
routine
checks,
3
hours
per
week
per
carbon
trap
system
(
8
minutes
per
8­
hour
shift).


For
carbon
trap
flow
calibration,
2
hours
per
quarter
per
carbon
trap
system
Accordingly,
it
is
estimated
there
are
744
technical
hours
associated
with
operation
of
two
carbon
trap
systems.

It
is
assumed
that
awarding
a
contract
for
carbon
trap
analysis
will
require
eight
technical
hours.
It
will
also
require
eight
hours
to
compile
an
emissions
report.
As
this
will
occur
four
times
annually,
it
will
result
in
an
additional
32
technical
hours.
Finally,
it
will
require
one
technical
hour
to
submit
the
report,
but
again,
this
will
occur
four
times
annually,
resulting
in
another
four
hours.
This
results
in
44
technical
hours
for
reporting,
or
$
2,334
annually,
including
management
and
clerical
labor.
Routine
operations
are
therefore
estimated
to
require
788
technical
hours
annually
for
a
cost
of
$
41,812
including
management
and
clerical
costs.

Plant
labor
associated
with
RATA
testing
is
estimated
to
require
121
technical
hours.
It
will
require
40
hours
to
obtain
a
contract
for
RATA
testing,
72
hours
for
oversight
of
nine
tests,
8
annual
hours
to
compile
and
review
analytical
data,
and
1
hour
annually
to
submit
test
results.
Plant
labor
costs
for
verification
testing
are
estimated
at
$
6,420
Thus,
combined
labor
costs
for
routine
operation
and
verification
are
estimated
at
$
48,232
19
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
3.4.2
Subcontract
and
ODC
Basic
analytical
costs
performed
by
the
manufacturer
to
analyze
and
replace
one
carbon
trap
are
$
200.22
These
costs
include
packing
the
traps
with
gloves
for
sample
handling
(
installation
and
removal),
data
collection
and
chain
of
custody
forms,
and
post­
test
analysis.
These
costs
do
not
include
any
additional
analyses,
and
the
special
tip
for
wet­
stack
applications.
For
cost
estimation
purposes,
all
traps;
3­
h,
weekly,
or
spiked,
are
assumed
to
cost
the
same
$
200
for
complete
turnaround.
In
addition
to
the
above,
$
20
per
week
is
assumed
for
shipping
and
handling
(
two
ways).
Based
on
the
assumed
115
annual
trap
requirement,
analytical
costs
are
estimated
at
$
24,040
annually.

A
nine­
test
OH
sampling
event
is
estimated
to
cost
$
37,000
to
subcontract.

Estimated
total
subcontracting
costs
and
ODC
are
$
61,040.

3.4.3
Annualized
Capital
costs
Based
on
a
three­
year
straight­
line
depreciation,
annualized
capital
costs
are
estimated
at
$
5737.

3.4.4
Total
Annualized
Costs
Total
annualized
costs
for
mercury
emissions
determination
with
the
carbon
trap
method
are
estimated
at
$
115,009.

This
may
be
compared
to
the
preliminary
cost
estimate
submitted
for
the
prototype
by
Walter
S.
Smith
and
Associates.
No
tests
have
been
performed,
but
total
annual
costs,
including
a
lease,
are
estimated
to
be
less
than
$
50,000.
Technical
specifics
are
confidential
at
this
point.
23
Assuming
this
technique
would
be
performed
in
duplicate
and
would
require
a
RATA
as
well,
total
annual
costs
for
the
Walter
S.
Smith
and
Associates
monitoring
technique
would
be
less
than
$
137,000.

22
Personal
communication
from
Sharon
Sjostrom,
ADAES,
to
Michiel
Doorn,
ARCADIS,
by
email,
June
24,
2003.

23
Personal
communication
from
Walter
Smith,
Walter
Smith
Associates,
Inc.
to
Michiel
Doorn,
ARCADIS,
by
email,
June
26
and
email
June
24,
2003.
20
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
3.5
Uncertainties
Application
of
carbon
traps
for
mercury
emissions
measurements
will
be
dependent
on
continued
development
and
acceptance
of
the
technique
as
a
method.
Methods
are
often
validated
against
various
spiking
techniques
or
against
a
standard
method.
EPA
acceptance
may
require
such
validation
on
multiple
sources,
depending
on
how
EPA
views
source
categories.

Implementation
of
carbon
traps
for
mercury
emissions
will
likely
require
the
development
of
QA/
QC
procedures
and
limits
as
part
of
published
procedures.
The
procedures
to
be
defined
include:


Leak­
check
documentation
and
limits

Sample
metering
calibration

Proportional
sampling
validation

Chain
of
Custody

Back­
half
capture
limits

Laboratory
calibration,
continuing
calibration
verification,
independent
calibration
verification

Laboratory
duplicates

Blanks

Carbon
tube
analytical
hold
times.

In
addition,
methods
to
audit
both
sample
collection
and
analysis
are
likely
to
be
crucial
to
acceptance
of
the
technique.
One
route
to
auditing
both
collection
and
analysis
is
the
development
of
a
standard
technique
to
spike
mercury
onto
a
carbon
trap.
Techniques
for
spiking
the
carbon
traps,
either
in
the
field
or
in
the
laboratory,
have
been
implemented
by
method
developers.

The
effectiveness
of
the
carbon
traps
depends
on
various
parameters,
such
as
sorbent
properties,
coal
type,
flue
gas
conditions
and
composition,
and
reaction
kinetics.
These
conditions
will
vary
with
site
configuration
and
coal
composition.
Although
the
21
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
method
does
not
distinguish
between
the
elemental
and
oxidized
forms
of
mercury,
the
efficiency
of
adsorption
for
both
oxidized
and
elemental
mercury
must
be
validated
over
a
variety
of
stack
conditions,
as
well
as
fly
ashes.
One
method
of
validation
would
be
analysis
of
duplicates
samples,
one
collected
on
a
clean
carbon
trap
and
one
collected
on
a
mercury
spiked
carbon
trap.
22
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
4.
CEMS
Analysis
4.1
Background
Hg
CEMS
are
undergoing
continuing
development,
with
current
technology
being
tested
under
various
programs.
Few,
if
any,
applications
at
coal­
fired
power
plants
currently
exist
in
the
United
States,
but
the
technology
is
developing
rapidly.
Many
CEMS
have
been
undergoing
field
tests,
and
some
have
participated
in
several.
Most
manufacturers
are
in
the
process
of
making
significant
improvements
or
refinements
to
their
monitors
as
time
proceeds.
Consequently,
the
CEMS
information
in
this
report,
and
similar
reports,
may
need
to
be
updated
in
the
relatively
short
term.

In
Germany,
automatic
mercury
sampling
devices
have
been
applied
since
the
early
1990s.
The
Federal
Republic
of
Germany
issued
regulations
for
both
emission
limits
and
continuous
monitoring
of
total
mercury
in
incinerator
flue
gases.
Under
the
regulation,
CEMS
need
certification
for
approval.
The
German
emission
limit
of
0.05
µ
g/
m3
was
adopted
for
the
European
Union
in
1994.24
CEMS
can
be
classified
according
to
their
method
of
converting
oxidized
Hg
to
elemental
Hg
as
dry
or
wet.
Wet
systems
convert
Hg
in
impinger
solutions,
while
dry
systems
thermally
or
catalytically
convert
the
Hg
species.
Most
CEMS
can
provide
speciated
Hg
emissions
values,
but
the
development
trend
in
industry
is
toward
total
Hg
measurement.

Based
on
operational
experience
and
industry
knowledge,
the
dry
systems
appear
to
offer
more
operational
availability
and
require
less
maintenance
compared
to
wet
systems.
They
provide
quick
real­
time
Hg
concentrations
and
many
have
inherent
auto
zeroing
and
system
checks
built
into
them.
The
dry
systems
are
easier
to
operate
overall
and
offer
a
more
reliable
Hg
number
because
of
not
having
any
chemical
interactions.
The
disadvantages
of
wet
systems
include
increased
maintenance
and
potential
problems
with
the
gas
conditioning
systems
malfunctioning.
Also,
the
solutions
used
in
wet
systems
may
react
with
flue
gas
components
in
unexpected
ways.
Another
disadvantage
of
current
wet
systems
is
that
they
only
yield
a
mercury
24
Sigl,
W.,
2000.
Continuous
monitoring
of
mercury
emissions
in
combustion
flue
gases
with
the
Hg­
CEMS
'
MERCEM'.
In
"
Recent
advances
in
the
science
and
management
of
air
toxics",
Proceedings
of
the
joint
international
specialty
conference,
exhibition
and
workshop
of
the
Air
and
Waste
Management
Association,
9
­
12
April
2000,
Banff,
Canada.
23
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
concentration
every
4­
6
minutes,
depending
on
the
cycle
set
up;
they
are
not
truly
"
continuous"
Hg
monitors.
For
the
purpose
of
further
discussion,
only
the
dry
systems
are
addressed.

It
is
apparent
through
the
results
from
each
subsequent
field
evaluation
that
the
reliability
and
accuracy
of
Hg
CEMS
has
greatly
improved.
Hg
CEMS
are
viewed
as
a
viable
choice
for
Hg
monitoring
because
of
the
continuous
data
delivery,
as
well
as
their
improving
accuracy.
It
is
anticipated
that
Hg
CEMS
would
be
a
preferred
method
of
Hg
monitoring,
provided
the
costs
are
comparable
to
other
viable
methods.

For
this
report
and
other
recent
associated
draft
reports,
technical,
performance,
and
cost
information
was
requested
from
12
vendors.
Eleven
vendors
provided
technical
and
performance
information,
while
seven
manufacturers
responded
with
cost
information.
Table
4­
1
summarizes
technical
information
of
12
different
Hg
CEMS.
Table
4­
2
includes
performance
information
related
to
operation
and
maintenance.
All
data
in
Tables
4­
1
and
4­
2
are
from
manufacturer
responses.

4.1.1
Scope
and
Applicability
There
is
currently
no
regulatory
driver
or
performance
standard
for
CEMS;
although
a
draft
performance
standard
has
been
developed25
against
which
Hg
CEMS
can
be
evaluated.
However,
quality
requirements
can
not
be
fully
defined
at
this
point.
A
regulation
for
Hg
could,
conceivably,
require
monitoring
of
elemental,
oxidized,
or
total
Hg,
resulting
in
different
potential
equipment
needs.

Hg
CEMS
could
find
wide
use
for
monitoring
flue
gas
emissions
at
coal­
fired
power
plants.
As
indicated
earlier,
there
is
widespread
use
of
Hg
CEMS
in
Europe,
but
mostly
at
municipal,
or
medical
waste
incinerators.

25
Draft
Performance
Specification
12
 
Specifications
and
test
procedures
for
total
mercury
continuous
monitoring
systems
in
stationary
sources.
http://
www.
epa.
gov/
ttnemc01/
propperf/
ps­

12.
pdf
24
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
Table
4­
1.
Overview
of
Mercury
CEMs
that
are
(
expected
to
be)
Commercially
Available
in
the
United
States
Model
Principles
of
operation
Speciation
Interferences
Zero
Drift
Calibration
Drift
Relative
Accuracy
Field
Application
US
Detection
Limit
Sampling/

Response
time
Cooper
X­
CMM
X­
ray
fluorescence
Total
none
No
zero
adjustment
is
necessary
Calibration
every
few
months
na
na
0.1
µ
g/
m3
10­
20
min
Durag
HM­
1400
TR
Dual­
Beam
CVAAS
Total
and
Elemental
None
below
1500
ppm
SO2
<
±
1.5%
F.
S.

per
week
<
±
2
%
F.
S.

per
week
±
5%
F.
S.

(
total
accuracy)
none
<
1
mg/
m3
180
s
Ecochem
HG
­
MKII
CVAAS
with
gold
trap
Total
and
Elemental
none
Uses
preset
intervals
with
alert
system
No
information
na
none
na
na
Envimetrics
Argus­
Hg
Plasma
source
spectrometer
Elemental,

Oxidized
na
No
zero
adjustment
is
necessary
Automatic
calibration
with
built­
in
Hg
standard
na
na
0.03
µ
g/
m3
na
Genesis
Quicksilver
Sky
Monitor
Dry
method
Particulate,

Oxidized,
Elemental
na
na
No
errors,
proprietary
technology
na
na
na
na
Nippon26
AM­
2/
AM­
3
CVAAS
with
gold
trap
Elemental
Cl2
and
to
a
lesser
extent
HCl
Hg0
0.96
µ
g/
m3
Hg0
2.93
µ
g
/
m3
Hg0
18%

Hg0
16%
Hg0
14%

Hg0
23%
none
0.1
µ
g/
m3
5­
13
min
26
Nippon
Instruments
Corporation
Model
AM­
2
Elemental
Mercury
Continuous
Emission
Monitor;
Environmental
Technology
Verification
Report.
U.
S.
Environmental
Protection
Agency
in
cooperation
with
Battelle,
Inc.
August
2001.
25
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
Model
Principles
of
operation
Speciation
Interferences
Zero
Drift
Calibration
Drift
Relative
Accuracy
Field
Application
US
Detection
Limit
Sampling/

Response
time
Nippon27
MS1/
DM5
CVAAS
and
wet
scrubbing
plus
chemical
for
total
Hg
Elemental
and
Oxidized
HCl
`
0.1
µ
g/
m3
`
0.1
ug/
m3
`
0.1
µ
g/
m3
`
0.1
µ
g/
m3
Hg0
14%

Hg+
2
13%

Hg0
12%

Hg+
2
11%
HgT
13%

Hg0
11%

Hg+
2
55%

HgT
39%

Hg0
49%

Hg+
2
50%
none
0.1
µ
g/
m3
Ionic
Hg
60
s.

Detector
35­

50
s.

Ohio
Lumex28
CVAAS,
with
Zeeman
highfrequency
polarization
background
correction
Total
and
Elemental
HCl
and
Cl2
0.20
HgT
0.37
Hg0
2.45
HgT
0.17
Hg0
62%
HgT
92%
Hg0
24%
HgT
18%
Hg0
HgT
58%

Hg0
50%

Hg+
2
99%

HgT
71%

Hg0
107%

Hg+
2
69%
none
na
20­
48
s
Opsis
HG200
Dual
beam
absorption
photometer
Total
and
Elemental
SO2
may
be
issue
na
na
na
Two
test
sites
0.5
pg
8
min
27
Nippon
Instruments
Corporation
Model
MS­
1/
DM­
5
Elemental
Mercury
Continuous
Emission
Monitor;
Environmental
Technology
Verification
Report.
U.
S.
Environmental
Protection
Agency
in
cooperation
with
Battelle,
Inc.
August
2001.

28
Lumex
Ltd.
Mercury
Continuous
Emission
Monitor;
Environmental
Technology
Verification
Report.
U.
S.
Environmental
Protection
Agency
in
cooperation
with
Battelle,
Inc.
August
2001.

Table
4­
1
continued.
26
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
Model
Principles
of
operation
Speciation
Interferences
Zero
Drift
Calibration
Drift
Relative
Accuracy
Field
Application
US
Detection
Limit
Sampling/

Response
time
PS
Analytical29
Sir
Galahad
II
Atomic
fluorescence
with
gold
trap
Total,

Oxidized,
and
Elemental
Elemental:
Cl2
0.003
µ
g/
m3
0.013
µ
g/
m3
12%
11%
HgT
21%

Hg0
23%

Hg+
2
27%

HgT
33%

Hg0
30%

Hg+
2
33%
18
units,

mainly
for
research
0.1
pg
6
min
ST2
SM­
3
CVAAS
Total
SO2
<
1
µ
g/
m3
after
1
week
of
testing
See
TüV
report
See
page
38
of
TüV
report
none
na
na
Tekran
Atomic
fluorescence
with
gold
trap
Total
and
Elemental
None
On­
line
automated
zero
and
multi­
level
span
checks
Automatic
recalibration
using
internal
permeation
source
na
none
<
0.05
µ
g/
m3
2.5
min
na
=
not
applicable
29
PS
Analytical
Ltd.
Sir
Galahad
II
Mercury
Continuous
Emission
Monitor;
Environmental
Technology
Verification
Report.
U.
S.

Environmental
Protection
Agency
in
cooperation
with
Battelle,
Inc.
August
2001.

Table
4­
1
concluded.
27
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
Table
4­
2.
Operation
&
Maintenance
Model
Principles
of
operation
Operation
continued
Conditioning
System
Filter
Trap
Other
Cooper
X­
CMM
X­
ray
fluorescence
Works
with
filter
tape
No
pre­
treatment
necessary
Automatic
No
No
calibration
gases
needed
Durag
HM­
1400
TR
Dual­
Beam
CVAAS
nav
Heated
lines;

Thermo­
catalytic
converter
Particle
filter
Gold
trap;
Cleaned
1x/
day
by
heat;

Aromatics,
SO2,
NO2
are
interferences
NH3
can
cause
flow
problems
Ecochem
HG
­
MKII
CVAAS
with
gold
trap
nav
Teflon
or
heated
parts;
Dry
catalytic
device
nav
Gold
trap;
Desorption
into
pure
N2
nav
Envimetrics
Argus­
Hg
Plasma
source
spectrometer
No
chemical
reagents
Solid
catalytic
converter;
Peltier
cooling
nav
­­
Automated
calibration;

Maintenance
only
every
6
months
Genesis
Quicksilver
Sky
Monitor
Dry
method
SOX
removal
by
scrubbing
­­
nav
­­
Calibration
1x
per
year;
Trade
secret
for
background
correction
Nippon
MS1/
DM5
CVAAS
and
wet
scrubbing
plus
chemical
for
total
Hg
nav
Distilled
water
scrubber;

Dehumidifier;
Two
coolers
Dust
filter,
membrane
filter
Gold
trap/

Heater;
Needs
purified
air
nav
OhioLumex
CVAAS,
with
Zeeman
highfrequency
polarization
background
correction
Two­
channel
instrument
"
Pyrolyzer"

converter
Particle
filter
and
zero
Hg
absorption
filter
No
gold
trap
Charge
battery;

Manual
available
on­
line
Opsis
HG200
Dual
beam
absorption
photometer
nav
Dilution
system;

Dry
thermocatalytic
converter;

Cooler
nav
Gold
trap
Two
calibration
gases
28
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
Model
Principles
of
operation
Operation
continued
Conditioning
System
Filter
Trap
Other
PS
Analytical
Sir
Galahad
II
Atomic
fluorescence
with
gold
trap
Converts
Hg
into
Hg0
with
aqueous
reagent
Scrubber;
Tubing
and
agents
replaced
every
2
weeks
Filter
Gold
trap,
Preconcentration
trap;

Needs
argon
gas
Heated
diaphragm
pump;

Built
in
calibration
ST2
SM­
3
CVAAS
nav
nav
nav
No
carrier
gas
Maintenance
free
pump
Tekran
Atomic
fluorescence
with
gold
trap
Inert
probe,
dry
converter
nav
nav
Gold
trap
Maintenance
1­
4
hours
per
4
weeks;
Built
in
automatic
calibration
nav
=
not
available
or
not
provided
by
the
vendor
Table
4­
2
concluded.
29
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
4.1.2
Technique
EPA
defines
the
CEMS
as
"
The
total
equipment
required
for
the
determination
of
a
pollutant
concentration.
30"
Total
equipment
includes
the
major
sub­
systems
such
as
the
sampling
interface,
pollutant
analyzer,
diluent
analyzer,
and
data
recorder.

Mercury
CEMS
typically
measure
only
Hg0
using
some
variation
of
CVAAS
or
CVAFS.
Many
incorporate
a
process
that
converts
gaseous
ionic
Hg
to
gaseous
Hg0,
resulting
in
measurements
of
total
gaseous
Hg.
PM
is
typically
filtered
from
the
sample
before
analysis,
which
removes
particle­
bound
Hg.
This
process
therefore
results
in
underestimates
of
total
Hg;
it
also
may
cause
bias
in
speciation
and
bias
in
total
gaseous
Hg.
Some
fly
ashes
react
with
Hg,
resulting
in
the
oxidation
of
Hg0
in
the
probe
and
on
filters,
which
in
turn
overestimates
ionic
Hg.
Fly
ash
may
also
adsorb
or
desorb
Hg,
which
biases
estimates
of
total
gaseous
Hg
because
the
ash
is
not
subsequently
analyzed
for
Hg.

The
measurement
techniques
used
by
CEMS
are
often
affected
by
common
coal
combustion
contaminants.
The
draft
performance
specification
lists
eight
potentially
interfering
components:
carbon
monoxide,
carbon
dioxide,
oxygen,
sulfur
dioxide,
nitrogen
dioxide,
water
vapor,
hydrogen
chloride,
and
chlorine.
Such
interferences
are
often
dealt
with
in
a
manner
specific
to
each
CEMS.
One
common
mechanism
is
to
amalgamate
the
Hg
on
a
gold
trap
and
regenerate
the
mercury
captured
on
the
gold
trap
into
a
carrier
gas
that
is
then
analyzed.

4.1.3
Accuracy
and
Precision
Evaluation
of
Accuracy
and
Precision
of
Hg
CEMS
is
confounded
by
the
rapid
and
continuing
development
of
Hg
CEMs
by
vendors
and
the
lack
of
published
information
on
CEM
performance.
Evaluation
of
the
performance
of
CEMs
will
improve
when
suitable
reference
materials
or
procedures
become
available.
A
National
Institute
of
Standards
&
Technology
(
NIST)
approved
elemental
mercury
gas
standard
is
still
under
development.
EPA
has
developed
a
draft
performance
specification,
PS­
12,
which
defines
relative
accuracy
requirements
for
Hg
CEMS
measurement
along
with
30
Draft
Performance
Specification
12
 
Specifications
and
test
procedures
for
total
mercury
continuous
monitoring
systems
in
stationary
sources.
Website:
http://
www.
epa.
gov/
ttnemc01/
propperf/
ps­
12.
pdf.
30
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
certain
other
QA
requirements;
a
relative
accuracy
of
20%
is
currently
specified
in
PS­
12.

EPA's
Office
of
Research
and
Development
conducted
Relative
Accuracy
tests
on
four
Hg
CEMs
under
the
Environmental
Test
Verification
(
ETV)
program
in
January
2001.
The
ETV
program
was
created
to
facilitate
the
use
of
innovative
environmental
technologies
[
including
continuous
emission
monitors
(
CEM)]
by
testing
their
performance
and
making
the
results
available
to
those
involved
in
their
design,
distribution,
permitting,
purchase,
and
use.
These
ETV
tests
were
performed
on
a
pilotscale
natural
gas­
fired
combustor.
31
As
shown
in
Table
4­
1,
a
vintage
2001
Nippon
CEMs
was
able
to
achieve
13%
relative
accuracy
during
the
ETV
tests.

Six
Hg
CEMs
were
also
recently
evaluated
under
the
ETV
program
at
a
Department
of
Energy
(
DOE)
Toxic
Substances
Control
Act
(
TSCA)
incinerator
in
Oak
Ridge,
Tennessee;
this
source
category
was
not
pursued
as
a
source
of
CEM
data
due
to
expected
differences
in
the
flue
gas
composition.
Results
from
the
TSCA
incinerator
were
presented
at
the
Air
and
Waste
Management
Association
(
AWMA)
conference
in
San
Diego.
32
EPA's
Office
of
Air
Quality
Planning
and
Standards
(
OAQPS)
is
undertaking
a
series
of
field
tests
of
several
mercury
CEMs
at
coal­
fired
power
plants.
The
first
series
of
Relative
Accuracy
tests
at
a
full­
scale
coal­
fired
power
plant
were
conducted
in
October
2002
but
are
not
yet
published.
OAQPS
is
currently
conducting
two
Hg
CEM
demonstrations/
evaluations
at
full­
scale
coal­
fired
power
plants.
The
results
of
these
tests
are
expected
to
be
available
in
October
2003.
EPRI
recently
published
CEMs
measurements
of
Hg
emissions
at
several
coal­
fired
power
plants;
QA
information
regarding
the
performance
of
the
CEMs
was
not
found
in
the
paper.
33
31
Environmental
Technology
Verification
Program.
Website:
http://
www.
epa.
gov/
etv/
verifications/
vcenter1­
11.
html
32
Dunn,
J.
E.,
et
al.
2003.
Evaluation
of
Mercury
Continuous
Emission
Monitors
at
the
U.
S.

DOE
TSCA
Incinerator.
AWMA,
96th
Annual
Conference,
June
22­
26,
2003,
San
Diego,
CA.

33
Chu,
Paul,
et
al.,
Characterization
of
"
Longer
Term"
Mercury
Emissions
from
Coal­
Fired
Power
Plants,
2003
MEGA
Symposium,
Washington,
DC.
31
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
4.2
General
Assumptions

A
three
year,
straight­
line
write­
off
for
capital
costs
is
assumed.


Capital
costs
do
not
include
installation
costs.


Technician
training
is
assumed
to
be
included
in
capital
costs.


Costs
for
utilities
(
water,
electricity,
heat,
cooling)
are
not
included.


Calibration
costs
have
not
been
included,
pending
guidance
from
EPA.
None
of
the
manufacturers
submitted
calibration
cost
information.


Because
of
the
trend
in
the
industry,
only
systems
that
can
measure
total
Hg
are
considered.
This
includes
systems
that
can
do
both
total
and
elemental
Hg.


The
trend
in
the
industry
seems
to
be
toward
dry
systems.
The
labor,
sampling
and
analysis
costs
reflect
estimates
for
dry
systems
only,
while
the
capital
cost
analysis
includes
all
systems.

Similar
to
the
performance
verification
of
SO2
monitors,
it
is
assumed
that
the
plant
would
perform
one
RATA
annually
using
the
OH
method,
consisting
of
nine
(
9)
tests,
as
well
as
perform
three
(
3)
cylinder
audits
on
the
Hg
CEMS.
The
cylinder
audits
are
assumed
to
be
performed
by
the
plant.
A
cylinder
audit
would
typically
require
using
a
certified
gaseous
elemental
Hg
standard
and
introducing
this
gas
through
the
probe
or
directly
to
the
detector
and
the
entire
sampling
system.
This
would
facilitate
determining
if
there
were
any
bias
being
imparted
by
the
sampling
systems
as
opposed
to
the
detector.
It
is
assumed
that
the
cylinder
audits
could
be
performed
by
the
plant
personnel
with
the
procurement
of
a
gas
standard.

4.3
Capital
Costs
Capital
costs
for
CEMS
that
are
available
on
the
U.
S.
market,
or
that
are
expected
to
be
available
soon,
are
listed
in
Table
4­
3.
This
information
was
collected
through
an
email
request
and
ensuing
telephone
conversations
with
individual
vendors.
Costs
have
been
standardized
to
the
extent
possible
to
reflect
a
complete
CEMS
and
to
exclude
installation.
Capital
costs
are
especially
dependent
on
the
configuration
of
the
system.
Also,
the
systems
differ
in
both
basic
and
ancillary
supporting
technology.
For
several
vendors,
these
costs
are
preliminary.
32
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
Table
4­
3
includes
15
entries
representing
CEMS
from
11
manufacturers.
When
a
manufacturer
provided
a
large
range
for
capital
cost,
as
was
the
case
with
Genesis
and
OhioLumex,
both
the
low­
end
and
the
high­
end
estimate
were
included
in
the
table
and
in
the
summary
calculations.
In
other
words,
the
submittals
were
regarded
as
separate
instruments.
The
Nippon
MS­
1A
instrument
was
excluded
because
it
can
not
measure
total
mercury.

Table
4­
3.
CEMS
Capital
Cost
(
March
2003)

Vendor
Name
Model
Capital
Cost
Notes
Mercury
System
Type
Dry
Systems:

Cooper
Environmental
Services
X­
CMM
$
100,000
Estimate
for
complete
system.
Filters
would
cost
around
$
5,000
per
year.
Total
Dry,
uses
filter
tape
Durag
HG­
1400
TR
$
45,000
Does
not
include
heated
sample
line.
Heated
probe
of
$
3,000
is
optional.
Speciating
Dry
Ecochem
HG­
MKII
$
59,400
­­
Speciating
Dry
Envimetrics
Argus­
Hg
1000
$
54,200
Estimate.
Includes
sample
probe,
heated
sample
line
and
controller.
Speciating
Dry
Genesis
Laboratory
Systems
Quicksilver
Sky
Monitor
$
35,000
Estimate
for
CEM
($
20,000)
plus
low­
end
gas
conditioning
system.
Speciating
Dry
Genesis
Laboratory
Systems
Quicksilver
Sky
Monitor
$
85,000
Estimate
for
CEM
($
20,000)
plus
high­
end,
specialized
gas
conditioning
system.
Speciating
Dry
OhioLumex
CEM
$
40,000
Low­
end
estimate.
"
CEM,
based
on
AA
with
Zeeman
correction
for
interferences,
dry
converter
system,
total
Hg,
detection
limit­
0.1
µ
g/
dscm.
Installation
is
not
included."
Speciating
Dry
OhioLumex
CEM
$
60,000
High­
end
estimate.
Speciating
Dry
Opsis
HG200
$
80,000
Basic
instrument:
$
30k,
Basic
instrument
(
30k)
with
converter
($
50k)
and
conditioning
system.
Speciating
Dry
ST2
Service
Technologies
SM­
3
$
47,666
Complete
system,
simple
probe.
Total
Dry
Tekran
$
85,000
Preliminary
estimate
Speciating
Dry
33
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
Vendor
Name
Model
Capital
Cost
Notes
Mercury
System
Type
(
Semi­)
wet
systems
Nippon
Instruments
DM­
6
$
38,000
Standard
gas
generator
and
KOH
scrubber
an
optional
Speciating
Semi­
dry
Nippon
Instruments
DM­
6A
$
34,000
­­
Speciating
Semi­
dry
PS
Analytical
Mercury
CEM
$
80,000
Low­
end
estimate.
"
The
PS
Analytical
Mercury
CEM
can
be
configured
in
a
number
of
ways
using
its
stream
selection
capability.
Systems
include
the
unique
calibration
device."
Speciating
Wet
Not
considered:

Nippon
Instruments
MS­
1A
$
13,000
Not
including
probe
unit
Elemental
Semi­
dry
The
average
capital
cost
of
the
speciating
and
total
Hg
CEMS
is
$
60,233,
while
the
median
value
is
$
56,800.
The
lowest
value
is
$
35,000
and
the
highest
value
is
$
100,000.
Additional
information,
such
as
copies
of
correspondence
with
manufacturers
and
copies
of
quotations
may
be
found
in
a
draft
EPA
report.
34
In
the
ensuing
cost
calculations,
the
average
capital
cost
of
$
60,233
is
used.

4.4
Annual
Costs
4.4.1
Annual
Labor
Costs
Based
on
the
ARCADIS
expert
review
and
review
of
the
data
submitted
by
the
vendors,
a
yearly
technical
labor
allocation
for
dry
systems
of
219
hours
is
estimated.
This
number
is
built
up
of
one
10­
minute
inspection
per
shift,
three
shifts
per
day,
and
365
days
per
year,
as
well
as
three
hours
of
maintenance
per
month.
Thirty­
two
technical
hours
are
assumed
for
compiling
and
validating
four
quarterly
emissions
reports.
Four
additional
hours
are
estimated
for
submission
of
these
four
reports.
Routine
operations
of
Hg
CEMs
are
therefore
estimated
to
require
255
technical
hours.

34
EPA,
2003.
Cost
of
Mercury
Measurement
Techniques.
Draft
Report
submitted
on
March
28,
2003,
by
Perrin
Quarles
Associates,
Inc.
34
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
As
mentioned,
three
cylinder
audits
per
year
are
assumed.
Four
technical
hours
are
estimated
for
the
performance
of
a
single
cylinder
audit
and
to
prepare
a
report,
a
total
of
12
hours
per
year.
Additional
labor
costs
associated
with
the
performance
of
three
cylinder
audits
is
therefore
estimated
at
$
637.

Plant
labor
costs
associated
with
relative
accuracy
verification
testing
(
one
events
with
nine
tests)
are
estimated
to
be
121
technical
hours
including:


40
hours
for
OH
contacting

72
hours
for
OH
oversight

8
hours
for
report
review

1
hour
for
report
submission
Including
managerial
and
clerical
support.
the
labor
costs
for
the
verification
component
for
this
program
are
estimated
to
be
$
6,420.

4.4.2
Subcontract
and
ODC
Based
on
the
expert
review
and
review
of
the
data
submitted
by
the
vendors,
ARCADIS
estimates
an
annual
materials
operations
and
maintenance
budget
associated
with
sampling
and
analysis
for
a
generic
dry
CEMS
of
$
2,500.
This
budget
includes
the
cost
for
replacement
parts,
including
the
catalyst,
tubing,
pumps,
and
fittings.
Because
many
CEMS
are
still
in
the
pilot
stage
and
the
technologies
are
often
proprietary,
this
estimate
is
preliminary.

It
is
assumed
that
the
procurement
of
the
gas
standard
associated
with
the
three
cylinder
gas
audits
the
standard
would
cost
$
5,000
annually.
35
One
nine­
test
OH
RATA
is
estimated
to
cost
$
37,000
to
subcontract.
This
number
consists
of
$
4,000
in
travel
and
$
33,000
in
testing
costs.

35
Current
rate
quoted
by
Spectrum
Gases.
35
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
Accordingly,
subcontracting
and
ODC
for
generic
a
dry
CEMS
are
estimated
to
be
$
44,500.

4.4.3
Annualized
Capital
Costs
Based
on
a
three­
year
straight­
line
depreciation,
annualized
capital
costs
are
estimated
to
be
$
20,077
based
on
the
capital
cost
average
for
both
wet
and
dry
systems,
as
submitted
by
the
vendors.

4.4.4
Total
Annualized
Costs
Total
annualized
costs
for
mercury
monitoring
with
a
CEMS
are
estimated
at
$
84,142.

Note
that
this
estimate
may
be
low,
in
regard
to
administrative,
sampling
and
analysis
cost,
for
the
first
year,
because
the
CEM
is
likely
to
undergo
extra
scrutiny
from
management.
Also,
additional
vendor
maintenance
calls
may
take
place
which
will
increase
the
vendor's
budget
as
well
as
require
time
from
plant
personnel.

4.5
Uncertainties

The
technology
is
still
under
development.


Limited
field
experience
has
been
collected.


No
performance
standard
has
been
defined
for
CEMS.


No
calibration
method
has
been
defined.


Demonstration
of
oxidized
Hg
conversion
efficiency
is
difficult
without
OH
testing.
Currently
envisioned
cylinder
audit
would
not
demonstrate
conversion.
36
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
5.
Fuel
Analysis
5.1
Background
and
Methods
Fuel
analysis
as
an
emissions
measurement
technique
implicitly
utilizes
an
emission
factor
approach;
mercury
emissions
are
some
fraction
of
all
the
mercury
inputs.
Mercury
in
the
fuel
is
generally
considered
the
sole
source
of
mercury
emissions
though
other
mercury
inputs
could
easily
be
incorporated
into
this
approach
if
deemed
necessary.
Mercury
content
in
the
fuel
and
the
amount
of
fuel
burned
are
used
to
determine
the
mercury
inputs
to
a
unit.
In
the
extreme
or
maximum
emissions
case,
mercury
emissions
from
a
unit
could
be
estimated
as
the
total
measured
mercury
inputs,
discounting
mercury
retained
in
residuals
such
as
fly
ash
or
gypsum.
An
alternate
approach
is
to
apply
an
emission
factor,
either
based
on
unit
specific
emission
measurements
or
some
classification
average
emission
measurements,
to
the
measured
mercury
inputs.

Measuring
mercury
content
in
the
fuel
and
fuel
inputs
to
units
appears
to
be
an
integral
part
of
proposed
mercury
control
legislation.
Mercury
inputs
provide
the
baseline
which
emissions
are
compared
to
in
control­
based
legislation.
Similarly,
the
proposed
Clear
Skies
Act
of
2003
reverts,
should
the
administrator
not
establish
the
emissions
trading
system,
to
a
control­
based
mercury
emissions
requirement
based
on
fuel
mercury
content36.

5.1.1
Scope
and
Applicability
Fuel
analysis
will
determine
the
total
amount
of
mercury
in
the
fuel.
Mercury
in
the
fuel
can
be
expressed
on
either
a
mass
basis,
either
wet
or
dry,
or
can
be
normalized
to
the
heating
value
of
the
fuel
and
be
expressed
on
an
energy
basis.
When
the
fuel
is
burned,
the
mercury
in
the
fuel
will
be
converted
to
elemental,
oxidized,
and
particlebound
mercury.
Though
the
fuel
analysis
is
most
directly
related
to
total
mercury
emissions,
it
may
be
possible
to
estimate
the
partitioning
of
the
mercury
based
on
other
fuel
components
and
fuel
firing
conditions.

36
"
Clear
Skies
Act
of
2003"
accessed
at
http://
www.
epa.
gov/
air/
clearskies/
Air_
005.
pdf
37
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
5.1.2
Accuracy
and
Precision
Accuracy
and
Precision
for
mercury
fuel
analysis
appear
to
be
dependent
upon
both
the
analytical
method
and
the
specific
matrix
being
analyzed.
ASTM
D­
3684­
01
is
reported
to
have
a
bias
with
respect
to
three
reference
materials
analyzed,
as
summarized
in
Table
5­
1.
Bias
was
significant
at
a
95
percent
confidence
level
for
all
three
reference
materials
but
appears
to
be
dependent
on
the
specific
matrix.
The
mean
Hg
content
from
ICR
phase
II
for
bituminous
coals
at
electric
utilities
was
0.11
ppm37.
This
was
quite
comparable
to
the
Certified
Reference
Material
(
CRM)
used
to
evaluate
bias
in
the
ASTM
method.
Repeatability
of
the
method
is
reported
as
0.036
ppm,
while
reproducibility
of
the
method
is
reported
as
0.054
ppm.
38
Table
5­
1.
ASTM
D­
3684­
01
Bias
39
CRM
Description
CRM
value,
ppm
Bias,
ppm
SRM
1630a
Bituminous
Coal,
PA40
0.0938
0.0081
SRM
2692b
Bituminous
Coal,
WV41
0.1333
­
0.0087
SARM
20
Subbituminous
to
Bituminous
Coal,

South
Africa42
0.25
­
0.06
37
National
Risk
Management
Research
Laboratory,
"
CONTROL
OF
MERCURY
EMISSIONS
FROM
COAL­
FIRED
ELECTRIC
UTILITY
BOILERS:
INTERIM
REPORT
INCLUDING
ERRATA
DATED
3­
21­
02,"
EPA­
600/
R­
01­
109,
April
2002.

38
D3684­
01
Standard
Test
Method
for
Total
Mercury
in
coal
by
the
Oxygen
Bomb
Combustion/
Atomic
Absorption
Method,
ASTM
International,
West
Conshohoken
Pennsylvania
39
D3684­
01
Standard
Test
Method
for
Total
Mercury
in
coal
by
the
Oxygen
Bomb
Combustion/
Atomic
Absorption
Method,
ASTM
International,
West
Conshohoken
Pennsylvania
40
http://
patapsco.
nist.
gov/
srmcatalog/
certificates/
view_
cert2gif.
cfm?
certificate=
1630a
41
http://
patapsco.
nist.
gov/
srmtcatalog/
certificates/
2692b.
pdf
42
http://
energy.
er.
usgs.
gov/
products/
papers/
B1823/
append.
htm
38
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
5.2
Assumptions
For
this
cost
estimate,
the
fuel
Hg
sample
is
assumed
to
be
derived
from
fuel
samples
already
being
collected
and
shipped
for
analysis.
The
equipment
and
techniques
required
to
take
the
sample
are
therefore
assumed
to
be
established
for
each
unit.
It
is
assumed
that
fuel
analysis
will
be
contracted.

The
form
of
any
regulation
is
expected
to
specify
the
sampling
and
analysis
frequency.
Under
ICR
phase
II,
over
152,400
shipments
were
recorded
and
over
39,500
analyses
performed43.
One
thousand
one
hundred
and
forty
(
1,140)
coal­
fired
units
were
identified
at
approximately
450
facilities44.
Assuming
a
similar
frequency
for
potential
regulation
will
result
in
approximately
35
analyses
per
unit
per
year.
Assuming
an
unbiased
measurement
technique
and
15%
RSD
with
respect
to
mercury
inputs,
35
analyses
should
provide
an
annual
estimate
within
5.2%
of
actual
mercury
inputs
with
95%
confidence.

There
are
several
ASTM
methods
available
for
measuring
mercury
including
D3684­
01,
D6414­
01,
and
D6722­
01.
Without
further
regulatory
guidance,
ASTM
D­
3684
is
assumed
to
be
the
utilities
method
of
choice
and
will
be
used
in
this
cost
estimate.
ASTM
D­
3684
appears
to
have
been
widely,
though
not
uniformly,
chosen
for
fuel
analysis
in
the
ICR.

Costs
for
emission
factor
development,
if
implemented,
are
assumed
to
be
on
a
unit
basis.
The
emission
factor
is
assumed
to
validate
the
percentage
of
mercury
inputs
emitted
from
the
stack.
The
emission
factor
could
similarly
be
required
to
demonstrate
no
significant
bias
in
the
mercury
analysis
technique.
While
procedures
for
implementing
such
a
validation
have
yet
to
be
developed,
this
cost
estimate
will
include
quarterly
(
4)
OH
sampling
events
with
three
tests
per
sampling
event.
No
additional
coal
samples
and
analysis
are
assumed
to
be
associated
with
the
emission
factor
validation.

43
U.
S.
EPA,
"
ICR
Data
Analysis
Presentation
for
NWF;
September
8,
2000,"

http://
www.
epa.
gov/
ttn/
atw/
combust/
utiltox/
nwf_
9_
8.
pdf
44
ibid.
39
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
5.3
Annual
Costs
5.3.1
Labor
Costs
Costs
for
administering
the
fuel
analysis
program
for
Hg
are
derived
using
estimates
from
the
ICR
supporting
statement.
45
Though
not
specified
in
the
ICR
Supporting
Statement,
two
technical
hours
per
year
have
been
included
to
support
contracting
for
additional
analysis.
Thirty­
two
technical
hours,
8
hours
per
quarter,
are
assumed
for
compiling
and
reviewing
analytical
data.
Four
technical
hours
are
estimated
for
quarterly
submission
of
analytical
results.
Routine
operations
are
therefore
expected
to
require
38
technical
hours
per
year.
Labor
costs
for
routine
operation
of
a
mercury
sampling
program
are
estimated
at
$
2,016,
including
management
and
clerical
costs.

If
an
emission
factor
verification
were
included,
additional
costs
would
be
associated
with
administering
this
program.
The
program
is
assumed
to
consist
of
four
(
4)
OH
sampling
events
with
three
(
3)
tests
per
sampling
event.
The
following
technical
hours
have
been
assumed
on
an
annual
basis
for
verification
testing:


For
obtaining
an
OH
sampling
contract,
40
technical
hours.


For
oversight
of
12
tests,
96
technical
hours.


To
compile
and
review
analytical
data,
32
technical
hours.


To
submit
test
results,
4
technical
hours.

Total
technical
hours
for
administration
of
the
emission
factor
verification
program
are
therefore
expected
to
be
172
per
year.
Including
management
and
clerical
hours
labor
costs
for
the
emission
factor
verification
program
at
the
stack
level
are
estimated
at
$
9,126
annually.

45
U.
S.
EPA,
"
Standard
Form
83­
1
Supporting
Statement
for
OMB
Review
of
ICR
No._____:[
sic]
Information
Collection
Request
for
Electric
Utility
Steam
Generating
Unit
Mercury
Emissions
Information
Collection
Effort,"
November
16,
1998.
available
at
http://
www.
epa.
gov/
ttn/
atw/
combust/
utiltox/
114_
ss19.
pdf
40
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
5.3.2
Subcontract
and
ODC
Cost
of
mercury
analysis
is
estimated
at
$
54
per
occurrence,
which
is
the
average
of
vendor
indicated
analysis
costs
of
$
5046
and
$
5847.
Shipping
to
the
contract
laboratory
is
expected
to
be
covered
by
current
analytical
requirements.
A
total
of
35
mercury
analysis
per
year
results
in
a
mercury
analysis
cost
of
$
1,890
per
year
per
unit.

As
mentioned
in
section
1.1.1,
costs
for
contracting
for
OH
sampling
and
analysis
are
estimated
a
$
15,000
per
sampling
event
with
three
tests
each.
An
emission
factor
verification
program
consisting
of
four
(
4)
sampling
events
is,
thus,
estimated
to
cost
$
60,000
per
year
for
each
unit.

5.3.3
Total
Annualized
Costs
Annual
costs
for
the
mercury
analysis
program
are
estimated
to
be
$
3,906
with
no
emission
factor
verification
program.
Cost
of
an
emission
factor
verification
program
is
estimated
to
be
$
69,126.
Whether
an
emission
factor
verification
program
would
be
required
for
all
units
or
just
for
those
wishing
to
demonstrate
a
level
of
mercury
control
would
have
to
be
defined
by
regulation.
Annual
cost
for
the
mercury
analysis
program
with
an
emission
factor
verification
program
is
estimated
at
$
72,926
per
unit.

5.4
Uncertainties
Use
of
Hg
fuel
analysis
for
quantifying
mercury
emissions
is
confounded
by
the
matrix
specific
bias
associated
with
the
analytical
technique.
Reducing
this
bias
or
quantifying
the
matrix
effects
may
greatly
influence
the
need
to
perform
emission
factor
verification
testing.
It
may
also
be
necessary
to
estimate
bias
associated
with
non­
coal
fuels
such
as
pet­
coke,
RDF
or
biomass
with
respect
to
the
chosen
analytical
technique.
Additionally,
QA
necessary
for
defensible
analysis
at
the
desired
detection
levels
must
be
established.

46
Personal
communication
from
Kim
Risi,
PSC
Analytical
Services
to
Carl
Singer,
ARCADIS
on
March
24,
2003.

47
Personal
communication
from
Ken
Munger,
SGS
Commercial
Testing
and
Engineering,
to
Carl
Singer,
ARCADIS,
on
March
12,
2003.
41
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
The
cost
of
Hg
fuel
analysis
for
quantifying
Hg
emissions
could
be
significantly
reduced
if
a
reliable
emission
factor
were
developed
for
important
equipment
and
fuel
classes.
This
approach
may
be
particularly
applicable
to
units
expecting
little
or
no
Hg
control.
Units
burning
sub­
bituminous
or
lignite
coals
controlled
only
by
ESPs
appear
to
exhibit
limited
control.
48
On
the
other
hand,
use
of
a
single
emission
factor
for
a
unit
may
not
be
appropriate
in
all
cases,
especially
for
units
which
change
fuel
or
fuel
blends,
because
the
emission
factor
may
change
significantly
dependent
on
the
fuel
due
to
changes
in
Hg
partitioning
between
elemental,
oxidized,
and
particle­
bound
forms.

The
use
of
fuel
analysis
for
quantifying
Hg
emissions
may
not
be
appropriate
for
all
units
or
within
and
market
based
trading
scenario.
The
Hg
emissions
can
be
influenced
by
a
number
of
factors,
which
may
vary
due
to
plant
operating
conditions
and
may
not
be
representatively
captured
during
verification
tests.
These
parameters
may
include
fuel
sources,
combustion
conditions,
and
control
device
operations.

48
EPA­
600/
R­
01­
109
42
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
6.
Discussion
and
Summary
6.1
Comparison
Carl:
please
update
this
section
with
revised
costs.

Table
6­
1
includes
an
overview
of
the
estimated
annual
costs
associated
with
the
four
methods
that
have
been
reviewed.
The
costs
for
OH
analysis
for
annual
emissions
estimation
is
$
203,134.
The
cost
for
carbon
trap
analysis
for
annual
Hg
emissions
estimation
is$
115,009
.
The
cost
for
CEMS
and
fuel
analysis
for
annual
Hg
emissions
estimation
are
$
85,164
and
$
73,032,
respectively.
A
large
portion
of
the
monitoring
costs
are
associated
with
validating
the
proposed
measurement
against
a
reference
method.
In
this
case,
the
reference
method,
OH,
samples
representatively
across
the
stack
as
opposed
to
point
sampling
used
for
Carbon
Traps
and
CEMS
or
inlet
Hg
based
fuel
analysis..

Table
6­
1.
Summary
of
Mercury
Annual
Analysis
Costs
OH
Method
Carbon
Trap
CEMS
Fuel
Analysis
Labor
Routine
Operations,
Technical
Hours
436
772
255
38
Verification
Testing,
Technical
Hours
n/
a
137
133
172
Total
Technical
Hours
436
909
388
210
Technical
Labor,
$
20,710
43,178
18,430
9,975
Management
Labor,
$
1,248
2,603
1,111
601
Clerical
Labor,
$
1,175
2,451
1,046
566
Total
Labor,
$
23,134
48,232
20,587
11,142
Subcontract
&
ODC
Verification
Testing
180,000
39,400
37,000
60,000
Other
Analysis
n/
a
21,640
n/
a
1,890
Audit
Cylinder
n/
a
n/
a
5,000
n/
a
Parts
&
Consumables
n/
a
n/
a
2,500
n/
a
Total
Subcontract
&
ODC,
$
180,000
61,040
44,500
61,890
Annualized
Capital
Cost
Annualized
Capital
Cost,
$
0
5737
20,077
n/
a
Total
Annual
Costs,
$
203,134
115,009
85,164
73,032
43
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
6.2
Barriers
to
Cost
Analysis
The
accuracy
of
cost
estimates
is
limited
by
the
developing
nature
of
the
mercury
emission
monitoring
industry.
The
accuracy
and
precision
of
the
measurement
methods
are
improving
but
are
largely
unquantified.
The
requirements
of
mercury
monitoring
are
similarly
undefined
by
regulation.
The
validation
requirements
must
also
be
developed
to
satisfy
the
accuracy
and
precision
requirements
defined
by
regulation.
Finally,
the
market
for
these
validations
will
mature
dependent
on
the
requirements
of
the
mercury
emission
monitoring
industry.

The
accuracy
and
precision
of
OH
testing,
carbon
trap
testing,
and
CEMS
are
currently
undergoing
evaluation
under
several
coal­
fired
plant
configurations.
While
OH
testing
is
widely
used
as
the
reference
method,
the
precision
of
the
method
is
poorly
established
confounding
comparisons
to
alternative
methods.
Alternatively,
generally
accepted
independent
standards
by
which
to
judge
accuracy
and
precision
are
not
available;
an
elemental
mercury
gas
standard
is
under
development
but
an
oxidized
or
total
mercury
gas
standard
is
unavailable.

Mercury
emissions
from
coal­
fired
boilers
are
not
currently
regulated
under
federal
authority.
The
measurement
and
reporting
requirements
of
any
proposed
regulation,
including
incentives
and
penalties,
will
shape
the
type
and
frequency
of
mercury
measurements.
The
marketplace
will
either
develop
based
on
industry
perception
of
risk
and
reward
or
based
on
regulatory
command
and
control
requirements.
Information
on
the
reliability,
O&
M
costs,
and
other
metrics
of
various
monitoring
techniques
are
largely
unavailable,
though
they
are
being
evaluated
by
EPA
during
short­
term
testing
at
coal­
fired
power
plants.

Validation
of
the
mercury
monitoring
can
have
a
large
impact
on
the
total
costs
of
a
mercury
monitoring
approach.
Validation
will
depend
on
the
regulatory
framework
required
for
mercury
monitoring.
Our
estimates
assumed
validation
techniques
to
provide
results
nominally
equivalent
to
a
RATA.

Finally,
the
cost
of
validation
of
mercury
monitoring
will
likely
change
as
the
market
for
these
services
develop.
These
costs
are
currently
based
on
OH
testing
at
an
individual
stack
in
the
current
market.
The
volume
of
contracting
required
may
significantly
affect
costs
as
the
amount
of
travel
and
number
of
deployments
are
reduced.
The
number
and
location
of
sampling
firms
providing
this
service
will
be
affected
by
the
amount
of
contracting
that
develops.
Competitive
pressures
are
also
likely
to
drive
down
costs.
44
Task
1:
Cost
of
Mercury
Measurement
Techniques
Draft
Report,
Revision
1
6.3
Suggestions
for
Improvements
and
Additional
Research
The
accuracy
and
precision
of
the
OH
tests
need
to
be
quantified
in
order
to
establish
appropriate
validation
procedures
with
this
method.
This
precision
may
define
the
number
of
tests
required
to
achieve
desired
accuracy.
It
is
particularly
important
to
understand
the
precision
at
low
levels
of
mercury
where
analytical
variability
may
become
a
large
fraction
of
the
total
variability.
Improvements
to
the
accuracy
and
precision,
especially
at
low
mercury
levels,
should
be
incorporated
into
the
method
prior
to
the
evaluation
of
precision.
A
sampling
program
to
evaluate
the
precision
and
accuracy
of
the
OH
method
is
therefore
recommended
to
evaluate
low,
medium,
and
high
concentration
sources.

A
more
accurate
estimate
of
annual
monitoring
costs
will
require
additional
information
regarding
continuous
application
of
each
method.
While
short­
term
evaluations
of
CEMS
and
a
version
of
carbon
traps
are
ongoing
or
planned,
these
evaluations
are
not
likely
to
extend
for
a
year,
the
benchmark
typically
encountered
in
Part
75.
Quantitative
information
regarding
operation
and
maintenance
and
reliability
will
be
necessary
for
method
comparison.
Preferably,
these
evaluations
should
encompass
a
wide
range
of
coals
and
plant
configurations.
End
of
Document
