United States
Environmental Protection
Agency
Air and Energy
Engineering Research Laboratory
Research Triangle Park NC 27711
Research and Development
EPA/600/S7-90/009 July 1990
&EPA Project Summary
Comparison of West
German and U.S. Flue Gas
Desulfurization and Selective
Catalytic Reduction Costs
T.E. Emmel, M. Maibodi, and J. A. Martinez
By the end of the 1980s, more than
45,000 MWe and, by early 1990, more
than 34,000 MWe of coal- and oil-fired
utility boilers in the Federal Republic:
of Germany (FRG) will have been
retrofitted with flue gas desulfuriza-
tion (FGD) and selective catalytic
reduction (SCR), respectively. This
report documents a comparison olf
the actual cost of retrofitting FGD and
SCR on FRG boilers to cost es-
timating procedures used in the U.S.
to estimate the retrofit of these con-
trols on U.S. boilers. The estimated
capital costs of FGD using the U.S.
procedures compared well to the
reported capital cost for the 13 FRG
boilers evaluated. The difference be-
tween the estimated and actual costs
was -8 to 12%. However, there are
significant design differences be-
tween U.S. FGD systems built to com-
ply with New Source Performance
Standards (NSPS) and the FRG
systems. These differences, which
result in significantly lower capital!
costs on a dollar per kilowatt basis
for the FRG systems, include: no
spare absorber modules, large scrub-
ber modules, and smaller sorbenl:
and waste handling systems due to
the low sulfur coals burned in the
FRG. The estimated capital cost olf
SCR using the U.S. procedures also
compared well to the reported capital
costs for the nine FRG boilers
evaluated. The difference was be-
tween -5 to 16%. However, the U.S.
procedures were modified to reflect:
the catalyst volume and cost used in
the FRG boilers. The previous U.S.
estimates used larger catalyst vol-
umes and higher catalyst costs, and
incorporated process contingences
that were not used in this study to
develop the SCR cost estimates.
This Project Summary was
developed by EPA's Air and Energy
Engineering Research Laboratory, Re-
search Triangle Park, NC, to announce
key findings of the research project
that is fully documented in a separate
report of the same title (see Project
Report ordering information at back).
Introduction
In the mid-1980s, the Federal
Republic of Germany (FRG) enacted leg-
islation requiring significant reductions in
sulfur dioxide (SOa) and nitrogen oxides
(NOx) from existing large utility boilers. As
a result, by 1988 more than 45,000 MWe
of conventional lime/limestone (L/LS) flue
gas desulfurization (FGD) and lime spray
drying (LSD) FGD systems had been
installed, and by 1990 more than 34,000
MWe of selective catalytic reduction
(SCR) systems will have been installed.
The reported capital costs for the L/LS-
and LSD-FGD systems appear to be
much lower than the actual costs of
similar systems in the U. S. The reported
capital costs of the SCR systems are also
much lower than the estimated cost of
applying SCR systems to U.S. utility
boilers.
This report documents the results of
an analysis for comparing estimated ver-
sus actual capital costs for FGD and SCR
systems installed at several utility boilers
in the FRG.
-------
Methodology
The effort was conducted in two
phases. In Phase 1, site visits were con-
ducted at five coal-fired utility boiler
power plants in the FRG. Two of the
plants (Walheim and Mannheim) were set
up by the Institute for Industrial Produc-
tion (IPP); two other plants (Nieder-
aussem and Scholven) were visited as
part of a North Atlantic Treaty Organiza-
tion (NATO) committee meeting on the
Control of Air Pollution from Coal Com-
bustion, and the fifth plant (Ibbenbueren)
visit was set up by EPA's contractor.
Table 1 summarizes the boiler and coal
characteristics for the five plants visited
and evaluated under this study.
Prior to the site visits, a questionnaire
was sent to each utility company with the
information needs of the study. This
questionnaire provided the basis for the
data gathering effort conducted during
the plant visits. This data collection effort
focused on obtaining capital cost informa-
tion, general design and operating
parameters, plot plans, and aerial photo-
graphs of the FGD and SCR systems.
In Phase 2, the collected information
was used to develop cost estimates
based on U.S. cost estimating proce-
dures. The procedures used were devel-
oped under a National Acid Precipitation
Assessment Program (NAPAP) project
that estimated the cost and performance
of SOg and NOX controls at existing coal-
fired utility boilers. The FGD procedures
were based on the Electric Power Re-
search Institute (EPRI) report. The SCR
cost estimating procedures were based
on an EPRI report and a Tennessee
Valley Authority (TVA) report funded by
U.S. EPA. The capital cost estimates
developed using the U.S. procedures
were then compared to the reported
costs for plants evaluated under this
study. This comparison was conducted to
identify capital cost differences, reasons
for the differences, and changes needed
to the cost estimating procedures.
Summary of FGD Results
Table 2 summarizes the capital cost
comparison for L/LS- and LSD-FGD for
four of the plants. Capital cost estimates
were not available for the individual
boilers at the Scholven plant. The
estimated capital costs for L/LS-FGD
versus the reported actual costs were
very close, having an absolute difference
between -8 and 12%. Likewise, the
average difference for the two LSD-FGD
systems was 12%. The conversion rate
used for this analysis was 2 deutsche
Table 1. German Plant Visits
Fuel Size Boiler
Plant/Units Typea (MWe) Typeb
SCR SCR On- FGDOn-
Typec line Date FGD Typed line Date
Walheim 1
2
Mannheim 3,4
7
Ibbenbueren B
HC
HC
HC
HC
HC
W3
153
2x220
475
770
WB
WB
WB
DB
WB
HD
TG
HD
TG
1988
1988
1988
1988
LSD
LSD
LS
LS
1987
1987
1988
1988
1987
Niederaussem
A-H
Scholven B
C
D
E
• F
G,H
LG
BC
BC
BC
BC
- HC -
Oil
9x300
370
370
370
370
•••-740
2X714
DB
DB
DB
DB
DB
DB
Oil
—
HD
HD
HD
HD
HD
HD
—
1989
1989
1989
1989
1989
1986-87
LS
LS
LS
LS
LS
LS
-
1988
1988
1988
1987
1987
1979-87
-
aCoal types: HC = hard coal, LG = lignite, BC = ballast coal.
bBoiler types: DB = dry bottom, WB = wet bottom.
°SCR types: HD = high dust, TG = tail gas.
dFGD types: LSD = lime spray drying, L = lime, LS = limestone.
marks (DM) to the U.S. dollar. The
following changes were made to the FGD
cost estimating model for this study.
Number of Spare Absorber
Modules
None of the FGD systems evaluated in
the FRG have spare absorber modules
because German legislation allows the
plant to be out of compliance for 240
hours a year. Therefore, the boiler does
not have to shut down due to FGD
system operation problems. In the U. S.,
the 1979 New Source Performance
Standard (NSPS) does not allow a boiler
to operate out of compliance unless a
spare absorber module is available.
Additionally, operating out of compliance
for any significant amount of time would
result in noncompliance with the 30-day
rolling average emission limit. As a result,
most U.S. utility companies have chosen
to have spare absorber modules rather
than reduce the load or shut down when
the FGD system is not operating
adequately. This increases capital costs
by 20% for a 500-MWe unit.
Scrubber Module Size
Many L/LS-FGD modules at the FRG
plants handled 300-500 MWe equivalent
of flue gas. Large module sizes reduce
capital costs due to economy of scale.
The size of most U.S. scrubber modules
is typically 100-150 MWe to minimize
spare module costs.
Sorbent and Waste Handling
Quantities
All FRG coals are low in sulfur (~1%),
which results in lower capital costs due to
smaller sorbent and waste handling sys-
tems. By contrast, U.S. boilers have coal
sulfur contents of 1 to 4%. Additionally,
most of the L/LS-FGD systems in FRG
receive the sorbent pulverized, and the
capital cost for pulverization is reflected
in the cost of the sorbent (consumables)
and not in the system capital costs.
General System Design
The FRG FGD systems represent
current FGD design concepts, which in
general are less complex and lower in
capital costs than those built in the-U.S.
before 1985. Process design simplifica-
tion examples include single-loop scrub-
ber with slurry addition and oxidation in
the scrubber bottom, and the use of
hydroclones instead of thickeners before
vacuum belt dewatering. These designs
represent state-of-the-art technology and
are used on new U.S. systems.
Combined Systems
Another factor which affects the capital
cost of FGD systems is the system size.
Larger systems cost less on a $/kW basis
because of economies of scale. Because
the FRG regulations required almost all
utility boilers to retrofit scrubbers, flue
gas from adjacent boilers at the same
plant were typically tied into a single
-------
large FGD system. Thus, economies of
scale resulted from having a single large
system, instead of multiple smaller ones,
and from having larger absorber mod-
ules. For example, at the Niederaussem
plant a 2700 MW system was
constructed instead of eight units ranging
in size from 150 to 600 MW.
Summary of SCR Results
Table 3 summarizes the capital cost
comparison for SCR at four of the plants.
Capital cost estimates were not possible
at the lignite-fired plant in Niederaussem
because these boilers are expected to
meet the NOX emission limit by retro-
fitting combustion modification controls.
As Table 3 shows, the difference
.between the actual cost and the
estimated capital cost varied between -5
and 16%. Catalyst volume and costs for
the SCR systems were based on informa-
tion from' the plants. Access and
congestion retrofit difficulty and scope
adders were estimated based on proce-
dures developed for FGD systems. The
study results confirm that the major
variables that affect the capital cost are
catalyst volume and cost, retrofit difficul-
ty, and scope adder costs. However, still
unknown is the expected catalyst life
when firing U.S. coals having high sulfur,
alkali metal and/or arsenic contents.
Access and congestion retrofit difficul-
ty and general facilities were estimated
using the methodology developed for
FGD systems. The following access and
congestion retrofit factor description was
found to give close approximation to the
actual reported SCR capital costs:
• Base - Similar to new plant where
there is free access for large cranes
and equipment near the boiler (hot
side) or chimney (cold side). The
Ibbenbueren plant with a cold side
•- SCR system behind the existing chim-
ney is representative of this situation.
• Low-Space is somewhat limited such
that a standard equipment layout is
not possible, but access exists for
large cranes on two sides. The Schol-
ven units B-E with the hot side SCR
reactors at ground level next to the
ESPs are representative of this
situation. Limited space existed be-
tween the units for locating the SCR
reactors and cranes.
• Moderate - Limited space requiring
special equipment designs and lay-
outs and crane access limited to one
side. The Scholven F and Mannheim 7
units represent this situation where the
SCR reactors are elevated between
the economizer and air heater, but
Table 2. Summary of FGD Capital Cost Comparison
Plant Name Mannheim Ibbenbueren Niederaussem
Walheim
Boiler/Block
Coal sulfur
FGD type
FGD size (MWe)
SO2 removal efficiency (%)
Number of absorbers
Estimate of retrofit difficulty
Access/Congestion factor
Scope adder costs ($/kWe)
General facilities (%)
Total capital cost ($/kWe)
EPA Contractor's Estimate
Actual reported3
Percent difference
7
1.0
LS-FGD
475
80
1
1.10
Low
0
8
140
140
0
B
1.0
L-FGD
770
85
2
1.10
Low
0
5
119
,130 ,
-8
A-H
1.0
LS-FGD
2700
90
9
1.87
Low
4.6
5
260
240
8
1-2
1.0
LSD-FGD
256
90
2
1.16
Low
0
10
190
170
12
aCosts are based on an exchange rate of DM 2 to the U.S. dollar.
outside of the boiler building. Both
units are end units allowing for
reasonable crane access.
• High - Severe space limitations with
access for large cranes blocked on all
sides. The Walheim 2 unit represents
this situation where the boiler building
wall was removed to allow access to
construct the SCR reactors between
the economizer and air heater.
Conclusions
Flue Gas Desulfurization
The results of this study show that
FRG capital costs for conventional
lime/limestone wet and dry FGD systems
are similar when differences in the scope
of supply (design) are taken into account.
The major design differences between
the FRG and U.S. designed systems are
due to: -
• Coal Sulfur Content - Higher coal
sulfur content of most U.S. coals
results in higher capital costs for
sdrbent and waste handling facilities
and use of spare absorbers to ensure
operating reliability.
• Combined Systems - Most FRG sys-
tems are large because of flue gas
from multiple units is processed in one
system. Combined systems have
lower capital cost requirements due to
economy of scale.
• Regulatory - NSPS bypass and aver-
aging provisions have resulted in use
of spare modules and small absorber
sizes to minimize cost of spare
absorbers.
• Technology Status - Most FRG
systems employ 1980s technology
whereas most U.S. systems employ
1970s technology because of when
the systems were built. Newer designs
are more reliable, reduce the need for
spare absorbers, and allow the use of
larger absorber sizes.
Results of this study indicate that:
1. The lower capital cost of FGD systems
in FRG relative to U.S. systems is due
to scope of the supply (design)
differences.
2. Future U.S. systems will have lower
capital cost requirements than past
systems due to technology improve-
ments.
3. Regulatory provisions having less
stringent bypass and averaging
requirements than NSPS can signif-
icantly reduce the capital cost of FGD
systems.
4. Combined handling of flue.,gas from
multiple units can significantly reduce
system capital costs.
Selective Catalytic Reduction
Study results show that FRG SCR
system capital cost can be accurately
estimated if catalyst cost and retrofit
difficulty are known. The retrofit difficulty
adjustment methodology found in an
EPRI report can be used to account for
different access and congestion situa-
tions in the FRG. Study results indicate
that:
1. The FRG SCR system capital costs
can be used to estimate U.S. retrofit
applications if adjustments for retrofit
difficulty and catalyst costs are made.
-------
2. Because of the differences in U.S. coal
trace element content versus that of
FRG, the catalyst life requirements of
the FRG applications are not directly
transferable to all U.S. applications. As
such, catalyst life assumptions used to
estimate annual costs must reflect this
uncertainty.
The following contaminants found in
utility boiler flue gases over time
deactivate the catalyst: SOX, particulate
matter, arsenic, and alkali metals. The
contaminants vary with the fuel and
reactor location. SCR has been com-
mercially applied in Japan and Germany
to the following utility boiler situations:
Table 3. Summary of SCR Capital Cost Comparison
Plant Name Mannheim Ibbenbueren Walheim Scholven
Scholven
Boiler/Block
Boiler type
SCR type
SCR size (MWe)
NOX removal efficiency (%)
Access/Congestion factor
Scope adder cost ($/kWe)
General facilities (%)
Catalyst cost
Dollarslftz*
DollarslkWe
Total capital cost ($/kWe)
7 8 2 B-E
Dry bottom Wet bottom Wet bottom Dry bottom
Hot
475
77-32
Moderate
3.0
13
504
18.7
Cold
770
91-92
Base
47.7
25
355
7.4
Hot
153
88
High
42.5
13
545
25.2
Hot
370 each
67
Low
3.1
13
283
16.2
F
Dry bottom
Hot
740
60
Moderate
3.4
13
283
7.8
Fuel
Type
Gas
Oil
Coal
Ash
Metals
None
Low
Low
Sulfur
Level
None
Low
1 Per-
Temperat-
ure
Hot
Hot
Hot/Cold
Dust
None
Low
Low/High
Radian estimate 82
Actual reported13 79
Percent difference 4
t>1 m3 = 35.3 ft3.
bCosts are based on an exchange rate of DM 2 to
cExcludes the costs of combust/on modifications.
98 _
703
-5
the U.S.
180 ^_87
789° 75
-5 16
dollar.
57
55
3
Coal
High 1 Per-
Arsenic cent
Hot/Cold Low/High
SCR has not been applied commercially
with high sulfur and high alkaline coals.
Many boilers in the U.S. burn high sulfur
coals (2-5%) and coals with highly
alkaline ash (lignites and subbituminous
coals). Also, many of the low sulfur
eastern coals have high arsenic contents.
To address the lack of data available on
catalyst life for hot side SCR for U.S.
coals, a number of SCR pilot programs
are planned.
I.E. Emmel, M. Maibodi, and J. A. Martinez are with Radian Corporation,
Research Triangle Park, NC 27709.
Norman Kaplan is the EPA Project Officer (see below).
The complete report, entitled "Comparison of West German and U.S. Flue
Gas Desulfurization and Selective Catalytic Reduction Costs," (Order
No. PB 90-206 319/AS; Cost: $17.00, subject to change) will be
available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Air and Energy Engineering Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
EPA'600/S7-90/009
------- |