Cost Estimates for Selected Applications of NOX Control Technologies on
Stationary Combustion Boilers
Draft Report, March 1996
and
Responses to Comments on the Draft Report
June 1997
Note to Reader: To satisfy reader requests most efficiently, EPA printed this document which
combines, in this single volume, the noted March 1996 draft report with the Responses to
Comments report prepared in June 1997. Following this page, each report is reproduced in its
entirety, with the original cover page and the associated table of contents. A solid blue divider
page separates the two reports.
U.S. Environmental Protection Agency
Acid Rain Division
501 Third Street
Washington, DC 20001
-------
430R96023
COST ESTIMATES FOR SELECTED APPLICATIONS OF
NOX CONTROL TECHNOLOGIES
ON STATIONARY COMBUSTION BOILERS
Draft Report
March 1996
Prepared for
U.S. Environmental Protection Agency
Acid Rain Division
501 Third Street
Washington, DC 20001
Bechtel Power Corporation
9801 Washingtonian Boulevard
Gaithersburg, MD 20878-5356
under subcontract to
The Cadmus Group, Inc.
135 Beaver Street
Waltham, MA 02154
Prime Contract No. 68-D2-0168
Work Assignment No. 4C-02
-------
COST ESTIMATES FOR NOX CONTROL TECHNOLOGIES
TABLE OF CONTENTS
1.0 PROJECT OVERVIEW
2.0 METHODOLOGY AND GENERAL ASSUMPTIONS
3.0 COAL-FIRED PLANT ASSUMPTIONS AND RESULTS
4.0 NATURAL-GAS-FIRED PLANT ASSUMPTIONS AND RESULTS
5.0 OIL-FIRED PLANT ASSUMPTIONS AND RESULTS
6.0 REFERENCES
APPENDIX A INVESTIGATION OF PERFORMANCE AND COST OF NOX
CONTROLS AS APPLIED TO GROUP 2 BOILERS, DRAFT REPORT,
AUGUST 1995
22885.008\Snj
-------
1.0 PROJECT OVERVIEW
This report presents the results of a study conducted by Bechtel to develop costs for NOX control
technologies for coal-, gas-, and oil-fired boilers. The types of boilers for each fuel along with
the size range and baseline NOX emission rate for each boiler type were identified by the United
States Environmental Protection Agency (EPA), as shown in Table 1-1.
The technical and economic evaluations conducted for this study used a consistent methodology
to develop costs for various NOX control technology applications. The costs are therefore
comparable between different boiler types and sizes.
1.1 Project Purpose
The primary objectives of this study were to:
• Develop costs for the NOX control technologies with a capability to reduce NOX emission
from the baseline NOX rate to 0.15 Ib/MMBtu for each study boiler
• Develop costs for the NOX control technologies with a capability to provide substantial NOX
emission reductions for the dry-bottom tangential and wall-fired boilers burning coal beyond
those required under 40 CFR Part 76
1.2 Major Results
The capital and levelized costs for each technology case are presented in the figures that are in-
cluded at the end of this report. The major costs from these figures are summarized in the fol-
lowing tables:
• Table 1-2 presents the fixed and variable costs for a 200 MW boiler for each technology
application. The variable costs are reported for both the 27 and 65 percent capacity factors.
Two types of variable costs have been included: one containing the carrying charges for the
capital expenditure and the other without this carrying charge (as reported in EPRI's TAG).
In addition, Table 1-2 also provides a mathematical relationship to facilitate estimation of the
capital cost for a given boiler size (MW).
• Tables 1-3 and 1-4 present the capital ($/kW) and levelized ($/ton of NOX removed) costs for
two selected sizes of boiler installations for each NOX control technology (for both 0.15
Ib/MMBtu and substantial reduction cases). These costs are reported for both the 27 and 65
percent capacity factors. Also provided are references to the figures from which these costs
have been obtained.
1.3 General Approach to Technical and Cost Analyses
The overall approach for both the technical and cost analyses was based primarily on the meth-
odology utilized in a previous Bechtel study that involved evaluation of NOX control tech-
22885.008\Study\Cost-Est.NOx
1-1
-------
nologies for the Group 2 boilers. A copy of the previous study is provided as Appendix A to this
report.
The major elements of the project approach and the areas where the approach differs from the
previous study are as follows:
• An evaluation of the commercially available NOX control technologies was made to deter-
mine feasibility for meeting the aforementioned project objectives. Table 1-5 lists these
technologies along with their NOX reduction effectiveness and applicability to each study
boiler type. The data presented in Table 1-5 were based on published information on a
variety of technology applications (References 1 through 17).
Based on the above evaluation, the following technologies are considered in this report:
* The selective catalytic reduction (SCR) technology was selected for its capability to
provide NOX reduction to the 0.15 Ib/MMBtu limit for all study boilers. For the oil- and
gas-fired boilers, both the selective noncatalytic reduction (SNCR) and gas reburning
technologies were also selected for the same purpose.
* The SNCR, gas reburning, and coal reburning technologies have been found to have a
capability to provide substantial NOX reduction for the tangential and wall-fired boilers
burning coal. Of these, the SNCR technology was selected for evaluation for this
project. Costs of gas and coal reburning applications on Group 2 boilers have been
examined in detail in the previous Bechtel study (Appendix A).
• The technical and economic evaluations were conducted on representative boiler installations
for each boiler category identified for this project. The design data for the representative
boiler installations were developed from Bechtel's in-house database.
• Both capital costs ($/kW) and levelized costs (mils/kWh and $/ton NOX removed) were
developed for the applicable boiler size range for each technology application.
• The capital cost estimates were developed by factoring from the 1994 cost data generated in
the previous Bechtel study (Appendix A) for each NOX control technology. The new esti-
mates were not based on detailed major equipment lists, as developed in the previous study.
Instead, appropriate power factors representing the general industry practice were applied to
the existing costs to obtain costs for this project. This method took into consideration the
differences in the overall system size and capacity between each technology application for
this project and the corresponding application in the previous study.
• All new costs were developed in 1995 dollars. The latest available Chemical Engineering
cost index for September 1995 was used to adjust the estimated 1994 costs to 1995.
• The levelized costs were based on the economic factors reported in the 1993 EPRI TAG
(Reference 18). They were developed using a constant dollar approach. Other economic
assumptions were the same as shown in Appendix A and detailed in Section 2.0.
2288S.008\Study\Cost-Est.NOx
1-2
-------
• All levelized costs were developed based on the following operating modes:
* The NOX control technology in operation for the entire 12-month period with a capacity
factor of 65 percent
* The NOX control technology in operation for 5 months in a year with a capacity factor of
65 percent (resulting in an effective yearly capacity factor of approximately 27 percent).
1.4 Clarifications
Note that alternate technologies other than those selected in this study may also be applied to
achieve the study objectives. For instance, a combination of some of the technologies, such as
hybrid SCR/SNCR, can be used for this purpose. However, these alternatives are beyond the
scope of the study. Additionally, this study does not imply that NOX emissions from every boiler
in the populations considered can be controlled to a 0.15 Ib/MMBtu level; some boilers may be
controlled to levels higher than 0.15 Ib/MMBtu and others to levels lower than 0.15 Ib/MMBtu.
22885.008\Study\Cosi-Est.NOx
1-3
-------
TABLE 1-1
STUDY BOILERS AND BASELINE NOV EMISSIONS1!)
Boiler Type
Dry bottom, wall-fired
Dry bottom, tangentially
fired
Cell
Cyclone
Wet bottom
Dry bottom, vertically
fired
Size Range, MW
30-1300
33-952
200-1300
25-1200
25-800
25-300
Fuel
Coal
Gas
Oil
Coal
Gas
Oil
Coal
Coal
Coal
Coal
Baseline NOX Rate
Ib/MMBtu
0.50 (Title IV limit)
0.25
0.30
0.45 (Title IV limit)
0.25
0.30
0.8-1.5
(1.00 average)
0.8-1.9
(1.1 7 average)
0.7-1.7
(1.13 average)
0.85-1.1
(1.08 average)
NOTE
For Group 1 boilers, the baseline NOX rates are the currently allowable emission limitations
under 40 CFR Part 76. For Group 2 boilers, the baseline NOX rates represent the average
uncontrolled NOX rates, per boiler type, as presented in Appendix A to "Investigation of
Performance and Cost of NOX Controls as Applied to Group 2 Boilers," August 1995,
prepared for the U.S. EPA.
2288S.008\Study\Cost-Ea NOx
-------
TABLE 1-2
CAPITAL AND O&M COSTS(I)
Case'4'
COAL- TANGEN- SCR
COAL-TANGEN-SNCR
COAL- WALL- SCR
COAL-WALL-SNCR
COAL- CELL- SCR
COAL- CYCLONE- SCR
COAL- WET BOTTOM- SCR
COAL- VERT-FIRED- SCR
GAS-SCR
GAS- REBURN
GAS-SNCR
OIL- SCR
OIL- REBURN
OIL-SNCR
Capital Cost (S)
66.824*(200/MW)A0.35
I5.551*(200/MW)A0.577
69.382*(200/MW)A0.35
I7.5II*(200/MW)A0.577
69.2I7*(200/MW)A0.324
69.55*(200/MW)A0.26I
70.57 I*(200/MW)A0.296
67.067*(200/MW)A0.391
27.483*(200/MW)A0.35
I9.025*(200/MW)A0.357
9.433*(200/MW)A0.577
39.975*(200/MW)A0.35
22.298*(200/MW)0.357
I0.638*(200/MW)A0.577
Fixed Cost for
200 MW
1
0.23
1.04
0.26
1.04
1.04
1.06
1.01
0.41
0.29
0.14
0.6
0.34
0.16
Variable Cost
for 200 MW
w/o Capital
65% Capacity
Factor
1.04
0.99
l.ll
1.02
1.35
1.38
1.41
1.33
0.17
0.03
0.42
0.36
0.51
0.58
Variable Cost
for 200 MW
(w/o Capital)
27% Capacity
Factor
2.2
0.99
2.33
1.02
2.67
2.72
2.76
2.6
0.28
0.03
0.42
0.7
0.51
0.58
Variable Cost
for 200 MW
(w/Capital)
65% Capacity
Factor
2.53
1.34
2.66
1.41
2.89
2.93
2.98
2.83
0.79
0.45
0.63
1.25
1.01
0.82
Variable Cost
for 200 MW
(w/Capital)
27% Capacity
Factor
5.78
1.82
6.05
1.95
6.38
6.44
6.54
6.19
1.75
1.04
0.92
2.84
1.7
1.15
NOTES
1. Fixed costs are reported in $/kW-yr. The variable costs are reported in mil/kWh.
2. The variable costs are reported both with and without the carrying charges for the capital costs. As per the EPRI's TAG, the variable costs do not include carrying
charges. Also, for this report, the costs associated with the changes in the fuel consumption rates because of the retrofit have been included in the variable costs.
EPRI does not include fuel costs in the variable cost component.
3. The capacity factor reflects the annual usage for which the NOX control technology is in operation.
4. Where the boiler firing type is not mentioned, the case applies to both the wall-fired and tangentially fired boilers.
3288S.008\Sludy\Cosl-Esl NOx
-------
TABLE 1-3
SUMMARY OF RESULTS
NOV CONTROL TECHNOLOGIES ACHIEVING 0.15 LB/MMBTU LIMIT
Boiler(1)
TN
WF
CELL
CYC
WB
VF
Fuel
Coal
Coal
Coal
Coal
Coal
Coal
NOX
Control
SCR
SCR
SCR
SCR
SCR
SCR
Boiler
Size,
MW
200
930
200
1030
200
1030
200
1030
200
730
70
200
65% Capacity
Factor(2)
S/kW
66.82
39.02
69.38
39.1
69.22
40.7
69.55
45.34
70.57
48.07
101.11
67.07
$/Ton
1935
1439
1670
1226
801
624
695
536
733
572
907
750
27% Capacity
Factor(2)
S/kW
66.82
39.02
69.38
39.1
69.22
40.7
69.55
45.34
70.57
48.07
101.11
67.07
$/Ton
4427
3238
3815
2748
1775
1351
1536
1125
1616
1231
2032
1654
Figures(3)
3-1,3,5
3-11,13,15
3-21,23,25
3-26,28,30
3-31,33,35
3-36,38,40
NOTES
1. The legend for the symbols used is:
CYC Cyclone-fired
TN Tangential
VF Vertically fired, dry bottom
WF Wall-fired, dry bottom
WB Wet bottom
2. The capacity factor reflects the annual duration for which the NOX technology is in operation.
3. The cost data presented are taken from the curves shown in the referenced figures included in
this report.
22885.008\Study\Co«-Esi.NOx
-------
TABLE 1-3 (Continued)
Boiler(1)
WF,TN
WF,TN
WF,TN
WF,TN
WF,TN
WF,TN
Fuel
Gas
Gas
Gas
Oil
Oil
Oil
NOX
Control
SCR
Reburn
SNCR
SCR
Reburn
SNCR
Boiler
Size,
MW
200
930
200
930
200
930
200
930
200
930
200
930
65% Capacity
Factor(2)
$/kW
27.48
16.05
19.03
10.99
9.43
3.66
39.98
23.34
22.30
12.88
10.63
4.38
$/Ton
2142
1429
1250
748
1632
1272
2263
1571
1776
1384
1407
1147
27% Capacity
Factor<2)
$/kW
27.48
16.05
19.03
10.99
9.43
3.66
39.98
23.34
22.30
12.88
10.63
4.38
$/Ton
4802
3091
2910
1706
2455
1592
5151
3492
3073
2122
2026
1402
Figures(3)
4-1,3,5
4-6,8,10
4-11,13,15
5-1,3,5
5-6,8,10
5-11,13,15
NOTES
1. The legend for the symbols used is:
CYC Cyclone-fired
TN Tangential
VF Vertically fired, dry bottom
WF Wall-fired, dry bottom
WB Wet bottom
2. The capacity factor reflects the annual duration for which the NOX technology is in operation.
3. The cost data presented are taken from the curves shown in the referenced figures included in
this report.
228S5.008\Study\Cost-Es.NOx
-------
TABLE 1-4
SUMMARY OF RESULTS
COST OF SNCR APPLICATIONS ON DRY-BOTTOM WALL- AND
TANGENTIALLY FIRED BOILERS
Boiler(1)
TN
WF
Fuel
Coal
Coal
NOX
Control
SNCR
SNCR
Boiler
Size,
MW
200
930
200
1030
65% Capacity
Factor(2)
$/kW
15.55
6.41
17.51
6.80
$/Ton
1378
1150
1210
988
27% Capacity
Factor(2)
$/kW
15.55
6.41
17.51
6.80
$/Ton
1921
1377
1720
1186
Figures(3)
3-6,8,10
3-16,18,20
NOTES
1. The legend for the symbols used is:
TN Tangential
WF Wall-fired, dry bottom
2. The capacity factor reflects the annual duration for which the NOX technology is in operation.
3. The cost data presented are taken from the curves shown in the referenced figures included in
this report.
22885.008\Study\Cosl-Est.NOx
-------
TABLE 1-5
APPLICABLE NOX CONTROL TECHNOLOGIES
Technology
Combustion
Controls(2)
Coal Reburning
Gas Reburning
Selective Catalytic
Reduction
Selective Non-
catalytic Reduction
NOX Reduction
Effectiveness
40 - 70%
35 -50%
40 -60%
80 - 90%
30 - 50%
Boiler Applications'1*
WF, TN, Cell, WB, VF
WF, TN, Cell, CYC
WB,VF
WF, TN, Cell, CYC
WB,VF
WF, TN, Cell, CYC
WB,VF
WF, TN, Cell, CYC
WB.VF
Primary Fuel(I)
C,0,G
C
C,0,G
C,0,G
C,0,G
NOTES
1. The legend for symbols used is:
C
O
G
WF
TN
CYC
WB
VF
Coal
Oil
Gas
Wall-fired dry bottom
Tangential
Cyclone-fired
Wet bottom
Vertically fired, dry bottom
Combustion controls include low-NOx burners, overfire air, and gas recirculation (for oil
or gas boilers only).
228!5.008\Study\Cosl-Est.NO*
-------
2.0 METHODOLOGY AND GENERAL ASSUMPTIONS
The methodology and assumptions used in selecting the applicable NOX control technologies and
conducting the technical and economic evaluations for this project are detailed in this section.
2.1 Technology Selections
Table 1-2 categorized the commercially available technologies and their NOX control potential
for various boiler types. As shown in this table, the NOX reduction effectiveness varies de-
pending on the site-specific conditions for any given application.
The study criteria define the baseline NOX rates for the dry-bottom wall-fired and tangential
boilers burning coal to be 0.45 and 0.5 Ib/MMBtu, respectively (these rates being required by 40
CFR Part 76). The baseline NOX rates for the same boilers on oil and gas are defined as 0.3 and
0.25 Ib/MMBtu, respectively, because these rates currently are being achieved on gas- and oil-
fired boilers. It is assumed that these NOX rates correspond to boilers equipped with low-NOx
burners only (no overfire air ports).
The above assumption implies that full credit can be taken for the NOX reduction potential of the
technologies (such as gas reburning) utilizing overfire air ports. Without this assumption, appli-
cation of these technologies to boilers with existing overfire air ports would be possible only if
the ports are replaced with the new ports associated with the technologies. Deletion of the
existing ports would have a corresponding impact of increasing the baseline NOX levels, thus re-
quiring a higher NOX reduction to achieve 0.15 Ib/MMBtu.
As per the study criteria, the NOX reduction efficiencies required to meet the 0.15 Ib/MMBtu for
the gas- and oil-fired boilers are 40 and 50 percent, respectively. For coal-fired boilers, these
efficiencies range from 66.67 to 87.18 percent.
Based on the above background information and assumptions, assessment of the feasibility of
applying various technologies to the study boilers is as follows:
• The various components of combustion controls include low-NOx burners, overfire air ports,
and gas recirculation fans. Where applicable, the study boilers are already equipped with
low-NOx burners. Since these burners reflect a major portion of the overall effectiveness of
combustion controls, installation of other technology components on these boilers to achieve
0.15 Ib/MMBtu does not appear possible.
• The coal reburning technology is not feasible for application on any coal-fired study boiler,
since the minimum required NOX reduction efficiency of 66.67 percent is still higher than the
maximum potential of this technology (50 percent reduction).
• The gas reburning technology can provide a NOX reduction ranging from 40 to 60 percent.
Since the reductions to achieve the 0.15 Ib/MMBtu level for the gas- and oil-fired boilers fall
within this range, this technology is considered to be a suitable candidate for these boilers. It
is to be recognized that site-specific factors for some plants may pose serious constraints
22885.00S\Study\Cost-Est.NOx
2-1
-------
either to achieve proper NOX reductions or to install the technology components. These fac-
tors include lack of sufficient space to install the rebum fuel injectors (or burners) and over-
fire air ports, lack of proper residence times, and unavailability of natural gas.
• The effectiveness of the SNCR technology can vary from 30 to 50 percent. Based on the
NOX reduction needs (0.15 Ib/MMBtu) of the study boilers, this technology can be applied
only to the gas- and oil-fired boilers. Similar to gas rebuming, this feasibility may be subject
to site-specific factors. The most important aspect of SNCR is the availability of a proper
residence time within the boiler in a required temperature zone, which varies with the type of
SNCR system used (ammonia- or urea-based). It is recognized that such residence times may
not be available in all gas- and oil-fired boilers.
• The NOX reduction needs of all study boilers fall within the potential effectiveness range (80
to 90 percent) for the SCR technology, which is therefore considered feasible for all of these
boilers.
Even for the SCR technology, the NOX reduction rates required for the cell, cyclone,, wet
bottom, and vertically fired boilers are relatively high. Such rates would require significantly
large amounts of catalyst. Other concerns, such as excessive SO3 conversion rates, may also
be applicable in some specific retrofits.
In some cases, the duty on the SCR systems could be reduced by applying more than one
NOX control technology. For instance, hybrid systems using SNCR and SCR could be used,
or SCR could be applied with combustion controls (applicable to cell, wet bottom, and verti-
cally-fired boilers). These applications are considered outside the scope of the study.
• Based on the above analyses, the technologies selected for meeting the 0.15 Ib/MMBtu limit
include SCR for all boiler categories and SNCR and gas rebuming for gas- and oil-fired
boilers only. Similarly, SNCR has been considered for achieving substantial NOX reduction
(50 percent) for the wall-fired and tangential boilers burning coal.
2.2 Technical Evaluations
The methodology for the technical evaluations is essentially the same as used in the previous
study (Appendix A, Section 2.0 of Appendix B). The highlights of this methodology are as fol-
lows:
• All design details pertaining to the representative boilers in the cyclone, cell, wet-bottom, and
vertically fired categories are the same as shown in the previous study.
• Since the tangential and wall-fired boilers burning coal, oil, or gas were not included in the
previous study, design details of representative boilers for these categories have been specifi-
cally developed for this project from the Bechtel in-house database. In the case of each boiler
category, the evaluations are performed using one representative boiler. It is assumed that
boiler design parameters vary in a direct proportion to the boiler size.
22885.008\Study\Cost-Est.NOx
2-2
-------
• The design and performance impacts of each technology application have been based on the
same assumptions as used in the previous study (Appendix A).
2.3 Economic Evaluations
Similar to the technical evaluations, the methodology in the previous study (Appendix A, Section
2.0 of Appendix B) has been utilized in most parts in conducting the economic evaluations for
this project. The major areas of differences are as follows:
• The costs for this study are in 1995 dollars compared to the 1990 dollars used in the previous
study. Because of this difference, the economic factors provided in Table B2-2 of Appendix
A have been revised for use in this study.
Table 2-1 shows the revised economic factors. The highlights of these revisions are
described below:
* The 1993 EPRI TAG has been used for establishing the carrying charge factor, leveliza-
tion factor, costs of consumables (ash and water), and cost of operating labor.
* The 1993 EPRI TAG shows a decline in the coal price from 1990 to 1995 period. For
conservatism, the 1990 coal price is used. Since the only study cases where coal con-
sumption is affected are the tangential and wall-fired boilers firing bituminous coals,
only the bituminous coal price is shown.
» The No. 6 oil and natural gas prices are based on a recent publication (Reference 19),
which is considered more current than the data presented in EPRI TAG.
* The SCR catalyst replacement costs and the SO2 allowance are assumed to be the same
as reported for the previous study.
* ' The urea and anhydrous ammonia costs are revised to reflect the 1995 costs reported in
Appendix A, Section C.2.
• For the tangential and wall-fired boilers, the costs have been developed using data for one
representative boiler. In establishing the costs for the boiler size range, it is assumed that
performance parameters vary in direct proportion to the boiler size. The capital costs for the
boiler size range have been developed by using the scaling methodology described in Section
2.4.1 of Appendix A. The same scaling factors have been used as determined for the various
technology cases evaluated in Appendix A.
• The capital costs have been adjusted to the study reference period by using the Chemical
Engineering cost index for September 1995.
22885 OOS\Study\Cos!-Est.NOx
2-3
-------
All levelized costs have been developed based on the following plant operating modes:
* The NOX control technology in operation for the entire 12-month period with a capacity
factor of 65 percent
* The NOX control technology in operation for 5 months in a year with a capacity factor of
65 percent (resulting in an effective yearly capacity factor of approximately 27 percent)
22885.008\Study\Cosl-Est.NOx
2-4
-------
TABLE 2-1
ECONOMIC FACTORS
Parameter
Value
Cost year
Useful life
Carrying charges
Levelization factor
Maintenance cost
Electrical power cost
Bituminous coal cost
Natural gas cost
No. 6 oil cost
Ash disposal cost
Anhydrous ammonia cost
Urea cost (50% solution)
SCR catalyst replacement cost
Operator cost
Water cost
SO2 allowance
September 1995
20 years
0.127
1.0
1.5% (of capital)/year
$0.05/kWh
$1.60/MMBtu
$2.27/MMBtu
$1.97/MMBtu
$11.28/ton
$202/ton
$0.80/gallon
$350/ft3
$24.82/person-hour
$0.0004/gallon
$150/ton
K885.008\Studv\Cost-EstNOx
-------
3.0 COAL-FIRED PLANT ASSUMPTIONS AND RESULTS
This section summarizes the technical and economic evaluations conducted for the coal-fired
boiler applications of NOX control technologies.
3.1 Tangential Boiler Applications
The NOX control technologies evaluated for this boiler type include SCR and SNCR. The design
data for the representative boiler selected for this evaluation are shown in Table 3-1. This boiler
is a balanced draft, forced circulation, reheat, single furnace boiler. It has four windboxes
located along the four corners of the furnace. There are a total of 20 coal burners, five per
comer. The boiler serves a 348 MW steam turbine generator and is equipped with two 50-per-
cent-capacity forced draft fans, two 50-percent-capacity induced draft fans, and an electrostatic
precipitator for removing dust from the flue gases exiting the boiler.
3.1.1 SCR Evaluation
The following major criteria and assumptions have been followed in evaluating the SCR tech-
nology for the coal-fired tangential boilers:
• The SCR system is designed to reduce NOX emission from a baseline level of 0.45 Ib/MMBtu
to the required limit of 0.15 Ib/MMBtu.
• Anhydrous ammonia is utilized as a reagent for the SCR system.
• The system is designed for an ammonia slip of 5 ppm.
• A 14-day storage is provided at the plant site for anhydrous ammonia. This storage capacity
is based on a full-load operation of the boiler.
• It is assumed that the existing plant setting allows installation of the SCR reactors between
the economizer and air heater without a need to relocate any major structure or equipment.
• The operating life of the SCR catalyst is assumed at 3 years. A catalyst life management
strategy is not used for this evaluation. It is also assumed that no appreciable difference in
the catalyst life occurs when the plant is operated at low capacity factors. This assumption
results in conservative cost estimates, since it is expected that a low-capacity factor may re-
sult in a net catalyst life increase.
• Other general SCR system design details, assumptions, and impacts on the existing equip-
ment outlined in Appendix A (Section 4.5 of Appendix B) also apply to this case.
The SCR technology is a postcombustion technology, in which the reagent is injected into the
flue gas stream at the economizer outlet upstream of the catalyst reactor. As such, SCR tech-
nology has no direct impact on the boiler performance. The boiler parameters shown in Table
3-1 would remain unchanged following a SCR retrofit. However, such a retrofit would impact
22885 008\Study\Cost-E«.NOx 3-1
-------
the plant's overall operating costs, because of the increased auxiliary power consumption.
anhydrous ammonia usage, and periodic catalyst replacement. For the study boiler, estimates of
these consumables associated with the SCR system are as follows:
Auxiliary power consumption 768 kW
Anhydrous ammonia consumption 365 Ib/hr
Average catalyst replacement 4,680 ft3/yr
Using the above parameters and the costing methodology described in Appendix A, both the
capital and levelized costs have been calculated for the entire size range (33 to 952 MW) of tan-
gential, coal-fired boilers. As shown hi Figure 3-1, the capital costs range from approximately
$35 to $130/kW. The levelized costs at a capacity factor of 65 percent range from 1.9 to 4.3
mils/kWh and $1,240 to $3,050/ton NOX removed (Figures 3-2 and 3-3). The levelized costs at a
capacity factor of 27 percent range from 4.25 to 9.9 mils/kWh and $3,100 to $7,100/ton NOX
removed (Figures 3-4 and 3-5).
3.12 SNCR Evaluation
The following major criteria and assumptions have been followed in evaluating the SNCR tech-
nology for the coal-fired, tangential boilers:
• The SNCR system is designed to provide a 50 percent NOX reduction from a baseline NOX
rate of 0.45 Ib/MMBtu.
• A urea-based SNCR technology is selected for this application.
• The system is designed for an ammonia slip of 10 ppm, selected to minimize impacts on the
equipment located downstream of the boiler (air heater, precipitator, etc.). Higher ammonia
slip may produce ammonium salts causing pluggage of air heater and contamination of ash
collected in the precipitator.
• A 14-day storage based on a full-load operation is provided at the plant site for the urea solu-
tion.
• For an effective reaction between the reagent and NOX, sufficient residence times must exist
within the boiler in a proper temperature zone (1,800 to 2,000 °F). It is assumed that such
residence times are available within the boilers being evaluated. It is to be noted that without
adequate residence tunes, it may not be possible to achieve a 50 percent NOX reduction while
maintaining the ammonia slip at 10 ppm.
• A reagent ratio of 1.75 is selected for the SNCR system design.
• Other general SNCR system design details, assumptions, and impacts on the existing equip-
ment outlined in Appendix A (Section 4.4 of Appendix B) also apply to this case.
22885 008\Study\Co5t-EslNCK
3-2
-------
Injection of the urea solution within the boiler does have an impact on the boiler performance.
because of the heat loss* associated with the moisture content of this solution. This heat loss
causes a slight reduction in the boiler efficiency, resulting in increased fuel flow, ash generation,
and combustion air and flue gas flow rates. The overall impacts of the SNCR system retrofit on
the study boiler are as follows:
• The boiler efficiency reduces from 88.39 to 88.00 percent. The boiler heat input increases
from 3,210 to 3,244 MMBtu/hr. The fuel flow, ash generation rate, and combustion and flue
gas flow rates increase in a direct proportion to the change in the heat input.
• There is an overall increase in the plant auxiliary power consumption due to the SNCR
equipment as well as the increased demand on the draft fans to accommodate the higher air
and flue gas flow rates. The estimated auxiliary power increase is 157 kW.
• The urea consumption requirement for the SNCR system is 350 gal./hr.
• The water consumption requirement for the SNCR system is 4,470 gal./hr.
Using the above parameters and the costing methodology described in Appendix A, both the
capital and levelized costs have been calculated for the entire size range (33 to 952 MW) of tan-
gential, coal-fired boilers. As shown in Figure 3-6, the capital costs range from approximately
$6 to $46/kW. The levelized costs at a capacity factor of 65 percent range from 1.13 to 2.15
mils/kWh and $1,140 to $2,130/ton NOX removed (Figures 3-7 and 3-8). The levelized costs at a
capacity factor of 27 percent range from 1.32 to 3.78 mils/kWh and $1,330 to $3,800/ton NOx
removed (Figures 3-9 and 3-10).
3.2 Wall-Fired Boiler Applications
The NOX control technologies evaluated for this boiler type include SCR and SNCR. The design
data for the representative boiler selected for this evaluation are shown in Table 3-1. This boiler
is a balanced draft, natural circulation, reheat, single furnace boiler. It has 24 burners located
four high and six wide on the front wall of the unit. The boiler serves a 381 MW steam turbine
generator and is equipped with two 50-percent-capacity forced draft fans, two 50-percent-
capacity induced draft fans, and an electrostatic precipitator for removing dust from the flue
gases exiting the boiler.
3.2.1 SCR Evaluation
The following major criteria and assumptions have been followed in evaluating the SCR tech-
nology for the wall-fired boilers:
• The SCR system is designed to reduce NOX emission from a baseline level of 0.5 Ib/MMBtu
to the required limit of 0.15 Ib/MMBtu.
• All of the other criteria and assumptions described in Section 3.1.1 apply equally to this case.
228S5.00S\Study\Cost-EstNOx
3-3
-------
The consumables associated with the SCR system retrofit for the study boiler are as follows:
Auxiliary power consumption 842 kW
Anhydrous ammonia consumption 476 Ib/hr
Average catalyst replacement 5417 ft3/yr
Using the above parameters and the costing methodology described in Appendix A, both the
capital and levelized costs have been calculated for the entire size range (30 to 1,300 MW) of
wall-fired boilers. As shown in Figure 3-11, the capital costs range from approximately $37 to
$134/kW. The levelized costs at a capacity factor of 65 percent range from 2.03 to 4.5 mils/kWh
and $1,180 to $2,600/ton NOX removed (Figures 3-12 and 3-13). The levelized costs at a
capacity factor of 27 percent range from 4.5 to 10.4 mils/kWh and $2,700 to $6,100/ton NOX re-
moved (Figures 3-14 and 3-15).
3.2.2 SNCR Evaluation
The following major criteria and assumptions have been followed in evaluating the SNCR tech-
nology for the wall-fired boilers:
• The SNCR system is designed to provide a 50 percent NOX reduction from a baseline NOX
rate of 0.50 Ib/MMBtu.
• All of the other criteria and assumptions described in Section 3.1.2 also apply equally to this
case.
The impacts of the SNCR technology retrofit on the study boiler are as follows (refer to Table
3-1):
• The boiler efficiency reduces from 88.39 to 87.96 percent. The boiler heat input increases
from 3,600 to 3,618 MMBru/hr. The fuel flow, ash generation rate, and combustion and flue
gas flow rates increase in a direct proportion to the change in the heat input.
• There is an overall increase in the plant auxiliary power consumption due to the SNCR
equipment as well as the increased demand on the draft fans to accommodate the higher air
and flue gas flow rates. The estimated auxiliary power increase is 193 kW.
• The urea consumption requirement for the SNCR system is 433 gal./hr.
• The water consumption requirement for the SNCR system is 5,570 gal./hr.
Using the above parameters and the costing methodology described in Appendix A, both the
capital and levelized costs have been calculated for the entire size range (30 to 1,300 MW) of
wall-fired boilers. As shown in Figure 3-16, the capital costs range from approximately $6.5 to
$52/kW. The levelized costs at a capacity factor of 65 percent range from 1.18 to 2.32 mils/kWh
and $980 to $l,920/ton NOX removed (Figures 3-17 and 3-18). The levelized costs at a capacity
22885 OOS\Stu6y\Cosl-Est.NOx
3-4
-------
factor of 27 percent range from 1.39 to 4.1 mils/kWh and $1,180 to $3,400/ton NOX removed
(Figures 3-19 and 3-20).
3.3 Cell-Burner Boiler Applications
Only the SCR technology was evaluated for NOX control on this boiler type. The design data for
the representative boilers of 300 and 600 MW sizes selected for this evaluation are shown in
Figures B3-3 and B3-4 of Appendix A.
The following major criteria and assumptions have been followed in evaluating the SCR tech-
nology for the wall-fired boilers:
• The SCR system is designed to reduce NOX emission from a baseline level of 1.0 Ib/MMBtu
to the required limit of 0.15 Ib/MMBtu.
• Similar to the methodology used in Appendix A, the evaluations are based on two represen-
tative boilers.
• All of the other criteria and assumptions described in Section 3.1.1 apply equally to this case.
The consumables associated with the SCR system retrofit for the study boilers are as follows:
300 MW 600 MW
Auxiliary power consumption, kW 716 1,431
Anhydrous ammonia consumption, Ib/hr 843 1,641
Average catalyst replacement, ft3/yr 4,556 8,773
Using the above parameters and the costing methodology described in Appendix A, both the
capital and levelized costs have been calculated for the entire size range (200 to 1,300 MW) of
cell-burner boilers. As shown in Figure 3-21, the capital costs range from approximately $38.5
to $69/kW. The levelized costs at a capacity factor of 65 percent range from 2.13 to 3.8
mils/kWh and $610 to $800/ton NOX removed (Figures 3-22 and 3-23). The levelized costs at a
capacity factor of 27 percent range from 4.6 to 6.8 mils/kWh and $1,305 to $l,780/ton NOX re-
moved (Figures 3-24 and 3-25).
3.4 Cyclone-Fired Boiler Applications
Only the SCR technology was evaluated for NOX control on this boiler type. The design data for
the representative boilers of 150 and 400 MW sizes selected for this evaluation are shown in
Figures B4-3 and B4-4 of Appendix A.
The following major criteria and assumptions have been followed in evaluating the SCR tech-
nology for the wall-fired boilers:
2:S85.00S\Snjdy\Cosi-Esi NO*
3-5
-------
• The SCR system is designed to reduce NOX emission from a baseline level of 1.17 Ib/MMBtu
to the required limit of 0.15 Ib/MMBtu.
• Similar to the methodology used in Appendix A, the evaluations are based on two represen-
tative boilers.
• All of the other criteria and assumptions described in Section 3.1.1 apply equally to this case.
The consumables associated with the SCR system retrofit for the study boilers are as follows:
150MW 400 MW
Auxiliary power consumption, kW 250 954
Anhydrous ammonia consumption, Ib/hr 490 1,380
Average catalyst replacement, ft3/yr 2,320 6,400
Using the above parameters and the costing methodology described in Appendix A, both the
capital and levelized costs have been calculated for the entire size range (25 to 1,200 MW) of
cyclone boilers. As shown in Figure 3-26, the capital costs range from approximately $44 to
$120/kW. The levelized costs at a capacity factor of 65 percent range from 2.8 to 4.3 mils/kWh
and $525 to $990/ton NOX removed (Figures 3-27 and 3-28). The levelized costs at a capacity
factor of 27 percent range from 5.9 to 9.8 mils/kWh and $1,080 to $2,270/ton NOX removed
(Figures 3-29 and 3-30).
3.5 Wet-Bottom Boiler Applications
Only the SCR technology was evaluated for NOX control on this boiler type. The design data for
the representative boilers of 100 and 259 MW sizes selected for this evaluation are shown in
Figures B5-3 and B5-4 of Appendix A.
The following major criteria and assumptions have been followed hi evaluating the SCR tech-
nology for the wall-fired boilers:
• The SCR system is designed to reduce NOX emission from a baseline level of 1.13 Ib/MMBtu
to the required limit of 0.15 Ib/MMBtu.
• Similar to the methodology used in Appendix A, the evaluations are based on two represen-
tative boilers.
• All of the other criteria and assumptions described in Section 3.1.1 apply equally to this case.
22885 008\Study\Co5t-Est.NCK
5-6
-------
The consumables associated with the SCR system retrofit for the study boilers are as follows:
100MW 259 MW
Auxiliary power consumption, kW 240 620
Anhydrous ammonia consumption, Ib/hr 320 840
Average catalyst replacement, ft3/yr 1,570 4,080
Using the above parameters and the costing methodology described in Appendix A, both the
capital and levelized costs have been calculated for the entire size range (25 to 800 MW) of wet-
bottom boilers. As shown in Figure 3-31, the capital costs range from approximately $46 to
$130/kW. The levelized costs at a capacity factor of 65 percent range from 2.65 to 4.65
mils/kWh and $560 to $l,100/ton NOX removed (Figures 3-32 and 3-33). The levelized costs at
a capacity factor of 27 percent range from 5.7 to 10.6 mils/kWh and $1,200 to $2,500/ton NOX
removed (Figures 3-34 and 3-35).
3.6 Vertically Fired, Dry-Bottom Boiler Applications
Only the SCR technology was evaluated for NOX control on this boiler type. The design data for
the representative boilers of 110 and 220 MW sizes selected for this evaluation are shown in
Figures B6-3 and B6-4 of Appendix A.
The following major criteria and assumptions have been followed in evaluating the SCR tech-
nology for the wall-fired boilers:
• The SCR system is designed to reduce NOX emission from a baseline level of 1.08 Ib/MMBtu
to the required limit of 0.15 Ib/MMBtu.
• Similar to the methodology used in Appendix A, the evaluations are based on two represen-
tative boilers.
• All of the other criteria and assumptions described in Section 3.1.1 apply equally to this case.
The consumables associated with the SCR system retrofit for the study boilers are as follows:
110MW 220 MW
Auxiliary power consumption, kW 260 525
Anhydrous ammonia consumption, Ib/hr 350 640
Average catalyst replacement, ft3/yr 1,750 3,200
Using the above parameters and the costing methodology described in Appendix A, both the
capital and levelized costs have been calculated for the entire size range (25 to 300 MW) of ver-
2288S.008\Study\Cost-Ejl NOx
3-7
-------
tically fired boilers. As shown in Figure 3-36, the capital costs range from approximately $57 to
$151/kW. The levelized costs at a capacity factor of 65 percent range from 2.65 to 5.25
mils/kWh and $720 to $l,170/ton NOX removed (Figures 3-37 and 3-38). The levelized costs at
a capacity factor of 27 percent range from 5.85 to 12.0 mils/kWh and $1,590 to $2,650/ton NOX
removed (Figures 3-39 and 3-40).
22885 OOS\Study\Cost-Es[NOx
3-8
-------
TABLE 3-1
*-
ORIGINAL DESIGN DATA
TANGENTIAL AND WALL-BURNER COAL-FIRED BOILERS
Parameter1"
Boiler size, MW
Boiler load, % MCR
Boiler type
Heat input, MMBtu/hr
Fuel consumption, ton/hr
Solid waste, ton/hr
Boiler efficiency
Fuel analysis (wt. %):
Ash
Moisture
Sulfur
HHV, Btu/lb
Tangential Boiler1"
348
100
Reheat
3,210
127
9.82
88.39
7.7
8.4
0.8
12,696
Wall-Fired Boilerw
381
100
Reheat
3,600
142
10.98
88.39
7.7
8.4
0.8
12,696
NOTES
1. Only data pertinent to the NOX control technologies are shown.
2. The same coal is fired hi both boilers. It is assumed that efficiency is the same for both boiler
types. In practice, there may be a small difference in the efficiencies; however, the difference
would be insignificant as long as the operating parameters, such as excess air levels, are the
same.
228SS.008\Study\G»l-Eu.NOx
-------
4.0 NATURAL GAS-FIRED PLANT ASSUMPTIONS AND
RESULTS
Both the tangential and wall-fired boilers firing natural gas have been considered in this evalua-
tion. The NOX control technologies evaluated for these boiler types include SCR, gas rebuming,
and SNCR. The design data for the representative boilers selected for this evaluation are shown
in Table 4-1. It is to be noted that the same design data apply to both the tangential and wall-
fired boilers.
The tangential boiler is a balanced draft, forced circulation, reheat, single furnace boiler. It has
four windboxes located along the four corners of the furnace. There are a total of 16 burners,
four per corner. The boiler serves a 350 MW steam turbine generator and is equipped with two
50-percent-capacity forced draft fans and two 50-percent-capacity induced draft fans.
The wall-fired boiler is a balanced draft, natural circulation, reheat, single furnace boiler. It is a
front wall-fired boiler with 20 burners arranged four high and five wide. The boiler serves a
350 MW steam turbine generator and is equipped with two 50-percent-capacity forced draft fans
and two 50-percent-capacity induced draft fans.
4.1 SCR Evaluation
The following major criteria and assumptions have been followed in evaluating the SCR tech-
nology for the tangential and wall-fired boilers:
• The SCR system is designed to reduce NOX emission from a baseline level of 0.25 Ib/MMBtu
to the required limit of 0.15 Ib/MMBtu.
• Since the flue gas flow conditions at the economizer outlet are the same for both the tangen-
tial and wall-fired boilers, the SCR system design would be extremely similar for these
boilers, which permits a joint presentation of the cost data for these boilers.
• Similar to the coal-fired tangential boiler case, the evaluation is based on one representative
boiler for each boiler type.
• A catalyst operating life of 5 years is assumed.
• All of the other criteria and assumptions described in Section 3.1.1 apply equally to this case.
The consumables associated with the SCR system retrofit for the study boilers are as follows:
Auxiliary power consumption 420 kW
Anhydrous ammonia consumption 130 Ib/hr
Average catalyst replacement 425 ft /yr
22885 008\Study\Cost-EstNOx
4-1
-------
Using the above parameters and the costing methodology described in Appendix A, both the
capital and levelized costs have been calculated for a boiler size range of 30 to 1,300 MW. As
shown in Figure 4-1, the capital costs range from approximately $14 to $54/kW. The levelized
costs at a capacity factor of 65 percent range from 0.55 to 1.5 mils/kWh and $1,350 to $3,750/ton
NOX removed (Figures 4-2 and 4-3). The levelized costs at a capacity factor of 27 percent range
from 1.15 to 3.44 mils/kWh and $2,900 to $8,600/ton NOX removed (Figures 4-4 and 4-5).
4.2 Gas Reburning Evaluation
The following major criteria and assumptions have been followed in evaluating the gas reburning
technology for the gas-fired boilers:
• The gas reburn system is designed to reduce the baseline NOX of 0.25 Ib/MMBtu to the re-
quired limit of 0.15 Ib/MMBtu.
• It is assumed that natural gas supply is available at the plant fence for both boilers.
• The rebum system design is based on a 25 percent heat input for the reburn injectors. Natural
gas is injected into the furnace along with gas recirculation (system designed for a 10 percent
recirculation rate). It is assumed that existing gas recirculation fans will be used for this pur-
pose. The overfire air system is designed for 20 percent of the full-load combustion air re-
quirement for the boiler.
• It is assumed that sufficient space is available in the boilers to add the reburn injectors and
overfire air ports. It is also assumed that the available space allows for an adequate residence
time for completing the combustion process for the rebum fuel. Lack of an adequate resi-
dence time may reduce the effectiveness of the gas rebum system or it may adversely affect
the feasibility of installing such a system.
• In some cases, capital cost of the rebum technology application may be lower for a tangential
boiler than for a wall-fired boiler. Because of the comer firing arrangement for the tangential
boiler, a potential may exist for effectively utilizing a smaller number of reburn injectors.
However, any cost difference is not expected to be significant. Therefore, for conservatism,
the same capital costs developed for the wall-fired boiler have been used for the tangential
boiler.
• Other general gas rebum system design details, assumptions, and impacts on the existing
equipment outlined in Appendix A (Section 4.3 of Appendix B) also apply to this case.
Reburn technology has a minimal impact on the performance of a gas-fired boiler. This applica-
tion involves withdrawal of a portion of the boiler fuel from the main combustion zone and in-
jection of this fuel above the top-most burners. Overfire air is injected further up in the furnace
to complete combustion of the reburn fuel. As long as the conditions permit proper combustion
of the rebum fuel, the boiler performance would not be affected. Operation of the reburn system
does result in an increased auxiliary power consumption (associated with the operation of the gas
recirculation fan). In the case of the study boilers, this increase is estimated at 176 MW.
22885.008\Sni
-------
Using the above parameters and the costing methodology described in Appendix A, both the
capital and levelized costs have been calculated for a boiler size range of 30 to 1.300 MW. As
shown in Figure 4-6, the capital costs range from approximately $10 to $37/kW. The levelized
costs at a capacity factor of 65 percent range from 0.28 to 0.95 mils/kWh and $700 to $2,400/ton
NOX removed (Figures 4-7 and 4-8). The levelized costs at a capacity factor of 27 percent range
from 1.32 to 3.78 mils/kWh and $1,330 to $3,800/ton NOX removed (Figures 4-9 and 4-10).
4.3 SNCR Evaluation
The following major criteria and assumptions have been followed in evaluating the SNCR tech-
nology for the gas-fired boilers:
• The SNCR system is designed to reduce the baseline NOX of 0.25 Ib/MMBtu to the required
limitof0.151b/MMBtu.
• A reagent ratio of 1.5 commensurate with the NOX reduction requirement is used.
;
• All of the other criteria and assumptions described in Section 3.1.2 also apply equally to this
case.
The impacts of the SNCR technology retrofit on the study boilers are as follows (refer to Table
4-1):
• The boiler efficiency reduces from 85.65 to 85.48 percent. The boiler heat input increases
from 2,980 to 2,986 MMBtu/hr. The fuel flow, ash generation rate, and combustion and flue
gas flow rates increase in a direct proportion to the change in the heat input.
• There is an overall increase in the plant auxiliary power consumption due to the SNCR
equipment as well as the increased demand on the draft fans to accommodate the higher air
and flue gas flow rates. The estimated auxiliary power increase is 80 kW.
• The urea consumption requirement for the SNCR system is 155 gal./hr.
• The water consumption requirement for the SNCR system is 1,980 gal./hr.
Using the above parameters and the costing methodology described in Appendix A, both the
capital and levelized costs have been calculated for a boiler size range of 30 to 1,300 MW. As
shown in Figure 4-11, the capital costs range from approximately $3.2 to $28/kW. The levelized
costs at a capacity factor of 65 percent range from 0.5 to 1.1 mils/kWh and $1,220 to $2,800/ton
NOX removed (Figures 4-12 and 4-13). The levelized costs at a capacity factor of 27 percent
range from 0.6 to 2.1 mils/kWh and $1,520 to $5,200/ton NOX removed (Figures 4-14 and 4-15).
228S5.008\Study\Cosl-Est.NOx
4-3
-------
TABLE 4-1
ORIGINAL DESIGN DATA
TANGENTIAL AND WALL-BURNER TYPE
GAS- AND OIL-FIRED BOILERS
Parameter*1'
Boiler size, MW
Boiler load, % MCR
Boiler type
Heat input, MMBtu/hr
Fuel consumption, ton/hr
Solid waste, Ib/hr
Boiler efficiency
Fuel analysis (wt. %):
Ash
Moisture
Sulfur
HHV,Btu/lb
CH4
C3H8
C2H6
HHV, Btu/ft3
Gas-Fired Boilers1"'
350
100
Reheat
2,980
64.1
0
85.65
Natural Gas
85.45
2.45
6.61
1,075
Oil-Fired Boilers'"
350
100
Reheat
2,895
79.5
303
88.15
No. 6 Oil
0.1
0.1
1.0
18,200
NOTES
1. Only data pertinent to the NOX control technologies are shown.
2. For each fuel, the same design data apply to both the tangential and wall-fired boilers. It is
assumed that efficiency is the same for both boiler types. In practice, there may be a small
difference in the efficiencies; however, the difference would be insignificant as long as the
operating parameters, such as excess air levels, are the same.
22885 OOS\Srudy\Cost-EM NO*
-------
5.0 OIL-FIRED PLANT ASSUMPTIONS AND RESULTS
Both the tangential and wall-fired boilers firing No. 6 oil have been considered in this evaluation.
The NOX control technologies evaluated for these boiler types include SCR, gas rebuming, and
SNCR. The design data for the representative boilers selected for this evaluation are shown in
Table 4-1. It is to be noted that the same design data apply to both the tangential and wall-fired
boilers.
The tangential boiler is a balanced draft, forced circulation, reheat, single furnace boiler. It has
four windboxes located along the four comers of the furnace. There are a total of 16 burners,
four per comer. The boiler serves a 350 MW steam turbine generator and is equipped with two
50-percent-capacity forced draft fans, two 50-percent-capacity induced draft fans, and an elec-
trostatic precipitator for removing dust from the flue gases exiting the boiler.
The wall-fired boiler is a balanced draft, natural circulation, reheat, single furnace boiler. It is a
front wall-fired boiler with 20 burners arranged four high and five wide. The boiler serves a
350 MW steam turbine generator and is equipped with two 50-percent-capacity forced draft fans,
two 50-percent-capacity induced draft fans, and an electrostatic precipitator for removing dust
from the flue gases exiting the boiler.
5.1 SCR Evaluation
The following major criteria and assumptions have been followed in evaluating the SCR tech-
nology for the tangential and wall-fired boilers:
• The SCR system is designed to reduce NOX emission from a baseline level of 0.30 Ib/MMBtu
to the required limit of 0.15 Ib/MMBtu.
• Since the flue gas flow conditions at the economizer outlet are the same for both the tangen-
tial and wall-fired boilers, the SCR system design would be extremely similar for these
boilers, which permits a joint presentation of the cost data for these boiler.
• Similar to the coal-fired tangential boiler case, the evaluation is based on one representative
boiler for each boiler type.
• All of the other criteria and assumptions described in Section 3.1.1 apply equally to this case.
The consumables associated with the SCR system retrofit for the study boilers are as follows:
Auxiliary power consumption 500 kW
Anhydrous ammonia consumption 170 Ib/hr
Average catalyst replacement 1,370 ft3/yr
22885 008\Studv\Co«-EstNOx
5-1
-------
Using the above parameters and the costing methodology described in Appendix A, both the
capital and levelized costs have been calculated for a boiler size range of 30 to 1,300 MW. As
shown in Figure 5-1, the capital costs range from approximately $21 to $77/kW. The levelized
costs at a capacity factor of 65 percent range from 0.87 to 2.27 mils/kWh and $1,500 to
$3,800/ton NOX removed (Figures 5-2 and 5-3). The levelized costs at a capacity factor of
27 percent range from 1.95 to 5.3 mils/kWh and $3,200 to $8,800/ton NOX removed (Figures 5-4
and 5-5).
5.2 Gas Reburning Evaluation
The following major criteria and assumptions have been followed in evaluating the gas reburning
technology for the oil-fired boilers:
• The gas rebum system is designed to reduce the baseline NOX of 0.3 Ib/MMBtu to the re-
quired limit of 0.15 Ib/MMBtu.
• Other criteria and assumptions outlined in Section 4.2 also apply to this case.
The performance impacts of the rebum technology on the oil-fired boilers are as follows:
• The boiler performance changes, because with the rebum system 20 percent of the heat input
is by natural gas and 80 percent is by oil. The boiler efficiency reduces from 88.15 to 87.38
percent. The levelized cost estimates must take into account the cost increases incurred in
firing natural gas rather than No. 6 oil.
• Firing of natural gas reduces the amount of ash generation by 58 Ib/hr and SO2 emission rate
by 620 Ib/hr. Both of these reductions benefit the operating costs.
Using the above parameters and the costing methodology described in Appendix A, both the
capital and levelized costs have been calculated for a boiler size range of 30 to 1,300 MW. As
shown in Figure 5-6, the capital costs range from approximately $12 to $44/kW. The levelized
costs at a capacity factor of 65 percent range from 0.8 to 1.6 mils/kWh and $1,350 to $2,650/ton
NOX removed (Figures 5-7 and 5-8). The levelized costs at a capacity factor of 27 percent range
from 1.2 to 3.1 mils/kWh and $2,000 to $5,200/ton NOX removed (Figures 5-9 and 5-10).
5.3 SNCR Evaluation
The following major criteria and assumptions have been followed in evaluating the SNCR tech-
nology for the oil-fired boilers:
• The SNCR system is designed to reduce the baseline NOX of 0.3 Ib/MMBtu to the required
limit of 0.15 Ib/MMBtu.
• All of the other criteria and assumptions described in Section 3.1.2 also apply equally to this
case.
22885.008\Study\Cost-Est.NOx
5-2
-------
The impacts of the SNCR technology retrofit to the study boilers are as follows (refer to Table
4-1):
• The boiler efficiency reduces from 88.15 to 87.88 percent. The boiler heat input increases
from 2,895 to 2,904 MMBtu/hr. The fuel flow, ash generation rate, and combustion and flue
gas flow rates increase in direct proportion to the change in the heat input.
• There is an overall increase in the plant auxiliary power consumption due to the SNCR
equipment as well as the increased demand on the draft fans to accommodate the higher air
and flue gas flow rates. The estimated auxiliary power increase is 115 kW.
• The urea consumption requirement for the SNCR system is 210 gal./hr.
• The water consumption requirement for the SNCR system is 2,690 gal./hr.
Using the above parameters and the costing methodology described in Appendix A, both the
capital and levelized costs have been calculated for a boiler size range of 30 to 1,300 MW. As
shown in Figure 5-11, the capital costs range from approximately $4.0 to $32/kW. The levelized
costs at a capacity factor of 65 percent range from 0.65 to 1.35 mils/kWh and $1,100 to
$2,300/ton NOX removed (Figures 5-12 and 5-13). The levelized costs at a capacity factor of
27 percent range from 0.8 to 2.46 mils/kWh and $1,350 to $4,100/ton NOX removed (Figures 5-
14 and 5-15).
22885.008\Studv\Cost.Est.NOx
5-3
-------
6.0 REFERENCES
1. U.S. EPA, "Alternative Control Techniques Document: NOX Emissions from Utility
Boilers," EPA-453/R-94-023, March 1994.
2. "Demonstration of Coal Rebuming for Cyclone Boiler NOX Control," Final Project Report,
prepared by Babcock & Wilcox for U.S. DOE/PETC, DOE/PC/89659-T16, February 1994.
3. R. Borio, et al., "Rebum Technology for NOX Control on a Cyclone-Fired Boiler," ABB
Combustion Engineering Services, Inc.
4. "White Paper for SNCR for Controlling NOX Emissions," Prepared by SNCR Committee
of 1C AC, July 1994.
5. L. Muzio, et al., "State-of-the-Art Assessment of SNCR Technology," Prepared for EPRI
by Fossil Energy Research Corp., Task 1 Report, RP2869-14, April 1993.
6. D. Hubbard, et al., "Long Term SNCR Demonstration at B. L. England Station - Unit 1,"
Prepared for Atlantic Electric by Carnot, August 1994.
7. F. Gibbons, "A Demonstration of Urea-Based SNCR NOX Control on a Utility Pulverized-
Coal Wet-bottom Boiler," EPRI Workshop onNOx Control for Utility Boilers, May 1994.
8. B. Folsom, et al., "Demonstration of Gas Rebuming-Sorbent Injection on a Cyclone-Fired
Boiler," Third Annual Clean Coal Technology Conference, September 1994.
9. C. P. Robie, et al., "Technical Feasibility and Cost of Selective Catalytic Reduction (SCR)
NOX Control," EPRI GS-7266, May 1991.
10. C. Pattersson, et al., "Alternative Solutions for Reducing NOX Emissions from Cell Burner
Boilers," EPRI/EPA Joint Symposium, May 1995.
11. K. Dresner, et al., "Low-NOx Combustion System Retrofit for a 630 MWe PC-Fired Cell
Burner Unit," EPRI/EPA Joint Symposium, May 1995.
12. T. J. May, et al., "Gas Rebuming hi Tangentially, Wall, and Cyclone-Fired Boilers," Third
Annual Clean Coal Technology Conference, September 1994.
13. B. Owens, et al., "SCR Retrofit for NOX Control at a Wet Bottom Boiler," EPRI/EPA Joint
Symposium, May 1995.
14. "Evaluation of NOX Control Removal Technologies, Volume 1, Selective Catalytic
Reduction," DOE Report, DE-AC22-94PC92100, Rev. 2, September 1994.
15. S. Khan, "NOX Reduction Technologies for Fossil Power Plants, Lessons Learned," Power-
Gen Europe'93 Conference, April 1993.
16. T. A. Laursen, et al., "Results of the Low NOX Cell Burner Demonstration at Dayton Power
& Light Company's J. M. Stuart Station," 1993 EPRI/EPA Joint Symposium.
17. "Phase II NOX Controls for Nescaum and Marama Region," ACUREX Final Report No.
95-102, May 10,1995.
22885 008\Study\CoM-EstNOx
6-1
-------
18. "Technical Assessment Guide," EPR1, Volume 1, Revision 7,1993.
19. "Energy Analysis: 1995-01," American Gas Association, January 13, 1995.
22885.008\Snj
-------
Figures 3-1 through 3-40
-------
$140
$130
$120
$110
$100
Figure 3-1
Coal-Fired, Tangential-Fired Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Capital Costs v. MW
$60
$50
$40
$30
130 230 330 430 530 630 730 830
Unit Size, MW
930
1030 1130 1230 1330
-------
Figure 3-2
Coal-Fired, Tangential-Fired Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levellzed Costs, mlls/kWh v. MW - 65% Capacity Factor Case
~ 2.80
2.50
2.20
1.90
1.60
1.30
1.00
30
130
230
330
430
530
630 730
Unit Size, MW
830
930
1030
1130
1230
1330
-------
Figure 3-3
Coal-Fired, Tangential-Fired Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levellzed Coats, $/ton NOx removed v. MW - 65% Capacity Factor Case
$3600
$3200
$2800
$2400
$2000
$1600
$1200
$800
$400
130 230 330 430
630 730 830 930 1030 1130 1230 1330
Unit Size, MW
-------
Figure 3-4
Coal-Fired, Tangential-Fired Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levellzed Costs, mlls/kWh v. MW - 27% Capacity Factor Case
30
230
330
430
530
630 730
Unit Size, MW
830
930
1030
1130
1230
1330
-------
x
O
i
$8000
$7600
$7200
$6800
$6400
$6000
$5600
$5200
$4800
$4400
$4000
$3600
$3200
$2800
$2400
30
Figure 3-5
Coal-Fired, Tangential-Fired Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levellzed Costs, Srton NOx removed v. MW - 27% Capacity Factor Case
130
230
330
430
530
630 730
Unit Size, MW
830
930
1030
1130
1230
1330
-------
Figure 3-6
Coal-Fired, Tangential Boiler, SNCR Retrofit
Base Capital Costs v. MW
$50
$30
$20
i
$o
230
330
430
530
630 730
Unit Size, MW
830
930
1030
1130
1230
1330
-------
2.40
Figure 3-7
Coal-Fired, Tangential Boiler, SNCR Retrofit
Base Levellzed Costs, mlls/kWh v. MW - 65% Capacity Factor Case
2.20
2.00
1.80
1.60
E 1.40
1.20
1.00
0.80
0.60
30 130 230 330 430
530 630 730 830
Unit Size, MW
930 1030 1130 1230 1330
-------
Figure 3-8
Coal-Fired, Tangential Boiler, SNCR Retrofit
Base Levellzed Costs, $rton NOx removed v. MW - 65% Capacity Factor Case
$2600
$2400
$2200
$2000 •
$1800
3
$1600
$1400
$1200
$1000
$800
$600
$400
\
\
\
\
X.
^».
— ' .
,
-*
-
.
^^^•""•^^^^
•••^
^^KMH^H
30 130 230 330 430 530 630 730 830 930 1030 1130 1230 13
Unit Size, MW
-------
Figure 3-9
Coal-Fired, Tangential Boiler, SNCR Retrofit
Base Levellzed Costs, mlls/kWh v. MW - 27% Capacity Factor Case
4.00
3.80
3.60
3.40
1.00
130
230
330
430
530
630 730
Unit Size, MW
830
930
1030
1130
1230
1330
-------
Figure 3-10
Coal-Fired, Tangential Boiler, SNCR Retrofit
Base Levellzed Costs, $/ton NOx removed v. MW -27% Capacity Factor Case
5
130
230
330
430
530
630 730
Unit Size, MW
830
930
1030
1130
1230
1330
-------
$140
Figure 3-11
Coal-Fired, Wall-Burner Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Capital Costs v. MW
$20
130 230 330 430 530 630 730 830
Unit Size, MW
930
1030 1130
1230 1330
-------
Figure 3-12
Coal-Fired, Wall-Burner Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levelized Costs, mils/kWh v. MW - 65% Capacity Factor Case
30
230
330
430
530
630 730
Unit Size, MW
830
930
1030
1130
1230
1330
-------
$2800
$2600
$2400
$2200
* $2000
$1800
$1600
$1400
$1200
$1000
Figure 3-13
Coal-Fired, Wall-Burner Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levellzed Costs, I/ton NOx removed v. MW - 65% Capacity Factor Case
30 130 230 330 430
530 630 730 830
Unit Size, MW
930 1030 1130 1230 1330
-------
Figure 3-14
Coal-Fired, Wall-Burner Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levelized Costs, mils/kWh v. MW - 27% Capacity Factor Case
m «in 1
m no '
Q fifl
Q on
Q on .
ft 4ft
ft nn
2 7 fin
1
5 7.20
E
6.80
6.40
6.00
5.60
5.20
4.80
4.40
4.00
3
\
\
\
\
\
\
k
\
>
X^
\
\,
^-~
•^-^
,
~~~— -.
— — .
.
— —
_
.
_
•y.
0 130 230 330 430 530 630 730 830 930 1030 1130 1230 1330
Unit Size, MW
-------
o
$6400
$6000
$5600
$5200
$4800
$4400
$4000
$3600
$3200
$2800
$2400
30
\
\
Figure 3-15
Coal-Fired, Wall-Burner Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levellzed Costs, $/ton NOx removed v. MW - 27% Capacity Factor Case
130
230
330
430
530
630 730
Unit Size, MW
830
930
1030
1130 1230
1330
-------
Figure 3-16
Coal-Fired, Wall-Burner Boiler, SNCR Retrofit
Base Capital Costs v. MW
$40
$30
I
130 230 330 430 530 630 730 830
Unit Size, MW
930 1030 1130 1230 1330
-------
2.80
2.60
2.40
2.20
2.00
1.80
1.60
1.40
1.20
1.00
0.80
Figure 3-17
Coal-Fired, Wall-Burner Boiler, SNCR Retrofit
Base Levellzed Costs, mlls/kWh v. MW - 65% Capacity Factor Case
230 330 430 530
630 730
Unit Size, MW
830 930 1030 1130 1230 1330
-------
Figure 3-18
Coal-Fired, Wall-Burner Boiler, SNCR Retrofit
Base Levellzed Costs, $fton NOx removed v. MW - 65% Capacity Factor Case
$2600
$2400
S77OO
tonOO
$1800 |
X
^ $1600
1
$1400
$1200
$1000
$800
$600
$400
3
\
\
N
X^
---^
—
— _
'
— — — _
— — — — .
•— ^^^^^
••••^•^MMass
^ ^— -~-^^
J
0 130 230 330 430 530 630 730 830 930 1030 1130 1230 1330
Unit Size, MW
-------
Figure 3-19
Coal-Fired, Wall-Burner Boiler, SNCR Retrofit
Base Levellzed Costs, mlls/kWh v. MW - 27% Capacity Factor Case
4.60
4.20
3.80
3.40
3.00
E 2.60
2.20
1.80
1.40
1.00
30
130 230 330 430
530 630 730 830
Unit Size, MW
930 1030 1130 1230 1330
-------
$4000
$3800
$3600
$3400
$3200
$3000
$2800
$2600
$2400
$2200
$2000
$1800
$1600
$1400
$1200
$1000
$800
Figure 3-20
Coal-Fired, Wall-Burner Boiler, SNCR Retrofit
Base Levellzed Costs, $/ton NOx removed v. MW - 27% Capacity Factor Case
\
30
130
230
330
430
530
630 730
Unit Size, MW
830
930
1030
1130
1230
1330
-------
Figure 3-21
Coal-Fired, Cell Burner Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Capital Costs v. MW
$80
$75
$70
$65
$60
$55
$50
$45
$40
$35
$30
$25
$20
\
200
300
400
500
600
700 800
Unit Size, MW
900
1000
1100
1200
1300
-------
Figure 3-22
Coal-Fired, Cell Burner Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levellzed Costs, mlls/kWh v. MW - 65% Capacity Factor Case
-------
Figure 3-23
Coal-Fired, Cell Burner Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levellzed Costs, S/ton NOx removed v. MW - 65% Capacity Factor Case
x
O
$850
$800
$750
$700
$650
$600
$550
$500
200
300
400
500
600
700 800
Unit Size, MW
900
1000
1100
1200
1300
-------
Figure 3-24
Coal-Fired, Cell Burner Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levellzed Costs, mlls/kWh v. MW - 27% Capacity Factor Case
720 •
7 00
680
6 60
6 40
A 2O
6 00
5 SRn
1
M *i fin
E
5 40
5 20
e nn
A on
A en
4.40
4 20
4.00
2(
\
\
N
"V
\
^N
^X_
"^,
•^^.^
^-^
•^^
~—
^-_
^^-J
' '
^— — 1
)0 300 400 500 600 700 800 900 1000 1100 1200 1300
Unit Size, MW
-------
X
O
Figure 3-25
Coal-Fired, Cell Burner Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levellzed Costs, $/ton NOx removed v. MW - 27% Capacity Factor Case
$2000
$1900
$1800
$1700
$1600
$1500
$1400
$1300
$1200
$1100
$1000
200
300
400
500
600
700 800
Unit Size, MW
900
1000
1100
1200
1300
-------
$125
$120
$115
$110
$105
$100
$95
$90
$85
$80
$75
$70
$65
$60
$55
$50
$45
$40
$35
$30
$25
Figure 3-26
Cyclone-Fired Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Capital Costs v. MW
\
25 125 225 325 425 525 625 725
Unit Size, MW
825
925
1025
1125
1225
-------
Figure 3-27
Cyclone-Fired Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levellzed Costs, mils/kWh v. MW - 65% Capacity Factor Case
440 •
4 30
A ?n -
4 10 1 1
A nn
3 90
Q on
3 70
•v^
•^-^
— — — _
•
^ -*
5 125 225 325 425 525 625 725 825 925 1025 1125 1225
Unit Size, MW
-------
Figure 3-28
Cyclone-Fired Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levellzed Costs, Sfton NOx removed v. MW - 65% Capacity Factor Case
$1100
$1050
$1000
$950
$900
$850
o
$450
$400
25
225
325
425
525 625 725
Unit Size, MW
825
925
1025
1125
1225
-------
Figure 3-29
Cyclone-Fired Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levellzed Costs, mlls/kWh v. MW - 27% Capacity Factor Case
in m
Q 7n
030 I
o an
8 50 •
8 10
5 7 7n
JC
•a
= 7.30
6.90
6.50
6.10
5.70
5.30
4.90
4.50
2
\
\
\
\
rv
^
^
^-~-
• — —— .
==
•
—
^— — ^-^— .
5 125 225 325 425 525 625 725 825 925 1025 1125 1225
Unit Size, MW
-------
$2800
$2600
$2400
$2200
* $2000
$1800
$1600
$1400
$1200
$1000
Figure 3-30
Cyclone-Fired Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levelized Costs, $/ton NOx removed v. MW - 27% Capacity Factor Case
125
225
325
425
525 625 725
Unit Size, MW
825
925
1025 1125
1225
-------
$145
Figure 3-31
Wet Bottom Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Capital Costs v. MW
$105
$95
$85
$75
$65
$55
$45
$35
\
25 75 125
175 225 275 325 375 425 475 525 575 625 675 725 775 825
Unit Size, MW
-------
Figure 3-32
Wet Bottom Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levellzed Costs, mils/kWh v. MW - 65% Capacity Factor Case
5.10
4.90
4.70
4.50
4.30
4.10
3.90
2.30
75
125
175
225
275 325
375 425 475 525 575
Unit Size, MW
625 675
725 775
825
-------
X
O
$1250
$1150
$1050
$950
$850
$750
$650
$550
$450
Figure 3-33
Wet Bottom Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levellzed Costs, $/ton NOx removed v. MW - 65% Capacity Factor Case
25 75 125 175 225 275 325 375 425 475 525 575 625 675 725 775 825
Unit Size, MW
-------
Figure 3-34
Wet Bottom Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levelized Costs, mlls/kWh v. MW - 27% Capacity Factor Case
11.00
10.60
10.20
9.80
9.40
9.00
8.60
£
3 8.20
E 7.80
7.40
7.00
6.60
6.20
5.80
5.40
5.00
\
25 75 125 175 225 275 325 375 425 475
Unit Size, MW
525
575
625
675
725
775
825
-------
Figure 3-35
Wet Bottom Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levellzed Costs, Srton NOx removed v. MW - 27% Capacity Factor Case
&?4nn -
tO7nn
tonoo
X
O
e $1800
1
$1600
$1400
$1200
$1000
2
\
\
\
\
X
\^
\
^-~~
-—».
-- s^_
•^
•— —
— —
— — .
,
— — _
-«
••
5 75 125 175 225 275 325 375 425 475 525 575" 625 675 725 775 825
Unit Size, MW
-------
$165
$155
$145
$135
$125
$115
$105
$95
$85
$75
$65
$55
$45
$35
Figure 3-36
Vertically-Fired Dry Bottom Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Capital Costs v. MW
25 50 75 100 125 150 175 200
Unit Size, MW
225
250
275
300
325
-------
Figure 3-37
Vertically-Fired Dry Bottom Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levellzed Costs, mlls/kWh v. MW - 65% Capacity Factor Case
5.60
5.40
5.20
5.00
4.60
4.60
4.40
4.20
5 4.00
n
= 3.80
E
3.60
3.40
3.20
3.00
2.80
2.60
2.40
2.20
25 50 75 100 125 150 175 200
Unit Size, MW
225
250
275
300
325
-------
I
O
$1200
$1150
$1100
$1050
$1000
$950
$900
$850
$800
$750
$700
$650
$600
$550
Figure 3-38
Vertically-Fired Dry Bottom Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levellzed Costs, $/ton NOx removed v. MW - 65% Capacity Factor Case
25 50 75 100 125 150 175 200
Unit Size, MW
225
250
275
300
325
-------
Figure 3-39
Vertically-Fired Dry Bottom Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levellzed Costs, mlls/kWh v. MW - 27% Capacity Factor Case
19 4f1
19 on .
11 60
11 20
10 80
1040
m on
9 60
5 9 20
- 880
8 40
8.00
r.eo
7.20
6.80
6.40
6.00
5.60
5.20
2
V
\
\
\
\
^
v
\
\
\
X
^x^
^*^^
"-^
-^
— --,
•^-^
->.
• — -^_
— • — --
5 50 75 100 125 150 175 200 225 250 275 300 325
Unit Size, MW
-------
Figure 3-40
Vertically-Fired Dry Bottom Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levelized Costs, Srton NOx removed v. MW - 27% Capacity Factor Case
$3000
•2.
$2000
$1800
$1600
$1400
$1200
25 50 75 100 125 150 175 200
Unit Size, MW
225
250
275
300
325
-------
Figures 4-1 through 4-15
-------
$80
$70
$60
$50
$40
$30
$20
$10
Figure 4-1
Gas-Fired, Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Capital Costs v. MW
$o
30 130 230 330 430 530 630 730 830 930 1030 1130 1230 1330
Unit Size, MW
-------
Figure 4-2
Gas-Fired, Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levellzed Costs, mlls/kWh v. MW - 65% Capacity Factor Case
= 0.90
0.80
0.70
0.60
0.50
OAO
0.30
30
130
230
330
430
530
630 730
Unit Size, MW
830
930
1030
1130
1230
1330
-------
Figure 4-3
Gas-Fired, Wall-Burner or Tangential, 0.1S Lb/MMBtu SCR Retrofit
Base Levellzed Costs, $/ton NOx removed v. MW - 65% Capacity Factor Case
$4200
$3800
$1400
$1000
30 130
230
330
430
530
630 730
Unit Size, MW
830
930
1030
1130
1230
1330
-------
Figure 4-4
Gas-Fired, Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levellzed Costs, mlls/kWh v. MW - 27% Capacity Factor Case
230
330
430
530
630 730
Unit Size, MW
830
930
1030
1130
1230
1330
-------
o
$9000
$8600
$8200
$7800
$7400
$7000
$6600
$6200
$5800
$5400
$5000
$4600
$4200
$3800
$3400
$3000
$2600
$2200
Figure 4-5
Gas-Fired, Wall-Burner or Tangential, 0.15 Lb/MMBtu SCR Retrofit
Base Levellzed Costs, Srton NOx removed v. MW - 27% Capacity Factor Case
30
130
230
330
430
530
630 730
Unit Size, MW
830
930
1030
1130
1230
1330
-------
$70
$60
$50
$40
$30
$20
$10
Figure 4-6
Gas-Fired, Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu Gas Reburnlng Retrofit
Base Capital Costs v. MW
$o
30 130 230 330 430
530 630 730 830
Unit Size, MW
930 1030 1130 1230 1330
-------
1.20
Figure 4-7
Gas-Fired, Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu Gas Reburnlng Retrofit
Base Levellzed Costs, mlls/kWh v. MW - 65% Capacity Factor Case
1.00
0.80
0.60
0.40
0.20
0.00
30 130 230 330 430 530 630 730 830 930 1030 1130 1230 1330
Unit Size, MW
-------
Figure 4-8
Gas-Fired, Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu Gas Reburnlng Retrofit
Base Levellzed Costs, $/ton NOx removed v. MW - 85% Capacity Factor Case
x
o
$3200
$2800
$2400
$2000
$1600
$1200
$800
$400
\
30 130
230
330
430
530
630 730
Unit Size, MW
830
930
1030
1130
1230
1330
-------
3.00
Figure 4-9
Gas-Fired, Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu Gas Reburnlng Retrofit
Base Levelized Costs, mlls/kWh v. MW - 27% Capacity Factor Case
2.80
2.60
2.40
2.20
2.00
1.80
1.60
1.40
\
1.20
0.60
0.40
0.20
30
130
230
330
430
530
630 730
Unit Size, MW
830
930
1030
1130
1230
1330
-------
Figure 4-10
Gas-Fired, Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu Gas Returning Retrofit
Base Levellzed Costs, $/ton NOx removed v. MW - 27% Capacity Factor Case
130
230
330
430
530
630 730
Unit Size, MW
830
930
1030
1130
1230
1330
-------
Figure 4-11
Gas-Fired, Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu SNCR Retrofit
Base Capital Costs v. MW
$40
$20
$0
130
230
330
430
530
630 730
Unit Size, MW
830
930
1030
1130
1230
1330
-------
1.60
Figure 4-12
Gas-Fired, Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu SNCR Retrofit
Base Levellzed Costs, mlls/kWh v. MW - 65% Capacity Factor Case
1.40
1.20
1.00
0.80
0.60
0.40
0.20
0.00
30 130 230 330 430
530 630 730 830
Unit Size, MW
930 1030 1130 1230 1330
-------
$3000
$800
Figure 4-13
Gas-Fired, Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu SNCR Retrofit
Base Levellzed Costs, $/ton NOx removed v. MW - 65% Capacity Factor Case
130
230
330
430
530
630 730
Unit Size, MW
830
930 1030 1130 1230
1330
-------
2.80
Figure 4-14
Gas-Fired, Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu SNCR Retrofit
Base Levellzed Costs, mils/kWh v. MW - 27% Capacity Factor Case
0.80
230
330
430
530
630 730
Unit Size, MW
830
930
1030
1130 1230
1330
-------
Figure 4-15
Gas-Fired, Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu SNCR Retrofit
Base Levellzed Costs, I/ton NOx removed v. MW - 27% Capacity Factor Case
$5600
$5200 -
$4800
$4400 •
$4000 •
X
$3600
$3200
$2800 •
$2400
$2000
$1600 •
Si?nn
I
\
\
\
\
\
\
-- — _
*^^
-— — ..
•
~
•
*
^KSS=^=
-^"^.^i^^—
— ^M^^HV
30 130 230 330 430 530 630 730 830 930 1030 1130 1230 13'
Unit Size, MW
-------
Figures 5-1 through 5-15
-------
$100
Figure 5-1
Oil-Fired, Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Capital Costs v. MW
30 130 230 330 430 530 630 730 830 930 1030 1130 1230 1330
Unit Size, MW
-------
Figure 5-2
OII-Flred. Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levellzed Costs, mlls/kWh v. MW - 65% Capacity Factor Case
1.60
1.40
1.20
1.00
0.80
0.60
0.40
0.20
0.00
30
130
230
330
430
530
630 730
Unit Size, MW
830
930
1030
1130
1230
1330
-------
$4600
$4400
$4000
$3600
$3200
Q $2800
Z
$2400
$2000
$1600
$1200
$800
$400
30
130
Figure 5-3
OII-Flred, Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levellzed Costs, $rton NOx removed v. MW - 65% Capacity Factor Case
230
330
430
530
630 730
Unit Size, MW
830
930 1030 1130 1230
1330
-------
Figure 5-4
Oil-Fired, Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levellzed Costs, mlls/kWh v. MW - 27% Capacity Factor Case
5.80
5.60
5.40
5.20
5.00
4.80 -
4.60
4.40
4.20
4.00
3.80
3.60
3.40
3.20
3.00
2.80
2.60
2.40
2.20
2.00
1.80
1.60
1.40
30
130
230
330
430
530
630 730
Unit Size, MW
830
930
1030
1130
1230
1330
-------
o
e
$9200
Figure 5-5
OII-Flred, Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu SCR Retrofit
Base Levellzed Costs, $/ton NOx removed v. MW - 27% Capacity Factor Case
130
230
330
430
530
630 730
Unit Size, MW
830
930
1030
1130
1230
1330
-------
$70
$60
$50
$40
$30
$20
$10
Figure 5-6
OII-Flred, Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu Gas Reburnlng Retrofit
Base Capital Costs v. MW
\
$0
30 130 230 330 430 530 630 730 830 930 1030 1130 1230 1330
Unit Size, MW
-------
Figure 5-7
Oil-Fired, Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu Oas Reburnlng Retrofit
Baae Levellzed Costs, mlls/kWh v. MW - 65% Capacity Factor Case
3.00
2.80
2.60
2.40
2.20
2.00
1.80
1.60
= 1.40
1.20
1.00
0.80
0.60
0.40
0.20
0.00
30
130
230
330
430
530
630 730
Unit Size, MW
830
930
1030
1130
1230
1330
-------
Figure 5-8
OII-Flred, Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu Gas Reburnlng Retrofit
Base Levellzed Costs, $/ton NOx removed v. MW - 85% Capacity Factor Case
o
$3200
$2800
$2400
$2000
$1600
$1200
\
$800
30
130
230
330
430
530
630 730
Unit Size, MW
830
930
1030
1130
1230
1330
-------
3.60
3.40
3.20
3.00
2.60
2.60
2.40
2.20
2.00
1.80
1.60
1.40
1.20
1.00
0.80
0.60
Figure 5-9
OII-Fired, Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu Gas Reburnlng Retrofit
Base Levellzed Costs, mlls/kWh v. MW - 27% Capacity Factor Case
30
130
230
330
430
530
630 730
Unit Size, MW
830
930
1030
1130
1230
1330
-------
$6000
$5600
$1600
30
Figure 5-10
Oil-Fired, Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu Gas Returning Retrofit
Base Levellzed Costs, $/ton NOx removed v. MW - 27% Capacity Factor Case
230
330
430
530
630 730
Unit Size, MW
830
930 1030 1130 1230
1330
-------
Figure 5-11
Oil-Flred, Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu SNCR Retrofit
Base Capital Costs v. MW
$40
$20
$10
130
230
330
430
530
630 730
Unit Size, MW
830
930
1030
1130
1230
1330
-------
Figure 5-12
Oil-Fired, Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu SNCR Retrofit
Base Levelized Costs, mlls/kWh v. MW - 65% Capacity Factor Case
1 80
1 60
1 40
1 7O
£z 1 nn
2
1
0 80
0.60
0.40
0.20
0.00
3
\
V
\
^"^
^^^
.
™* •
-
- - -
...
0 130 230 330 430 530 630 730 830 930 1030 1130 1230 1330
Unit Size, MW
-------
Figure 5-13
OII-Flred, Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu SNCR Retrofit
Base Levelized Costs, $/ton NOx removed v. MW - 65% Capacity Factor Case
$2800
$2400
$2000
x
O
$1600
$1200
$800
130
230
330
430
530
630 730
Unit Size, MW
830
930
1030
1130
1230
1330
-------
Figure 5-14
OII-Flred, Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu SNCR Retrofit
Base Levellzed Costs, mlls/kWh v. MW - 27% Capacity Factor Case
2 60
2 40
' \
220 I *
200
1 80
1 60
"5 1 4O
jj
1 20
1.00
0.80
0.60
0.40
0.20
0.00
3
\
\
\
X^
^^
[^^
===== —
1
—
~
-
-
-
—
0 130 230 330 430 530 630 730 830 930 1030 1130 1230 1330
Unit Size, MW
-------
$4600
$4200
$3800
$3400
>< $3000
Z
§
5 $2600
$2200
$1800
$1400
$1000
Figure 5-15
OII-Flred, Wall-Burner or Tangential Boiler, 0.15 Lb/MMBtu SNCR Retrofit
Base Levelized Costs, $/ton NOx removed v. MW - 27% Capacity Factor Case
230 330 430 530
630 730
Unit Size, MW
830 930 1030 1130 1230 1330
-------
EPA Contract Number 68-D2-0168
Work Assignment Numbers 5C-05 and 5C-09
RESPONSES TO COMMENTS ON THE MARCH
1996 DRAFT REPORT - COST ESTIMATES FOR
SELECTED APPLICATIONS OF NOX CONTROL
TECHNOLOGIES ON STATIONARY
COMBUSTION BOILERS
June 1997
Prepared for
U.S. Environmental Protection Agency
Acid Rain Division
501 3rd Street
Washington, DC 20001
by
Bechtel Power Corporation
9801 Washingtonian Boulevard
Gaithersburg, MD 20878-5356
subcontractor to
The Cadmus Group, Inc.
135 Beaver Street
Waltham, MA02154
-------
TABLE OF CONTENTS
Page
1.0 INTRODUCTION 1
2.0 RESPONSE TO COMMENTS FROM DOE 1
3.0 RESPONSE TO COMMENTS FROM ICAC 5
4.0 RESPONSE TO COMMENTS FROM BAKER & BOTTS - 7
5.0 RESPONSE TO COMMENTS FROM BLACK AND VEATCH 9
6.0 RESPONSE TO COMMENTS FROM NALCO FUELTECH 12
7.0 RESPONSE TO COMMENTS FROM NORTHEAST UTILITIES 14
8.0 RESPONSE TO COMMENTS FROM HUNTON & WILLIAMS 17
9.0 REFERENCES 26
ATTACHMENTS
ATTACHMENT 1: CORROBORATION OF THE CAPITAL COST FOR
MERRIMACK'S SCR INSTALLATION
ATTACHMENT 2: COPIES OF COMMENTS RECEIVED FROM PUBLIC
-------
1.0 INTRODUCTION
The purpose of this report is to record responses to the public comments
received on the EPA draft report, "Cost Estimates for Selected Applications of
NOX Control Technologies on Stationary Combustion Boilers." The Comments
as well as corresponding responses are listed for each commenter separately.
All of the appropriate editorial comments, although not listed below, will be
incorporated in the next revision to the report. Copies of the correspondence
containing these public comments are included as Attachment 2 in this report.
The report for another EPA study, "Investigation of Performance and Cost of NOX
Controls as Applied to Group 2 Boilers," was attached as Appendix A to this
draft report. The Group 2 boiler study was conducted in support of the Phase II
NOX control rule under Title IV of the 1990 Clean Air Act Amendment. Appendix
A was included with this draft report as a reference, because the general
technical and economic evaluation criteria and the design data for the typical
Group 2 boilers described in Appendix A formed a basis for the evaluations in
this report.
Some of the comments received on the draft report address items in Appendix A.
Any Appendix A comments were addressed separately as part of the EPA's work
on the Phase II rule. The discussion below, therefore, does not address any of
the comments received on Appendix A.
The report in Appendix A was revised as a result of the public comments and
reissued in August 1996. In the next revision to the study, the new, revised
report will be included in Appendix A.
2.0 RESPONSE TO COMMENTS FROM DOE
The comments were provided in a May 20, 1996, letter from DOE to EPA. These
comments along with the responses are provided below:
1. The report addresses both Groups 1 and 2 boilers. While Group 2 boilers
are covered by Appendix A, no documentation is provided for Group 1
boilers.
Response: Appendix A was attached to the report to provide a reference
for the general economic evaluation approach for the study as well as to
provide a reference for the design parameters of the Group 2 boilers used
as the basis for cost estimates. While the general economic approach
described in Appendix A applies equally to both Groups 1 and 2 boilers, the
design parameters for Group 1 boilers have been covered in the report
itself. Section 3.0 and Table 3-1 list design parameters for the tangential-
and wall-fired boilers burning coal. Sections 4 and 5 and Table 4-1 include
similar parameters for oil- and gas-fired boilers.
-------
2. No background documentation is provided for the costs for achieving 0.15
Ib/MMBtu of NOX. Furthermore, no basis is given for the selection of the
0.15 Ib/MMBtu target.
Response: Section 2.0 of the report discusses in detail the basis for
technology selection and economic evaluation criteria for achieving a NOX
emission rate of 0.15 Ib/MMBtu. As mentioned in Section 2.3 of the report,
the cost estimating methodology in most parts was the same as that used
for a previous 1995 study conducted by EPA for estimating the costs of NOX
controls for the Group 2 boilers. The report for the 1995 study published in
August 1995 was attached to the subject study as Appendix A.
A controlled NOX emission rate of 0.15 Ib/MMBtu was selected as a
reasonable level achievable with some of the commercially available NOX
control technologies. The capabilities of various NOX control technologies
are fully covered in Appendix A, which lists the experience to date with
each of these technologies. Section 2.1 of the study describes as how the
selection of the 0.15 Ib/MMBtu NOX limit was supported by this experience.
3. The statement in the first bullet item of Section 2.1 regarding Iow-N0x
burners (LNBs), overfire.air (OFA) ports, and gas recirculation (GR) fans is
not clear. If the combustion modifications are already in place for some
boilers, are other technologies not technically or economically feasible?
Response: The first bullet item of Section 2.1 is a summation of what is
discussed in this paragraph prior to this item. As stated in the second
paragraph of this section, the wall- and tangential-fired boilers being
evaluated were assumed to be equipped with LNBs (one of the components
of combustion controls). The first bullet item of this section is referring to
the feasibility of applying other components (OFA and GR) of combustion
controls to achieve the 0.15 Ib/MMBtu limit on these boilers. As stated in
this item, it is not technically feasible to achieve 40 to 70 percent of NOX
reduction by adding OFA and/or GR to the boilers already equipped with
LNBs.
4. Inclusion of overfire air ports with Iow-N0x burners has been disallowed in
the past NOX control rules. Should it be considered now?
Response: The purpose of the study was to consider all of the
commercially available NOX control technologies for NOX emission limits
being considered for future rules. Consideration of OFA is important and
justified because it is a relatively low capital-cost approach that can provide
a substantial NOX reduction.
5. The report appears to be biased in favor of SNCR. Technologies other
than those evaluated in the study could also be used. There is insufficient
-------
discussion in the study to inform the readers of reasons for the study
conclusions.
Response: The evaluation of NO* control technologies considered in the
study has been based strictly on the technology merits and capabilities.
One of the conclusions in the study was that SNCR is not a viable
technology for achieving the 0.15 Ib/MMBtu NOX emission limit on coal-fired
boilers (this cannot be termed as a favorable conclusion). It was, however,
concluded that SNCR could be applied to gas- and oil-fired boilers with
proper requisites, such as adequate residence time in the flue gas
temperature zone to support the reaction between the reagent and NOX.
SNCR was also evaluated for coal-fired boilers based on its ability to
provide substantial NOX reduction. As clearly stated in the study,
application of SNCR would depend on site-specific factors, one being the
presence of the aforementioned residence time (refer to Section 2.1).
The study has evaluated all of the available commercial technologies that
could be considered for achieving the 0.15 Ib/MMBtu emission limit (refer to
Section 2.0). The merits of these technologies have been covered in detail
in Appendix A of the study. The study does mention the possibility of
utilizing a combination of technologies to achieve the 0.15 Ib/MMBtu limit,
such as a hybrid system using both SCR and SNCR or a system using SCR
in conjunction with portions of combustion controls (refer to Section 2.1).
These systems are, however, considered an optimization of individual
technologies, and they were not evaluated because of the limited scope of
the study.
The study draws from the extensive work done concerning the technical
and economic evaluation of NOX controls in the aforementioned Group 2
boiler report. The Group 2 boiler report was attached to the study report as
Appendix A specifically to provide the readers with the background
information on the approach and methodology used in the study.
6. The levelized costs are reported (Table 1 -2) both with and without the
capital charge component. The levelized costs should only be reported
with the capital charge.
Response: The total levelized costs, including the capital charge, for each
technology application are presented in Tables 1-3 and 1-4 as well as in
Figures 3-1 through 3-40, Figures 4-1 through 4-15, and Figure 5-1 through
5-15. Additional information in Table 1-2 was specifically included to
differentiate between various technology applications based on the
contribution of the capital and operating costs to the total levelized costs.
This information is useful in that it readily identifies the technologies that
-------
are capital cost intensive as well as those that are operating cost intensive.
Therefore, we recommend keeping the information in Table 1-3 unchanged.
7. There is no supporting documentation for some of the cost estimates
presented in Tables 1-2 through 1-4.
Response: It is not clear what additional supporting documentation, other
than that provided in the study, is required. The evaluation criteria,
assumptions, and economic factors are detailed in Sections 2.0 through 5.0
and Table 2-1. The overall cost estimating approach is presented in
Appendix A, which also lists the boiler design parameters for coal-fired,
cell-burner, cyclone, wet bottom, and vertically fired boilers.
The design parameters for wall- and tangential-fired boilers burning coal,
gas, and oil are presented in Tables 3-1 and 4-1. In addition, the
consumables associated with each technology application are presented in
the various sections of the study. All of the cost estimates presented in the
study can be verified by using the information on consumables and the
economic parameters presented in Table 2-1.
8. The cost figures presented in the study are not consistent. For example,
Table 1-3 shows a levelized cost of $695/ton for SCR applied to a 200 MW
cyclone boiler at a 65 percent capacity factor. Figure 4-21 in Appendix A
shows a value of about $625/ton for the same boiler at the same capacity
factor.
Response: Appendix A was attached to the study report to provide the
reader information on the evaluation criteria, technical background on each
technology, and design information on certain coal-fired boilers used in the
cost estimates for the study. The design basis for the NOX control
technologies in Appendix A was different from that used in this study. For
example, the SCR system in Figure 4-21 of Appendix A was designed for a
NOX removal rate of 50 percent. In comparison, the cyclone-fired SCR cost
presented in Table 1-3 of the study is based on a NOX reduction rate of
approximately 87 percent.
The design basis for each technology application is clearly stated in both
the study and Appendix A. The costs are, therefore, not comparable when
a difference exists between the two design bases.
9. On Page 1-2 of the study, a reference to economic factors "reported" in the
EPRI TAG has been provided. Since the economic factors in the EPRI
TAG are only examples and are not meant to be recommendations, the
word "reported" should be replaced with the term "listed" or "given".
Response: The recommended change will be incorporated in the report.
-------
10. The basis for the two capacity factors (27 and 65 percent) mentioned in
Section 1.3 of the report should be explained in greater detail for the
levelized costs in mill/kWh reported in Table 1-2.
Response: The notes provided at the bottom of Tables 1-2 through 1-4
will be expanded to further explain the basis for the capacity factors.
11. Revise Table 1-1 to identify Group 1 and 2 boilers separately.
Response: The recommended change will be incorporated.
3.0 RESPONSE TO COMMENTS FROM ICAC
The comments were provided in a May 28, 1996, letter from ICAC to EPA. The
comments along with the responses are provided below:
1. The report should note that cost effectiveness value (expressed in $/ton of
NOX removed) will decrease as capacity factor increase over 65 percent.
Response: A typical capacity factor of 65 percent was selected for the
power plants in general. It is recognized that there will be plants operating
above and below the 65 percent factor. A note will be added to reflect that
cost effectiveness values can vary with the capacity factor, i.e., increasing
when the capacity factor decreases and decreasing when the capacity
factor increases above the 65 percent value.
2. The report should include calculated SNCR costs for all boiler/fuel
combinations. While SNCR alone may not be sufficient to reduce NOX
emissions to 0.15 Ib/MMBtu in all cases, SNCR may be part of the
combinations of control technologies.
Response: SNCR has been evaluated for all of the Group 1 boilers
considered for the study. Further, Appendix A evaluates SNCR application
on cyclone, vertically fired, and wet-bottom boilers. For these boilers with
relatively low baseline NOX emissions, SNCR can reduce NOX to
significantly low levels. Viability of using SNCR along with another
technology (as a hybrid system) for Group 1 and 2 boilers was recognized
in the report (refer to Section 1.4). Because of the limited nature of the
study, evaluation of such hybrid systems was considered outside of the
scope.
3. The report overestimates initial catalyst charges and catalyst replacement
rates for SCR, and thus overestimates SCR costs. Actual original installed
catalyst volumes (cubic meters of catalyst per MW of plant capacity) are 20
to 75 percent lower than the volumes used in the report. Improvements in
-------
catalyst technology and experience over time have allowed installation of
smaller catalyst volumes.
The report also conservatively assumes total replacement of the catalyst
bed every 3 years for coal-fired boilers. This assumption inflates actual
catalyst replacement costs by a factor of 1.3 to 3, depending on the boiler
type, and therefore introduces unacceptable errors into the cost
calculations. Industry experience universally supports a staged addition-
replacement strategy for extending catalyst life. No SCR system will
require total catalyst change-out at the end of the guarantee period. (The
commenter has provided data from certain operating installations and from
quotes by one supplier.)
Response: One of the criteria for SCR evaluation in the study was not to
use staged catalyst addition/replacement (refer to Section 3.1.1). This was
a conservative approach, resulting in conservative cost estimates.
EPA recognizes that, for most power plants, a catalyst life management
strategy would be desirable to extend the catalyst life. However, a system
designed for staged catalyst addition/replacement would result in a larger
reactor volume because of the presence of additional reactor layers
required for such a design. This system may also have a relatively high
pressure drop because of the increased pressure drop resulting from the
loading of catalyst into the initially empty layers. For retrofit applications,
the merits of a longer catalyst life would have to be evaluated against the
potential need for additional space to accommodate a larger reactor and
consequences of higher pressure drops.
Regardless of the above issues, EPA believes that exclusion of the life
management strategy represents a more conservative approach that
addresses the site-specific needs of a variety of SCR applications. Since
this approach results in conservative cost estimates, its use is justifiable.
The SCR catalyst life and volume estimates for various applications in the
study were conservatively based on the recent coal-fired experience in the
U.S. and coal characteristics that require more conservatively sized catalyst
volumes. Recent information from two SCR suppliers confirms the use of a
3-year guaranteeable catalyst life.
A review of the data presented by the commenter-specifically from
operating installations-shows that the catalyst volumes in the EPA study
generally agree with those in the data. EPA notes that the nonoperating
catalyst volume data provided by the commenter appears to be low only
because of the use of a catalyst life management strategy (which was not
the basis for the EPA's figures).
-------
.4. In reference to Page 2-2, fourth paragraph, of the report, SCR system will
not necessarily lead to excessive S02 to SO3 conversion rates; SCR
catalysts are available that oxidize less than 1 percent of the S02 to S03
Response: EPA agrees that new catalysts have now become
commercially available that minimize oxidization of S02 to S03. The
excessive S02 to S03 conversion rate was specifically mentioned in the
report in conjunction with applications requiring very high NOX reduction
rates (e.g., 87 percent for cyclone-fired boilers) and those where relatively
high flue gas concentrations of S02 are present. For such applications,
design measures, such as use of the aforementioned special catalysts,
would become necessary. The section of the report referenced by the
commenter will be revised to further explain this issue.
5. In reference to Pages 3-2 and 3-4 of the report, catalyst replacement
volume rates shown on this page are high.
Response: The catalyst replacement volume rates shown in the report are
based on a catalyst life of 3 years for coal-fired applications. As mentioned
in the response to Comment 3 above, the design basis for the SCR systems
in the EPA's study did not utilize a catalyst life management strategy. This
approach does result in conservative catalyst replacement rates.
6. In reference to Page 4-1 of the report, a catalyst operating life of 5 years is
low for natural gas service; a life of 8 to 10 years would be more
representative of actual operating experience.
Response: EPA used an operating catalyst life of 5 years for which
commercial guarantees can be obtained from different suppliers. EPA
agrees that actual experience shows an operating life of longer than 5
years. However, unless such data can be backed up by commercial
guarantees, using it to establish technology costs does not appear to be
justified.
4.0 RESPONSE TO COMMENTS FROM BAKER & BOTTS
The comments were provided in a May 20, 1996, letter from Baker & Botts to
Perrin Quarles Associates, Incorporated. The comments along with the
reponses are provided below.
1. Use of the power factor scaling methodology in the EPA's report to project
capital costs from the known cost of an SCR application to other different
size applications is not valid. This methodology is viable for estimating
costs for complete power plants, and not for single technology applications,
such as SCR. EPA's costs show a substantial difference between the costs
of SCR retrofits on 200 and 900 MWe plants. Since SCR lacks the
-------
assumed economy of scale, there should be no cost difference between the
200 and 900 MWe systems.
Furthermore, the cost estimates available for SCR retrofit applications in
the U.S. (at Memmack and Mercer Stations) do not support the EPA's
estimates. Even though the $/kW costs reported for Mercer (larger of the
two installations) is lower than that for Merrimack, which tends to support
EPA's methodology, this cost difference was for reasons other than the
installation sizes. For example, the outdoor construction for Merrimack
made it relatively easy to retrofit the SCR reactor into the proper location in
the flue gas duct.
Response: EPA believes that the power factor scaling methodology used
in its study reflects a practice commonly employed by the utility industry in
determining capital costs for both complete power plants and individual
equipment and technologies. As a result of investigations, EPA also finds
that the cost models used in the EPA's estimates result in conservative
costs for Group 2 NOX controls. The following address various issues
raised by the commenter:
• EPA notes that use of power factor scaling to develop costs for
individual components or systems has been reported in numerous
publications. One source has listed data from several publications (a
total of 15) that confirms use of power law scaling factors for a large
number of individual components and systems used in the chemical and
pollution control systems (Ref. 1). This article alone provides
substantial information to nullify the concern raised by the commenter.
Another industry source has addressed the scaling factor issue
specifically with regards to SCR costs (Ref. 2). According to this
source, factors of 0.3 to 0.4 ($/kW basis) can be used to scale up the
SCR costs. The SCR scaleup factors used in the EPA study are
generally towards the lower end of the range recommended by this
source, thus resulting in more conservative cost estimates for larger
plants. This source clearly upholds EPA's use of the power factor
scaling methodology.
SCR retrofit costs for 122 to 750 MWe plants have been reported by
another source (Ref. 3). This data shows a reduction in the SCR retrofit
cost of approximately 48 percent between 122 and 750 MWe plants.
The same source provides another example of SCR costs 100 to 375
MWe plants, showing a cost savings of approximately 41 percent for the
larger plant. These reported cases show a greater reduction in the SCR
retrofit costs for the larger plants than what would be obtained using the
EPA's methodology. The commenter's claim that the SCR cost would
not reduce for larger plants is therefore not valid.
8
-------
• EPA does not agree with the commenter's interpretation of the cost
difference between the Merrimack's and Mercer's SCR installations.
Contrary to the favorable retrofit conditions at Merrimack alleged in the
comment, the SCR retrofit at Merrimack has been reported to be a
difficult retrofit, which required addition of extensive flue gas ductwork to
accommodate the SCR reactor. In addition, the baseline NO* emission
at Merrimack was 2.66 Ib/MMBtu (compared with 1.8 Ib/MMBtu for
Mercer), and the system is designed to reinject 100 percent of ash to
the fumace(4). All of these factors resulted in an increased capital cost;
the high baseline emission required a large ammonia storage and
injection system and the concerns with high flue gas arsenic
concentrations because of ash reinjection resulted in a larger and
arsenic-resistant catalyst.
EPA performed an analysis to compare the retrofit cost reported for
Merrimack with those obtained from EPA's costing methodology:
Attachment 1 presents the results of this comparison, which
corroborates EPA's methodology.
2. EPA's report shows a pressure drop increase of 5 inches of water with the
addition of SCR, without addressing any capacity derates associated with
fan capacity limitations.
Response: All of the SCR cases in the EPA's report have assumed
replacement of existing draft fans with larger fans to accommodate the additional
gas side pressure drop. Therefore, a plant capacity derate would not be required
with the larger fans. EPA recognizes that fan replacement would not be required
for all retrofits. However, the cost for new fans was added as part of the SCR
retrofit as a conservative measure.
5.0 RESPONSE TO COMMENTS FROM BLACK AND VEATCH
The comments were provided in a May 23, 1996, letter from Black & Veatch to
Perrin Quarles Associates, Incorporated. The comments along with the
responses are provided below.
1. The report should assume the use of a catalyst management plan for SCR
systems. Using this plan reduces annual catalyst replacement cost by at
least 65 percent. Such a plan can be incorporated by providing an extra
layer in the SCR reactor for future catalyst addition. Even in the unlikely
event of the inability to include a spare layer in the design, an effective
catalyst management plan can be incorporated replacing individual layers.
This also leads to substantial saving when compared with complete
replacement at the end of catalyst life.
-------
Response: As mentioned in the EPA's response to the ICAC's Question 3,
the decision to adopt an SCR design basis without a catalyst management
plan was made on conservative grounds. Specifically, this design
approach was used in consideration of those retrofit installations where the
SCR reactor size and pressure drop may be important considerations.
EPA agrees that a catalyst management plan would be feasible for new
installations and for many retrofit installations. Such a plan is expected to
result in significant cost savings associated with longer catalyst life.
However, EPA believes that an SCR design basis without this plan results
in conservative cost estimates that reflects more-difficult-to-control retrofits.
2. Published data reporting results of SNCR installations does not support the
report's assumption that SNCR has a NOX reduction capability of 50
percent. SNCR has demonstrated capability for reliably removing 20 to 40
percent NOX reduction on small to medium PC boilers while maintaining
ammonia slip in acceptable ranges. (The commenter has quoted data from
studies done by the commenter as well as published data from a 1996
ICAC forum held in Baltimore that show SNCR performance ranging from
30 to 40 percent.)
Response: EPA disagrees that SNCR has not been demonstrated at NOX
reduction levels above 40 percent with acceptable ammonia slip levels.
One source lists several coal-fired applications of SNCR where NOX
reduction levels of 50 percent and above were achieved (Reference 6).
One of the listed coal-fired installations (WEPCO's Valley Power Plant) has
been reported to have achieved 60 percent NOX reduction with an ammonia
slip of 5 ppm. The data quoted by the commenter also includes a coal-fired
plant where NOX reductions of up to 50 percent were demonstrated.
EPA's report states that the SNCR system is capable of NOX reductions of
30 to 50 percent. EPA agrees that not all candidate plants would be able to
achieve NOX reductions toward the higher end of this range with SNCR.
For retrofit applications, this technology is heavily dependent on the
existing boiler design and operating conditions, such as the flue gas
residence time within an appropriate temperature range for reaction
between the reagent and NOX, temperature gradient at the reagent injection
plane, baseline NOX, etc. Any one of these factors can affect the
effectiveness of SNCR.
Experience with SNCR shows that substantial NOX reductions (30 to 50
percent) are possible with this technology for coal-fired retrofit installations.
EPA selected a 50 percent NOX reduction level for this study to show the
costs associated with the higher end of the SNCR performance. The next
revision of the study will be based on an average SNCR NOX removal
efficiency of 40 percent.
10
-------
EPA analyzed the effect of lowering NOX reduction from 50 to 40 percent on
the study costs. For this purpose, the operating costs were revised for the
case with tangential-fired boilers firing coal. Even though the capital costs
for the SNCR retrofit would decrease with a reduction from 50 to 40 percent
NOX, these costs were not changed (resulting in conservative levelized
costs). The results of the analysis showed that, with an NOX reduction from
50 to 40 percent, the cost effectiveness increased from $1,378 to
$1,543/ton NOX removed for a 200 MWe boiler and from $1,150 to
$1,262/ton NOX removed for a 900 MWe boiler.
The above comparison shows an increase in the total levelized costs of
approximately 12 percent for the 200 MWe boiler and 10 percent for the
900 MWe boiler with the lower NOX reduction level (40 percent). This is not
a significant change in the total levelized cost.
3. The report does not discuss and reflect potential economic impacts caused
by the ammonia slip from SNCR systems, such as forced outages and
boiler load limitations. Experience has shown that numerous SNCR
installations need relatively frequent off-line cleanings of the air heater
when using SNCR with sulfur bearing fuels. Forced outages would be very
expensive to accommodate especially during the summer peak season.
(The commenter provides a reference to one SNCR installation and
mentions another installation without a reference, in support of the
comment.)
Response: The commenter has not provided any proof of the claim that
"numerous" SNCR installations have reported forced outages due to
problems specific to this technology. Only one verifiable reference has
been provided by the commenter to support this claim.
One source lists several references of SNCR installations where this
technology has been successfully applied (Ref. 6). As mentioned by this
source and also indicated in references provided with the report (refer to
Appendix A), the operating problems indicated by the commenter can be
controlled by minimizing ammonia slip levels. Based on reported
experience from many installations, EPA does not agree that forced
outages due to ammonia slip are an inherent part of SNCR installations.
4. The capacity factor used (65 percent) in the economic analysis of the report
is too low. Likely target caseload units operating during the 5-month "NOX
season" are likely to have very high capacity factors (85 to 95 percent)
during this summer peak period. A misrepresentative value of 65 percent
has a punitive effect on capital intensive technologies such as SCR.
Response: EPA agrees with the commenter that plant capacity factors
higher than 65 percent are likely during the summer peak period (5-month
11
-------
"NOX season"). The use of a 65 percent factor is appropriate, however,
because it represents a basis for comparison of the cost estimates.
5. Currently, this draft version appears to be heavily biased towards SNCR
and against SCR when discussing post-combustion NOX control systems.
We believe that it is misleading to imply that the installation of an SNCR
system will reliably lead to 50 percent NOX reduction with ammonia slip less
than 10 ppm with no potential for significant detrimental impact on plant
operation.
The bias against SCR is demonstrated in paragraphs such as the third
complete paragraph on Page 2-2 where catalyst volume requirements are
described as "significantly large" and S03 conversion rates are described
as "excessive."
Response: EPA does not agree that the study reflects any bias towards
SNCR or against SCR. For coal-fired applications, the study clearly shows
the SNCR technology to not be comparable to SCR. The SCR technology
has been evaluated as the only technology capable of reducing NOX
emissions to 0.15 Ib/MMBtu. The SNCR technology has been evaluated
only as a technology that can provide substantial NOX reductions. Such an
evaluation cannot be termed as biased towards SNCR.
The comment regarding the 50 percent NOX reduction with SNCR has been
addressed in Item 2 above. The last item in the comment regarding
catalyst volume and S03 conversion rates is a misinterpretation of the
statements in the EPA's report. The term "significantly large" has been
used in conjunction with the catalyst volume requirements for relatively high
NOX reduction efficiencies required, especially for Group 2 boilers (e.g., an
efficiency of 87 percent for cyclone-fired boilers). Increased catalyst
volumes do result in increased capital costs. The S03 conversion rates
have been mentioned in conjunction with high concentrations of S02 in the
flue gas, which would require consideration of catalyst materials that
minimize such conversion. Both of these items have been mentioned in
terms of their impact on the cost. In the next revision of the report, these
items would be clarified.
6.0 RESPONSE TO COMMENTS FROM NALCO FUELTECH
The comments were provided in a May 7,1996, letter from Nalco Fueltech to
Perrin Quarles Associates, Incorporated. The comments along with the
reponses are provided below.
1. The treatment of capital costs may not appropriately reflect the cost to the
utility plants subject to NOX control regulations. If a capital carrying charge
of 0.115 was used in the report (as in the Group 2 boiler report attached as
12
-------
Appendix A), it would be too low compared with the commenter's
experience of carrying charges from 0.145 to 0.200 for post combustion
NOX controls. In the deregulation and enforced competition environment,
the utilities may have to carry capital for only 5 years, by the years 1999
and 2000, which would lead to higher carrying charges.
Response: The basis for the carrying charge is provided in the report in
Table 2-1. As shown in this table, a carrying charge of 0.127 was used,
based on information provided in the 1993 EPRI TAG (Ref. 7). This
carrying charge is higher than 0.115 mentioned by the commenter. EPA
recognizes that the carrying charge as well as other economic factors may
differ for different applications. It would not be prudent to consider only one
of these factors specific to certain installations and base other factors from
different sources. For this study, it was necessary to utilize criteria that fit
typical applications of NOX controls. Therefore, the criteria presented in the
EPRI TAG was used as the basis for this study.
EPA cannot agree with the commenter on the issue of the future direction of
carrying charges in the power plant industry. An economic life of only 5
years as mentioned by the commenter is considered to be only a
speculation, without any basis.
2. Since one premise for all the data in the report is that LNB or combustion
modifications have already been employed, the gas reburning data may
need to be revised in Table 1-5. The Acurex report entitled "Phase II NOX
Controls for the NESCAUM and MARAMA Region" states that cost
effectiveness diminishes significantly for this add-on control because the
NOX reduction is only 20 percent when LNB is already installed.
Response: Experience does exist with gas reburning application on a
boiler equipped with LNBs. At Public Service Company's Cherokee Unit 3,
gas reburning was applied along with LNBs (Ref. 8). NOX reduction
associated with gas reburning alone was approximately 46 percent, well
above the 20 percent level claimed by the commenter. The NOX reduction
achieved at this installation falls within the levels used for the study.
Therefore, EPA cannot agree with the commenter's concern regarding the
nonapplicability of this technology to units equipped with LNBs.
3. In Section 3.1.1 regarding SCR, the statement "It is assumed that the
existing plant setting allows installation of the SCR reactors between the
economizer and air heater without a need to relocate any major structure or
equipment" is such an egregious leap, it is better to qualify the statement
with the admission that installation on a number of sites would be
impossible or imprudently costly.
13
-------
Response: EPA is opposed to adding any statement in the report without
a basis. Although it was impossible to evaluate the SCR retrofit potential to
the entire boiler population considered in the study, data from several
published sources indicates that SCR technology can be readily retrofitted .
to the majority of these boilers. EPA is not aware of any published source
that has referenced power plants for which SCR retrofit is impossible (the
commenter has not provided any references either).
During the evaluation of the rule for Group 2 boilers, public comments were
received that provided an indication of the applicability of SCR to the
boilers in general. One commenter provided results of an SCR retrofit
feasibility survey done on cyclone-fired (Ref. 9) boilers. Of the 28 boilers in
the survey, feasibility of SCR retrofit was confirmed on 25 boilers without
relocation of major equipment or structures. Even for the remaining three
boilers, it was reported that SCR retrofit was possible, although requiring
long duct runs. EPA notes that the Merrimack installation (refer to
Attachment 1) required extensive duct runs between the economizer outlet
and air heater inlet, yet resulted in reasonable capital costs that
corroborate the estimates provided in the EPA's study.
Another commenter (Tampa Electric Company) on the rule for Group 2
boilers provided results of a study (Ref. 10) that covered SCR application
on a large number of boilers. Based on this study, SCR retrofit was
feasible on all of these boilers. A large number of other commenters on the
rule presented examples of SCR retrofittability to their boilers. Where cost
data was provided by these commenters, EPA determined that any
differences between this data and the costs reported by EPA were mostly
due to different economic assumptions used by these commenters. In light
of data available from so many different sources, EPA cannot agree with
the commenter that SCR retrofit would be impossible or imprudently costly
on a significant number of installations.
7.0 RESPONSE TO COMMENTS FROM NORTHEAST UTILITIES
The comments were provided in a May 24, 1996, letter from Northeast Utilities
System to EPA. The comments along with the responses are provided below.
1. In general, the report is quite reasonable and complete. Some of the cost
estimates are lower than we have used; some are higher.
Response: EPA acknowledges the commenter's agreement with the study
results. The cost differences are addressed below in the responses to
specific comments on costs.
2. The SCR retrofit capital costs assume no allowance for relocating any
existing structures or equipment. In general, this is a bad assumption.
14
-------
Response: The study assumes that relocation of any major equipment or
structure would not be required as part of the SCR retrofit. Cost
allowances have been provided for relocation of minor equipment,
ductwork, and piping. This assumption is considered reasonable, since the
majority of new equipment required for SCR retrofit (tanks, pumps,
vaporization system, piping, etc.) does not have to be installed in any
specific location within the plant. The SCR reactor is located between the
economizer outlet and air heater. For the majority of power plants, this
reactor can be accommodated in the space above or on the side of the air
heater (refer to the response to Comment 3 from Nalco Fueltech).
3. The report does not mention including cWs for wastewater treatment
facility modifications that might be required to handle SCR ammonia plant
wastes and washwater wastes.
Response: In the extensive published data available for operating SCR
facilities, wastewater treatment problems due to ammonia contamination
have not been raised as an issue for this technology. Even without SCR,
ammonia storage and handling would generally be part of a power plant for
water treatment purposes. Ammonia system for SCR would therefore pose
no new issues for power plants in terms of handling any occasional
spillages or leakages.
The SCR systems are designed to minimize ammonia slip. The design
basis used in the EPA's study was a maximum ammonia slip level of 5 ppm.
At these levels, it can be expected that minimal amounts of ammonium salts
would end up in the wastewater streams from washdown of equipment in
the flue gas path. The reported SCR experience confirms this.
4. The coal unit sulfur content assumed in the report is only 0.8 percent by
weight. Higher sulfur coal applications may result in air heater pluggages,
incurring downtime costs and air heater capital work. S03 formation in the
reactor can accelerate downstream corrosion and produce opacity
plume/acid fallout problems.
Response: The SCR system design in the report is based on restricting
the ammonia slip to a level below 5 ppm. At this level of ammonia slip,
problems due to formation of ammonium bisulfate, such as air heater
pluggages, are not expected to occur (Ref. 6). In addition, catalysts are
now commercially available that minimize oxidation of SO2 to SOa. EPA
therefore does not believe that SCR poses a problem with regards to
corrosion or opacity plume/acid fallout.
5. The levelized carrying charge factor assumed by EPA is only 60 percent of
the value commenter would use.
15
-------
Response: The carrying charge factor (0.127) used in the EPA's study has
been taken directly from the 1993 EPRI TAG. This factor is based on a
useful plant life of 20 years and a constant dollar approach. It should be
noted that, based on a current dollar approach, the equivalent carrying
charge factor would be 0.179, which is approximately 41 percent higher
than the constant dollar factor used by EPA. Other economic assumptions,
such as useful plant life, can also make a significant difference in the
carrying charge factor value. Since the commenter has not provided the
basis for commenter's carrying charge factor, EPA is not in a position to
elaborate further.
The EPA's study applies to a variety of boiler applications. It is recognized
that differences would exist in the economic factors applicable to different
applications. However, EPA believes that the factor chosen from EPRI
TAG yields reasonably accurate cost estimates for typical NOX control
installations.
6. The anhydrous ammonia cost assumption is about 20 percent less than
experienced at Merrimack.
Response: Ammonia costs used by EPA were derived from reliable
sources that provided the average ammonia costs within the U.S. for the
evaluation periods used in the study (refer to Section C.2 of Appendix A).
Use of this average cost is considered more prudent for the study that
covers a large population of boilers in this country.
7. No mention is made of the disposal cost of used SCR catalyst. Ash
disposal costs are about 33 percent less than the commenter's experience.
Response: The disposal cost of used SCR catalyst are part of the catalyst
replacement cost (which will be noted in the next revision to the study).
The ash disposal costs for the study are based on the cost data in the 1993
EPRI TAG.
8. The reported technology costs in the study do not match the commenter's
experience; EPA's SCR costs are 20 to 25 percent lower for coal
applications and about one-third lower for oil or oil/gas applications. The
reported SNCR costs for oil or oil/gas units are more than 50 percent higher
compared with the commenter's estimates.
Response: The technology costs are highly dependent on the system
design basis, equipment redundancy, and contingency factors used in the
estimates. Since such information is not available for the commenter's cost
estimates, EPA is not in a position to further address the cost differences
quoted by the commenter.
16
-------
9. Natural gas assumptions for reburn on oil unit should reflect the higher
pricing more representative of noninterruptible gas contracts (typically 20
percent of total unit heat input).
Response: The natural gas price used in the study reflected the 1995
price, and it was taken from a reliable published source (Ref. 11) This
price reflects average price for gas paid by utilities.
10. Large variations in NOX reduction equipment capital cost estimates on many
of commenter's smaller units can be seen in the asymptotic scaling factors,
which the report applies to units less than 200 MW. The same effect can
be seen for level ized annual costs.
Response: EPA agrees with the commenter that the costs increase at a
higher rate as the unit size reduces, especially below 200 MWe. However,
this is strictly because, for small units, even a modest reduction in size
represents a significant percentage.
11. The report summary table for oil and gas fuels shows that SNCR is
50 percent or less of the cost of SCR on units of the commenter's size.
This cost comparison is generally true for low and high capacity
assumptions and on a basis of $/kW and $/ton of NOX.
Response: EPA acknowledges the commenter's agreement with the
comparison shown in the study between SNCR and SCR.
12. Adding natural gas reburn to a pressurized unit can be a safety hazard.
The report should mention such limiting application factors.
Response: EPA disagrees with the commenter that gas reburn cannot be
applied to pressurized units. Natural gas-fired boilers are generally
designed for pressurized operation. Established industrial codes and
standards exist to ensure that these boilers are designed and operated
without compromising safety. Furthermore, gas reburning has already
been successfully applied to a pressurized boiler (Ref. 8).
8.0 RESPONSE TO COMMENTS FROM HUNTON & WILLIAMS
The comments were provided in an August 7, 1996, letter from Hunton &
Williams to EPA. The comments in this letter as well as in the report (Ref. 12)
attached to this letter have been addressed below.
1. The key assumption in the EPA's analysis is the use of a power-law scaling
relationship to project capital cost over a wide range of generating capacity
and process conditions. Using this approach can introduce significant error
if the range over which cost is projected is too large. Generally, the range
17
-------
of extrapolation should be within a factor of two so as not to require
changes in process design; otherwise an inappropriate design is
considered as the cost basis.
Response: EPA has utilized a power-law scaling methodology for its cost
estimates that is commonly used throughout the power and other industries.
For all of the Group 2 boilers, the scaling factors were developed from cost
estimates generated for two boiler sizes in each Group 2 category (for
Group 1 boilers, appropriate factors were selected based on Group 2 boiler
factors). A comparison with cost data from other published sources shows
that these scaling factors result in more conservative cost projections for
the boiler size range (refer to the response to Comment 1 from Baker &
Botts).
The commenter's premise in restricting the range of cost extrapolation to a
factor of two is based on the concern that for larger boilers the technology
design basis would be different than that used for the smaller boilers. EPA
does not share this concern. For larger boilers, any changes in the
technology design basis would account for the increased boiler size (or flue
gas volume); the equipment associated with the technology would increase
in proportion to the increase in the boiler size. This does not amount to a
change in the fundamental design basis for the technology retrofit.
The cost estimates presented in the EPA's study do not reflect the concern
raised by the commenter. As shown in the cost curves attached to the
study, the capital costs for the technology retrofits do not drop off rapidly for
larger boilers. For example, the largest cyclone boiler size used in the
estimation of SCR scaling factor was 400 MWe. Using the extrapolation
range of two as recommended by the commenter, this scaling factor can be
applied up to a boiler size of 800 MWe.
As shown in the study, the SCR capital cost for 800 MW is approximately
$48/kW. For the largest cyclone boiler of 1,200 MW, using the scaling
factor developed in the study, the capital cost projection is approximately
$44/kW-a cost reduction of only 8.3 percent. It is obvious that, even if it
were to be assumed that the SCR cost did not vary between 800 and
1,200 MWe boilers, there would be a negligible effect on the levelized
costs. This same analogy applies to all of the cost estimates presented in
the EPA's study.
2. The baseline NOX rates assumed for the oil- and gas-fired boilers are 0.3
and 0.25 Ib/MMBtu, respectively. Based on these levels, the study
assumes that gas reburn and SNCR can be applied to achieve
0.15 Ib/MMBtu NOX level. However, the national boiler population may
contain a significant number of units that produce NOX in excess of the
assumed rates. Also, reburn may be limited to 35 percent NOX reduction for
18
-------
those units where LNB and OFA are already present. SNCR may be
limited to 35 percent NOX reduction on sulfur bearing fuels. Subsequently,
it can be concluded that SNCR and gas reburn are not capable of achieving
the 0.15 Ib/MMBtu limit on all oil- and gas-fired boilers.
Response: The possibility of selected NOX controls not achieving the
0.15 Ib/MMBtu limit on some boilers has been clearly recognized in the
EPA study. As stated in Section 1.4, it has been noted that some boilers
may be controlled to levels higher than 0.15 Ib/MMBtu while others to levels
below this limit.
3. Additional information is requested regarding key assumptions for the
study: basis for 20 years remaining life, specification of space velocity for
SCR, design concepts assumed to provide load following capability for
SNCR, residence time assumed for the boiler population for rebum, and
specifics of calculating levelized cost and cost per ton from information
presented in the summary table.
Response: Responses to various comments raised are provided below.
• A remaining life of 20 years was assumed based on data reported in
published sources as well as the power industry trend towards
prolonged operation of older plants. High costs of new power plants
and moves towards deregulation are expected to maintain a competitive
environment that would make extended operations of older plants
economically attractive.
Use of a remaining life of 20 years is fully supported by published data.
In one study, an economic life of 20 years was used for SCR retrofits at
several plants in the Pacific Gas and Electric's system (Ref. 13). In
another SCR study, the average economic life used for several plants in
the Tampa Electric's system was 19.2 years (Ref .1).
• For all Group 2 boilers, sufficient information is provided in the study for
estimation of space velocities; flue gas flowrate and the temperature at
the economizer outlet are provided, and catalyst volume can be
estimated from the annual replacement requirement multiplied by the
3-year life. For Group 1 boilers, catalyst volumes have already been
provided. Flue gas flowrates and temperatures will be added in the next
revision of the report.
• Design concepts for load following with SNCR are covered in
Appendix A. Specifically, several levels of reagent injections are
included for each application to provide the capability for varying the
location of reagent injection with changes in the flue gas temperature
profile as the boiler load changes. A microprocessor-based control
19
-------
system with boiler load feedback from the plant controls is included with
each SNCR application. Automatic load following can be affected by
these controls.
• The EPA's study assumes that gas residence times in excess of
0.25 second will be available for gas rebum.
• The methodology used for estimating the levelized costs has been
described in detail in Appendix A. For each technology application,
estimates of consumables are clearly listed that have been used in the
estimation of levelized costs. Table 2-1 lists the economic factors for
these costs.
4. Recognize and account for an approximately 20 percent of the oil- and gas-
fired boiler inventory that will not be operating at RACT NOX limits of 0.3
and 0.25 Ib/MMBtu used in the study for these boilers, respectively.
Response: Refer to the response to Comment 2 above.
5. The Title IV rule for Group 2 boilers specifies NOX emission limits of 0.94
and 0.68 Ib/MMBtu for cyclone and cell burner boilers, respectively.
However, the study uses limits of 1.17 and 1.0 Ib/MMBtu for these boilers.
These higher rates increase capital cost and lower cost per ton of NOX
removed. Recommend use of Title IV limits for all boilers.
Response: The baseline NOX rates (1.17 and 1.0 Ib/MMBtu) used for
cyclone and cell-burner boilers represent the current average emission
rates for these boilers. EPA will revise the costs to reflect the Title IV,
Phase 2 NOX limits.
6. The study covers the entire national utility boiler population, which makes it
impossible to provide a detailed evaluation of each candidate boiler. EPA
should recognize the uncertainty regarding application of the results of this
study to the entire boiler population, and assign an appropriate margin of
error. At a minimum, the study should also include specific process design
(equipment lists/layout drawings) for each boiler category, and use of
power-law derived costs should be limited to only small capacity changes.
Response: EPA agrees that site-specific differences for the candidate
boilers would affect the design requirements and costs for each technology
application. However, the study was conducted with sufficient
conservatism built into the technology design bases and cost estimates to
address the variations expected between different boiler sites. EPA
believes that the margin of error sought by the commenter is already built
into this study. EPA also believes that use of the methodology and
information generated for the Title IV, Phase 2 NOX rule provided a credible
20
-------
and effective base for the study to uphold accuracy of the study results.
The concern regarding the power-law derived costs has already been
addressed in the response to Comment 1 above.
7. EPA should further define the boilers that served as "baseline" for the
analysis, disclosing site features, balance-of-plant equipment, and the
anticipated process impacts.
Response: EPA believes that the report includes sufficient information on
each boiler to inform the reader of the basis for cost estimates. The
information provided for each boiler includes design data, fuel analysis, and
description of boiler backend equipment. In addition, needs for existing
equipment modifications or replacements and other process impacts, such
as changes in fuel and auxiliary power consumption, are. clearly identified.
The comment is not clear as to what additional information is being sought
and for what purpose.
8. EPA should recognize a certain fraction of boilers feature site conditions
that present a greater challenge than assumed for this analysis, and thus
will incur higher costs. Accordingly, some boilers may require a premium
for further "scope adders" beyond that assumed by the Bechtel database.
Response: The commenter has not provided any example of requirements
beyond those represented by "scope adders" listed in the study. Without
specific reference to such requirements, it is not possible for EPA to
address them. It should also be noted that contingency allowances have
been included in each cost estimate for the study, which would provide
coverage for additional requirements at certain sites (refer to the discussion
in Attachment 1).
9. Even though the study claims use of the costing methodology in the EPRI
TAG, it has ignored the requirement for AFDC (allowance for funds used
during construction) included in the TAG. Since the construction period for
SCR is anticipated to be 1 year, ignoring AFDC amounts to understating
capital costs by nominally 5 percent.
Response: The construction period for SCR retrofits would depend on the
plant size as well as other site-specific conditions. For a relatively difficult
SCR retrofit at the 330 MWe Merrimack installation, the total project
schedule from authorization to proceed to completion of construction took
only 11 months (Ref. 11) This schedule covered design, fabrication,
delivery, and construction. Based on this experience, it is obvious that the
1 -year construction period quoted by the commenter is excessive.
Furthermore, even if 1 year is assumed to be applicable to certain special
situations, the 1993 EPRI TAG used for the study recommends no AFDC
allowance for construction periods of up to 1 year.
21
-------
10. A capital recovery factor of 0.127 is appropriate for a remaining plant life of
20 years, and it is in accordance with the recommendation in the EPRI
TAG. However, a remaining life of 20 years appears to be too high for the
national boiler population. Recommend using a remaining life factor of 17
to 18, which would also require increasing the capital carrying charge from
0.127 to 0.14.
11. Response: The response to Comment 3 above has already addressed the
comment on the remaining plant life. EPA fails to understand the
recommended increase in carrying charge from 0.127 to 0.14 for a change
in remaining life from 20 to 17/18 years. The commenter agrees with the
EPRI's carrying charge factor of 0.127 for a 20-year life. However, the 0.14
factor is well beyond what the EPRI TAG shows for a remaining life of 17 or
18 years (0.130 and 0.129, respectively). It is to be noted that even if
factors in the EPRI TAG for a 17- or 18-year life are used, they would not
make any appreciable difference in the EPA's cost estimates.
12. In general, higher SCR retrofit costs would be expected for an application
whose inlet NOX concentration is greater than for another application. The
cost estimates presented in the study do not follow this well accepted trend.
The reported costs for wall-fired boilers are the same as those for cyclone
or cell burner boilers, despite a significant difference in the inlet NOX
concentration for these boilers. This implies that the costs are under-
reported for boilers with higher NOX concentrations, especially the cyclone
boilers. EPA should recognize the potential for errors in capital cost due to
the selection of reference site and extrapolation from the Bechtel database
over generating capacity and process conditions. These results further
support UARG-suggested capital cost estimates (~ $18/kW higher).
Response: The EPA agrees that inlet NOX concentration would affect the
SCR catalyst volume requirement and, therefore, the associated capital
costs. However, the costs generated for one boiler category should not be
compared with those for another category. For such a comparison to be
valid, it would become necessary to use an exactly similar design basis for
all boiler categories, including fuel, excess air rates, boiler efficiency,
turbine configuration, main steam conditions, turbine heat rate, etc. In
reality, differences exist between boilers in all of these factors, each one of
which can affect the SCR design (and therefore the associated costs).
The differences quoted by the commenter in the reported cost estimates
are due to the differences in the selected boiler designs and in the cost
scaling factors. It is apparent that cost impacts of these differences have
compensated for the cost differences due to inlet NOX concentrations so
that the costs reported for 200 MWe cyclone, cell burner, and wall-fired
boilers are fairly close to each other. This also became possible because
the total installed costs for SCR do not vary by large amounts between
22
-------
different inlet NOX concentrations (or NOX reduction efficiency
requirements).
One source reporting the SCR costs for different NOX reduction efficiencies
shows a difference of approximately $2/kW between efficiencies of 70 and
80 percent and $4/kW between efficiencies of 80 and 90 percent (Ref. 14)
for a 200 MWe plant. The differences are even smaller for a larger plant,
which were used in EPA's estimates. As explained above, the effect of
even small differences in the boiler designs and scaling factors on the SCR
costs can easily equate to that caused by the differences in NOX reduction
efficiencies.
The objective of the EPA's study was to develop reasonable, representative
costs for SCR retrofits. For this purpose, several conservative design
assumptions were made and contingency factors were added to provide
costs that cover a wide variety of conditions expected to be prevalent at
various sites. Because of a variety of applications and design conditions, it
was not possible to maintain exactly the same amount of conservatism in
each cost. This led to apparent differences in technology costs when
compared based on a single system parameter, such as reflected in the
commenter's comment regarding the SCR costs for wall-fired, cell burner,
and cyclone boilers.
EPA believes that the design basis used in its study for SCR applications
has resulted in conservative costs. One factor used in this design basis
was to exclude consideration of a catalyst life management strategy. As
pointed out by other commenters on this report (ICAC and Black and
Veatch), use of this strategy in the study could reduce the initial catalyst
charge by as much as 20 to 75 percent and the overall catalyst
replacement cost by at least 65 percent. In light of these comments, EPA
cannot agree with the cost increase of $18/kW suggested by the
commenter.
13. NOX reduction efficiencies of higher than 80 percent are not appropriate for
coal- and oil- fired applications. Several factors may limit NOX reduction
capability to 80 percent, such as a need to maintain a strict limit (<3 ppm) of
ammonia slip, achieving uniform NOX and NH3 mixing, and managing
maldistribution in flue gas velocity.
Response: The commenter has failed to cite any references to support the
80 percent limit. EPA considers the concerns raised by the commenter to
be speculations. Application of SCR to achieve greater than 80 percent
and as high as 90 percent NOX reduction efficiency has been reported for a
large number of operating units (Ref. 15).
23
-------
14. EPA should recognize that the increase in complexity of SNCR technology
with greater generating capacity will negate any economies of scale, and
employ the capital requirement developed at 200 MWe for all capacities.
Response: In the EPA's study, the costs were developed for boilers larger
than 200 MWe. For example, the boiler size used for tangential boiler firing
coal was 348 MWe (refer to Table 3-1). The design basis used for SNCR
takes into consideration the items quoted by the commenter for larger
boilers. It is therefore not understandable why the economies of scale
would not apply to SNCR. Additionally, the scaling factors used in the
study do not result in significant cost reductions for larger boilers, similar to
the case described earlier (refer to the response to Comment 1 above).
15. EPA should recognize that the SCR costs projected are applicable to
natural gas firing. A cost premium should be included for applications to
sulfur containing fuel oil.
Response: The study has addressed the oil and gas firing applications for
SCR separately. Appropriate design factors have been used for these two
fuels, with consideration given to sulfur content in oil. The reported capital
costs for SCR application on oil are significantly higher than those on gas.
16. For both SNCR and reburn on gas and oil firing, the potential for requiring
increased complexity for either reagent or reburn fuel injectors with higher
capacity may negate any economies of scale. Thus, capital requirements
developed at 200 MWe should be applied to all capacities.
Response: Both boilers used for gas- and oil-fired applications were
350 MWe. (Refer to the response to Comment 14 above for further
justification of EPA's approach.)
17. The assumptions that coal reburn, SNCR, and natural gas reburn on coal
firing cannot provide sufficient NOX reductions to achieve 0.15 Ib/MMBtu
limit are appropriate.
Response: EPA acknowledges the commenter's agreement with the study
approach regarding coal and gas reburn and SNCR applications on coal-
fired boilers.
18. On coal and oil applications, SNCR should be limited to 25 to 30 percent
NOX reduction.
Response: SNCR technology has been demonstrated to achieve NOX
reduction levels of 30 to 50 percent (Ref. 6) which was used as a criterion
for this study. EPA cannot agree with an arbitrary limit of 25 to 30 percent
24
-------
set by the commenter (refer to the response to Comment 2 from Black and
Veatch).
19. The report should identify the following:
• Specified limit (if any) on conversion of S02 to S03
• Presence of an economizer bypass
• Fraction of the boiler population that must address unusual site
features, and significant equipment location
Response: The study was based on the methodology used in deriving the
NOX control costs for the Group 2 boiler rule. The Group 2 boiler report
was attached to this study as Appendix A. This report does address the
above comments: the design basis for SCR restricted S02 to S03
conversion to a maximum of 1 percent and an economizer bypass is
provided with each SCR application. The design basis for the study did not
assume any major equipment relocation as part of SCR retrofit. In light of
extensive published literature showing no need for such relocations, the
EPA considers this assumption to be valid (refer to the response to
Comment 3 from Nalco Fueltech).
25
-------
9.0 REFERENCES
1. Remer, et. al., "Air Pollution Control: Estimate The Cost Of Scaleup,"
Chemical Engineering, November 1994,
2. Cochran, et. al., 'The Effect of Various Parameters on SCR System Cost,"
Power-Gen' 93, Dallas, Texas.
3. Cichanowicz, et. al., "Factors Affecting SCR Capital Costs For Utility
Boilers," Handout, Meeting with EPA, December 20,1993.
4. Owens, et. al., "SCR Retrofit for NOX Control at a Wet Bottom Boiler," 1995
Joint Symposium on Stationary Combustion NOX Control, Kansas City,
Missouri.
5. Philbrick, et. al., "SCR System at Merrimack Unit 2," March 1996 ICAC
Forum, Baltimore, Maryland.
6. SNCR Committee, Institute of Clean Air Companies, Inc., "White Paper:
Selective Non-Catalytic Reduction (SNCR) for Controlling NOX Emissions,"
July 1994.
7. "Technical Assessment Guide," EPRI, Volume I, Revision 7, 1993.
8. Folsom, et. al., 'Three Gas Reburning Field Evaluations: Final Results and
Long Term Performance," 1995 Joint Symposium on Stationary Combustion
NOx Control, Kansas City, Missouri.
9. "Supplemental Comments for Group 2 Boiler NOX Emission Limits," by
J. E. Cichanowicz, prepared for UARG, June 1996.
10. "Nitrogen Oxide Limitation Study," by Carnot/Sargent & Lundy, prepared for
Tampa Electric Company, March 15, 1996.
11. "Energy Analysis: 1995-01," American Gas Association, January 13, 1995.
12. "Summary of Comments For the Draft Report Prepared for US EPA by
Bechtel/Cadmus," by J. E. Cichanowicz, prepared for UARG, July 13, 1996.
13. Veerkamp, et. al., "Evaluation of SCR as a NOX Control Option for Pacific
Gas and Electric," 1993 Joint Symposium on Stationary Combustion NOX
Control, Miami Beach, Florida.
14. "Evaluation of NOX Removal Technologies - Volume 1, Selective Catalytic
Reduction," by Bums and Roe Services Corp., Prepared for US DOE,
September 1994.
26
-------
15. SCR Committee, Institute of Clean Air Companies, Inc., "White Paper:
Selective Catalytic Reduction (SCR) To Abate NOX Emissions," October
1994.
27
-------
ATTACHMENT 1
CORROBORATION OF THE CAPITAL COST
FOR
MERRIMACK'S SCR INSTALLATION
-------
CORROBORATION OF THE CAPITAL COST
FOR
MERRIMACK'S SCR INSTALLATION
For the 330 MWe Merrimack installation designed for a 65 percent NOX removal
efficiency, the total capital cost was reported at $56/kW (Ref. 1, 2). Also, space
limitation at this site required addition of a significant amount of additional
ductwork and support steel for this retrofit. The baseline NOX emission for this
unit was also unusually high (2.66 Ib/MMBtu), thus requiring a relatively large
and expensive ammonia handling system.
The information available from Merrimack was used to corroborate the costing
methodology used in the EPA study. A comparison of the Merrimack cost with
the EPA-reported costs requires some adjustments in EPA's costs, because of
the differences in the design NOX reduction efficiency (65 versus 50 percent) and
the baseline NOx emission levels (2.66 versus 1.3 to 1.4 Ib/MMBtu). The
comparison strategy consisted of developing a capital cost based on design
criteria similar to Merrimack while using the EPA costing methodology (Ref. 3).
The capital cost developed with this approach could then be compared with the
actual Merrimack cost for validation purposes.
Table 1 shows an equipment list for the Merrimack installation. This list has
been prepared from published information (Ref. 1, 2) and information received
by EPA from the system supplier (Ref. 4). It should be noted that this installation
did not require some of the existing plant modifications that were included for the
boilers used in the EPA study (e.g., replacement of the existing draft fans and an
economizer bypass). However, being a moderately difficult installation,
Merrimack did require extensive flue gas ductwork to accommodate the SCR
within the existing setting and a bypass around the SCR reactor.
Table 2 shows the capital cost estimate for the Merrimack retrofit. This estimate
utilizes the same cost model that was used to generate costs for the EPA study.
As shown in Table 2, the total plant capital requirement is $68.53/kW, which is
higher than the actual cost reported for Merrimack of $56/kW. Thus, this
comparison confirms the conservatism used in the cost methodology utilized in
the EPA study.
It should be noted that the estimated cost in Table 2 is higher than the reported
cost by approximately $12.5/kW. This difference is greater than the combined
-------
value of the process and project contingencies (which is $11.3/kW). This
comparison supports the EPA's belief that the contingencies used in the EPA's
cost estimates can cover any additional costs that might, in rare cases, be
incurred at certain atypical installations because of site-specific factors.
-------
REFERENCES
1. B. Owens, et. al., "SCR Retrofit for NOx Control at a Wet Bottom Boiler",
1995 Joint Symposium on Stationary Combustion NOx Control, Kansas
City, Missouri.
2. Philbrick, et. al., "SCR System at Merrimack Unit 2", March 1996 ICAC
Forum, Baltimore, Maryland.
3. "Investigation of Performance and Cost of NOx Controls as Applied to
Group 2 Boilers", Draft Report, August 1995, EPA Contract No. 68-D2-
0168.
4. Conversations between R. Srivastava of EPA and S. Khan of Bechtel,
August 1996.
-------
Table 1
MAJOR EQUIPMENT LIST
MERRIMACK SCR
ANHYDROUS AMMONIA-BASED
BOILER SIZE: 330 MW
#
1
Item
SCR reactor
3
2
1
1 Lot
1 Lot
Anhydrous ammonia storage
Compressors
Electric vaporizer
Mixing chamber
Ammonia injection grid
Ammonia supply piping
1 Lot
Air ductwork
Description/Size
Vertical flow type, 1,615,350 acfm
capacity, equipped with a plate
type catalyst with 14,124 ft3 volume
placed in two layers, insulated
casing with two empty layers for
future catalyst addition,
sootblowers, hoppers, and hoisting
mechanism for catalyst
replacement
Horizontal tank, 250 psig pressure;
87.5-ton storage capacity
Rotary type, rated at 400 scfm and
10 psig pressure
Horizontal vessel, 450 kW
capacity
Carbon steel vessel
Stainless steel construction
Piping for ammonia unloading and
supply, carbon steel pipe: 4.0 in.
diameter, 600 ft long, with valves,
and fittings
Ductwork between air heater,
mixing chamber, and ammonia
injection grid, carbon steel, 400 ft
long, with two isolation butterfly
dampers, and expansion joints
-------
Table 1 (Continued)
MAJOR EQUIPMENT LIST
MERRIMACK SCR
ANHYDROUS AMMONIA-BASED
BOILER SIZE: 330 MW
#
1 Lot
Item
Sootblowing steam piping
1 Lot
1 Lot
Flue gas ductwork
modifications
SCR bypass
1 Lot
Ash handling modifications
1 Lot
Controls and instrumentation
Description/Size
Steam supply piping for the reactor
sootblowers, consisting of 200 feet
of 2" diameter pipe with an on-off
control valve and drain and vent
valved connections
Ductwork modifications to
install the SCR reactors, consisting
of insulated duct, isolation damper,
turning vanes, and expansion joints
Ductwork consisting of insulated
duct, 12'x24' double-louver isolation
damper with air seal, and
expansion joints
Extension of the existing fly ash
handling system modifications,
consisting of one slide gate valves,
one material handling valves, one
segregating valve, and ash
conveyor piping, 180 ft long with
couplings
Stand-alone microprocessor based
controls for the SCR system with
feedback from the plant controls for
the unit load, NOX emissions, etc.,
including NOX and ammonia
analyzers, air and ammonia flow
monitoring devices, and other
miscellaneous instrumentation
-------
Table 1 (Continued)
MAJOR EQUIPMENT LIST
MERRIMACK SCR
ANHYDROUS AMMONIA-BASED
BOILER SIZE: 330 MW
# Item Description/Size
1 Lot Electrical supply Wiring, raceway, and conduit to
connect the new equipment and
controls to the existing systems
1 Lot Foundations Foundations for the equipment and
ductwork/piping, as required
1 Lot Structural steel Steel for access to and support of
the SCR reactors and other
equipment, ductwork, and piping
-------
Table 2
RETROFIT CAPITAL COST ESTIMATE SUMMARY FOR
SCR MODIFICATIONS MERRIMACK BOILER
NOX Control Technology SCR
Boiler Size (MW) 330
Cost Year 1994
Direct Costs ($/kW):
SCR Reactors/Ammonia Storage 313
Piping/Ductwork 131
Electrical/PLC 3 1
Draft Fans - fj.O
Platform/lnsulation/Ertclosure 11
Total Direct Costs ($/kW): 48.6
Scope Adder Costs ($/kW)
Asbestos Removal 0.0
Transformer 0.0
Air Heater Modifications 0.0
Boiler System Structural Reinforcement 0.0
Total Scope Adder Costs ($/kW): 0.0
Total Direct Process Capital ($/kW): 48.6
Indirect Costs:
General Facilities 5.0% 2.4
Engineering and Home Office Fees 10.0% 49
Process Contingency 5.0% 2.4
Project Contingency 15.0% Q 7
Total Plant Cost (TPC) ($/kW): 67.1
Construction Years 0.0
Allowance for Funds During Construction 0.0
Total Plant Investment (TPI) ($/kW): 67.1
Royalty Allowance 0.00% Q.O
Preproduction Cost 2.00% 13
Inventory Capital Note Q. 13
Initial Catalyst and Chemicals 0.00% Q.O
Total Plant Requirements ($/kW): 68.53
NOTE: Cost for anhydrous ammonia stored at site.
-------
ATTACHMENT 2
COPIES OF COMMENTS RECEIVED FROM PUBLIC
-------
Department of Energy
Pittsburgh Energy Technology Center
P.O. Box 10940
Pittsburgh, Pennsylvania 15236-0940
May 20, 1996
Ravi Srivastava
U.S. EPA
Mail Drop 6204J
401 M Street SW
Washington DC 20460
Subject: Comments on EPA Draft Report, "Cost Estimates for Selected
Applications of NOX Control Technologies on Stationary Combustion
Boilers"
Dear Mr. Srivastava:
Attached are our comments on the subject report, along with marked-up pages.
Please feel free to discuss these comments with us.
Sincerely,
3W
Tfrttwr^ Baldwin
Program Coordinator
NOX Control Technology
Office of Clean Coal Technology
A-*/VNJY\jsE> Vl •
Dennis N. Smith
Technical Analyst
Office of Clean Coal Technology
Enclosure
-------
COMMENTS ON DRAFT REPORT TITLED
"COST ESTIMATES FOR SELECTIVE APPLICATIONS
OF NOX CONTROL TECHNOLOGIES
ON STATIONARY COMBUSTION BOILERS," Dated March 1996
This draft report was prepared by Bechtel Corporation for the U.S.
EPA. Attached to the draft is Appendix A, titled "Investigation of
Performance and Cost of NOX Controls as Applied to Group 2 Boilers,
Draft Report," dated August 1995. The present review covers both
the new text and Appendix A.
We had commented previously (August 1995) on the July 1995 draft of
Appendix A. The current version reflects most of the changes we
proposed at that time. As such, it represents a great improvement
over the earlier draft. Although we feel it would have been pref-
erable to reorganize Appendix A along the lines we suggested previ-
ously for easier access to the material, it appears that this is
not likely to happen. On the one hand there is still considerable
duplication, while on the other hand it is necessary to search in
more than one place for information on a given topic.
A related problem is the numbering of the pages. Taken out of con-
text, one cannot readily determine where a page numbered, for ex-
ample, 3-2, belongs, since both the new main text and Appendix A
follow the same numbering system. It would be helpful to insert an
A in front of each page number in Appendix A.
Adding to the complication is the existence of two Appendix A's.
As stated above, Appendix A is attached to the new document dated
March 1996. Appendix A, however, also contains an Appendix A con-
sisting of the EPA database for Group 2 boilers. A simple way to
correct this might be to rename the EPA database Appendix AA, and
number the pages accordingly. There are only a few pages in the
-------
EPA database, and there is little if any crossrreferencing to this
material in the remainder of the text.
Despite these shortcomings, the August 1995 version of Appendix A,
along with Appendices B and C, provides useful information on a
variety of NOX reduction technologies, and should be helpful in the
rule making process as applied to Group 2 boilers.
Scope of Main Report
A more important issue, however, is the scope of the March 1996
report, "Cost Estimates for Selected Applications of NOX Control
Technologies on Stationary Combustion Boilers," referred to in the
present review as the main text. As stated in the second bullet
item in Section 1.1, p. 1-1, one of the objectives of the present
study is to develop costs for NOX control technologies applicable
to tangentially-fired and dry bottom wall-fired boilers, which are
Group 1 boilers. However, Appendix A, by definition, deals only
with Group 2 boilers. Thus one of the shortcomings of the main
text is that there is no documentation for the treatment of Group
1 boilers.
f I Moreover, the first bullet item in Section 1.1, p. l-l, addresses
: more stringent NOX controls than had been promulgated in Phase I,
I namely an emissions level of 0.15 lb/106 Btu, again without back-
ground documentation for the costs of achieving this degree of
control. Furthermore, no basis is given for the selection of the
0.15 lb/106 Btu target.
The problem of overall orientation becomes critical on p. 2-1,
where the first bullet item near the bottom of the page refers to
the various components of the systems studied, including low-NOx
burners (LNB), overfire air (OFA), and gas recirculation fans. The
statement, "Where applicable, the study burners are already equip-
ped with low-NOx burners" is unclear. What constitutes cases where
-------
i
LNBs are applicable? Where are they not applicable? If such com-
bustion modifications are already in place, is the application of
other technologies not "possible,11 as stated in the draft, or not
economically justified?
Related to these questions of scope is the whole issue of including
OFA as a combustion modification technology. The EPA was severely
criticized in the past for assuming that OFA should be included
with LNBs for NOX control. By appearing to link these technologies
now, will the EPA be subjected to the same criticism again? Thus
this potentially very significant bullet item is misleading as now
.Worded; it needs clarification and amplification.
Throughout the new text, technologies for NOX removal are selected
for a given boiler application without much explanation. Presum-
ably other technologies would also be appropriate in some instanc-
es, and in fact some of these are analyzed in the Appendices.
Without additional information, it would appear that the report is
biased in favor of SNCR. While we do not necessarily disagree with
the choices made, we feel there is insufficient discussion to in-
form the reader regarding the chosen technologies. If certain
technologies were considered and rejected, reasons should be given.
If, on the other hand, time or budget constraints precluded exami-
nation of other technologies, this should be stated.
Because of questions such as these, it would be helpful to include
an introductory paragraph putting the whole document into perspec-
tive and orienting the reader as to the scope and purpose of the
current report.
Specific Comments
Economies Methodology Appendix A ("Investigation of Performance
and Cost of NOX Controls as Applied to Group 2 Boilers," dated
August 1995) now incorporates the correct method of developing
-------
process economics. This treatment includes all the relevant com-
ponents of levelized costs, as discussed in our review of the
earlier draft. However, the new main text refers to calculations
for levelized costs both with and without the capital charge com-
ponent, although it is not clear where these calculations are
presented other than in Table 1-2.
All of the costs for Group 2 boilers reported in Appendix A cor-
rectly include capital charges in the levelized costs, as presum-
ably do the costs given in Tables 1-3 and 1-4 of the main text.
Unless there is some compelling reason to the contrary, we recom-
mend eliminating all references to two versions of levelized costs.
We would be glad to discuss this question with the authors of the
document.
Tables 1-3 and 1-4 contain some cost estimates for which there is
no supporting documentation. This presents a problem for readers
who wish to verify or adjust the figures based on alternative input
data. Moreover, to the extent that it is feasible to find cases
documented in the Appendices, the figures do not seem consistent.
For example, Table 1-3 shows a levelized cost of $695/ton for SCR
as applied to a cyclone boiler at a capacity of 200 MW and a 65%
:apacity factor. Figure 4-21 in Appendix A shows a value of about
>625/ton at the same capacity and capacity factor. What are the
Reasons for the difference? If these cases are not meant to be
compared, what is the difference in methodology? Which figures
should the reader use in planning a NOXcontrol strategy?
Technology Selections The Table of Contents for Appendix B (p. B-
i) is quite useful, since it presents at a glance the NOX reduction
technologies evaluated for the separate boiler types. Thus, for
cell-burner boilers, the technologies are plug-in low-NOx burners
and non plug-in low-NOx burners. For cyclone boilers, the technolo-
gies are coal rebuming, gas returning, SNCR, and SCR. It would be
helpful if the NOX control strategy were also given for wet-bottom
-------
boilers and vertical, dry-bottom boilers (SNCR in both cases) . The
reader could thereby quickly find a discussion of any particular
technology of interest.
Capital Costs The capital cost for the SCR process given in Table
B4-18 (p. B4-72) is of the correct order of magnitude, but the pre-
sentation is potentially misleading in that there is no cost given
for the catalyst installed in the reactors. While the cost for
reactors and ammonia storage is given as $22/kW (for a 150 HW
plant) , our work on SCR has shown that the major portion of that
cost is in fact the initial catalyst inventory, .which is a large
figure due to the high unit price of the catalyst. While the
bottom line would not change, it would be more accurate to show the
catalyst inventory as a separate line item at the appropriate place
in Table B4-18 and reduce the capital cost for reactors/ammonia
storage by the same amount.
As pointed out above, background information on NOX control for
Group 1 boilers is missing. This includes capital costs.
In Table l-f> the second column lists capital costs in $. Should
this be $/kW?
Levelized Costs Appendix B now presents a breakdown into the major
components of levelized costs, which had been missing previously.
While the format does not show sufficient detail to permit verifi-
cation or adjustment of the individual figures, it is helpful to
have even this much of a breakdown, showing at a glance the rela-
tive contributions of the several cost components. Note that the
tables giving levelized costs, including the one on p. B3-6 and
continuing throughout the document, would be more readable if the
columns under $/ton NOX were right justified.
On p. 4-3 of Appendix A, item 6 refers to the fact that the level-
ized cost estimates "consider" several factors. This sounds too
-------
casual for a clearly defined calculation. A more definitive word
such as "include" is recommended since it would be entirely unam-
biguous in this context. Otherwise the reader cannot be sure that
all the factors "considered" are actually included, as indeed they
must be in order to perform the calculations correctly.
On p. 1-2 of the main text, last bullet item, there is reference to
economic factors "reported" in the EPRI TAG. As we pointed out
; previously, any economic factors given by EPRI are only examples
j and are not meant to be recommendations. In the context of the
\ present report, to avoid any implication that these numbers are
i j anything other than representative values, we suggest simply using
\ the term "listed" or "given."
On p. B2-8, the first paragraph in Section 2.4.2 contains a sen-
tence about estimating the capital-related components of level ized
costs "along with the predicted capital costs for the boilers with-
in the corresponding range." This statement is very confusing.
The economics of NOX reduction technologies are independent of the
cost of the boiler on which the technology is installed. If some-
thing else is meant here, it should be clarified. If not, the
statement should be eliminated.
Capacity Factor In the main text, Section 1.3 mentions two assumed
values for capacity factor: 65% and 27%. While the economic calcu-
lations reported in Tables 1-3 and 1-4 appear to handle these two
capacity factors properly in terms of $/kW and $/ton of NOX re-
moved, it is not clear how the mills/kWh figures in Table 1-2 re-
flect these different assumptions. Even if the NOX control tech-
nology is operated only five months out of the year, the power
plant generates electricity at its normal rate for the entire year,
To avoid possible misinterpretation or confusion on the part of
utilities seeking guidance from this document, the basis for the
mills/kWh calculations should be checked carefully and explained in
greater detail in this section.
-------
Sensitivity Analyses As we pointed out previously, the sensitivity
analyses for critical process variables given in Appendix B are
useful in principle but are, in fact, of limited value because the
conclusions drawn are a function of the ranges of the variables
investigated rather than reflecting their true significance.
For example, it is stated in several places that reagent cost is
not an important variable in SNCR, although it can easily be seen
that reagent cost is a large component of SNCR levelized cost.
Likewise, it is stated in a number of places (for a variety of
technologies) that capital cost variation has a minor impact on
capital (and levelized) costs. This is fallacious reasoning, and
results from the fact that only a narrow range of capital costs
(±5%) was investigated. Similarly, catalyst replacement cost is a
significant component of SCR economics, as shown on p. B4-24, but
the statement on p. B4-25 does not support this obvious fact.
The impact of any variable can be made unimportant if a narrow
enough range is evaluated. In the case of reagents, it is of
course reasonable to explore only the range of prices likely to be
encountered. A more meaningful approach would be to say that an x
percent range of reagent prices was selected for the sensitivity
calculations, representing a realistic degree of variation, and
that within these limits, the effect on levelized cost was found to
be y percent.
In principle, cost sensitivity effects should be compared on a rel-
ative basis. What is important is the percent increase in process
capital or levelized cost for a given percent increase in a par-
ticular process parameter. If costs go up 5% for a 5% increase in
a parameter, then that parameter is important even though a 5% in-
crease is small.
As pointed out previously, the graphs presenting the sensitivity
analyses, showing the variation in costs at two levels of a par-
-------
ticular variable, lack labels on the two individual curves. Thus,
a reader looking at the graphs is forced to refer to information
located elsewhere in the report to determine the values of the
variables being studied. This is true for Figures B3-6 through B3-
11, B3-17 through B3-22, B4-6 through B4-13, B4-19 through B4-30,
B4-36 through B4-45, B4-51 through B4-61, B5-6 through B5-15, and
B6-6 through B6-15. Adding labels to the individual curves would
be a simple matter and would greatly increase the usefulness of
these graphs and of the document as a whole.
SNCR Process The section on chemical type and stoichiometry, p. C-
39, is much improved over the previous version. However, it is
still not clear why the use of urea has an advantage over ammonia
in large boilers. Isn't the same stoichiometric effect valid for
all boiler sizes? Are there other factors related to boiler size?
Definitions Since the main text deals with both Group 1 and Group
2 boilers, it would be helpful to explain these terms. Group 2
boilers are defined by reference to the Appendices, but there is no
definition of Group 1. A simple way to do this would be to insert
\ subheadings in Table 1-1, showing both groups of boilers.
Likewise, there are no references to Phase I or Phase II of the NOX
control implementation plan, although the target NOX levels can be
related to the two phases.
I Throughout the report, costs are variously given in mills/kWh and
/ mils/kWh. The dictionary definition calls for the use of mills to
represent 1/1000 of a dollar. We recommend making the appropriate
changes, using a search function in your word processing program.
We did not mark every page where this inconsistency occurs. Note
also that the captions on some of the figures use the term mils and
\should be corrected if possible.
-------
On p. B3-2, the word "mils" is used in reference to pulverizers.
Clearly, this needs to be corrected to say "mills."
! Significant Figures In Tables 1-3 and 1-4, the capital costs show
too many significant figures. Considering the accuracy of the cal-
culations, two significant figures would be maximum. Making this
• change would also make these tables more readable. Likewise, at
I various points in the text, boiler efficiencies could be reported
/ to the first decimal; the tables from which these values are
derived should, of course, continue to show two decimal accuracy.
' Miscellaneous Items in Main Text In Section 2.1, first paragraph,
there is reference to Table 1-2. This should be Table 1-5. The
second sentence states that this table shows variations in NOX
; reduction effectiveness on a site specific basis. This is not
quite true. The table shows a range of effectiveness for each
; category. The text would be more clear and accurate if it said,
: "As shown in this table, the NOX reduction effectiveness for each
; technology varies over a significant range. These variations are
i a result of site specific factors."
I Section 2.3, second bullet item, refers to Section 2.4.1 of Appen-
•; dix A. There is no such section. /
Other editorial comments are noted on the marked-up sheets.
-------
1.0 PROJECT OVERVIEW
This report presents the results of a study conducted by Bechtel to develop costs for NOX control
technologies for coal-, gas-, and oil-fired boilers. The types of boilers for each fuel along with
the size range and baseline NOX emission rate for each boiler type were identified by the United
States Environmental Protection Agency (EPA), as shown in Table 1-1.
The technical and economic evaluations conducted for this study used a consistent methodology
to develop costs for various NOX control technology applications. The costs are therefore
comparable between different boiler types and sizes.
1.1 Project Purpose
The primary objectives of this study were to:
• Develop costs for the NOX control technologies with a capability to reduce NOX emissions
from the baseline NOX rate to 0.15 Ib/MMBtu for each study boiler
• Develop costs for the NOX control technologies with a capability to provide substantial NOX
emission reductions for the dry-bottom tangential and wall-fired boilers burning coal beyond
those required under 40 CFR Part 76
12 Major Results
The capital and leveUzed costs for each technology case are presented in the figures that are in-
cluded at the end of this report The major costs from these figures are summarized in the fol-
lowing tables:
• Table 1-2 presents the fixed and variable costs for a 200 MW boiler for each technology
application. The variable costs are reported for bom the 27 and 65 percent capacity factors.
| Two types of variable costs have been included: one containing the carrying charges for the
I capital expenditure and the other without this carrying charge (as reported in EPRI's TAG).
In addition, Table 1-2 also provides a mathnn?TJeal relationship to facilitate estimation of the
capital cost for a given boiler size (MW).
• Tables 1-3 and 1-4 present the capital (S/kW) and leveUzed ($/ton of NOX removed) costs for
two selected sizes of boiler installations for each NOX control technology (for bom 0.15
Ib/MMBtu and substantial reduction cases). These costs are reported for both the 27 and 65
percent capacity factors. Also provided are references to the figures from which these costs
have been obtained.
13 General Approach to Technical and Cost Analyses
The overall approach for both the technical and cost analyses was based primarily on the meth-
odology utilized in a previous Bechtel study that involved evaluation of NOX control tech-
1-1
-------
oologies for the Group 2 boilers. A copy of the previous study is provided as Appendix A to this
report.
The major elements of the project approach and the areas where the approach differs from the
previous study are as follows:
• An evaluation of the commercially available NOX control technologies was made to deter-
mine feasibility for meeting the aforementioned project objectives. Table 1-5 lists these
technologies along with their NOX reduction effectiveness and applicability to each study
boiler type. The data presented in Table 1-5 were based on published information on a
variety of technology applications (References 1 through 17).
Based on the above evaluation, the following technologies are considered in this report:
• The selective catalytic reduction (SCR) technology was selected for its capability to
provide NOX reduction to the 0.15 Ib/MMBtu limit for all study boilers. For the oil- and
gas-fired boilers, both the selective noncatalytic reduction (SNCR) and gas rebuming
technologies were also selected for the same purpose.
• The SNCR, gas rebuming, and coal rebuming technologies have been found to have a
capability to provide substantial NOX reduction for the tangential and wall-fired boilers
burning coal. Of these, the SNCR technology was selected for evaluation for this
project Costs of gas and coal rebuming applications on Group 2 boilers have been
examined in detail in the previous Bechtel study (Appendix A).
• The technical and economic evaluations were conducted on representative boiler installations
for each boiler category identified for this project. The design data for the representative
boiler installations were developed from Bechtel's in-house database.
• Both capital costs (S/kW) and levelized costs (mils/kWh and S/ton NOX removed) were
developed for the applicable boiler size range for each technology application.
• The capital cost estimates were developed by factoring from the 1994 cost data generated in
the previous Bechtel study (Appendix A) for each NOX control technology. The new esti-
mates were not based onudetailed major equipment lists, as developed hi the previous study.
Instead, appropriate powcifffactors representing the general industry practice were applied to
the existing costs to obtain costs for this project. This method took into consideration the
differences in the overall system size and capacity between each technology application for
this project and the corresponding application in the previous study.
• All new costs were developed in 1995 dollars. The latest available Chemical Engineering
cost index for September 1995 was used to adjust the estimated 1994 costs to 1995.
The levelized costs were based on the economic factors reported in the 1993 EPRI TAG
(Reference 18). They were developed using a constant dollar approach. Other economic
assumptions were the same as shown in Appendix A and detailed in Section 2.0.
2»S oaitetfMCeB.EB.NOx
1-2
-------
TABLE 1-1
STUDY BOILERS AND BASELINE NOX EMISSIONS0'
NOTE
Boiler Type
'Dry bonom, wall-fired
Dry bottom, tangentialiy
fired
/^ ~
&l***6 "Z—
Cell'
Cyclone
Wet bottom
Dry bottom, vertically
fired
Size Range, MW
30-1300
33-952
200-1300
25-1200
25-800
25-300
Fuel
Coal
Gas
Oil
Coal
Gas
Oil
Coal
Coal
Coal
Coal
Baseline NOX Rate
Ib/MMBtu
0.50'(Title IV limit)
0.25"
030"
f
0.45vfTitle IV limit)
0.25*
030 J
0.8-1.5
(1.00 average)
0.8-1.9
(1.1 7 average)
0.7-1.7
(1.13 average)
0.85-1.1
(1.08 average)
For Group 1 boilers, the baseline NOX rates are the currently allowable emission limitations
under 40 CFR Pan 76. For Group 2 boilers, the baseline NOX rates represent the average
uncontrolled NOX rates, per boiler type, as presented in Appendix A to "Investigation of
Performance and Cost of NOX Controls as Applied to Group 2 Boilers," August 1995,
prepared for the U.S. EPA.
at MftSad*Ceae« NO.
-------
TABLE 1-2
CAPITAL AND O&M COSTS(IM2UJ)
Case"'
COAL- TANOEN- SCR
COAL-TANGEN-SNCR
COAL. WALL- SCR
COAL-WALL-SNCR
COAL- CELL- SCR
COAL- CYCLONE- SCR
COAL- WET DOT TOM- SCR
COAL- VERT-FIRED- SCR
GAS-SCR
GAS- REBIIRN
CAS- SNCR
Oil.- SCR
OIL- REBURN
OIL- SNCR
Capital Cost ($)
66.824»(200/MW)A0.3S
!5.55l»(200/MWr0.577
69.382»(200/MWr0.35
l7.3ll*(200/MWr0.577
69.2I7((200/MW)A0.324
69.55«(200/MWr0.26l
70.57I»(200/MW)A0.296
67.067*(200/MWT0.39I
27.483«(200/MW)A0.35
I9.025M200/MW)A0.357
9.433«(200/MWr0.577
39.975«(200/MWr0.35
22.298*(200/MW)0.357
I0.638M200/MWT0.577
Fixed Cost for
200 MW
1
0.23
1.04
0.26
1.04
1.04
1.06
1.01
0.41
0.29
0.14
0.6
0.34
0.16
Variable Cost
for 200 MW
w/o Capital
65% Capacity
Factor
1.04
0.99
l.tl
1.02
1.35
1.38
1.41
1.33
0.17
003
0.42
0.36
051
0.58
Variable Cost
for 200 MW
(w/o Cnpllal)
27% Capacity
Factor
2.2
0.99
2.33
1.02
2.67
2.72
2.76
2.6
0.28
0.03
0.42
0.7
. 0.51
0.58
Variable Cost
for 200 MW
(w/Capllal)
65% Capacity
Factor
2.53
1.34
2.66
1.41
2.89
2.93
2.98
2.83
0.79
0.45
0.63
1.25
1.01
0.82
Variable Cost
for 200 MW
(w/Capllal)
27% Capacity
Factor
5.78
1.82
6.05
1.95
6.38
6.44
6.5-1
6.19
1.75
1.04
•
0.92
2.R4
1.7
1.15
MQIES
I. Fixed costs are reported in $/kW-yr. The variable costs ore reported In niil/kWh.
/ 2. The variable costs are reported both with and without the carrying charges Tor the capitnl costs. As per the El'RI's TAG, the varinblc costs do not include cnrrying
' I charges. Also, Tor this report, the costs associated with the changes in the fuel consumption rates because of (lie retrofit have been included in the varinblc costs.
* I EPRI does not include fuel costs In the variable cost component.
3. The capacity factor reflects the annual usage Tor which the NO, control technology is In operation.
4. Where the boiler firing type is not mentioned, the case applies to both the wall-fired and tangenllally fired boilers.
-------
TABLE 1-3
SUMMARY OF RESULTS
NOX CONTROL TECHNOLOGIES ACHIEVING 0.15 LB/MMBTU LIMIT
Boiler(1)
TN
. WF
CELL
CYC
WB
VF
Fuel
Coal
Coal
Coal
Coal
Coal
Coal
NOX
Control
SCR
SCR
SCR
SCR
SCR
SCR
Boiler
Size,
MW
200
930
200
1030
200
1030
200
1030
200
730
70
200
65% Capacity
Factora)
S/kW
ff 0^ 4
^UTOZw
.39J82-
69 3*T
39JT
69.32"
^fTf?T^J
49^7<
45>r
3&S?7t
48 JW
10LW
67jyr
S/Ton
r 1935
1439
1670
1226
801
624
695
536
733
572
907
750
27% Capacity
Factor™
S/kW
£&£36
39JBT
692*
39^
69J22"
A&t4(
&&3K
453*
J&£?7
48J9T
101.W
67^T
S/Ton
?4427
3238
3815
2748
1775
1351
1536
1125
1616
1231
2032
1654
Figured
3-U,5
3-11,13,15
3-21,23,25
3-26^830
3-3U3^5
3-36^8,40
NOTES
1. The legend for the symbols used is:
CYC Cyclone-fired
TN Tangential
VF Vertically fired, dry bottom
WF Wall-fired, dry bottom
WB Wet bottom
2. The capacity factor reflects the annual duration for which the NOX technology is in operation.
3. The cost data presented are taken from the curves shown in the referenced figures included in
this report.
CnNO»
-------
TABLE 1-3 (Continued)
Bofler(1)
WF,TN
WF,TN
WF,TN
WF?TN
WF.TN
WF,TN
Fuel
Gas
Gas
Gas
Oil
Oil
Oil
NOX
Control
SCR
Rebum
SNCR
SCR
Rebum
SNCR
Boiler
Size,
MW
200
930
200
930
200
930
200
930
200
930
200
930
65% Capacity
Factor®
S/kW
21JHT
16JW
19J93"
l&99n
9Af
3663.1
39£ft4
23.34-
22.33"
&MI3
4fc63/«
4££*
S/Ton
2142
1429
1250
748
1632
1272
•> 2263
1571
1776
1384
1407
r 1147
27% Capacity
Factor0'
S/kW
27.4T
16JXT
\9jyy
10#9a
9.4?
^46i;
49^fr^
23>r
22je^
i*» p.fif
16^3- «
4^€^/
S/Ton
4802
3091
2910
1706
2455
1592
5151
3492
3073
2122
2026
1402
Figures'3'
4-13,5
4-6,8,10
4-11,13.15
5-1,3,5
5-6,8,10
5-11,13,15
X
X
y
X
x
X
NOTES
1. The legend for the symbols used is:
CYC Cyclone-fired
TN Tangential
VF Vertically fired, dry bottom
WF Wall-fired, dry bottom
WB Wet bottom
2. The capacity factor reflects the annual duration for which the NOX technology is in operation.
3. The cost data presented are taken from the curves shown in the referenced figures included in
mis report
-------
TABLE 1-4
SUMMARY OF RESULTS
COST OF SNCR APPLICATIONS ON DRY-BOTTOM WALL- AND
TANGENTIALLY FIRED BOILERS
Boiler(l)
TN
WF
Fuel
Coal
Coal
NOX
Control
SNCR
SNCR
Boiler
Size,
MW
200
930
200
1030
65% Capacity
Factor®
S/kW
15.5$;
6.4/
\13ff.
S/Ton
1378
1150
1210
988
27% Capacity
Factor®
S/kW
\*xr
6.4/
n&r
6.8/8^
S/Ton
1921
1377
. 1720
1186
Figures0'
3-6,8,10
3-16,1820
NOTES
1. The legend for the symbols used is:
TN Tangential
WF Wall-fired, dry bottom
2. The capacity factor reflects the annual duration for which the NOX technology is in operation.
3. The cost data presented are taken from the curves shown in the referenced figures included in
this report
-------
2.0 METHODOLOGY AND GENERAL ASSUMPTIONS
The methodology and assumptions used in selecting the applicable NOX control technologies and
conducting the technical and economic evaluations for this project are detailed in this section.
2.1 Technology Selections
Table 1 ^categorized7 the commercially available technologies and their NOX control potential
for various boiler types. As shown in mis table, the NOX reduction effectiveness varies de-
pending on the site-specific conditions for any given application.
The study criteria define th^Jjaseline^NOx rates for the dry-bottom wall-fired and tangential
boilers burning coal to be &4-5 and &S Ib/MMBtu, respectively (these rates being required by 40
CFR Pan 76). The baseline NOX rates for the same boilers on oil and gas are defined as 0.3 "and
0.25lb/MMBtu, respectively, because these rates currently are being achieved on gas- and oil-
fired boilers. It is assumed that these NOX rates correspond to boilers equipped with low-NOx
burners only (no overfire air ports).
The above assumption implies that full credit can be taken for the NOX reduction potential of the
technologies (such as gas rebuming) utilising overfire air ports. Without this assumption, appli-
cation of these technologies to boilers with existing overfire air ports would be possible only if
the ports are replaced with the new ports associated with the technologies. Deletion of the
existing ports would have a corresponding impact of increasing the baseline NOX levels, thus re-
quiring a higher NOX reduction to achieve 0.15 Ib/MMBtu. •+-
As per the study criteria, the NOX reduction efficiencies required to meet the 0.15 Ib/MMBtu Jbr
the gas- and oil-fired boilers are 40 and 50 percent, respectively. For coal-fired boilers, these
efficiencies range from 66.67 to 87.44T percent
7
-------
either to achieve proper NOX reductions or to install the technology components. These fac-
tors include lack of sufficient space to install the rebum fuel injectors (or burners) and over-
fire air ports, lack of proper residence times, and unavailability of natural gas.
• The effectiveness of the SNCR technology can vary from 30 to 50 percent. Based on the
NOX reduction needs (0.15 Ib/MMBtu) of the study boilers, this technology can be applied
only to the gas- and oil-fired boilers. Similar to gas rebuming, this feasibility may be subject
to site-specific factors. The most important aspect of SNCR is the availability of a proper
residence time within the boiler in a required temperature zone, which varies with the type of
SNCR system used (ammonia- or urea-based). It is recognized that such residence times may
not be available in all gas- and oil-fired boilers.
• The NOX reduction needs of all study boilers fall within the potential effectiveness range (80
to 90 percent) for the SCR technology, which is therefore considered feasible for all of these
boilers.
Even for the SCR technology, the NOX reduction rates required for the cell, cyclone, wet
bottom, and vertically fired boilers are relatively high. Such rates would require significantly
large amounts of catalyst. Other concerns, such as excessive SO3 conversion rates, may also
be applicable in some specific retrofits.
In some cases, the duty on the SCR systems could be reduced by applying more than one
NOX control technology. For instance, hybrid systems using SNCR and SCR could be used.
or SCR could be applied with combustion controls (applicable to cell, wet bottom, and verti-
cally-fired boilers). These applications are considered outside the scope of the study.
• Based on the above analyses, the technologies selected for meeting the 0.15 Ib/MMBtu limit
include SCR for all boiler categories and SNCR and gas rebuming for gas- and oil-fired
boilers only. Similarly, SNCR has been considered for achieving substantial NOX reduction
(50 percent) for the wall-fired and tangential boilers burning coal.
22 Technical Evaluations
The methodology for the technical evaluations is essentially the same as used in the previous
study (Appendix A, Section 2.0 of Appendix B). The highlights of mis methodology are as fol-
lows:
• All design details pertaining to the representative boilers hi the cyclone, cell, wet-bottom, and
vertically fired categories are the same as shown in the previous study.
• Since the tangential and wall-fired boilers burning coal, oil, or gas were not included in the
previous study, design details of representative boilers for these categories have been specifi-
cally developed for this project from the Bechtel in-house database. In the case of each boiler
category, the evaluations are performed using one representative boiler. It is assumed that
boiler design parameters vary in^ direct proportion to the boiler size.
2-2
-------
3.0 COAL-FIRED PLANT ASSUMPTIONS AND RESULTS
This section summarizes the technical and economic evaluations conducted for the coal-fired
boiler applications of NOX control technologies.
3.1 Tangential Boiler Applications
The NOX control technologies evaluated for this boiler type include SCR and SNCR. The design
data for the representative boiler selected for this evaluation are shown in Table 3-1 . This boiler
is a balanced draft, forced circulation, reheat, single furnace boiler. It has four windboxes
located along the four comers of the furnace. There are a total of 20 coal burners, five per
comer. The boiler serves a 348 MW steam turbine generator and is equipped with two 50-per-
cent-capacity forced draft fans, two 50-percent-capacity induced draft fans, and an electrostatic
precipiiator for removing dust from the flue gases exiting the boiler.
3.1.1 SCR Evaluation
The following major criteria and assumptions have been followed in evaluating the SCR tech-
nology for the coal-fired tangential boilers:
y s
• The SCR system is designed to reduce NOX emission from a baseline level of 0.45 Ib/MMBtu
to the required limit of 0.15 Ib/MMBtu.
• Anhydrous ammonia is va^-nA as a reagent for the SCR system.
• The system is designed for an ammonia slip of 5 ppm.
• A 14-day storage is provided at the plant site for anhydrous ammonia. This storage capacity
is based on a full-load operation of the boiler.
• It is assumed that the existing plant setting allows installation of the SCR reactors between
the economizer and air heater without a need to relocate any major structure or equipment
• The operating life of the SCR catalyst is assumed at 3 years. A catalyst life management
strategy is not used for this evaluation. It is also assumed that no appreciable difference in
the catalyst life occurs when the plant is operated at low capacity factors. This assumption
results in conservative cost estimates, since it is expected that a lowrcapacity factor may re-
sult in a net catalyst life increase. ^-t_
• Other general SCR system design details, assumptions, and impacts on the existing equip-
ment outlined in Appendix A (Section 4.5 of Appendix B) also apply to this case.
The SCR technology is a postcombustion technology, in which the reagent is injected into the
flue gas stream at the economizer outlet upstream of the catalyst reactor. As such, SCR tech-
nology has no direct impact on the boiler performance. The boiler parameters shown in Table
3-1 would remain unchanged following a SCR retrofit. However, such a retrofit would impact
3-1
-------
Injection of the urea solution within the boiler does have an impact on the boiler performance.
because of the heat loss associated with the moisture content of this solution. This heat loss
causes a slight reduction in the boiler efficiency, resulting in increased fuel flow, ash generation.
and combustion air and flue gas flow rates. The overall impacts of the SNCR system retrofit on
the study boiler are as follows:
X, • The boiler efficiency reduces from 88.3# to 88.Q£f percent. The boiler heat input increases
from 3,210 to 3,244 MMBtu/hr. The fuel flow, ash generation rate, and combustion and flue
/^- gas flow rates increase inXmrect proportion to the change hi the heat input.
• There is an overall increase in the plant auxiliary power consumption due to the SNCR
equipment as well as the increased demand on the draft fans to accommodate the higher air
and flue gas flow rates. The estimated auxiliary power increase is 157 kW.
• The urea consumption requirement for the SNCR system is 350 galThr.
• The water consumption requirement for the SNCR system is 4,470 galVhr.
Using the above parameters and the costing methodology described in Appendix A, both the
capital and levelized costs have been calculated for the entire size range (33 to 952 MW) of tan-
gential, coal-fired boilers. As shown in Figure 3-6, the capital costs range from approximately
$6 to $46/kW. The levelized costs at a capacity factor of 65 percent range from 1.13 to 2.15
mils/kWh and $1,140 to $2,130/ton NOX removed (Figures 3-7 and 3-8). The levelized costs at a
capacity factor of 27 percent range from 132 to 3.78 mils/kWh and $1,330 to $3,800/ton NOx
removed (Figures 3-9 and 3-10).
3.2 Wall-Fired Boiler Applications
The NOX control technologies evaluated for this boiler type include SCR and SNCR. The design
data for the representative boiler selected for this evaluation are shown in Table 3-1. This boiler
is a balanced draft, natural circulation, reheat, single furnace boiler. It has 24 burners located
four high and six wide on the front wall of the unit. The boiler serves a 381 MW steam turbine
generator and is equipped with two 50-percent-capacity forced draft fans, two 50-percent-
capacity induced draft fans, and an electrostatic precipitator for removing dust from the flue
gases exiting the boiler.
3.2.1 SCR Evaluation
The following major criteria and assumptions have been followed in evaluating the SCR tech-
nology for the wall-fired boilers:
y
• The SCR system is designed to reduce NOX emission from a baseline level of 0.5 Ib/MMBtu
to the required limit of 0.15 Ib/MMBtu.
• All of the other criteria and assumptions described in Section 3.1.1 apply equally to this case.
-------
The consumables associated with the SCR system retrofit for the study boiler are as follows:
Auxiliary power consumption 842 kW
Anhydrous ammonia consumption 476 Ib/hr
Average catalyst replacement 54 1 7 ft J/yr
Using the above parameters and the costing methodology described in Appendix A, both the
capital and levelized costs have been calculated for the entire size range (30 to 1300 MW) of
wall-fired boilers. As shown in Figure 3-1 1, the capital costs range from approximately $37 to
$134/kW. The levelized costs at a capacity factor of 65 percent range from 2.03 to 4.5 mils/kWh
and $1,180 to $2.600/ton NOX removed (Figures 3-12 and 3-13). The levelized costs at a
capacity factor of 27 percent range from 4.5 to 10.4 mils/kWh and $2,700 to $6,100/ton NOX re-
moved (Figures 3-14 and 3-15).
322 SNCR Evaluation
The following major criteria and assumptions have been followed in evaluating the SNCR tech-
nology for the wall-fired boilers:
• The SNCRjSystem is designed to provide a 50 percent NOX reduction from a baseline NOX
rate of 0.50 Ib/MMBtu.
• All of the other criteria and assumptions described in Section 3.1.2 also apply equally to this
case.
The impacts of the SNCR technology retrofit on the study boiler are as follows (refer to Table
3-1):
• The boiler efficiency reduces from -8&3S to 87^96 percent The boiler heat input increases
from 3,600 to 3,618 MMBtu/hr. The fuel flow, ash generation rate, and combustion and flue
gas flow rates increase in a direct proportion to the change in the heat input
• There is an overall increase in the plant auxiliary power consumption due to the SNCR
equipment as well as the increased demand on the draft fe"* to accommodate the higher air
and flue gas flow rates. The estimated auxiliary power increase is 193 kW.
• The urea consumption requirement for the SNCR system is 433 galThr.
• The water consumption requirement for the SNCR system is 5,570 gaL/hr.
Using the above parameters and the costing methodology described in Appendix A, both the
capital and levelized costs have been calculated for the entire size range (30 to 1300 MW) of
wall-fired boilers. As shown in Figure 3-16, the capital costs range from approximately $6.5 to
$52/kW. The levelized costs at a capacity factor of 65 percent range from 1.18 to 232 mils/kWh
and $980 to $l,920/ton NOX removed (Figures 3-17 and 3-18). The levelized costs at a capacity
3-4
-------
TABLE 3-1
ORIGINAL DESIGN DATA
TANGENTIAL AND WALL-BURNER COAL-FIRED BOILERS
Parameter"'
Boiler size, MW
Boiler load, %MCR
Boiler type
Heat input, MMBtu/hr
Fuel consumption, ton/hr
Solid waste, ton/hr
Boiler efficiency ^
Fuel analysis (WL %):
Ash
Moisture
Sulfur
HHV,Btu/lb
Tangential Boiler*"
348
100
Reheat
3,210
127
9.82
8839
7.7
8.4
0.8
12,696
Wall-Fired Boiler"
381
100
Reheat
t
3,600
142
10.98
8839
7.7
8.4
0.8
12,696
NOTES
1. Only data pertinent to the NOX control technologies are shown.
2. The same coal is fired in both boilers. It is assumed that efficiency is the same for both boiler
types. In practice, there may be a small difference in the efficiencies; however, the difference
would be insignificant as long as the operating parameters, such as excess air levels, are the
same.
-------
TABLE 3-7
NO, REDUCTION PERFORMANCE OF COMBUSTION CONTROLS ON
VERTICALLY-FIRED BOILERS [323334]
Source
AEP Tanner's Creek 1 (152 MWe)
Duquesne Light Elrama Unit 1 (100 MWe)
Duquesne Light Elrama Unit 2 (100 MWe)
Duquesne Light Elrama Unit 3 (125 MWe)
Performance
% Reduction
40 (estimated)
42
>40
>40
Controlled Emission Rate
0.57 (estimated)
0.45
-0.45
-0.45
Combustion controls have not yet been applied to wet-bottom boilers in the U.S.
However, as shown above, a major utility has announced plans to retrofit a wet-bottom wall-
fired boiler in the fall of 1995 with combustion controls, specifically a two-level overfire air
(OF A) system. According to the utility's engineering estimates, the two-level OF A system
will achieve an overall 50 percent reduction from uncontrolled levels and will allow the wet-
bottom boiler to have a NO, emission rate of 0.675 Ib/MMBtu.
32.6.4 Tmpacts on Boiler Operation. The staged combustion approach has the
potential to change the UBC, CO, excess air, and furnace exit gas temperature relative to pre-
retrofit levels; thereby potentially affecting boiler combustion efficiency and plant economics.
Presented below is information on the possible variation of these parameters.
UBC. CO. and Excess Air [33]
Post-retrofit results from Elrama Unit 2 indicate that UBC levels decreased across the
load range. Although some of this decrease may be due to elimination of boiler casing in-
leaks, still the results indicate that the retrofit had no negative impacts on UBC. Post retrofit
results also indicate that post-retrofit CO levels were maintained at or below 100 ppmv across
the load range.
At the Elrama retrofits, the pre- and post-retrofit excess air levels remained relatively
constant No data are currently available on CO, UBC, and excess air changes from the
American Electric Power retrofits:
3-19
-------
Using the above parameters and the costing methodology described in Appendix A. both the
capital and levelized costs have been calculated for a boiler size range of 30 to 1.300 MW. As
shown in Figure 4-1, the capital costs range from approximately $14 to $54/kW. The levelized
costs at a capacity factor of 65 percent range from 0.55 to 1.5 mils/kWh and $1J50 to $3.750/ton
NOX removed (Figures 4-2 and 4-3). The levelized costs at a capacity factor of 27 percent range
from 1.15 to 3.44 mils/kWh and $2,900 to $8,600/ton NOX removed (Figures 44 and 4-5).
42 Gas Reburning Evaluation
The following major criteria and assumptions have been followed in evaluating the gas rebuming
technology for the gas-fired boilers:
./
• The gas rebum system is designed to reduce the baseline NOX of 025 Ib/MMBtu to the re-
quired limit of 0.15 Ib/MMBtu.
• It is assumed that natural gas supply is available at the plant fence for both boilers.
• The rebum system design is based on a 25 percent heat input for the rebum injectors. Natural
gas is injected into the furnace along with gas recirculation (system designed for a 10 percent
recirculation rate). It is assumed that existing gas recirculation fay* will be used for this pur-
pose. The overfire air system is designed for 20 percent of the full-load combustion air re-
quirement for the boiler.
• It is assumed that sufficient space is available in the boilers to add the rebum injectors and
overfire air ports. It is also assumed that the available space allows for an adequate residence
time for completing the combustion process for the rebum fuel. Lack of an adequate resi-
dence tune may reduce the effectiveness of the gas rebum system or it may adversely affect
the feasibility of installing such a system.
• In some cases, capital cost of the rebum technology application may be lower for a tangential
boiler than for a wall-fired boiler. Because of the comer firing arrangement for the tangential
boiler, a potential may exist for effectively utilising a smaller number of rebum injectors.
However, any cost difference is not expected to be significant Therefore, for conservatism,
the same capital costs developed for the wall-fired boiler have been used for the tangential
boiler.
• Other general gas rebum system design details, assumptions, and impacts on the existing
equipment outlined in Appendix A (Section 43 of Appendix B) also apply to this case.
Rebum technology has a minimal impact on the performance of a gas-fired boiler. This applica-
tion involves withdrawal of a portion of the boiler fuel from the main combustion zone and in-
jection of this fuel above the top-most burners. Overfire air is injected further up in the furnace
to complete combustion of the rebum fuel. As long as the conditions permit proper combustion
of the rebum fuel, the boiler performance would not be affected. Operation of the rebum system
does result in an increased auxiliary power consumption (associated with the operation of the gas
recirculation fen). In the case of the study boilers, this increase is estimated at
4-2
-------
Using the above parameters and the costing methodology described in Appendix A, both the
capital and levelized costs have been calculated for a boiler size range of 30 to 1300 MW. As
shown in Figure 4-6, the capital costs range from approximately $10 to S37/kW. The levelized
costs at a capacity factor of 65 percent range from 028 to 0.95 mils/kWh and S700 .to $2.400/ton
NOX removed (Figures 4-7 and 4-8). The levelized costs at a capacity factor of 27 percent range
from 1.32 to 3.78 mils/kWh and $1,330 to $3,800/ton NOX removed (Figures 4-9 and 4-10).
43 SNCR Evaluation
The following major criteria and assumptions have been followed in evaluating the SNCR tech-
nology for the gas-fired boilers:
^
• The SNCR system is designed to reduce the baseline NOX of 025 Ib/MMBtu to the required
limit of 0.15 Ib/MMBtu.
• A reagent ratio of 1.5 commensurate with the NOX reduction requirement is used.
• All of the other criteria and assumptions described in Section 3.12 also apply equally to this
case.
The impacts of the SNCR technology retrofit on the study boilers are as follows (refer to Table
4-1):
-7 y
• The boiler efficiency reduces from 85.65'to 85.48 percent. The boiler heat input increases
from 2,980 to 2,986 MMBtu/hr. The fuel flow, ash generation rate, and combustion and flue
gas flow rates increase in a direct proportion to the change in the heat input.
• There is an overall increase in the plant auxiliary power consumption due to the SNCR
equipment as well as the increased demand on the draft fen< to accommodate the higher air
and flue gas flow rates. The estimated auxiliary power increase is 80 kW.
• The urea consumption requirement for the SNCR system is 155 gaL/hr.
• The water consumption requirement for the SNCR system is 1,980 gaL/hr.
Using the above parameters and the costing methodology described in Appendix A, bom the
capital and levelized costs have been calculated for a boiler size range of 30 to 1300 MW. As
shown in Figure 4-11, the capital costs range from approximately $32 to $28/kW. The levelized
costs at a capacity factor of 65 percent range from 0.5 to 1.1 mils/kWh and $1220 to $2,800Aon
NOX removed (Figures 4-12 and 4-13). The levelized costs at a capacity factor of 27 percent
range from 0.6 to 2.1 mils/kWh and $1,520 to $5200/ton NOX removed (Figures 4-14 and 4-15).
4-3
-------
TABLE 4-1
ORIGINAL DESIGN DATA
TANGENTIAL AND WALL-BURNER TYPE
GAS- AND OIL-FIRED BOILERS
Parameter'"
Boiler size, MW
Boiler load, %MCR
Boiler type
Heat input, MMBtu/hr
Fuel consumption, ton/hr
Solid waste, Ib/hr
Boiler efficiency ? /
-------
Using the above parameters and the costing methodology described in Appendix A. both the
capital and levelized costs have been calculated for a boiler size range of 30 to UOO MW. As
shown in Figure 5-1, the capital costs range from approximately S21 to $77/kW. The levelized
costs at a capacity factor of 65 percent range from 0.87 to 227 mils/kWh and SI.500 to
53,800/ton NOX removed (Figures 5-2 and 5-3). The levelized costs at a capacity factor of
27 percent range from 1.95 to 53 mils/kWh and $3200 to $8,800/ton NOX removed (Figures 5-4
and 5-5).
53, Gas Retraining Evaluation
The following major criteria and assumptions have been followed in evaluating the gas rebuming
technology for the oil-fired boilers:
. y
• The gas rebum system is designed to reduce the baseline NOX of 0.3 Ib/MMBru to the re-
quired limit of 0.15 Ib/MMBtu.
• Other criteria and assumptions outlined in Section 42 also apply to this case.
The performance impacts of the rebum technology on the oil-fired boilers are as follows:
• The boiler performance changes, because with the rebum system 20 percent of the heatii,
is by natural gas and 80 percent is by oil. The boiler efficiency reduces from 88.jo to 87.!
percent The levelized cost estimates must take into account the cost increases incurred in
firing natural gas rather than No. 6 oiL
• Firing of natural gas reduces the amount of ash generation by 58 Ib/hr and SCX emission rate
by 620 Ib/hr. Both of these reductions benefit the operating costs.
Using the above parameters and the costing methodology described in Appendix A, both the
capital and levelized costs have been calculated for a boiler size range of 30 to 1300 MW. As
shown in Figure 5-6, the capital costs range from approximately S12 to $44/kW. The levelized
costs at a capacity factor of 65 percent range from 0.8 to 1.6 mils/kWh and SI350 to S2.650/ton
NOX removed (Figures 5-7 and 5-8). The levelized costs at a capacity factor of 27 percent range
from 12 to 3.1 mils/kWh and $2,000 to $5200/ton NOX removed (Figures 5-9 and 5-10).
53 SNCR Evaluation
The following major criteria and assumptions have been followed in evaluating the SNCR tech-
nology for the oil-fired boilers:
y
• The SNCR system is designed to reduce the baseline NOX of 03 Ib/MMBtu to the required
limit of 0.15 Ib/MMBtu.
• All of the other criteria and assumptions described in Section 3.12 also apply equally to this
case.
5-2
-------
The impacts of the SNCR technology retrofit to the study boilers are as follows (refer to Table
4-1):
^ «»
>C • The boiler efficiency reduces from 88.tf to 87.8S percent. The boiler heat input increases
from 2.895 to 2,904 MMBtu/hr. The fuel flow, ash generation rate, and combustion and flue
gas flow rates increase in direct proportion to the change in the heat input
• There is an overall increase in the plant auxiliary power consumption due to the SNCR
equipment as well as the increased demand on the draft fans to accommodate the higher air
and flue gas flow rates. The estimated auxiliary power increase is 115 kW.
• The urea consumption requirement for the SNCR system is 210 gal Jbi.
• The water consumption requirement for the SNCR system is 2,690 galThr.
Using the above parameters and the costing methodology described in Appendix A, both the
capital and levelized costs have been calculated for a boiler size range of 30 to 1,300 MW. As
shown in Figure 5-11, the capital costs range from approximately $4.0 to $32/kW. The levelized
costs at a capacity factor of 65 percent range from 0.65 to 135 mils/kWh and Si,100 to
$2JOO/ton NOX removed (Figures 5-12 and 5-13). The levelized costs at a capacity factor of
27 percent range from 0.8 to 2.46 mils/kWh and $1,350 to $4,100/ton NOX removed (Figures 5-
14 and 5-15).
5-3
-------
TABLE 4-1
FACTORS FOR INDIRECT COSTS(1)
£
COST ITEM
General Facilities
Engineering*2*
Project Contingency
Process Contingency
Royalty Allowance
Pre-production Cost
Inventory Capital
Initial Catalyst/Chemicals'"
Allowance for Funds During
Construction
NOTES
COMBUSTION
CONTROLS
0.0
10.0
15.0
5.0
0.0
2.0
0.0
0.0
0.0
COAL REBURNING
5.0
10.0
15.0
5.0
0.0
2.0
0.0
0.0
0.0
GAS REBURNING SNCR SCR
2.0 5.0 5.0
10.0 10.0 10.0
15.0 15.0 15.0
5.0 5.0 5.0
0.0 0.0 0.0
2.0 2.0 2.0
0.0 (3) (3)
0.0 0.0 0.0
0.0 0.0 0.0
1. The indirect costs are listed as a percentage of direct capital and other costs. (Refer to Appendix B for further
explanation.)
2. Where the capital costs differed significantly between the two boilers evaluated for each technology, the engineering cost
factor was modified to compensate for this difference. (Refer to Appendix B for details.)
3. The inventory capital represents the cost of a 14-day storage at full-load demand of urea and anhydrous ammonia for the
SNCR and SCR systems, respectively.
4. Formal! technologies, the royalties and initial catalyst and chemicals are considered part of the direct capital costs. The
1 is assumed to be negligible, because of a relatively short construction schedule for all technologies.
-------
bed design [7]. An additional categorization of these boilers is based on operation at
atmospheric or pressurized conditions. An atmospheric FBC (AFBC) system is similar to a
pulverized coal-fired boiler in that the furnace operates at close to atmospheric pressure and
depends upon heat transfer of a working fluid (water) to recover the heat released during
combustion [1]. Pressurized FBC (PFBC) operates at elevated pressures and recovers energy
through both heat transfer to a working fluid (water) and the use of the pressurized gas to
power a gas turbine [1]; this pressurized design approach also can be classified as a combined
power cycle.
23 Characterization of Group 2 Boilers
Table 2-1 lists the Group 2 boiler types with respect to population, nameplate capacity.
size and estimated uncontrolled NOX emissions. This table has been developed using
information on the boilers in the EPA Group 2 Boiler Database (Appendix A).
Table 2-1. Characterization of Group 2 Boilers
Boiler Type
Population
(Units) (%)
Cyclone- 89 39.9
Fired j
Cell-burner
Fired
Wet-
bottom1
Dry-Bottom
Vertically-
Fired
Stoker-
Fired
FBC
36 16.1
39 17.5
33 14.8
21 9.4
5 2.3
Total 223 100
Nameplate Size
Capacity Mean Range
(MWe) ; (%) : (MWe) > (MWe)
! •
27,562 40.9 j 310 j 33-
1 I 1150
24,572 36.4 ' 683 82-
1300
8,626 12.8
4,779 7:1
1,083 ; 1.6
814 : 1.2
"
221
145
52
29-544
35-254
32-79
163 114-
194
67,436 j 100
• Estimated
Uncontrolled
(Tpy)
. 732,300
682,000
277,000
99,700
3,400
3,900
1/798,304
NO,
(%)
40.7
37.9
15.4
5.6
02
02
! 100
NO, controls for wet bottom boilers of any firing design have to be designed to not disturb
slag tapping capability.
A 2-3
-------
TABLE 3-1
SUMMARY OF GROUP 2 BOILER/NO, CONTROL TECHNOLOGY
DEMONSTRATIONS AND COMMERCIAL RETROFITS
Group 2 Boiler
Types
Boilers with Cell
Burners
Cyclone-Fired Boilers
Wet-bottom Boilers
Dry-bottom,
Vertically-Fired
Boilers
Selected NO,
Control
Technologies
Plug-in Combustion
Controls
Non Plug-in
Combustion
Controls
Coal Reburning
Natural Gas
Reburning
SNCR
SCR
SNCR
SCR
Combustion
Controls
SNCR
Combustion
Controls
Number of Fnll-
Scale or
Commercial
Retrofits
7
^
J *
1
2
1
1
1
1
1
1
4
Retrofit Size
Range
(MWe)
555 - 780
630 - 760
110
33-114
138
320
320
80 (321)'
217
100
100-152
SCR system was installed only in one of four ducts of the 321 MWe boiler, and only one
quarter of the total unit's flue gas volume passes through the SCR system (equivalent to 80
MWe).
-------
TABLE 3-2
NO. CONTROL PERFORMANCE FOR CELL BURNER RETROFITS
Low-NO, Burner Retrofit Project
J. M. Slnart Station Unit #4
W. H. Sammis Power Station, Unit #6
Four Corners Steam Electric Station Unit
#4;
Muskingum River Unit #5
Hatfield's Ferry Unit #2
Monroe Unit #1*
Load,
MWe or %
MCR
Pre/
Post Retrofit
-/605
-/460
-/350
627/630
362/377
100%
600/607
-/454
-/368
555
730 (93. 6%)
Average
Baseline
NO, Emissions
Ib/MMBlu
1. 16
N/A
N/A
1. 15- 1.40
0. 49
1.15
1.2
1. 17
0. 93
Average
Post-retrofit
NO, Emissions
Ib/MMBtu
0. 53
N/A
N/A
0. 43 - 0. 481
0. 31
0.49
0.59
0.53
0.51
0.58
0.52
Avg. NO,
Emissions
Reduction
%
55
54
48
58-69
37
58
51
50
44
I/I
Nominal test results achieved during a 61 hour test period one year after retrofit witli low-NO, burners
Data for a performance at full load were not available.
2- ' 3 , -»6».< -I* I ?
-------
the existing cyclone units can be successfully retrofitted by the coal rebuming technology,
except for the small, single wall-fired units less than 80 MWe; these units, which represent
less than 1% of the cyclone population, lack sufficient furnace height to provide adequate gas
residence time.
3.2.2.2 Applications and Demonstration . There has been one coal -rebuming
demonstration project on a coal-fired cyclone boiler in the U.S. This is the long-term, U.S.
DOE Clean Coal Program demonstration project at Wisconsin Power and Light Company's
Nelson Dewey Station, Unit #2. The Babcock & Wilcox Company's (B&W's) coal-
rebuming system was installed on the 1 10 Mwe cyclone-fired boiler. The demonstration
project included three steps: l)mathematical simulation using B&W in-house models, 2) tests
on B&W's Small Boiler Simulator (SBS), and 3)installation and testing on the 1 10 MWe
boiler using two types of coal - Lamar bituminous coal and Powder River Basin (PRB)
subbituminous coal. Pertinent project information and data were collected from reference
[11].
3.2.2.3 NO. Reduction Performance. As discussed in Appendix C, given
sufficient time hi the rebum zone, rebum zone stoichiometry is the critical parameter that
influences NOX reduction. However, rebum zone stoichiometry is directly related to the
percentage of rebum heat input (or the fuel split between the cyclones and rebum burners).
In general, an increase in the rebum heat input and commensurate decrease hi cyclone heat
input will decrease the stoichiometry hi the rebum zone and improve NOX reduction
efficiency. Although not readily apparent, this fuel split parameter is constrained in a number
of ways, viz.,: 1) diminished flame stability in the rebum zone due to insufficient oxygen
concentration, 2) minimum cyclone coal flow rates that must be maintained to control slag
tapping, 3) potential for increased boiler tube corrosion within the rebum zone, and 4)
increased fly ash load in the furnace which may increase UBC and cause fouling of heat
exchange surfaces. These limitations will vary with load, coal type, and unit-specific design.
Due to these considerations, NOX reductions achieved using coal rebuming in cyclone boilers
will, in general, be dependent on cyclone load and fuel split between cyclones and rebum
injectors (or burners).
As discussed in Appendix C, average NOX reductions at Nelson Dewey retrofit ranged
between 52.4 % (full load) and 33.3% (33% of load) for Lamar bituminous and 55.4% (full
load) to 52.6% (55% of load) for PRB subbituminous coals. In general, as per the vendor of
this technology, "nominal 50 to 60% reductions can be expected from existing cyclone-
equipped boilers."[10].
Based on this data, a range of 40% to 60% NO, redaction is recommended for
cost sensitivity analyses.
A 3-8
-------
SNCR process is capable of load following through adjustment of Normalized Stoichiometric
Ratio (NSR).
Coal Sulfur Content
As discussed in Appendix C, ammonia slip needs to be controlled in SNCR
applications to minimise formation of ammonium salts and subsequent boiler impacts. An
"acceptable" value of NH3 slip will probably be a balance of optimum NOX reduction
performance and minimum sulfate/bisulfate formation (as determined by the specific
application and fuel).
32.5 Application of Selective Catalytic Reduction CSCR) to Group 2 Boilers
3.2.5.1 Description of Control Technology . Selective ..catalytic reduction (SCR)
is a post-combustion, dry NOX control technology which is typically applied after the boiler
economizer (hot-side configuration) or after the ESP (cold-side configuration). The SCR
process employs a catalyst, which in the presence of ammonia (NH3) and oxygen (O^,
reduces NOX to free nitrogen (N,) and water (H,O). The reduction reactions are promoted by
heterogeneous catalysts and occur at temperatures below 800° F.
3.2.5.2 Demonstrations and Applications. First patented by a U.S. company in
1959, SCR is a proven technology used to significantly reduce NOX emissions from over 200
sources in the U.S., and over 500 sources worldwide. Table 3-6 depicts SCR applications,
currently operating or planned, on U.S. coal fired boilers. Pilot and demonstration projects in
the U.S., as well as extensive experience abroad, suggest that SCR is a commercially viable
control technology option for Group 2 boilers.
The following foil-scale SCR installations are currently operational at U.S. coal
fired boilers:
• Southern Company Services/demonstration project being conducted at Gulf
Power Company's Plant Crist The objective is to evaluate the performance of
commercially available SCR catalysts when applied to operating conditions found in
U.S. pulverized coal-fired utility boilers. The testing program on this project was
started in July of 1993 and will bo finished- in July of 1995.
4t*£ &+«*p&£Zt **. «& *^(
US Generating Company's Chambers Works, Carneys Point Station, where Foster
Wheeler Energy Corporation's SCR system was installed on two 140 MWe boilers
which fire an Eastern medium-sulfur coal (with up to 2% sulfur and an ash content of
6 to 10%) [29].
Public Service Electric & Gas Mercer Generating Station (321 MWe wet-bottom
boiler). At this installation, SCR system has been installed in one of four ducts, and
only one quarter of the total unit's flue gas volume passes through the SCR system
(equivalent to 80 MWe).
^3-15
-------
Gas Rebuming Applied to Cyclone-Fired Boilers - test results from the demonstration
project at Ohio Edison's Niles Station Unit #1 showed insignificant change in
precipitator outlet paniculate loadings.
SNCR Applied to All Group 2 Boilers - little quantitative experimental data was found
in the literature to characterize SNCR's impact on paniculate emissions. Generally,
SNCR does not increase paniculate emissions.
SCR Applied to All Group 2 Boilers - SCR does not impact paniculate emissions.
Impact of NO. Controls on CO? Emissions
Implementation of NOX controls on Group 2 boilers can yield simultaneous impacts on
CO, emissions if: 1) boiler thermal efficiency changes so as to alter coal feed, 2) a portion of
the baseline coal feed is replaced by a fuel with a different carbon content per unit of heat
input (e.g., natural gas), and 3) post-combustion control technology chemical reagents (i.e.
urea and cyanuric acid) react to generate CO,. The first is unlikely to have any significant
impact on C02 emissions since, on average, the NOX control technologies do not have a
significant impact on combustion efficiency. The second item characterizes the application of
gas reburning, which can displace up to 25% of the baseline boiler coal input with natural
gas, thereby yielding about a 10% reduction in CO, emissions (based on a 1.7 ratio of coal-
produced CO2 versus methane-produced CO2. The third item characterizes the application of
urea- or cyanuric acid-based SNCR whose chemical reactions produce modest amounts of
added CO,.
3.3.1.2 Impact of NO. Controls on Secondary Air Emissions. Application of the
alternative NOX control technology options may yield changes to baseline quantities of
secondary air emissions generated by Group 2 boilers. These secondary air emissions may
include carbon monoxide (CO), nitrous oxide (N2O), and total hydrocarbons (THC).
Both SCR and SNCR also generate small amounts of ammonia (NH,).
Table 3-9 qimmariTes the typical baseline and "controlled" emissions of all secondary
air pollutants for the Group 2 boiler/technology combinations under consideration. The details
on these pollutants can be found in Appendix C. As seen in Table 3-9, all the control
technologies result in negligible CO, THC, and, with the exception of SNCR, N2O impacts.
emissions are maintain^ at minimal levels.
3.3.2 Impacts on Solid Waste Disposal
In general, the application of the various NOX control options is not expected to
increase or decrease the quantities of solid wastes generated by the Group 2 boiler population.
However, the potential exists to change the characteristics of the solid waste thereby
impacting its disposal. There are two ways in which this can happen: 1) a significant increase
or decrease in the unburned carbon (UBC) in the boiler fly ash and bottom ash and 2)
A3-28
-------
3.3.4.2 Ancillary Power Requirements. The ancillary power requirements
associated with the NOX control technologies result from the electric power required to operate
associated new equipment, such as fans and pumps. Also, as a result of the retrofit, existing
plant equipment such as forced draft or induced draft fans, may also cause changes in power
consumption due to incremental changes in air- and gas-side pressure drops. Ancillary power
requirements are given below for the various NOX control technologies.
Low-NO., Cell Burner Replacement
As discussed in Section 3.2, minor changes in fan energy consumption may occur.
Coal Rebuming on Cvclone-Fired Boilers
The implementation of rebuming on a cyclone-fired boiler results in the addition of
new combustion system components and modified operation of existing components. The
modified unit, with addition of primary air fan, pulverizer, piping and burners, overfire air
system, and flue gas recirculation, must be compared to the original unit with regard to the
fan power requirement to transport fuel and air. This will likely vary from unit to unit based
on site-specific factors (e.g., availability of an existing FGR system).
Gas Rebuming on Cvclone-Fired Boilers
No data -was available in the literature on incremental changes to ancillary power
requirements. However, only minor changes would be expected.
SNCR Applications
Energy consumption by the SNCR process is related to pretreatment and injection of
ammonia-based reagents and their carrier gas and liquids. Anhydrous ammonia, aqueous
ammonia, or a urea solution are injected in liquid form at high pressure to ensure efficient
droplet atomization and dispersion. When anhydrous ammonia is used as the reagent (e.g.,
Thermal DeNOx installations), the ammonia is stored in liquid form under pressure. This
liquid ammonia must be vaporized via energy addition, mixed with a carrier gas (air or
steam), and then injected for adequate mjying The amount of electricity consumed depends
on whether the process uses air or g»»am for carrier gas. If steam is used, less electricity is
consumed by fans but the steam which is taken from the plant will reduce turbine output [42];
the specific energy impact will depend on the location in the power cycle from where the
steam is withdrawn.
The Thermal DeNOx process will consume approximately 1.0 to 1.5 kW for each
MWth of boiler capacity (or 029 to 0.44 kW/MMBtu/h) when using compressed air as the
carrier medium [42]. The actual amount of electricity consumed will ultimately depend on
the baseline NOX emissions, the NH3/NOX stoichiometry, and the NO, reduction goal. For
steam-assisted ammonia injection, power consumption is reduced to about 0.2 to 0.3
kW/MWth (0.05 to 0.08 kW/MMBtu/h) of boiler capacity. The amount of steam used is
about 25 to 75 Ib/h/MWth, but use of compressed air is typically more cost-effective.
A 3-34
-------
APPENDIX AA
EPA GROUP 2 BOILER DATABASE
AUGUST 1995
-------
APPENDIX
LIST OF TABLES
TABLE PAGE
/IA-1 Cell Burner Fired Boilers M-l
AA-2 Cyclone Fired Boilers M-2
AA-3 Wet Bottom Fired Boilers M-5
»A-4 Vertically Fired Boilers AA-6
/JA-5 Stoker Fired Boilers , 4A-7
M-6 FBC Boilers / A-8
-------
TableM-1
Cell Burner Fired Boilers
STATE
ALABAMA
GEORGIA
GEORGIA
GEORGIA
INDIANA
IOWA
IOWA
MASSACHUSETTS
MICHIGAN
MICHIGAN
MICHIGAN
MICHIGAN
MICHIGAN
MINNESOTA
NORTH CAROLINA
NORTH CAROLINA
OHIO
OHIO
OHIO
OHIO
OHIO
OHIO
OHIO
OHIO
OHIO
OHIO
OHIO
OHIO
OHIO
PENNSYLVANIA
PENNSYLVANIA
PENNSYLVANIA
TENNESSEE
TENNESSEE
WEST VIRGINIA
WEST VIRGINIA
PLANT
GREENE COUNTY
HARLIEE BRANCH
HARLLEE BRANCH
HARllEE BRANCH
WARRICK
DUBUQUE
DUBUQUE
BRAYTON POINT
JH CAMPBELL
MONROE
MONROE
MONROE
MONROE
FOX LAKE
BELEWS CREEK
BELEW8 CREEK
AVON LAKE
CARDINAL
CARDINAL
EABTLAKE
GEN JM GAVIN
GEN JM GAVIN
JM STUART
J M STUART
JM STUART
JM STUART
MIAMI FORT
MU8KINOUM RIVER
W H 8AMMI8
HATFIEltre FERRY
HATFIElffS FERRY
HATFIELOT8 FERRY
CUMBERLAND
CUMBERLAND
FORT MARTIN
JOHN E AMOS
UTILITY
ALABAMA POWER CO
GEORGIA POWER CO
GEORGIA POWER CO
GEORGIA POWER CO
SOUTHERN INDIANA GAS « ELEC CO
NTERSTATE POWER CO
NTERSTATE POWER CO
NEW ENGLAND POWER CO
CONSUMERS POWER CO
DETROIT EDISON CO
DETROIT EDISON CO
DETROIT EDISON CO
DETROIT EDISON CO
INTERSTATE POWER CO
DUKE POWER CO
DUKE POWER CO
CLEVELAND ELECTRIC ILLUM CO
CARDINAL OPERATING COMPANY
CARDINAL OPERATING COMPANY
CLEVELAND ELECTRIC ILLUM CO
OHIO POWER CO
OHIO POWER CO
DAYTON POWER A LIGHT CO
DAYTON POWER * LIGHT CO
DAYTON POWER & LIGHT CO
DAYTON POWER A LIGHT CO
CINCINNATI OA8 * ELECTRIC CO
OHIO POWER CO
OHIO EDISON CO
WEST PENH POWER CO
WEST PENN POWER CO
WEST PENN POWER CO
TENNESSEE VALLEY AUTHORITY
TENNESSEE VALLEY AUTHORITY
MONONOAHELA POWER CO'
APPALACHIAN POWER CO
BOILER
3
4
4
I
8
3
2
1
2
3
4
3
1
2
12
1
2
S
FIRING
TYPE
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CELL
OPPOSED/CEIL
OPPOSED/CELL
OPPOSED/CELL
BOTTOM
TYPE
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
NAMEPLATE
CAPACITY
(MW«)
299.20
230.00
480.70
490.00
323.00
37.50
20.75
642.60
385.00
•17.20
822.60
822.60
817.20
81.60
1080.07
1060.07
680.00
615.20
615.20
680.00
1300.00
1300.00
610.20
610.20
610.20
610.20
957.10
615.23
623.00
576.00
576.00
576.00
1300.00
1300.00
576.00
1300.00
COM
DATE
1969
1985
1968
1969
1970
1959
1952
1969
1967
1971
1972
1973
1974
1962
1974
1975
1970
1986
1987
1972
1974
1978
1971
1970
1972
1974
1975
1968
1971
1969
1970
1971
197]
1973
1668
1973
1990 HEAT
INPUT
(QBIu)
10.6502270
18.6651380
30.2172470
31.8027730
14.7474960
1.6582570
0.0896960
29.4908590
21.6168990
45.4068130
43.4412050
51.5113540
43.6503710
2.4662680
39.0053500
60.3982330
35.4069970
25.8145450
36.2087610
29.3478360
87.0884360
71.4697730
38.8768560
36.1870630
38.4387740
40.2369050
31.3106060
27.3070570
34.9729710
34.304196C
36.2576510
27.8638570
81.3561830
49.9625510
34.6464200
62.3038020
NCONTROl
Ib/MMBIu)
0.919
1.178
1.037
1.037
1.000
0/689
0.796
1.300
1.000
0.655
0.655
0.779
0.778
0.763
1.455
1.384
0.960
0.900
1.020
0.674
1.160
1.160
1.110
1.050
0.950
1.110
1.070
1.097
1.080
0.760
0.760
0.600
1.570
1.330
1.070
1.048
.LEO NOD
ourc*
CREV
CREV
CREV
CREV
ETS
CREV
CREV
UARG
ETS
CREV
CREV
CREV
CREV
CREV
CREV
CREV
ETS
ETS
ETS
CREV
ETS
ETS
ETS
ETS
ETS
NURF
ETS
CREV
ETS
ETS
ETS
ETS
ETS
ETS
ETS
CREV
-------
Tabl^A-2
Cyclone Fired Boilers
STATE
:LORIDA
FLORIDA
FLORIDA
FLORIDA
ILLINOIS
ILLINOIS
ILLINOIS
LLINOIS
ILLINOIS
ILLINOIS
ILLINOIS
LLINOIS
ILLINOIS
ILLINOIS
ILLINOIS
ILLINOIS
ILLINOIS
ILLINOIS
ILLINOIS
ILLINOIS
ILLINOIS
ILLINOIS
ILLINOIS
ILLINOIS
ILLINOIS
ILLINOIS
INDIANA
INDIANA
INDIANA
INDIANA
INDIANA
INDIANA
IOWA
IOWA
IOWA
KANSAS
KANSAS
KANSAS
PLANT
: J GANNON
F J GANNON
F J GANNON
F J GANNON
BALDWIN
BALDWIN
COFFEEN
COFFEEN
DALLMAN
DALLMAN
JOLIET 0
KINCAID
KINCAID
LAKESIDE
LAKESIDE
MARION
MARION
MARION
MARION
POWERTON
POWERTON
POWERTON
POWERTON
WAUKEGAN
WILL COUNTY
WILL COUNTY
BAILLY
BAILLY
MICHIGAN CITY
R M SCHAHFER
STATE LINE
TANNERS CREEK
GEORGE NEAL
MUSCATINE
SUTHERLAND
KAW
LACYGNE
QUINDARO
UTILITY
TAMPA ELECTRIC CO
TAMPA ELECTRIC CO
TAMPA ELECTRIC CO
TAMPA ELECTRIC CO
LLINOIS POWER CO
ILLINOIS POWER CO
CENTRAL ILLINOIS PUB SERV CO
CENTRAL ILLINOIS PUB SERV CO
SPRINGFIELD CITY OF (IL
SPRINGFIELD CITY OFJIL
COMMONWEALTH EDISON CO
COMMONWEALTH EDISON CO
COMMONWEALTH EDISON CO
SPRINGFIELD CITY OF (IL)
SPRINGFIELD CITY OF (IL)
SOUTHERN ILLINOIS POWER COOP
SOUTHERN ILLINOIS POWER COOP
SOUTHERN ILLINOIS POWER COOP
SOUTHERN ILLINOIS POWER COOP
COMMONWEALTH EDISON CO
COMMONWEALTH EDISON CO
COMMONWEALTH EDISON CO
COMMONWEALTH EDISON CO
COMMONWEALTH EDISON CO
COMMONWEALTH EDISON CO
COMMONWEALTH EDISON CO
NORTHERN INDIANA PUB
NORTHERN INDIANA PUB
NORTHERN INDIANA PUB
SERV. CO
SERV CO
SERV CO
NORTHERN INDIANA PUB SERV CO
COMMONWEALTH EDISON CO IN INC
INDIANA MICHIGAN POWER CO
IOWA PUBLIC SERVICE CO
MUSCATINE CITY OF
IOWA ELECTRIC LIGHT &
KANSAS CITY CITY OF
POWER CO
KANSAS CITY POWER & LIGHT CO
KANSAS CITY CITY OF
BOILER
GB01
GB02
GB03
GB04
1
2
01
02
31
32
6
1
2
7
a
1
2
3
A
61
62
61
62
17
1
2
7
8
12
14
4
U4
1
8
3
3
1
1
FIRING
TYPE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
BOTTOM
TYPE
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
NAMEPLATE
CAPACITY
(MWa)
125.00
125.00
179.52
187.50
623.05
634.50
388.06
616.50 .
90.25
90.25
360.40
659.70
659.70
37.50
37.50
33.00
33.00
33.00
173.00
892.80
892.80
892.80
892.80
121.00
^ 187.50
183.75
194.00
421.60
540.00
540.00
388.96
579.70
147.05
75.00
81.60
65.28
893.40
81.60
COM
DATE
1957
1950
1960
1963
1970
1973
1965
1972
1968
1972
1959
1987
1968
1961
1985
1963
1963
1963
1978
1972
1972
1975
1975
1951
1955
1955
1962
1968
1974
1976
1962
1964
1964
1969
1961
1962
1973
1965
1690 HE AT
INPUT
(OBtu)
6.5604890
6.8700440
8.7183550
9.8375710
30.2082700
33.6716390
12.7391450
22.6853900
4.2457250
2.9833730
3.5967990
24.6687340
29.0791550
0.7565660
0.5852170
0.3738360
0.4163270
1.3159010
12.1554550
8.5587950
10.4857770
16.1212330
13.3083340
0.4566470
2.5372980
4.5667510
11.3507960
15.0107630
25.7698190
13.7693910
6.7045790
30.7332060
5.3215280
3.0698110
2.7113180
0.4816790
29.369357!
3.5499800
NCONTROLLEO NOx
Ib/MMBIu)
1.327
1.272
1.519
1.482
1.700
1.470
1.260
1.260
0.928
1.118
0.907
1.300
1.300
0.733
0.733
1.320
1.320
1.320
1.320
0.915
0.915
0.915
0.915
0.807
0.894
0.871
1.500
1.500
1.320
1.328
0.748
1.910
0.940
0.978
0.717
0.766
1.088
0.970
Souic*
CREV
CREV
CREV
CREV
ETS
ETS
ETS
ETS
CREV
CREV
CREV
ETS
ETS
CREV
CREV
UARG
UARG
UARG
UARG
CREV
CREV
CREV
CREV
CREV
CREV
CREV
ETS
ETS
ETS
CREV
CREV
ETS
ETS
CREV
CREV
CREV
CREV
CREV
-------
Table/)A-2 (Continued)
Cyclone Fired Boilers
STATE
KENTUCKY
KENTUCKY
KENTUCKY
KENTUCKY
MARYLAND
MARYLAND
MINNESOTA
MINNESOTA
MISSOURI
MISSOURI
MISSOURI
MISSOURI
MISSOURI
MISSOURI
MISSOURI
MISSOURI
MISSOURI
MISSOURI
MISSOURI
MISSOURI
NEBRASKA
NEBRASKA
NEW HAMPSHIRE
NEW HAMPSHIRE
NEW JERSEY
NEW JERSEY
NORTH DAKOTA
NORTH DAKOTA
NORTH DAKOTA
NORTH DAKOTA
OHIO
OHIO
OHIO
OHIO
OHIO
OHIO
SOUTH DAKOTA
TENNESSEE
PLANT
ELMER SMITH
3ARADISE
PARADISE
PARADISE
C P CRANE
C P CRANE
ALLEN S KINO
RIVERSIDE
ASBURY
CHAMOIS
LAKE ROAD
NEW MADRID
NEW MADRID
SIBLEY
SIBLEY
SIBLEY
SIOUX
SIOUX
THOMAS HILL
THOMAS HILL
SHELDON
SHELDON
MERRIMACK
MERRIMACK
B L ENGLAND
B L ENGLAND
COYOTE
LELANDOLDS
MILTON R YOUNG
MILTON R YOUNG
CONESVILLE
CONESVILLE
MUSKINGUM RIVER
MUSKINGUM RIVER
NILES
NILES
BIG STONE
ALLEN
UTILITY
OWENSBORO CITY OF
TENNESSEE VALLEY AUTHORITY
TENNESSEE VALLEY AUTHORITY
TENNESSEE VALLEY AUTHORITY
BALTIMORE GAS & ELECTRIC CO
BALTIMORE GAS & ELECTRIC CO
NORTHERN STATES POWER CO
NORTHERN STATES POWER CO
EMPIRE DISTRICT ELECTRIC CO
CENTRAL ELECTRIC POWER COOP
ST JOSEPH LIGHT & POWER CO
ASSOCIATED ELECTRIC COOP INC
ASSOCIATED ELECTRIC COOP INC
UTILICORP UNITED INC
UTILICORP UNITED INC
UTILICORP UNITED INC
UNION ELECTRIC CO
UNION ELECTRIC CO
ASSOCIATED ELECTRIC COOP INC
ASSOCIATED ELECTRIC COOP INC
NEBRASKA PUBLIC POWER DISTRICT
NEBRASKA PUBLIC POWER DISTRICT
PUBLIC SERVICE CO OF NH
PUBLIC SERVICE CO OF NH
ATLANTIC CITY ELECTRIC CO
ATLANTIC CITY ELECTRIC CO
MONTANA-DAKOTA UTILITIES CO
BASIN ELECTRIC POWER COOP
MINNKOTA POWER COOP INC
MINNKOTA POWER COOP INC
COLUMBUS SOUTHERN POWER CO
COLUMBUS SOUTHERN POWER CO
OHIO POWER CO
OHIO POWER CO
OHIO EDISON CO
OHIO EDISON CO
OTTER TAIL POWER CO
TENNESSEE VALLEY AUTHORITY
BOILER
1
1
2
3
1
2
1
a
1
.2
6
1
2
1
2
3
1
2
MB1
MB2
1
2
1
2
1
2
B1
2
B1
B2
1
2
3
4
1
2
1
1
FIRING
TYPE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
BOTTOM
TYPE
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
NAMEPLATE
CAPACITY
(MWe)
151.00
704.00
704.00
1150.20
100.40
209.44
508.40
23B.65
231.55
44.00
90.00
600.00
600.00
55.00
§0.00
418.50
549.80
549.80
180.00
285.00
108.80
119.85
113.60
345.60
136.00
• 163.20
450.00
440.00
257.00
477.00
148.00
136.00
237.50
237.50
125.00
125.00
456.00
330.00
COM
DATE
1964
1963
1963
1969
1961
1962
1968
1964
1970
1060
1966
1972
1977
1980
1962
1969
1987
1968
1966
1969
1960
1964
1960
1968
1962
1964
1981
1975
1970
1977
1959
19b/
1957
1958
1953
1954
1976
1951
1990 HEAT
INPUT
(OBIu)
8.1617950
45.1297220
30.1492820
50.4944440
8.9415260
9.9828680
32.1317670
12.3620520
14.2313340
2.4340000
2.3499020
32.4647340
33.3642780
0.9578420
1.5850440
15.0176580
19.5632420
25.0897970
6.0113280
7.5607320
6.3003910
5.2712410
8.7181080
17.6801220
6.7533290
B.0946150
24.1816990
27.8957000
18.6890430
36.6861900
1.7122600
5.4652000
8.4730910
10.9051150
6.7483510
5.6055230
25.6133091
16.3418091
NCONTROLLEO NOx
Ib/MMBlu)
1.450
1.833
1.722
1.940
1.270
1.460
1.222
0.979
1.090
1.233
1.144
1.470
1.320
1.190
1.190
1.190
1.070
1.210
0.900
0.900
0.825
0.828
1.170
1.960
0.890
0.960
0.811
1.034
0.811
1.046
1.040
1.040
1.090
1.090
0.980
0.930
0.808
1.950
Source
ETS
CREV
CREV
ETS
ETS
ETS
CREV
CREV
ETS
CREV
CREV
ETS
ETS
ETS
ETS
ETS
ETS
ETS
ETS
ETS
CREV
CREV
ETS
ETS
ETS
ETS
CREV
CREV
CREV
CREV
ETS
ETS
CREV
CREV
ETS
ETS
CREV
ETS
-------
f
TableM-2 (Continued)
Cyclone Fired Boilers
STATE
TENNESSEE
TENNESSEE
WEST VIRGINIA
WEST VIRGINIA
WEST VIRGINIA
WEST VIRGINIA
WISCONSIN
WISCONSIN
WISCONSIN
WISCONSIN
WISCONSIN
WISCONSIN
WISCONSIN
PLANT
ALLEN
ALLEN
KAMMER
KAMMER
KAMMER
WILLOW ISLAND
BAY FRONT
EDGEWATER
EDGEWATER
NELSON DEWEY
NELSON DEWEY
ROCK RIVER
ROCK RIVER
UTILITY
TENNESSEE VALLEY AUTHORITY
TENNESSEE VALLEY AUTHORITY
OHIO POWER CO
OHIO POWER CO
OHIO POWER CO
MONONGAHELA POWER CO
NORTHERN STATES POWER CO
WISCONSIN POWER & LIGHT CO
WISCONSIN POWER & LIGHT CO
WISCONSIN POWER a LIGHT CO
WISCONSIN POWER &. LIGHT CO
WISCONSIN POWER & LIGHT CO
WISCONSIN POWER & LIGHT CO
BOILER
2
3
1
2
3
2
5
3
A
1
2
1
2
FIRING
TYPE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
CYCLONE
BOTTOM
TYPE
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
NAMEPLATE
CAPACITY
(MW«)
330.00
330.00
237.50
237.50
237.50
163.20
67.00
66.00
351.00
113.60
113.60
75.00
75.00
COM
DATE
1959
1959
1958
1956
1959
1960
1949
1951
1969
1959
1962
1953
1955
1990 HEAT
INPUT
(OBIu)
11.6469000
6.6339480
16.4168070
14.6240860
13.9501920
5.9307270
0.0000000
3.3468000
16.9814880
4.7544380
5.1293390
2.4068960
1.9435030
UNCONTROLLED NOx
(Ib/MMBlu)
1.910
1.870
1.340
1.340
1.340
1.264
0.826
0.770
1.170
0.690
0.690
1.000
1.000
Sourc*
ETS
ETS
ETS
ETS
ETS
CREV
CREV
ETS
ETS
ETS
ETS
UARG
UARG
-------
Table/lA-3
Wet Bottom Boilers
STATE
COLORADO
COLORADO
FLORIDA
FLORIDA
:LORIDA
:LORIDA
FLORIDA
INDIANA .
NDIANA
NDIANA
INDIANA
INDIANA
INDIANA
KANSAS
KENTUCKY
KENTUCKY
KENTUCKY
MICHIGAN
NEW JERSEY
NEW JERSEY
NEW YORK
MEW YORK
NEW YORK
NEW YORK
OHIO
OHIO
OHIO
OHIO
OHIO
OHIO.
OHIO
OHIO
OHIO
OHIO
OHIO
OHIO
OHIO
UTAH
PLANT
MYDEN
'AWNEE
JIG BEND
310 BEND
310 BEND
: J GANNON
: J GANNON .
CLIFTY CREEK
CLIFTY CREEK
CLIFTY CREEK
CLIFTY CREEK
CLIFTY CREEK
CLIFTY CREEK
KAW
CANE RUN
MILL CREEK
MILL CREEK
JAMES DE YOUNG
MERCER
MERCER
C R HUNTLEY
C R HUNTLEY
C R HUNTLEY
C R HUNTLEY
KYGER CREEK
KYGER CREEK
KYGER CREEK
KYGER CREEK
KYGER CREEK
MUSKINGUM RIVER
MUSKINGUM RIVER
R E BURGER
R E BURGER
R E BURGER
R E BURGER
R E BURGER
R E BURGER
BONANZA
UTILITY
COLORADO-UTE ELECTRIC ASSN INC
HJBLIC SERVICE CO OF COLORADO
TAMPA ELECTRIC CO
TAMPA ELECTRIC CO
TAMPA ELECTRIC CO
TAMPA ELECTRIC CO
TAMPA ELECTRIC CO
NDIANA-KENTUCKY ELECTRIC CORP
NDIANA-KENTUCKY ELECTRIC CORP
NDIANA-KENTUCKY ELECTRIC CORP
NDIANA-KENTUCKY ELECTRIC CORP
NDIANA-KENTUCKY ELECTRIC CORP
NDIANA-KENTUCKY ELECTRIC CORP
KANSAS CITY CITY OF
LOUISVILLE GAS & ELECTRIC CO
LOUISVILLE GAS & ELECTRIC CO
LOUISVILLE GAS & ELECTRIC CO
HOLLAND CITY OF
PUBLIC SERVICE ELECTRIC&GAS CO
PUBLIC SERVICE ELECTRIC&GAS CO
NIAGARA MOHAWK POWER CORP
NIAGARA MOHAWK POWER CORP
NIAGARA MOHAWK POWER CORP
NIAGARA MOHAWK POWER CORP
OHIO VALLEY ELECTRIC CORP
OHIO VALLEY ELECTRIC CORP
OHIO VALLEY ELECTRIC CORP
OHIO VALLEY ELECTRIC CORP
OHIO VALLEY ELECTRIC CORP
OHIO POWER CO
OHIO POWER CO
OHIO EDISON CO
OHIO EDISON CO
OHIO EDISON CO
OHIO EDISON CO
OHIO EDISON CO
OHIO EDISON CO
DESERET GENERATION & TRAN COOP
BOILER
H1
1
BB01
BB02
BB03
GBOS
GB06
1
2
3
4
5
6
1
4
3
4
6
1
2
63
64
65
66
1
2
3
4
5
1
2
1
2
3
4
5
6
1-1
FIRINO
TYPE
FRONT
FRONT
OPPOSED/TURBO
OPPOSEDfTURBO
OPPOSEDrtURBO
OPPOSED
OPPOSED
FRONT
FRONT
FRONT
FRONT
FRONT
FRONT
FRONT
FRONT
OPPOSED
OPPOSED
FRONT
FRONT
FRONT
ARCH
ARCH
ARCH
ARCH
FRONT
FRONT
FRONT
FRONT
FRONT
FRONT
FRONT
ROOF
ROOF
ROOF
ROOF
ROOF
ROOF
OPPOSED
BOTTOM
TYPE
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
WET
NAMEPLATE
CAPACIVY
(MWc)
190.00
500.00
445.50
445.50
445.50
239.36
414.00
217.26
217.26
217.26
217.26
217.28
217.28
46.00
163.20
462.60
543.60
26.75
326.40
326.40
60.00
60.00
60.00
60.00
217.26
217.26
217.26
217.26
217.28
219.69
219.69
62.50
62.50
62.50
62.50
100.00
100.00
400.00
COM
DATE
1968
1981
1970
1973
1976
1985
1967
1954
1955
1955
1955
1955
1956
1954
1982
1978
1982
1969
1960
1901
1942
1946
1953
1954
1955
1955
1955
1955
195!)
1953
1954
1944
1944
1947
1947
1950
19S>0
198b
1890 HEAT
INPUT
(OBIu)
13.3067590
34.1895360
16.4036840
23.7656440
27.3514490
15.0333430
9.2538380
12.4347740
17.1267610
16.1717330
15.6984550
17.1484240
15.9761690
0.8045750
7.3922350
19.2378010
32.0244950
1.8080370
15.4533950
10.8890100
6.6464390
6.8837910
5.3529740
8.1400000
16.5446340
16.8041350
15.9875470
15.8744420
15.3839760
9.9537390
11.1628340
1.1410640
1.2417790
1.2052110
1.1982330
.1.9375720
2.0177160
27.0444140
UNCONTROLLED NOx
(Ib/MMBIu)
0.220
0.955
1.260
1.260
0.640
1.263
1.608
1.680
1.680
1.680
1.710
1.710
1.710
0.623
0.985
0.594 •
0.367
0.994
1.354
1.899
0.908
0.908
0.908
0.908
1.410
1.410
1.410
1.410
1.410
1.090
1.090
0.630
0.900
0.940
1.090
0.750
0.730
0.550
Some*
NURF
NURF
ETS
ETS
ETS
CREV
CREV
ETS
ETS
ETS
ETS
ETS
ETS
CREV
CREV
CREV
CREV
CREV
CREV
CREV
CREV
CREV
CREV
CREV
ETS
ETS
ETS
ETS
ETS
CREV
CREV
ETS
ETS
ETS
ETS
ETS.
ETSU
UARG
-------
Table/A-4
Vertically Fired Boilers
STATE
COLORADO
COLORADO
COLORADO
COLORADO
COLORADO
COLORADO
INDIANA
INDIANA
INDIANA
OHIO
OHIO
OHIO
OHIO
PENNSYLVANIA
PENNSYLVANIA
PENNSYLVANIA
PENNSYLVANIA
PENNSYLVANIA
PENNSYLVANIA
PENNSYLVANIA
PENNSYLVANIA
VIRGINIA
VIRGINIA
VIRG MIA
WES" VIRGINIA
WEST VIRGINIA
WEST VIRGINIA
WEST VIRGINIA
WEST VIRGINIA
WEST VIRGINIA
WEST VIRGINIA
WEST VIRGINIA
WEST VIRGINIA
PLANT
ARAPAHOE
ARAPAHOE
ARAPAHOE
ARAPAHOE
CHEROKEE
CHEROKEE
TANNERS CREEK
TANNERS CREEK
TANNERS CREEK
BAY SHORE
BAY SHORE
MIAMI FORT
MIAMI FORT
ELRAMA
ELRAMA
ELRAMA
HOLTWOOD
SUNBURY
SUNBURY
SUNBURY
SUNBURY
CLINCH RIVER
CLINCH RIVER
CLINCH RIVER
KANAWHA RIVER
KANAWHA RIVER
PHIL SPORN
PHIL SPORN
PHIL SPORN
PHIL SPORN
RIVESVILLE
RIVESVILLE
WILLOW ISLAND
UTILITY
PUBLIC SERVICE CO OF COLORADO
PUBLIC SERVICE CO OF COLORADO
PUBLIC SERVICE CO OF COLORADO
PUBLIC SERVICE CO OF COLORADO
PUBLIC SERVICE CO OF COLORADO
PUBLIC SERVICE CO OF COLORADO
INDIANA MICHIGAN POWER CO
INDIANA MICHIGAN POWER CO
INDIANA MICHIGAN POWER CO
TOLEDO EDISON CO
TOLEDO EDISON CO
CINCINNATI GAS & ELECTRIC CO
CINCINNATI GAS & ELECTRIC CO
DUQUESNE LIGHT CO
DUQUESNE LIGHT CO
DUQUESNE LIGHT CO
PENNSYLVANIA POWER & LIGHT CO
PENNSYLVANIA POWER & LIGHT CO
PENNSYLVANIA POWER & LIGHT CO
PENNSYLVANIA POWER & LIGHT CO
PENNSYLVANIA POWER & LIGHT CO
APPALACHIAN POWER CO
APPALACHIAN POWER CO
APPALACHIAN POWER CO
APPALACHIAN POWER CO
APPALACHIAN POWER CO
CENTRAL OPERATING CO
CENTRAL OPERATING CO
CENTRAL OPERATING CO
CENTRAL OPERATING CO
MONONGAHELA POWER CO
MONONGAHELA POWER CO
MONONGAHELA POWER CO
BOILER
1
2
3
4
1
2
U1
U2
U3
1
2
5-1
5-2
1
2
3
17
1A
1B
2A
2B
1
2
3
1
2
11
21
31
41
7
8
1
FIRING
TYPE
VERTICAL
VERTICAL
VERTICAL
VERTICAL
VERTICAL
VERTICAL
VERTICAL
VERTICAL
VERTICAL
VERTICAL
VERTICAL
ROOF
ROOF
ROOF
ROOF
ROOF
ARCH
ARCH
ARCH
ARCH
ARCH
ROOF
ROOF
ROOF
VERTICAL
VERTICAL
ROOF
ROOF
ROOF
ROOF
TOP
TOP
TOP
BOTTOM
TYPE
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
NAMEPLATE
CAPACITY
(MWe)
44.00
44.00
44.00
100.00
100.00
110.00
152.50
152.50
215.40
140.62
140.62
100.00
100.00
100.00
100.00
125.00
75.00
253.53
253.53
253.53
253.53
237.50
237.50
237.50
'219.69
219.69
152.50
152.50
152.50
152.50
35.00
74.75
50.00
COM
DATE
1950
1951
1951
1955
1957
1959
1951
1952
1954
1955
1959
1949
1949
1952
1953
1954
1954
1949
1949
1949
1949
1956
1958
1961
1953
1953
1949
1950
1951
1952
1944
1951
1949
1990 HEAT •
INPUT
(OBIu)
2.1823950
1.9238510
2.3245480
5.6034210
7.4214160
6.2873530
6.9099000
6.9750700
8.2409640
8.0714440
8.0672570
0.2064160
0.2064160
3.4592260
3.3554500
4.1995280
5.8314960
2.3458920
2.3464740
2.3464740
2.3464740
13.2137470
15.1694070
12.7258590
7.0583640
5.2641650
4.8176170
5.9663940
6.7833270
5.6310440
1.1801750
3.2031420
1.7776870
UNCONTROLLED NOx
(Ib/MMBtu)
0.850
0.850
0.850
1.060
1.006
1.100
1.141
1.141
1.141
1.081
1.081
0.859
0.859
1.000
1.000
1.000
1.072
1.057
1.057
0.847
0.847
1.335
1.335
1.417
1.231
1.231
1.207
1.207
1.207
1.207
0.840
0.840
0.881
Source
UARG
UARG
UARG
UARG
UARG
UARG
CREV
CREV
CREV
CREV
CREV
CREV
CREV
UARG
UARG
UARG
CREV
CREV
CREV
CREV
CREV
CREV
CREV
CREV
CREV
CREV
CREV
CREV
CREV
CREV
NURF
NURF
CREV
-------
TableOA-5
Stoker Fired Boilers
STATE
IOWA
IOWA
KENTUCKY
MICHIGAN
MINNESOTA
MINNESOTA
MISSOURI
MISSOURI
NEW YORK
NEW YORK
NEW YORK
NEW YORK
NEW YORK
NEW YORK
NEW YORK
NEW YORK
WISCONSIN
WISCONSIN
WISCONSIN
WISCONSIN
WISCONSIN
PLANT
PELLA
PELLA
HENDERSON 1
WYANDOTTE
M L HIBBARD
M L HIBBARD
COLUMBIA
COLUMBIA
HICKLINO
HICKLINO
HICKLING
HICKLING
JENNISON
JENNISON
JENNISON
JENNISON
BAY FRONT
BAY FRONT
BAY FRONT
MANITOWOC
MANITOWOC
UTILITY
PELLA CITY OF
PELLA CITY OF
HENDERSON CITY UTILITY COMM
WYANDOTTE MUNICIPAL SERV COMM
MINNESOTA POWER & LIGHT CO
MINNESOTA POWER & LIGHT CO
COLUMBIA CITY OF
COLUMBIA CITY OF
NEW YORK STATE ELEC & GAS CORP
NEW YORK STATE ELEC & GAS CORP
NEW YORK STATE ELEC & GAS CORP
NEW YORK STATE ELEC & GAS CORP
NEW YORK STATE ELEC & GAS CORP
NEW YORK STATE ELEC & GAS CORP
NEW YORK STATE ELEC & GAS CORP
NEW YORK STATE ELEC & GAS CORP
NORTHERN STATES POWER CO
NORTHERN STATES POWER CO
NORTHERN STATES POWER CO
MANITOWOC CITY OF
MANITOWOC CITY OF
BOILER
6
7
6
5
3
4
6
7
1
2
3
4
1
2
3
4
1
2
4
6
7
FIRING
TYPE
STOKER/SPR
STOKER/SPR
STOKER/SPR
STOKER/SPR
STOKER/SPR
STOKER/SPR
STOKER/SPR
STOKER/SPR
STOKER/SPR
STOKER/SPR
STOKER/SPR
STOKER/SPR
STOKER/SPR
STOKER/SPR
STOKER/SPR
STOKER/SPR
STOKER/SPR
STOKER/SPR
STOKER/SPR
STOKER/SPR
STOKER/SPR
BOTTOM
TYPE
DRY
DRY
DRY
DRY,
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
NAMEPLATE
CAPACITY
. (MWe)
38.00
38.00
32.30
73.00
35.00
37.50
73.50
73.50
37.50
37.50
49.00
49.00
37.50
37.50
' 37.50
37.50
67.00
67.00
67.00
79.00
79.00
COM
DATE
1964
1964
1968
1958
1949
1951
1957
1957
1948
1948
1952
1952
1945
1945
1950
1950
1949
1949
1949
1935
1935
1990 HEAT
INPUT
(OBIu)
0.3176750
0.6465380
0.4825320
0.9372060
0.5667950
0.3475240
0.1601240
0.6072390
1.6016710
1.5928490
1.9472650
2.2057190
1.2857800
1.2298170
1.3538600
1.3422850
1.2999630
0.0000000
0.0000000
1.0558550
0.9341210
UNCONTROLLED NOx
(Ib/MMBIu)
0.496
0.496
0.517
0.400
0.400
0.400
0.400
0.400
0.377
0.377
0.261
0.261
0.300
0.300
0.286
0.286
0.403
0.400
0.657
0.314
0.314
Source
CREV
CREV
CREV
UARG
UARG
UARG
UARG
UARG
CREV
CREV
CREV
CREV
CREV
CREV
CREV
CREV
CREV
UARG
CREV
CREV
CREV
-------
Table/K-6
FBC Boilers
STAT6
COLORADO
KENTUCKY
MINNESOTA
NORTH DAK
TEXAS
TEXAS
PLANT
NUCLA
SHAWNEE
BLACK DOG
R M HESKETT
TNP ONE
TNP ONE
UTILITY
COLORADO-UTE ELECTRIC ASSN INC
TENNESSEE VALLEY AUTHORITY
NORTHERN STATES POWER CO
MONTANA-DAKOTA UTILITIES CO
TEXAS-NEW MEXICO POWER CO
TEXAS-NEW MEXICO POWER CO
BOILER
1
10
2
B2
U1
U2
FIRING
TYPE
CFB
AFB
AFB
AFB
CFB
CFB
BOTTOM
TYPE
DRY
DRY
DRY
DRY
DRY
DRY
NAMEPLATE
CAPACITY
. (MWe)
113.88
175.00
137.00
75.00
194.00
194.00
COM
DATE
1990
1956
1954
1963
1990
1991
1990 HEAT
INPUT
(QBIu)
4.7857640
3.5407200
1.1586080
4.6598200
8.1054060
0.0000000
UNCONTROLLED NOx
(Ib/MMBIu)
0.170
0.230
0.258
0.286
0.169
0.153
Souice
CREV
ETS
MODEL
CREV
CREV
CREV
-------
The use of low NOX retrofits on cell burners could change the burner pressure drop relative to the
original boiler operation and, thereby, impact the required forced draft fan power.
Similar to low-NO, burners used on Group 1 boilers, cell burner retrofits typically have a higher
pressure drop than regular burners because of their basic design. However, for some retrofit
situations, it may be difficult to observe changes because, in addition to burner replacement, the air
distribution system may also be modified. This observation can be supported by comparison of two
of the cell burner retrofit projects being cited. The retrofit at J. M. Stuart Station was conducted by
replacing all of the existing 24 cell burners with 24 low-NOx cell burners without significant changes
in the boDert air distribution system. As a result, the average burner pressure drop increased by 23
inches H2O. AttheW.H. Sammis Power Station, where the retrofit included significant changes in
the air distribution system, the pressure drop from windbox-to-furnace was actually reduced from its
original level by 0.64 inches H2O. No data are currently available for the other projects.
As with Group 1 LNBs, the approach qf replacing original equipment cell burners with low-NO,
"retrofit burners is likely to increase windbox-to^pressure drop by several inches H2O based on the
Stuart Station results. However, unfike the Sammis Station retrofit, the burner air distribution system
is unlikely to be modified for most retrofits of this type, and the distribution system pressure drop
should not Change Therefore, based on the limited data available from current retrofit projects, the
total windbox-to-furnace pressure drop may increase by up to several inches H2O. While this increase
should be incorporated into the boiler retrofit cost assessment, no quantitative sensitivity analysis is
warranted.
C.2.1.2 NQX Reduction Performance
Table C-2 shows full- and partial-load NOX reduction performance for all six cell burner retrofit
projects previously identified. It is evident from this table that all four low-NOx plug-in retrofits
provide significant NOX reduction, as compared to baseline operation. For the projects with
installation of plug-in retrofits, the range of NO, reduction at full load is between 50% and 55%
(since it is not known whether Detroit Edison's Monroe Unit #1 was tested at its nominal full load
and whether this unit had a tow baseline NOX level). [Note also that these projects do not include
overfireair.] For plug-in burners with overfire air the expected NOx reductions are 50-65%. [The
"non plug-in" type retrofits of ceH burners has yielded up to 69% NOX reduction at full load.]
[A smaller percent reduction of 37% was obtained at 60% load, but was most likely due to a
relatively low NOX production level during the pre-retrofit conditions, rather than a result of
degradation of retrofitted boiler performance at partial load. Notice that NOX emissions for the
Sammig plant were similar fyf fill) and partial loads, alriymgh the pfflWUt reductions WCTC signifiipmrty
different]
Continuous emissions monitoring data collected for JJvL Stuart Unit 4 and Muskingum River Unit
5 show consistent long-term NOX reductions of 55%.
C-8
-------
1 ) Gas temperature at the injection location
2) Urea/ammonia injection location and injector design
3) Chemical reagent type and stoichiometry
4) Uncontrolled NOX emissions
• Gas Temperature at the Injection Location
As mentioned in almost all SNCR technical references, the SNCR process operates optimally over
a relatively narrow temperature range with either ammonia- or urea-based reagents. This range is
from 1600° F to 1850 F with peak removals nominally occurring at 1730 F. However, data
presented in [9] have shown that reagents can be successfully injected in furnace locations with
temperatures as higb as 2400° F, if the baseline NOZ level is relatively high. This concept, described
in some detail in [8,9], opens new options for SNCR implementation on Group 2 boilers, which
typically exhibit a furnace exit gas temperature (FEGT) above 2100° F, as well as high boiler outlet
NOX concentrations (> 1 Ib/MMBtu). Test data for the three projects mentioned previously are
presented in Table C-10.
These data confirm the conclusions of ICAC member companies [8,9] regarding the effectiveness of
urea injection in flue gas with temperatures higher than 2100° F. However, they also show that NOZ
reduction is greater for operation within the optimum temperature range, and the required urea
stoichiometry is lower.
• Urea/ Ammonia Tnieion Location and
The appropriate choice for SNCR injection location and injector technology will be based on two
considerations, namely 1) providing a uniform, reagent distribution in the gas flow, and 2) providing
a proper time-temperature regime. Test data for 13 large-scale SNCR applications provided in [ 1 ]
show mat the highest efficiency can be achieved by injecting the reagent into me furnace, even if the
temperature at the injection location is higher than optimum. It was mentioned in [9] mat the lower
efficiency that occurs when the injection system is installed in cavities of the convective pass is
caused by lower residence time and non-uniform distribution of the reagent Also, a convective pass
injection system is usually more complicated and, accordingly, more costly. Therefore, injection into
the furnace is the most common design; bothNFT and all three Group 2 boiler applications discussed
above have used multi-level in-fumace injection through wall-mounted injectors.
• Chemic i
Either urea or ammonia can be used as the reagent in an SNCR application. Kinetically, one mole
of urea will react with two moles ofNO. while one mole of ammonia wffl react with one mole of NO..
Thus, in large boilers, the use of urea instead ammonia may result m lowering the required reagent
quantity. Additionally, urea is safer to handle and its use can lead to a lower cost for reagent
handling and supply equipment Urea was used on both NFT process applications at the BX.
England and the Mercer stations. A system designed by NOELL for the Arapahoe station is also
C-39
-------
Appendix B
, *r
1.0 INTRODUCTION
This appendix documents detailed results of a study conducted by Bechtel Power
Corporation to develop costs associated with various NOX control technology applica-
tions for the coal-fired, Group 2 boilers. For each Group 2 boiler category, one or
more applicable NOX control technologies were studied. The technology selection
was provided by the U.S. Environmental Protection Agency (EPA), as discussed in the
main report. The Group 2 boilers and the Iow-N0x technologies covered in the study
are as follows:
a. Cell-Burner Boilers
• Combustion modifications (plug-in Iow-N0x burners)
• Combustion modification (non plug-in Iow-N0x burners)
b. Cyclone-Fired Boilers
• Coal reburning
• Gas reburning
• Selective non-catalytic reduction (SNCR)
• Selective catalytic reduction (SCR)
c. Wet-Bottom Boilers
• SNCR
d. Vertical-Fired, Dry-Bottom Boilers
• SNCR
The primary objective for this study was to develop costs that accurately represent
typical low-NOx technology retrofit applications for the Group 2 boilers. The costs
developed included both capital and levelized costs. To facilitate comparisons
between various technology cases, the levelized costs were estimated both in mils per
killowatthour (mils/kWh) and dollars per ton ($/ton) of NOX removed. *
A
The study activities included selection of boilers for each technology application,
determination of performance and equipment impacts of the technology retrofits,
estimation of capital and levelized costs, and development of cost algorithms to cover
the plant size range existing within each boiler category population. In addition,
22M6.007Slu4y>Gi»2Boil
-------
Appendix B
* is
h. The costs for inventory capital apply only to the SNCR and SCR technologies
where chemical reagents will be stored on-site. These costs are based on a
14-day storage of the reagent in each case.
i. The allowance for funds during construction (AFDUC) was assumed to be
zero for the following reasons:
• The AFDUC depends on the payment terms agreed upon between the
buyer and the equipment suppliers, which may vary considerably
between various applications.
• The construction period for all technologies is estimated to be less than
a year.
•
• Even for the technologies with significant capital investments, such as
SCR and coal reburning, any AFOUC would be negligible, especially
because of the short construction period.
The technology retrofit design requirements are expected to vary with different
installations. Site-specific factors may influence not only the design requirements
pertaining to the technology components, but also the required modifications of
existing equipment.
The conceptual designs developed for each study case were based on site conditions
applicable to a typical Group 2 boiler. In some cases, additional requirements may be
imposed because of special limitations prevailing at certain sites. Based on the
experience with these technologies at existing installations, these additional
requirements have been anticipated and shown in the cost estimates as scope adder
items.
The scope adder costs have not been included in the costs shown on the figures. The
contingencies provided in the cost estimates generally cover these additional costs.
It is, therefore, assumed that the costs incurred by a few plants requiring any of the
scope adder items would still fall within the predicted ranges in this report.
232 Levelized Cost Estimates
For each technology retrofit, the ievelized cost estimates -consider the following
components:
• Carrying charges for the capital costs
• Increases in fuel costs associated with the retrofit technology
Study Methodology • B2-5
-------
Appendix-B-
2.4.2 Scaling of Levelized Costs
The levelized costs are made up of various components. Of these, the carrying
charges and maintenance costs are a direct function of the capital costs. These
components can, therefore, be estimated using the pertinent economic factors along
with the predicted capital costs for the boilers within the corresponding range.
For the other levelized cost components, a statistical curve fitting approach was
followed to define the variation in each component value with respect to changes in
the boiler size. Two points for this curve fitting exercise were available for each
component from the data generated for the two generic boilers. A third point on the
.curve was assumed to be the zero point (i.e., for a unit size of zero MW, each
component assumed a value of zero).
2.4.3 Reported Cost Data
The scaling factors were used to generate curves showing the variation in the capital
and levelized costs with respect to changes in the boiler size. For each study case,
the following three curves were developed:
• Capital cost in $/kW versus unit size
t
• Levelized cost in mils/kWh versus unit size
4
• Levelized cost in $/ton of NOX removed versus unit size
2.5 SENSITIVITY EVALUATIONS
The costs reported in this study were based on certain assumptions and specific
economic factors described earlier. Sensitivity evaluations were conducted to analyze
the impact of varying some of the critical system design parameters and economic
factors.
Appendix C of the main report provides detailed discussions on various parameters
and factors whose variations can have an impact on the design and performance of
the NOX control technologies. Table B2-3 lists some of the critical parameters and
factors for which sensitivity evaluations were performed. The range values assigned
to each parameter for this evaluation are also provided in Table B2-3.
The ranges for the parameter values in Table B2-3 were established as follows:
a. The capital cost estimates for each technology include a 15 percent margin
for project contingency and a 5 percent margin for process contingency.
Since these margins already cover possible cost increases for the Group 2
boiler applications, there is a low potential for variations beyond these
Study Methodology • B2-8
-------
Appendix B
TABLE B2-2
ECONOMIC FACTORS
Cost Year
Useful Life
Plant Capacity Factor
Carrying Charges
Levelization Factors
Maintenance Cost
Electrical Power Cost
Coal Cost: Western Subbituminous
Eastern Bituminous
Midwestern Bituminous
Lignite
Natural Gas Cost
Ash Disposal Cost
Anhydrous Ammonia Cost
Urea Cost (50% solution)
SCR Catalyst Replacement Cost
SCR Catalyst Operating Life
Operator Cost
Water Cost
SO2 Allowance
November 1990
20 years
65%
0.115
1.0
1.5% (of capital)/year
$0.05/kWh
$1.06/MMBtu
$1.60/MMBtu
$1.45/MMBtu
$1.58/MMBtu
$2.68/MMBtu
$9.0/ton
$162/ton of dry NH3
$0.75/gallon
saso/ft3
3 years
$21.00/person hour
$0.0004/gallon
$150/ton
22886.0O2\Snjdv\Gn>28«»l
52-77
-------
Appendix B
The maximum continuous rating (MCR) conditions for the boiler consist of a main
steam flow of 2,200,000 Ib/h, a reheat steam flow of 1,770,000 Ib/h, and pressure/
temperatures of 2,625 psig/1005 °F/1005 °F.
Eight pulverizers are provided, each supplying coal to the three burners in one cell.
Full-load operation can be achieved with one pulverizer offline. The system design is
based on a minimum coal fineness of 70 percent through 200 mesh.
Two regenerative type air heaters are provided at the economizer outlet. An
electrostatic precipitator (ESP) located downstream of the air heater provides control
of paniculate emissions. Two half-capacity forced draft (FD) fans deliver combustion
air to the boiler. The flue gases from the boiler .pass through the ESP and are
discharged to the atmosphere through a stack by two half-capacity induced draft (ID)
fans.
The 600 MW unit consists of a supercritical, once-through, balanced-draft, double-
reheat, single-furnace boiler. The boiler is equipped with 20 burner cells arranged in
a two high and five wide, opposed-fired arrangement. Each cell contains two burners.
The MCR conditions for the boiler consist of a main steam flow of 4,050,000 Ib/h,
a high-pressure reheat steam flow of 3,350,000 Ib/h, a low-pressure reheat steam
flow of 3,050,000 Ib/h, and pressure/temperatures of 3,800 psig/1005 °F/1030
°F/1055 °F.
Five pulverizers are provided, each servjng eight burners in four cells. Full-load
operation can be supported with four mils. The design coal fineness is 70 percent
through 200 mesh. A
The boiler backend equipment configuration is similar to the 300 -MW boiler. There
are two regenerative air heaters, an ESP, two half-capacity FD fans, and two half-
capacity ID fans.
3.2 PLUG-IN LOW-NOx BURNER APPLICATIONS
The plug-in burner technology used for this study entails direct replacement of each
individual burner in a cell with a new burner. Overfire air (OFA) ports are added above
the top-most burner elevation. This technology has the potential for retrofit in both
two- and three-cell burner boilers.
The plug-in type burners used in the study can be supplied by most of the major boiler
suppliers. One supplier, Babcock and Wilcox (B&W), offers a different plug-in burner
design. To date, this design, referred to as the Low-NOx Cell Burner (LNCB)
technology, has been applied only to boilers with two-burner cells.
Cell-Burner Boilers • B3-2
-------
Appendix B
3.2.4 Levelized Cost Estimates
The capital cost and plant performance impacts identified in this study were used to
develop the overall levelized costs for the 300 and 600 MW units. The costs were
estimated both in mils/kWh and $/ton of NOX removed. For the study boilers, the
cost components are as follows:
Mils/kWh
300 MW 600 MW
$/ton NOX
300 MW 600 MW
Coal consumption
Power consumption
Ash disposal
General maintenance
Capital cost charge
Total levelized costs
0.043
0.007
0.009
0.034
0.260
0.353
0.043
0.007
0.009
0.024
0.183
0.266
13.75
2.26
3.02
10.92
83.70
113.65
11,80
1.87
2.54
' 6.60
50.63
73.44
The scaling methodology defined in Section 2.4 of this appendix was used to extend
the above costs to cover the boiler size range in the cell-burner category. Figures B3-
4 and B3-5 show the levelized costs in relation to the unit size. As shown for the 200
to 1,300 MW unit size range, the ievelized costs vary from approximately 0.42 to
0.195 mils/kWh and $163 to $48/ton of NOX removed.
3.2.5 Sensitivity Analyses
The sensitivity evaluations were performed by varying the following parameters for
the range values shown in parentheses. (Refer to Section 2.5 of this appendix for the
methodology.)
• Capital costs (+. 5 percent)
• Capacity factor (50-85 percent)
• NOX reduction (50 - 65 percent)
Figures B3-6 through B3-11 show the results of these variations on the technology's
capital and levelized costs. The results are summarized below:
a. The capital cost variation has a minor impact on the capital and levelized
costs.
b. The capacity factor variation has the greatest impact on the levelized costs
(both mils/kWh and $/ton of NOX). There is no impact on the capital cost.
2288S.002\Studv\Gn>2Mi
Cell-Burner Boilers • B3-6
-------
TABLE B3-3 (Continued)
Appendix B
Air for Combustion, klb/hr
Flue Gas Leaving Boiler, klb/hr
Total Solid Waste, klb/hr
COAL ANALYSIS:
Proximate Analysis, %
Moisture
Ash
Fixed Carbon
Volatile Matter
Ultimate Analysis, %
Carbon
Hydrogen
Nitrogen
Sulfur
Ash
Oxygen
Moisture
HHV, Btu/lb
Grindability
BOILER CHARACTERISTICS AND NOX EMISSIONS:
Number of Burners
Pulverizer Performance
Number of Pulverizers
Burners per Pulverizer
Average Excess Air, %
Ignitors
Original Modified
2.372 2.379
2,564 2,572
25.7 26.3
Eastern Bituminous
Original Modified
24 24
70% through 200 mesh
8 8
3 3
20 20
24 24
B3-16
-------
TABLE B3-8 (Continued)
Appendix.
Air for Combustion, klb/hr
Flue Gas Leaving Boiler, klb/hr
Total Solid Waste, klb/hr
Oriainal
2,372
2,564
25.7
Modified
2.379
2,572
26.3
COAL ANALYSIS:
Proximate Analysis, %
Moisture
Ash
Fixed Carbon
Volatile Matter
Ultimate Analysis, %
Carbon
Hydrogen
Nitrogen
Sulfur
Ash
Oxygen
Moisture
HHV, Btu/lb
Grindability
BOILER CHARACTERISTICS AND NOX EMISSIONS:
Number of Burners
Pulverizer Performance
Number of Pulverizers
Burners per Pulverizer
Average Excess Air. %
Eastern Bituminous
8.0 .-
i
11.5 I
46.0 j
34.5 !
67.7
4.3
1.3
0.7
11.5
6.5
8
12,200)
46 '
Original Modified
24 16
70% through 200 mesh
8 8
.3 2
20 20
2288B.002>Slwlv\C«aa<»l
B3-27
-------
TABLE B3-9 (Continued)
Appendix^}
Air for Combustion, klb/hr
Flue Gas Leaving Boiler, klb/hr
Total Solid Waste, klb/hr
COAL ANALYSIS:
Proximate Analysis, %
Moisture
Ash
Fixed Carbon
Volatile Matter
Ultimate Analysis, %
Carbon
Hydrogen
Nitrogen
Sulfur
Ash
Oxygen
Moisture
HHV, Btu/lb
Grindability
BURNER CHARACTERISTICS AND NOX EMISSIONS:
Number of Burners
Pulverizer Performance
Number of "Pulverizers
Burners per Pulverizer
Average Excess Air, %
Igniters
Original Modified
4,619 4,633
4.993 5.008
50.1 51.3
Eastern Bituminous
8.0
i
11.5 ;
46.0
34.5
67.7
4.3
1.3
0.7
11.5
6.5
8
12.200
46
Original Modified
40 40
70% through 200 mesh
5 5
8 8
20 20
40 40
228S6.002\Sudv\G.e2BaJ
B3-30
-------
Appendix B
• New coal reburn silo, feeder, pulverizer mill with classifier and
inerting/clearing system, primary air fan, and seal air fan
• Reburn burners with retractable oil lighters, spark igniters, scanners, scanner
cooling air fans
• OFA ports with flow adjustment dampers and drives
• Piping systems for pulverized coal, scanner air, and oil
• Ductwork for primary air, secondary air, gas recirculation, tempering air, and
seal air
• Pulverizer pyrites removal system
• Burner management system (BMS) for the pulverizer, reburn burners, and
OFA, with an interface to the existing BMS system
• Electrical equipment, including switchgear circuit breaker, step-down
transformer, switchgear, and motor control center
• Building enclosure for the silo, feeder, and pulverizer mill
• Platforms and stairways for^beaccess to the new equipment
The above equipment/material quantities vary between the 150 MW and 400 MW
units. (Refer to Tables B4-1 and B4-2.) For example, for the 150 MW unit, only one
pulverizer mill along with a coal silo, feeder, and primary air fan are required. To
accommodate the large reburn fuel use rate for the 4OO MW unit, two half-capacity
pulverizer mills must be provided. This requires use of two coal silos, feeders, and
primary air fans.
For the 150 MW boiler, the coal reburn burners and the OFA ports are instajjetfon the
boiler rear wall opposite the cyclones (Figure B4-1). For the 400 MWMhe reburn
burners and the OFA are installed on both the front and rear walls of the*boiler above
the cyclones (Rgure B4-2).
The following potential scope adder items have been identified for the coal reburn
retrofit. (Refer to Section 2.3.1 of this Appendix for definition.)
a. A stand-alone BMS system for the coal reburn equipment with an interface
to the existing BMS system was used in this study. For certain applications,
it may be necessary to replace the entire existing BMS system to make it
workable with the new equipment.
Cyclone-Fired Boilers • B4-4
-------
Appendix B
b. One potential impact of the coal reburning retrofit is an increase in the
amount of fly ash exiting the boiler. For Nelson Dewey,181 the inlet ash
loading to ESP increased from 1 .3 Ib/MMBtu for the baseline conditions to
2.52 Ib/MMBtu for the post-retrofit conditions. This increase resulted from
more of the total ash converting to fly ash.
Despite the ash loading increase, Nelson Dewey did not report any adverse
impact on the ESP performance. In fact, the limited testing done showed an
improvement in the average ESP outlet emissions. There was no increase in
the stack opacity levels. Thisjserformance was attributed to an increase in
>< the particle size distribution^ the ESP inlet.
For the study, it was assumed that coal reburning has no impact on the ESP
performance. To cover the potential for such ah impact, costs were
developed for an extension of the existing ESP surfaces to provide two
additional fields. This cost adder item also includes the modifications
required to the existing ash handling system to cover the additional ESP
fields.
c. For the plants where asbestos-laden insulation exists on the surfaces affected
by the coal reburn retrofit, a cost adder item has been shown for the
asbestos removal and reinsulation of the affected surfaces.
4.2.2 Performance Impacts
Tables B4-3 and B4-4 present the design and performance ratings of the 1 50 and
400 MW boilers, respectively. Both the original design and the post-retrofit conditions
are shown. The analyses of the coal fired and baseline NOX emissions are also
included. The coal reburning performance is based on the long-term results from
Nelson Dewey. m The highlights of the data presented are as follows:
a. It is assumed that the iow-NOx retrofit has no impact on the boiler's
capability to maintain the original MCR steam flow conditions, including the
steam flow rates, temperatures, and pressures.
b. The coal reburning impact on the flue gas temperatures in the boiler backpass
and at the air heater outlet is expected to be minimal. These temperatures
are, therefore, assumed to be the same for both pre- and post-retrofit
situations for the study boilers.
c. The NOX reduction for both boilers is assumed to be 50 percent. The
experience at Nelson Dewey showed reduction efficiencies as high as 60
percent. The long-term operating data for this installation when firing
bituminous and western coals showed an average reduction of over 50
percent.
Cyclone-Fired Boilers • B4-5
-------
Appendix B
• Existing BMS modifications to incorporate the reburn system
• Platforms and stairways for access to the new equipment
For the 150 MW boiler, the gas injectors and the OFA ports are Installed on the boiler
rear wall opposite the cyclones (Figure B4-14). For the 400 Mwf the reburn injectors
and the OFA are installed on both the front and rear walls of the boiler above the
cyclones (Figure B4-15).
The following potential scope adder items have been identified for the gas reburn
retrofit. (Refer to Section 2.3.1 of this Appendix for definition.)
a. It is assumed that the existing BMS system can be modified to incorporate
the reburn system. For certain applications, such a modification may not be
technically feasible and a new BMS system may be required.
b. A gas recirculation system exists for both the study boilers. As discussed
previously, a gas recirculation system may not be required for the gas
reburning technology. However, a cost estimate is provided for the addition
of a gas recirculation system for the Group 2 boilers where such a system is
not present. This cost adder item includes a gas recirculation fan, dust
collector, ductwork, existing ash handling system modifications to serve the
dust collector, and other accessories.
c. For the plants where asbestos-laden insulation exists on the surfaces affected
by the gas reburn retrofit, a cost adder item has been shown for the asbestos
removal and reinsulation of the affected surfaces.
4.3.2 Performance Impacts
Tables B4-8 and B4-9 present the design and performance ratings of the 150 and
400 MW boilers, respectively. Both the original design and the post-retrofit conditions
are shown. The analyses of the coal-fired and baseline NOX emissions are also
included. The gas reburning performance is based on the long-term operating results
of existing installations.'9'131 The highlights of the data presented are as follows:
a. It is assumed that the low-NOx retrofit has no impact on the boiler's
capability to maintain the original MCR steam flow conditions, including the
steam flow rates, temperatures, and pressures.
b. The gas reburning impact on the flue gas temperatures in the boiler backpass
and at the air heater outlet is expected to be minimal. These temperatures
are, therefore, assumed to be the same for both pre- and post-retrofit
situations for the study boilers.
Cyclone-Fired Boilers • B4-10
-------
Appendix B
The SNCR technology's effectiveness generally has been tied to the ability to inject
the reagent ina proper flue gas temperature zone (1 ,800 to 2,000 °F for urea). Also,
significant NOX reductions are considered possible only if an appropriate residence
time exists in this effective temperature zone. For many boilers, this temperature
zone occurs in the area of high-temperature surfaces where limited residence times
exist.
Experience tw" some operating installations now shows that significant NOX
reductions are possible with the reagent injected at temperatures exceeding the above
effective temperature zone.1181 The temperatures at the injection points for these
installations have been as high as 2,200 to 2,300 °F. These high temperatures allow
reagent injection within the furnace.
The critical NOx can be defined as the minimum NOx emission achievable with SNCR
for a given set of flue gas conditions. This minimum NOx is a function of the baseline
NOx concentration and the flue gas temperature. It can be calculated by assuming
that all nitrogen reactions have infinite time to complete with the result that the
reaction product species exist in their equilibrium concentrations.
The critical NOx calculations show that at high baseline NOx concentrations
significant NOx reductions are still theoretically possible, even though the reagent is
injected outside of the effective temperature zone. There is still a need to select this
temperature carefully, because the critical NOx emission increases with increasing
injection temperatures.
As an example, at a baseline NOx of 900 ppm, the critical NOx emission is
approximately 325 ppm at 2,200F injection temperature. This still represents a
significant NOx reduction potential, even though it is low in comparison to injection
in the effective temperature zone (at 1 ,900F, the critical NOx is approximately 50
ppm). At an injection temperature of 2,400F, the calculated critical NOx is
approximately 600 ppm, which implies a low NOx reduction potential.
In the above example, the reagent can be injected outside of the effective temperature
zone at 2.200F with a sizable NOx reduction. At this temperature, concerns regarding
ammonia slip are minimized because reagent decomposition to ammonia should not
occur until gas temperatures are below 2,200F.
For achieving reasonable NOx reductions, it is necessary that the baseline NOx be
relatively high. At a baseline NOx of 500 ppm, the calculated critical NOx is
approximately 230 ppm at an injection temperature of 2.200F, which shows a lower
NOx reduction potential (54%) compared to the potential (64%) at 900 ppm of
baseline NOx.
Since the effective temperature zone residence time is expected to be limited for many
Group 2 boilers, the critical NOx phenomenon allows injection of the reagent at a
Cyclone-Fired Boilers • B4-14
-------
Appendix B
temperature beyond the effective temperature zone. In practice, this method may not
achieve the maximum NOx reduction represented by the critical NOx; however, as
experience shows, reasonable NOx reduction efficiencies are possible.
The baseline NOX levels for typical cyclone boilers range from 0.9 to 1 .8 lb/MMBtu.(8)
With these high levels, urea can be injected in the furnace to achieve substantial NOX
reductions, even if the required residence times are not available in the effective
temperature zone.
The SNCR systems for the study boilers were designed based on this critical NOX
phenomenon. The design and location of the reagent injectors follow the experiences
from the above operating installations, especially the one at B. L. England. For the
larger boiler, a higher number of injector levels is used. It is assumed that multiple
level injection can be used for larger boilers to achieve the same NOX reduction as for
the smaller boilers.
There is^ limited SNCR experience with large size boilers. Proper mixing of the
reagent with the flue gas would be a concern with these boilers. However, it is felt
that with proper flow modeling, injector designs and locations can be selected for a
viable SNCR application.
The NOX reduction efficiency for this study was selected to maintain a maximum
ammonia slip of 10 ppm. The source of urea was assumed to be the NOXOUT
reagent commercially supplied by Nalco Fuel Tech. The 10 ppm ammonia slip is
selected to minimize concerns regarding adverse impacts of ammonium salts on the
boiler backend equipment. The ammonium salts can form via reaction between the
unreacted ammonia and sulfur-trioxide present in the flue gas stream.
The results of the SNCR technology evaluations for this study are presented below.
4.4.1 Equipment and Material Modifications
Tables B4-11 and B4-12 list the major new equipment and materials required for
retrofitting the SNCR technology for the 150 MW and 400 MW boilers, respectively.
These tables also include descriptions of the major modifications required to the
existing equipment. The equipment additions and modifications include:
• Urea solution storage tank
• Urea circulation module consisting of pumps, electric heaters, and piping to
maintain urea in storage at a proper temperature
• Metering module consisting of urea metering pumps, dilution water pumps,
and piping to provide metered flow of urea and dilution water to the injectors
Cyclone-Fired Boilers • B4-15
-------
Appendix B
• Urea price ($0.7 - 0.8/gallon)
• NSR(O.S-I.O)
Figures B4-36 through B4-45 show the results of these variations on the technology's
capital and levelized costs. The results are summarized below:
a. The capital cost variation has a minor impact on the capital and levelized
costs.
b. The capacity factor variation has a relatively low impact on the levelized
costs (both mils/kWh and $/ton of NOXK There is no impact on the capital
cost.
c. NOX reduction variation has no impact on the capital cost and levelized cost
in mils/kWh. It has the greatest impact on the $/ton of NOX costs.
d. The urea price variation has a relatively low impact on the levelized costs.
The capital costs are not affected.
e. The NSR variation has a significant impact on the levelized costs and no
impact on the capital cost.
4.5 SELECTIVE CATALYTIC REDUCTION (SCR) APPLICATIONS
The SCR technology has been applied extensively to fossil boilers. The majority of
this experience to date has been on low to medium sulfur fuels. Also, coal-fired
installations have generally consisted of pulverized coal (PC) boilers. One installation
on a cyclone-fired boiler is at the Merrimack Power Station of Public Service of New
Hampshire. The SCR system installed at the 320 MW Unit 2 of this station re-
scheduled to complete startup in June 1995.119'
The design and performance estimates for the SCR system in this study were based
on the published data119"211 and information received from the suppliers.(X **•M1 In
addition, Bechtel inhouse information from various past studies and three coal-fired
SCR installations (two of them operational) was used.
The SCR system design is greatly influenced by site-specific factors, which vary
between power plants. In general, the most important factors are common to all
types of coal-fired boilers. They include the fuel characteristics, flue gas
temperatures, and space available for the SCR reactors. The cyclone boilers do have
certain specific features that require consideration in the SCR design. However, the
majority of the technology experience on PC boilers is applicable to the cyclone
boilers.
Cyclone-Fired Boi/ers • B4-19
-------
Appendix B
Air for Combustion, klb/h
Flue Gas Leaving Boiler, klb/h
Total Solid Waste, klb/h
1,152 1,152
1,317 1,317
12.9 13.1
COAL ANALYSIS:
Proximate Analysis, %
Moisture
Ash
Fixed Carbon
Volatile Matter
Ultimate Analysis, %
Carbon
Hydrogen
Nitrogen
Sulfur
Ash
Oxygen
Moisture
HHV, Btu/lb
Grindability
Midwestern
Bituminous
BOILER CHARACTERISTICS AND NOX EMISSIONS:
Number of Cyclones
Number of Reburn Burners/OFA Ports
Number of Pulverizers
Burners per Pulverizer
Average Excess Air, %
NOX Emission, Ib/MMBtu (100% Load)
increased Auxiliary Power, kW
Original
4
NA
NA
NA
16
1.40
Base
Modified
4
4
1
4
16
0.70
476
22S8S.002\SaMY\C>B2Boil
B4-35
-------
Appendix B
Air for Combustion, klb/h
Flue Gas Leaving Boiler, klb/h
Total Solid Waste, klb/h
3,227 3,227
3,692 3,692
35.9 36.7
COAL ANALYSIS:
Proximate Analysis, %
Moisture
Ash
Fixed Carbon
Volatile Matter
Ultimate Analysis, %
Carbon
Hydrogen
Nitrogen
Sulfur
Ash
Oxygen
Moisture
HHV, Btu/lb
Grindability
BOILER CHARACTERISTICS AND NOX EMISSIONS:
Number of Cyclones
Number of Reburn Burners/OFA Ports
Number of Pulverizers
Burners per Pulverizer
Average Excess Air, %
NOX Emission, Ib/MMBtu (100% Load)
Increased Auxiliary Power, kW
Midwestern
Bituminous
Original
12
NA
NA
NA
16
1.3
Base
Modified
12
12
2
6
16
0.65
1,462
2288S.002\SMvttfp2Boil
B4-38
-------
Appendix B
Coal Consumption, tons/h
Air for Combustion, klb/h
Flue Gas Leaving Boiler, klb/h
Total Solid Waste, klb/h
54
1,152
1,317
13
44
1,148
1,306
10
COAL ANALYSIS:
Proximate Analysis, %
Moisture
Ash
Fixed Carbon
Volatile Matter
Ultimate Analysis, %
Carbon
Hydrogen
Nitrogen
Sulfur
Ash
Oxygen
Moisture
HHV, Btu/Ib
Grindability
Natural Gas Analysis, % by volume
CH4
C2H6
N2
HHV, Btu/scf
Midwestern
Bituminous
5.8
11.7
44.5
38.0
66.4 /
4.5
1.3
2.7
11.7
7.6
5.8
11,900
50
90
5
5
1,002
•J
2288S.002\SnidT\Gm2Boil
B4-48
-------
Appendix
Coal Consumption, tons/h
Air for Combustion, klb/h
Flue Gas Leaving Boiler, klb/h
Total Solid Waste, klb/h
152
3,227
3,692
36
123
3,215
3,660
29
COAL ANALYSIS:
Proximate Analysis, %
Moisture
Ash
Fixed Carbon
Volatile Matter
Ultimate Analysis, %
Carbon
Hydrogen
Nitrogen
Sulfur
Ash
Oxygen
Moisture
HHV, Btu/lb
Grindability
Natural Gas Analysis, % by volume
CH4
C2H6
N2
HHV, Btu/scf
Midwestern
Bituminous
5.8
11.7
44.5
38.0
66.4
4.5
1.3
2.7
11.7
7.6
5.8
11;900
50
90
5
5
1,002
VI
I /' *•
r
B4-51
-------
Appendix B
Air for Combustion, klb/h
Flue Gas Leaving Boiler, klb/h
Total Solid Waste, klb/h
1,152 1,160
1,317 1,351
13 13.1
COAL ANALYSIS:
Proximate Analysis, %
Moisture
Ash
Fixed Carbon
Volatile Matter
Ultimate Analysis, %
Carbon
Hydrogen
Nitrogen
Sulfur
Ash
Oxygen
Moisture
HHV, Btu/lb
Grindability
BOILER CHARACTERISTICS AND NOX EMISSIONS:
Number of Cyclones
Average Excess Air, %
NOX Emission, Ib/MMBtu (100% Load)
Increased Power Consumption, kW
Midwestern
Bituminous
11,900,-Btuflb--G__
2288S.OOZISlwfv\Ga»2ae*
B4-61
-------
Appendix
Air for Combustion, klb/h
Flue Gas Leaving Boiler, klb/h
Total Solid Waste, klb/h
3,227 3,249
3,692 3,779
36 36.2
COAL ANALYSIS:
Proximate Analysis, %
Moisture
Ash
Fixed Carbon
Volatile Matter
Ultimate Analysis, %
Carbon
Hydrogen
Nitrogen
Sulfur
Ash
Oxygen
Moisture
HHV, Btu/lb
Grindability
BOILER CHARACTERISTICS AND NOX EMISSIONS:
Number of Cyclones
Average Excess Air, %
NOX Emission, ib/MMBtu (100% Load)
Increased Power Consumption, kW
Midwestern
Bituminous
Original
12
16
1.3
Base
Modified
12
16
0.85
205
22S8S.OO2\Saidv\GfD2Boil
B4-64
-------
Appendix B
Air for Combustion, klb/h
Flue Gas Leaving Boiler, klb/h
Total.Solid Waste, klb/h
800
860
6.77
804
864
6.8
COAL ANALYSIS:
Proximate Analysis, %
Moisture
Ash
Fixed Carbon
Volatile Matter
Ultimate Analysis, %
Carbon
Hydrogen
Nitrogen
Sulfur
Ash
Oxygen
Moisture
HHV
BOILER CHARACTERISTICS AND NOX
EMISSIONS:
Average Excess Air, %
NOX Emissions, Ib/MMBtu (100% Load)
Increased Power Consumption, kW
Urea Consumption, gal/h
Water Consumption, gal/h
Eastern Bituminous
5.0
10.0
53.5
31.5
71.8
4.7
1.2
2.1
10.0
5.2
5.0
13,100
22
0.95
Base
Base
Base
22
0.62
25
100
1,320
BOILER EFFICIENCY BY HEAT LOSS:
Dry Gas Loss
4.07
4.07
2288S.OOTOBJdv\Gfp2Boil
B5-12
-------
Appendix B
Air for Combustion, klb/h
Flue Gas Leaving Boiler, klb/h
Total Solid Waste, klb/h
2,082 2,091
2,240 2,250
17.95 18.04
X
COAL ANALYSIS:
Proximate Analysis, %
Moisture
Ash
Fixed Carbon
Volatile Matter
Ultimate Analysis, %
Carbon
Hydrogen
Nitrogen
Sulfur
Ash
Oxygen
Moisture
HHV, &Z+JsJf
)
BOILER CHARACTERISTICS AND NOX EMISSIONS:
Average Excess Air, %
NOX Emission, Ib/MMBtu (100% Load)
Increased Power Consumption, kW
Urea Consumption, gal/h
Water Consumption, gal/h
Eastern Bituminous
5.0
10.0
53.5
31.5 i
71.8
4.7
1.2
2.1
10.0
5.2
5.0
13,100
,
>
if
20
0.92
Base
Base
Base
20
0.60
109
260
3,370
BOILER EFFICIENCY BY HEAT LOSS:
Dry Gas Loss
4.00
4.00
B5-15
-------
Appendix B
Air for Combustion, klb/h
Flue Gas Leaving Boiler, klb/h
Total Solid Waste, klb/h
901
971
13.6
907
977
13.7
COAL ANALYSIS:
Proximate Analysis, %
Moisture
Ash
Fixed Carbon
Volatile Matter
Ultimate Analysis, %
Carbon
Hydrogen
Nitrogen
Sulfur
Ash
Oxygen
Moisture
HHV, Btu/Ib
Eastern Bituminous
BOILER CHARACTERISTICS AND NOX EMISSIONS:
Average Excess Air, %
NOX Emissions, Ib/MMBtu (100% Load)
Increased Power Consumption, kW
Urea Consumption, gal/h
Water Consumption, gal/h
Oriainal
15
1.20
Base
Base
Base
Modified
15
0.78
51
150
1,930
2288S.002\Sutfy\Gfp2Boil
B6-12
-------
Appendix B
Air for Combustion, klb/h
Flue Gas Leaving Boiler, klb/h
Total Solid Waste, klb/h
1,649 1,659
1,777 1,788
24.9 25.1
COAL ANALYSIS:
Proximate Analysis, %
Moisture
Ash
Fixed Carbon
Volatile Matter
Ultimate Analysis, %
Carbon
Hydrogen
Nitrogen
Sulfur
Ash
Oxygen
Moisture
HHV, Btu/lb
Eastern Bituminous
4.5
15.8
50.5
29.2
69.3
4.3
1.2
0.6
15.8
4.3
4.5
12,100
y
BOILER CHARACTERISTICS AND NOX EMISSIONS:
Average Excess Air, %
NOX Emissions, Ib/MMBtu (100% Load)
Increased Power Consumption, kW
Urea Consumption, gal/h
Water Consumption, gal/h
Oriainal
15
1.20
Base
Base
Base
Modified
15
0.78
95
275
3,530
2288S.OO2\SludvWir«2Boil
B6-15
-------
INSTITUTE OF
CLEAN
1707 L Street, NW
Suite 570
Washington, DC 20036-4201
202.457.0911
Fax: 202.331.1388
Internet: icac@tmn.com
COMPANIES JEFJREYC. SMITH
Executive Director
MlCHAJELj. WAX, PH.D.
Deputy Director
May 28,1996
Mr. Ravi K. Srivastava
U.S. Environmental Protection Agency (6204J)
401M Street, SW
Washington, DC 20460
Dear Mr. Srivastava:
The Institute of Clean Air Companies, Inc. (ICAC) is pleased to submit the
following comments on the draft report, "Cost Estimates for Selected Applications of NO,
Control Technologies on Stationary Combustion Boilers," and on appendix to the report,
"Investigation of Performance and Cost of NO, Controls as Applied to Group 2 Boilers." As
you know, ICAC is the national association of companies which supply stationary source
air pollution monitoring and control systems, equipment, and services. Our members
include leading suppliers of selective catalytic and non-catalytic reduction (SCR and
SNCR), and also of low-NO, burners, reburn systems, and NO, and ammonia monitoring
systems.
The report should note that cost effectiveness values (expressed in $/ton
of NO, removed) will decrease as capacity factors increase above 65 percent.
Given the likelihood that some sort of trading scheme will accompany tightened NO,
limits, boiler owners can be expected to install controls on the highest capacity factor
boilers: doing so will spread capital and fixed operating costs over the greatest number of
tons of averted emissions, thus minimizing total costs per ton.
The report should include calculated SNCR costs for all boiler/fuel
combinations. While SNCR alone may not be sufficient to reduce NO, emissions to 0.15
Ib/MMBtu in all cases, SNCR may be part of combinations of control technologies which
do.
The report overestimates initial catalyst charges and catalyst
replacement rates for SCR, and thus overestimates SCR costs. Actual original
installed catalyst volumes (cubic meters of catalyst per MW of plant capacity) are 20-75%
lower than the volumes used in the report Improvements in catalyst technology and
experience over time have allowed installation of smaller catalyst volumes.
The report also conservatively assumes total replacement of the catalyst bed every
three years for coal-fired boilers. This assumption inflates actual catalyst replacement
costs by a factor of 1.3 to 3, depending on boiler type, and therefore introduces
Formerly Industrial Gas Cleaning Institute. Inc.
Recycled
-------
Mr. Ravi K. Srivastava
May 28,1996
Page 2
unacceptable errors into the cost calculations. Industry experience universally supports a
staged addition-replacement strategy for extending catalyst life. No SCR systems on coal-
fired boilers will require total catalyst change-out at the end of the guarantee period.
Table 1 gives initial catalyst volumes and catalyst replacement rates for existing
high-dust SCR systems on coal-fired boilers, as well as predicted rates given in several
commercial bids made by one catalyst supplier. Table 2 contains recommendations by that
same catalyst supplier for initial catalyst volumes and average replacement rates.
Specific comments:
p. 2-2, fourth paragraph: SCR systems will not necessarily lead to excessive SO2 to SO3
conversion rates; SCR catalysts are available which oxidize less than 1% of the SO2 to SO3.
p. 3-2: An average catalyst replacement rate of 4,680 fts/yr is high, as noted above.
p. 3-2: An average catalyst replacement rate of 5,417 fta/yr is high, as noted above.
p. 4-1: A catalyst operating life of 5 years is low for natural gas service; a life of 8-10 years
would be more representative of actual operating experience
Appendix, p. B4-22: other SCR system components not listed are reactor structural
support steel and NO, analyzers and miscellaneous instrumentation
Appendix, p. B4-67, B4-69: 200 feet of ductwork between the air blowers and ammonia
injection grid seems high
Appendix, p. B4-68: we question whether two new forced draft fans appropriate to a
pressurized unit would be necessary
Appendix, p. B4-70: we question whether two new induced draft fans would be necessary
Appendix, p. B4-72: we question the inclusion of the cost of "draft fans as" as a necessary
part of the cost of an SCR system
Appendix, p. C-49: The SCR design NO, removal efficiency is 47% at Stanton 2, and 53%
at Birchwood. The actual removal efficiency is 63% at Logan (formerly Keystone), 59% at
Indiantown, and 65% at Merrimack. (Note that the 1999 removal efficiency target at
Merrimack on the installation of additional catalyst is 90%.)
Appendix, p. C-55, second paragraph: Designing the SCR system for uniform mixing of
ammonia in the flue gas upstream of the reactor helps to achieve low ammonia slip.
-------
Mi. Ravi K. Srivastava
May 28,1996
Page3
Please let us know if you have any questions regarding our comments or wish
additional information.
Sincerely,
Michael J. Wax
cc: Perrin Quarles Associates, Inc.
501 Faulconer Drive, Suite 2-D
Charlottesvflle,VA 22903
-------
Mr. Ravi K. Srivastava
May 28,1996
Page 4
Table 1. High-Dust SCR Catalyst Requirements in Units in Service or Designed for Coal-Fired Boilers
Boiler "type
dry bottom
dry bottom
dry bottom
dry bottom
dry bottom
dry bottom
dry bottom
dry bottom
cyclone
cyclone
Size
(MW)
75
410
760
700
S50
500
234
162
480
330
Commercial
Operation
1983
1988
1989
1989
1989
1991
bid
bid
bid
bid
Uncontrolled NO,
Emissions (Ib/MMBtu)
0.23
0.51
0.61
0.36
0.75
0.29
0.39
0.5
1.35
2.4
Removal
Efficiency (%)
65
69
75
73
79
67
80
70
86
68-91
Initial Catalyst
Volume (m'/MW)
1.00
1.86
1.05
0.96
1.21
0.83
0.70
0.73
0.84
0.92
Annualized Catalyst
Replacement
13%
6%
5%
11%
10%
<8%
12%
10%
22%
29%
-------
Mr. Ravi K. Srivastava
May 28,1996
Paged
Table 2. High-Dust SCR Catalyst Requirements for Coal-Fired Service, By Boiler Type
Boiler Type
tangential
wall-fired
cell burner
cyclone
wet bottom
vertically fired
Size,
MW
848
881
600
400
259
220
Uncontrolled NO,
Emissions, Ib/MMBtu
0.45
0.50
1.00
1.17
1.13
1.08
Removal
Efficiency, %
67
70
85
87
87
86
Initial Catalyst
Volume, m8 (m8/MW)
230 (0.66)
260 (0.68)
600 (1.00)
445 (1.11)
285 (1.10)
210 (0.95)
Annualized Catalyst
Replacement, m8 (%)
23 (10)
26 (10)
120 (20)
111 (26)
71 (25)
31 (15)
-------
(i
\Faxrfeeiptwillnoii* Baker & BottS. I+LJf. Othtr Offices:
i >• - - -» JL.. ..I im. .mf.j •*•—*
i ujMUf JMH py prune tanm
requested. TheWamer Austin
1299 Pennsylvania Ave.,NW Dallas
Washington, DC 20004-2400
(202)639-7700
Fax (202) 639-7890
To: Perrin Quarles Associates. Inc.
CharioOesvifle, VA
From: William Bumpers, Esq. Attorney/Employee No: 2953
Return transmitted fax to:
Fax No: 804-296-2860 Voice/Dlfficntty No: 804-979-3700
Date: May 20,1996 Total* of Pages: _4 + Corer
•MESSAGE*
NOTICE OF CONFIDENTIALITY
The infonnatioo cntttnd in and transmitted witfatbta tacsuOt is
1. SORJECTTOTCffi ATTORNEY-CLIENT PRIVILEGE;
2. Ai'lOKNEY WORK PROPDCT; OR
3.
. You«mb*»»byBOttfi*d th«t «ny dhrnirniMrinn. diHAotioo. copyag, or
aw of or IB&MCO apao the infoniwtioa cootuaed in end tnnsmittai with tfaa fecsnnilebyortotnyoneodjertbaBlfaereeainetideitgnated
ribcM6bythoBtodergtg»mirta>u^miu &ccinjile incnar, plow* notify B*kBr & BottB, ULP.
by tofaphonomt (202) 639-7996 innnedtttEly. Aor fi«in^em»eoo^ fawwnitted to yoa«faoakibe mnnedi^
Mail or, if Jtmbonaoion is gnated by dw >Mdgr. dettuyod.
If you do not receive all pages, please call: (202) 639-7996 for assistance.
BffiingNo: 062927.0101
Critical Deadline, Send By: ASAP AIWFM
18'd 01 0681 6E9 202 S11OS GNU aa»OS dd IC:iI 36.02 AbW
-------
BAKER & BOTTS
L.L.F?
MIST] N A MCOISTBieO UMITCO UAIIUTV MKTNCIISHIP
OALI>S THE WARNER BUILDING
TCUtrHONC.aoZ, C3.-7700
NEW YORK WASHINGTON, D.C. 2OOO4-S4OO rACSlMiLC:zoa
May 20, 1996
Perrin Qnarles Associate, Inc.
501 Fanlconer Dove, Suite 2-D
Chariottcsvillc, VA 22903
Re: Comments of the Oass of '85 Regulatory Response Group on
the Draft Report Entitled "Cost Estimates far Selected
Applications of NOx Control Technologies on Stationary
. Bouers
Dear Sir or Madam:
The Class of '85 Regulatory Response Group appreciates the opportunity to
review and comment on the draft report entitled "Cost Estimates for Selected
Applications ofNOx Control Technologies on Stationary Combustion. Boilers" that* was
prepared by Becntel Power Corporation *m^ The faArmc Group for die U.S.
Environmental Protection Agency. The Class of '85 does not have extensive comments
on me draft report, but would Kke to address one issue: me evaluated cost of retrofitting
selective catatytic reduction (SCR) technology on coal-fired boflcrs. The Class of '85
believes mat the analysis is biased due to me use of an improper evaluation methodology,
optimistic cosi ^fftifir**t^^ giiii Ttift i^|ii^|ffy fQ tnclude lilcerv canacxty derates*
The draft report uses a power factor seating methodology to derive estimated costs
for a system of one size based on known costs of a second system of a different size.
Power factor scaling is a generally accepted methodology for powerplant cost estimating,
since it addresses economies of scale associated with permuting, land, and other
infrastructure. However, me use of power factor scaling for SCR estimating is not
appropriate because SCR is not a powerplant, but a system component that inherently
lacks the assumed economies of scale. For example, unit costs (m$/Ib or S/MW) of a
SCR catalyst for a 200 MW powerplant should be me same as for a 950 MW powerplant
The net effect of using power factor seating in the draft report was to reduce units costs
of SCR from $68/kW for a 200 MW powerplant to $3 9/kW for a 950 MW plant The
DCDfclCQM&l
20-d 01 068i sea zvz SJ.IOH any gaxus ad tc-.ii 95,02
-------
BAKER & BOTTS
UL.P.
May 20,1996
Page 2
Class of '85 does not believe that the SCR cost estimate of S39/kW for a 950 MW plant
is credible bin\ tngf^g jnifrsl8****^"y nmfarft^tifnotte^ tne tme costs of SCR retrofits.
Second, the cost history of domestic coal-fired SCR retrofits does not support the
draft report's engineering cost estimate for typical coal-fired SCR retrofits when power
factor scaling is ^srnfssftd from Ag evaluation, rvtfimi^rr.ial
-------
BAKER & BOTTS
L.L.R
May 20,1996
PageS
The Class of '85 Regulatory Response Group appreciates your consideration of its
comments. For your information, a list of the members of me dM9 of '85 is enclosed. If
yon have any questions about these comments, please do not hesitate to contact us.
Respectfully submitted,
William M. Bumpers
Debra J. Jczomt
Counsel to the Class of'85
Regulatory Response Group
EncL
01 068i 6C9 202
SilOS QNU d3»tia
96.02 AtlW
-------
** S00'39bd 1U101 **
CLASS OF <85 REGUIATORY RESPONSE GROUP
Arizona Pnbfic Service Company
AritaasaiElefArica»op«ratto Corporation
Arkansas Power & Light
CeaO^Loofclama Electric Company
Csotnl A Sootfa Wort Serried
OtyofTalbuauM
rnmolMatrd Mhrni Cbmpaay of New Yorit
EXergy Servlea»lBe.
CTotiifa M«mgpalIVini4 Aypcy
norida Power
GmYStatet DiStic* Compaay
JackMovflte Efectric Authority
Lkltdawl Department of Electric and Wi
MiMiwipiH Power & light
New Orieant PnbBc Stnrfee Company
Mla^jma. Mfliiawit Inunm Cuipun^iini
Nortben States Power Company
OriandotJtOitfefl
PBcflk Gai A Electric Company
Wisconsin Poww A lisfct Company
90 'd 01 368i BC9 202 S110S QNU a3Mfc)9 ad SE:iI 96.02 AtlW
-------
BLACK &VEATCH
8400 Word Parkway, P.O. Box No. 8405, Kansas City, Missouri 64114, (913)339-2000
May 23, 1996
Ms. Peggy Quarles
Perrin Quarles Associates, Inc.
501 Faulconer Drive
Suite 2-D
Charlottesville, Virginia
Dear Ms. Quarles:
Thank you for the opportunity to review the draft report entitled "Cost
Estimates for Selected Applications of NOX Control Technologies on
Stationary Combustion Boilers." The following are our comments regarding
the draft report.
This report is comprehensive, and will serve a valuable function to
utilities attempting to select an appropriate NOX reduction technology.
Due to the timely importance of the information and the influence of the
EPA documents regarding the selection of appropriate compliance
technologies, it is critical that the information within the report be
accurate and reflect current knowledge and experience. We would like to
address five critical issues that are not correctly represented in the
draft of this report:
1: The report should assume the use of a catalyst management plan for
SCR systems. The use of management plans reduce annual catalyst
replacement costs by at least 65 percent. Not assuming the use of a
catalyst management plan results in the inaccurate, nonrepresentative
characterization of SCR costs.
2: Published data reporting results of SNCR installations does not
support the report's assumption that SNCR has a NOX reduction
capability of 50 percent. SNCR has demonstrated capability for
reliably removing 20 to 40 percent NOX reduction on small to medium PC
boilers while maintaining ammonia slip in acceptable ranges.
3: The report does not discuss or reflect potential economic impacts
caused by the ammonia slip from SNCR systems such as forced outages
and boiler load limitations. Experience has indicated that numerous
SNCR installations need relatively frequent offline cleanings of the
air heater when using SNCR with sulfur bearing fuels. Forced outages
would be very expensive to accommodate especially during the summer
peak season.
4: The capacity factor used (65 percent) in the economic analysis of the
report is too low. Likely target baseload units operating during the
-------
Ms. Peggy Quarles Page 2
May 23, 1996
5 month "NOX season" are likely to have very high capacity factors (85
to 95 percent) during this summer peak period. Assuming a
misrepresentative value of 65 percent has a punitive effect on
capital intensive technologies such as SCR.
5: Currently, this draft version appears very heavily biased towards
SNCR and against SCR when discussing post-combustion NOX control
systems. We believe that it is misleading to imply that the
installation of an SNCR system will reliably lead to 50 percent NOX
reduction with ammonia slip less than 10 ppm with no potential for
significant detrimental impact on plant operation. Experience has
demonstrated that this performance level to be the exception, not the
rule.
We realize that it may be difficult to make significant changes to the
report, but to not reflect these comments regarding SCR and SNCR
performance and cost will mis-inform the users of this report with respect
to the performance, cost, and plant impacts of post-combustion NOX control
technologies.
Further discussion of these comments is included as an attachment to this
letter. If you have any questions regarding these comments, please call
either me (913-339-7785) or John Cochran (913-339-2190). Our fax number is
913-339-2934. We look forward to discussing these comments with you.
Sincerely, Sincerely,
John R. Cochran Michael G. Gregory
Manager-Air Quality Control Section NOX Control Unit Leader
Attachment
cc: Ravi Srivastava - US EPA
Volker Rummenhohl - STEAG
Jeff Smith - ICAC
-------
BLACK & VEATCH COMMENTS REGARDING DRAFT REPORT OF "COST ESTIMATES FOR
SELECTED APPLICATIONS OF NOX CONTROL TECHNOLOGIES ON STATIONARY COMBUSTION
BOILERS"
SCR CATALYST MANAGEMENT:
On page 3-1 of the draft report, the assumption is stated that "a catalyst
management strategy is not used for this evaluation." This assumption is
used for all boiler types. This is not a valid assumption considering the
current design philosophy for SCR systems. The catalyst life guarantee is
a measure of how long the catalyst will meet both NOX reduction
requirements and ammonia slip limits. However, when the catalyst is unable
to meet these requirements, there is still 70 to 80 percent of original
catalyst activity remaining. A catalyst management plan will increase the
effective utilization of this remaining activity. To not assume a catalyst
management plan results In inaccurate and non-representative economics to
be presented for SCR.
The design basis used for the tangential boiler example in this report has
an average catalyst replacement value of 4,680 ft3/yr. Therefore, it
appears that the total catalyst volume is approximately 14,040 ft3 (3 yrs x
4,680 ft3/yr). Assuming a two-layer reactor design, 3 year catalyst life,
20 year remaining plant life, and no management plan, the total catalyst
replacement would be a total of 12 layers (2 layers x 6 replacements) with
a total cost of $32,760,000. Using the stated catalyst replacement cost of
$350/ft3 (Table 2-1), the annual cost catalyst replacement without a
management plan is $l,638,000/yr ($350/ft3 x 4,680 ft3/yr).
In a modern SCR system (new or retrofit), at least one extra layer will be
included in the reactor design for future addition of catalyst. This layer
has a significant impact on the catalyst replacement cost over the life of
the plant. An SCR system with the same catalyst volume (assume two layers)
with a spare layer installed in the original design will have a much lower
annual cost than an equivalent system with no spare layer. When the
remaining active material within the catalyst is allowed to be utilized by
adding catalyst rather than complete replacement, the total catalyst volume
required over a 20 year life becomes 5 layers at a total cost of
$12,285,000. Use of a catalyst management plan leads to an average
catalyst replacement cost of $614,250/yr, or only 37 percent of the cost
without the management plan111. This will have a significant impact on the
operating cost and cost effectiveness of SCR measured in $/ton of NOX
removed. Accordingly, we strongly believe that the report should assume
the use of a catalyst management plan.
Even in the unlikely event of the inability to include a spare layer in the
design, an effective catalyst management plan can be incorporated replacing
individual layers. This also leads to substantial savings when compared to
complete replacement at the end of catalyst life.
May 23, 1996
-------
SNCR PERFORMANCE ON COAL FIRED BOILERS:
On page 3-2, the report states the assumption that "the SNCR system is
designed to provide a 50 percent NOX reduction from a baseline NOX rate of
0.45 Ib/MBtu." Except for smaller CFB boilers, we have not been able to
verify this performance capability on large coal fired boilers. In
Appendix A of this draft report (page 3-14 of "Investigation of Performance
and Cost of NOX Controls as Applied to Group 2 Boilers"), the SNCR
performance test results are summarized as follows: "In light of the ...
data, a 30% to 40% NOX reduction range can be recommended for sensitivity
analyses." These SNCR performance results are based on the high NOX inlet
loading of Group 2 boilers. It is very unlikely that this performance can
be exceeded with the relatively low inlet NOX loading of a Group 1 boiler.
In the course of site specific NOX evaluation studies we have conducted for
numerous utilities (in excess of 10,000 MW), we have received quotes from
the leading supplier of urea SNCR systems-Nalco Fuel Tech. These quotes
are based on the best removal rates that they felt could be provided while
maintaining a "reasonable" ammonia slip. The best of these performance
quotes have been 50% reduction but with an unacceptably high ammonia slip
value of 20 ppm. On these specific units, Nalco Fuel Tech indicated a
maximum capability of 30% to 37% reduction when the ammonia slip is limited
to 10 ppm or less. Even at an ammonia slip of 10 ppm, air heater fouling
can be experienced when using SNCR with sulfur bearing fuels.
This is further evidenced by papers describing urea-based SNCR performance
presented at the recent Institute of Clean Air Companies (ICAC) forum in
Baltimore. NOX reduction capabilities within the 30% to 40% range are
described at two PC boilers. One of the papers'21 described a 112 MW boiler
(with baseline NOX emissions of 0.49 Ib/MBtu) reported test results with
NOX reductions up to 50%, but normally operates at 40% NOX reduction to
meet regulatory requirements. The other paper stated that the capability
of the SNCR portion of a catalyst/SNCR hybrid system could achieve a
maximum NOX reduction of 37% while burning coal.131
There is no evidence to support the design basis of this report that 50%
reduction can be consistently met by SNCR while maintaining ammonia slip at
10 ppm or less. Although this performance may be met for short periods of
time on small boilers, the 50% NOX reduction basis cannot be presented to
the readers of this report as a design basis for all SNCR systems. Quite
simply, there is no experience with reliably obtaining these NOX reduction
levels on large coal fueled boilers. We believe that SNCR systems are very
effective for installations on CFB's, and can be effective for NOX
reductions of 20% to 40% on small-to-medium sized PC boilers. However, the
use of 50% as the design value of SNCR performance will skew the reader's
performance expectations and artificially deflate relative life-cycle costs
of the SNCR system compared to other NOX reduction alternatives.
FORCED OUTAGES FROM SNCR OPERATION:
In the past several years, there has been much experience gained on post-
combustion control methods. SNCR suppliers have been very active in test
May 23, 1996 2
-------
programs, research projects, and actual commercial installations. The
plant impact of SNCR systems has received mixed reviews. Although there
are some reported results of SNCR systems operating on small boilers for
one year without any significant problems on downstream equipment'21, there
is a very high risk of forced outages when using SNCR on coal fueled units.
One IPP with SNCR installed on eight 50 MW stoker boilers has had a great
deal of trouble keeping their units operating because of the impacts of
ammonia slip141. This installation describes air heater plugging, fabric
filter bag fouling, and waste water treatment problems due to the high
concentrations of ammonia. All this occurred while little or no ammonia
slip was being measured at the stack, indicating that knowing ammonia slip
at the boiler outlet is much more important than stack measurements. Also,
another recent use of SNCR on a 140 MW eastern U.S. bituminous fueled
pulverized coal unit has reportedly resulted in off-line air heater
washings every 2 weeks or less despite an average ammonia slip of
approximately 5 ppm. These forced outages would be very expensive to the
utility requiring the purchase of power during the peak season.
These results indicate that the potential for problems due to SNCR systems
cannot be ignored. A report issued by the EPA should address both the
benefits and potential problems of each technology. For this report, that
would involve adding a discussion of the potential downside of SNCR
operation and fairly evaluating the likely costs for forced outages related
to use of this technology. No similar likelihood for forced outages can be
identified or justified for SCR systems.
LOW CAPACITY FACTORS IN ECONOMIC CAPARISON:
The draft report assumes a 65% capacity factor as the basis of the economic
evaluation. We believe that this value does not accurately represent the
units that a utility would consider for cost effective post-combustion NOX
reduction. Although a seemingly minor item, the capacity factor is a
critical component of a utility's system-wide NOX compliance evaluation. A
large utility system will have units with a wide variety of capacity
factors. In an effort to maximize the cost effectiveness of the control
systems selected, these utilities will no doubt first concentrate on the
base loaded units, with decreasing consideration as the capacity factor
decreases.
This is especially true for the 5-month "NOX season". The report uses 27%
as the annual capacity factor for this time period (65% x 5/12). This time
period by definition is the peak season for power generation, during which,
65% would probably not be considered acceptable to the system owners. To
maximize generation, the utility would expect 85% to 95% capacity factors
for its large generating units during this "NOX season". The use of this
higher factor would lead to an annual average capacity factor of
approximately 37.5% (90% x 5/12). The use of a higher, more representative
capacity factor in your analysis will lead to a more accurate
representation of SCR cost effectiveness.
May 23, 1996
-------
BIAS TOWARD SNCR:
We believe that the report's discussion of post-combustion NOX control is
heavily biased toward SNCR and away from SCR. The bias toward SNCR is
evidenced by the complete lack of discussion of system variability and the
associated high degree of supervision required to achieve acceptable
performance. SNCR performance is critically affected by site specific
design constraints and normal boiler operation transients such as load
changes and heat transfer surface fouling/slagging. It is noteworthy that
papers describing acceptable NOX reduction capabilities indicate that
changes in unit operation lead to large changes in urea utilization and
difficulty in automatic control(2). To imply that the installation of an
SNCR system will reliably lead to 50% NOX reduction with ammonia slip less
than 10 ppm with no potential for significant detrimental impact on plant
operation will mislead many readers into thinking it is the perfect high-
efficiency control solution. In reality it also has the potential to
achieve only 20% reduction while forcing frequent outages for equipment
cleaning, lost fly ash sales, and limit the boiler's turn-down
capability151. Please make sure that the report presents an accurate
portrayal of SNCR capabilities and limitations which indicate that numerous
recent SNCR installations have demonstrated unacceptable as well as
acceptable performance. The user of this report must be informed that their
actual results would probably fall somewhere within this SNCR performance
spectrum.
The bias against SCR is demonstrated in paragraphs such as the third
complete paragraph on page 2-2 in which catalyst volume requirements are
described as "significantly large" and S03 conversion rates are described
as "excessive". In addressing the catalyst volume requirements, rather
than using "significantly large" (compared to what?), the report could
state that improvements in catalyst formulation and system design have
reduced the amount of catalyst volume to 65% to 70% of the volume required
by pre-1990 systems. Also, S03 oxidation is not "excessive". The
oxidation rate is a function of catalyst formulation, which is a design
variable. In systems burning sulfur-bearing fuel, the catalyst formulation
can keep oxidation to levels to 1% or less, which is not excessive by any
definition.
REFERENCES:
(1) Cochran, John R.; Gregory, Michael G.; Rummenhohl, Volker; "SNCR,
SCR, and Hybrid Systems Capabilities, Limitations, and Cost".
Presented at the EPRI/EPA Joint Symposium on Stationary Combustion NOX
Control, May 16-19, 1995.
(2) Tsai, Thomas et. al.; "Living with Urea Selective Non-Catalytic NOX
Reduction at Montaup Electric's 112 MWe PC Boiler". Presented at the
ICAC Forum '96, March 19-20, 1996; Baltimore, MA.
(more References on following page)
May 23, 1996 4
-------
REFERENCES (Cont.)
(3) Wallace, A.J.; Gibbons, F.X.; Roy, R.O.; O'Learv, J.H.; Knell, E.W.;
"Demonstration of SNCR, SCR, and Hybrid SNCR/SCR NOX Control
Technology on a Pulverized Coal. Wet-Bottom Utility Boiler",
Presented at the ICAC Forum '96, March 19-20, 1996; Baltimore, MA.
(4) Hall, David; Bonner, Thomas J.; "SNCR Experience With Coal Fired
Boilers and Fabric Filters". Presented at the ICAC Forum '94,
November 1-2, 1994, Washington, D.C.
(5) Gregory, Michael G.; Cochran, John R.; Rummenhohl, Volker; "The
Impact of SCR and SNCR Systems on Plant Equipment and Operations".
Presented at the ICAC Forum '94, November 1-2, 1994, Washington, D.C.
May 23, 1996
-------
Technology for a renewed environment
Vincent M.AIbanese
'.' 1:1} -''.'Sici.MV
'.^ S i.iies aw M.irkeiinr.
Mav 7,1996
Perrin Quarles Associates, Inc.
501 Falconer Drive, Suite 2-D
Charlottesville, VA 22903
Subject Comments on Draft Report
Thank you for the opportunity to present comments on the draft report entitled "Cost
Estimates for Selected Applications of NOx Control Technologies on Stationary
Combustion Boilers". The stated purposes of the subject draft report were to develop
costs for NOx control technologies to reduce NOx emissions from baseline to 0.15
Ib/MBtu (the putative emissions limit after Phase HE of the Ozone Transport Region's
MOU is implemented), and to develop costs for NOx control technologies providing
substantial NOx reductions beyond emissions limits mandated in Phase I for Group 1
boilers subject to acid rain NOx limits.
As a general comment, the treatment of capital costs may not appropriately reflect the
.cost to utility plants subject to NOx control regulations. If the subject report used the
same methodology as Bechtel used in the referenced Group 2 boiler report to the EPA,
the capital carrying charge utilized was a modest value of 0.115. Our recent experience
in installing post-combustion NOx controls indicates carrying charges of 0.145-.200. As
utilities prepare for deregulation and enforced competition in generation, the planning
horizon for capital outlay has shortened considerably. By 1999 and 2000 the real
financial world for utilities may be to carry capital for only 5 years. We encourage that
cost estimates for NOx control be recalculated utilizing more representative higher
carrying charges. Perhaps in recognition of the need to minimize capital outlay, the
authors did well to point out that combinations of technologies might well be cost
effectively applied, but that details for potential NOx reduction combinations were
beyond the scope of the draft report
? 0. Box 3031 a Naoerviile. Illinois 60566-7031 a Area 708-963-324> c rax 708-983-3240
- 1 - ...\vma\perrinqu.L01
-------
Technology lot a renames tmnronmem ™
The technical data presented in the report seem to be an accurate recapitulation.
However, two small comments do arise:
(1) Since one premise for all the data in the report is that LNB or combustion'
modifications have already been employed, the Gas Reburning data may need to
be revisited in Table 1-5. The Acurex report entitled "Phase n NOx Controls for
the NESCAUM and MARAMA Region" states cost effectiveness is significantly
diminished for this add-on control because the NOx reduction is only 20% when
LNB is already installed.
(2) In Section 3.1.1 regarding SCR, the statement "It is assumed that the existing
plant setting allows installation of the SCR reactors between the economizer and
the air heater without a need to relocate any major structure or equipment" is
such an egregious leap it is better to qualify the statement with the admission
that installation on a number of sites would be impossible or imprudently costly.
Of perhaps greater concern to NFT than any comments apropos to the subject report is
there are several factual errors and misleading premises in the appendaged report
"Investigation of Performance and Cost of NOx Controls as Applied to Group 2
Boilers" prepared for EPA by the Cadmus Group, Bechtel and SAL Apparently this
report served as a basis for EPA's proposed Group 2 boiler rule FR 1442, Jan. 19, 1996.
NFT had provided both background information and formal comments on the
proposed rule, but may have been disadvantaged by never having seen the Draft
Group 2 report NFT trusts this oversight will be rectified by incorporating the
following comments to that report
Section 3-1 Selection of Control Technologies for Evaluation
Report Statement "For each Group 2 boiler type, this selection was based on
availability of control technologies as established by at least one
full-scale demonstration or commercial application."
NFT Comment
It is inappropriate to include thinly demonstrated technologies like
coal reburn and combustion modifications on wet-bottom boilers,
while maintaining that hybrid SNCR/SCR is outside the scope of
the report The successful demonstration of the technology on a
Group 2 boiler at Public Service Electric and Gas along with the
potentially broad applicability demand inclusion of the technology
with greater focus than currently in the draft
NFT-17
- 2 -
..\vma\perrinqu.L01
-------
Technology for a renewed e
Another caveat regarding the basic premise of including full scale
demonstrations is that demonstration costs are often far higher
than application of the technology on a commercial basis. Overall
scope is far more confined in commercial installations, and reagent
prices will be more favorable.
Section 3.2.4.4 Factors Affecting Performance
Report Statement "...SNCR process is capable of load following through adjustment
ofNSR".
NFT Comment
The statement is quite incomplete and does not represent the
technology. It may be corrected to -
"SNCR process is capable of load following through automated
control of injection level use, reagent dilution through the various
injectors, and reagent flow rate at a given injection level(s), all of
which usually changes NSR as load changes. The automatic
control is setup on a feed-forward basis."
Sections 3.2.4 and 3.2.5 re Coal Sulfur Content
Report Statement
NFT Comment
In describing the potential for ammonium salt formation in the flue
gas, the authors state "...ammonia slip needs to be controlled in
SNCR..." whereas in Section 3.2.5.4 regarding SCR, the authors
state "...ammonia slip is controlled in SCR application to
minimize..."
The difference is subtle but the implication is SNCR doesn't control
slip and SCR does. The fact is ammonia slip is an operational
limitation of both technologies as it limits NOx reduction (for
SNCR) or causes more capital expense (for SCR). U.S. Generating
Company revealed pictures of NHs slip induced air heater
pluggage in its ICAC Forum presentation entitled "Multiple Coal
Plant SCR Experience - a. U.S. Genco. Perspective". NFT asks that
the subject of ammonium salt formation and subsequent air heater
deposition be treated in a more equitable tone.
NFT-17
-3-
...\vma\perrinqu.L01
-------
Technology tor a ren
Appendix B, Section 4.4 SNCR Application
Report Statement "The source of urea was assumed to be the NOxOUT reagent
commercially supplied by Nalco Fuel Tech."
NFT Comment
NFT does not supply urea or any other reagents used in air
pollution control systems. The urea at the referenced plant was
supplied by one of five commodity suppliers of urea that offer
appropriate quality material to be used in conjunction with the
NOxOUT process.
Section C2.4.5 Reagent Cost
Report Statement "A significant portion of the SNCR O.& M. costs is tied to the
consumption of the chemical reagent Therefore, the annualized
cost of an SNCR application is particularly sensitive to the market
price of the reagent used."
NFT Comment
This is not borne out by the sensitivity graphs on pg. B4-42. The
parameter to which SNCR cost is more sensitive is % NOx
reduction to be achieved by the SNCR system.
NFT will be happy to supply further comments if required. Please feel free to call me
at (708) 983-3254.
Sincerely,
Vincent M. Albanese
VMA/mjb
cc: S.C Argabright
R.A. Johnson
J.E. Hofmann
NFT-17
-4-
...\vroa\perrinqu. L01
-------
Northeast t07 Selden Streei- Berlin- CT °6037
Utilities System Northeast Utilities Service Company
P.O. Box 270
Hartford. CT 06141-0270
(860) 665-5000
Charles F. Carlin. Jr.
Principal Engineer
Environmental Affairs
(860) 665-5344
(860) 665-3777 FAX
May 24, 1996 ni_ntn
RECD MAY 2 8 1996
Mr. Ravi Srivastava
Environmental Engineer
Acid Rain Division
U.S. EPA
401 M Street, SW
Mail Code 6204J
Washington, DC 20460
Dear Mr. Srivastava,
Thank you for the opportunity to comment on the Bechtel report on NOx control costs
for utility boilers. I forwarded the report to our engineering folks, and they reviewed it
in detail. In general, the report is quite reasonable and complete. Some of the cost
estimates are lower than we have used; some are higher. These comparisons are
attached, along with some specific suggestions for improving the report.
We look forward to your final rule on NOx emission controls for Title IV sources. If
you would like to discuss our comments, please call Mr. Robert H. Thomas at (860) 665-
3793.
Sincerely,
- -/r^•!•
'
OS3422 REV. 8-95
-------
ATTACHMENT
, K Asset Management has reviewed the EPA draft report entitled "Cost
Estimates for Selected Applications of NOx Control Technologies on
Stationary Combustion Boilers". The final version of this report will be
used by EPA to evaluate the costs of retrofit NOx controls applicable to
fossil fuel-fired boilers in the U.S. The report makes the assumption that
the controls will be required to limit NOx emissions to 0.15 Ib/MMBtu and
includes essentially all basic boiler designs and oil/gas/coal fuels.
The technologies evaluated in the report were limited to Selective Catalytic
Reduction (SCR), Selective Non-Catalytic Reduction (SNCR) and Gas Reburning.
To evaluate the report, cost projections were compared to the costs that we
have experienced in the NO system installing similar equipment and to NU
cost estimates prepared for 1999 NOx compliance.
~~r>eral report comments are immediately below, followed by comments on the
-.ort details,typos, and specifics.
GENERAL COMMENTS
o The report recognizes that some "Hybrid systems" are more
cost-effective for certain applications. These could include SNCR with
a shallow-bed catalyst or additional combustion controls with SCR.
However, these options were not in the scope of analysis.
n SNCR alone was not considered for coal units because the limit of 0.15
is considered beyond the reduction capability of SNCR alone from a
baseline of 0.45-0.50.
o Only natural gas as a reburn fuel is considered applicable because coal
or oil reburn fuel does not have enough reduction capability.
The SCR retrofit capital costs assume no allowance for relocating any
existing structures or equipment. In general this is a bad assumption.
o The report does not mention including costs for wastewater treatment
facility modifications which might be required to handle SCR ammonia
plant wastes and washwater wastes.
o The coal unit sulfur content assumed in the report is only 0.8 wt%.
Bisulfate formation on air heaters due to the SCR/ammonia combination
could become a much greater factor in downtime costs and air heater
capital work if the higher sulfur coals were assumed. SO3 formation
from the SCR on higher sulfur coals can accelerate downstream corrosion
and produce opacity plume/acid fallout problems. These conditions
should be factored into the report cost estimates for coal.
o The levelized carrying charge factor assumed is only 60% of the value
NU would use.
o The anhydrous ammonia cost assumption is about 20% less than
experienced at Merrimack.
o No mention is made as to disposal cost of used SCR catalyst, and ash
-------
disposal costs are about 33% less than NU's experiences.-
The reported SCR capital costs for oil or oil/gas units are about 1/3
lower than the latest NO estimates.
The reported SCR capital costs for coal units are about 20-25% lower
than NU would estimate.
The reported SNCR capital costs for oil or oil/gas units are more than
50% higher than NU estimates.
Natural gas assumptions for oil unit reburn should reflect the higher
pricing more representative of non-interruptible gas contracts
(typically 20% of total unit heat input).
Large variations in NOx reduction equipment capital cost estimates on
many of NU's smaller units can be seen in the asymptotic scaling
factors which the report applies to units less than 200 MW. The same
effect can be seen for levelized annual costs.
The report summary table for oil and gas fuels shows that SNCR is 50%
or less of the cost of SCR on units of NU's size. This cost
comparison is generally true for low and high capacity assumptions and
on a basis of $/KW or $/Ton NOx.
DETAIL COMMENTS
Table 1-2 should have a note that these are "$1995" and that the
formulas for capital cost are "$/kw" and not "$".
Table 1-3 should have a note that the baseline NOx levels are as shown
in table 1-1.
Table 1-4 should have a note reflecting that the SNCR assumption for
coal units is that 0.15 Ib/MMBtu cannot be achieved.
Section 2.1, first para.—Seems to erroneously refer to table 1-2.
Sect. 2.1, sec. para.—0.45 and 0.50 are in reverse order,
respectively.
Sect. 2.1, third bullet—Adding natural gas reburn to a pressurized
unit can be a safety hazard. The report should mention limiting
application factors such as this.
Sect. 2.1, fourth bullet—The cost and NOx reduction assumptions for
for SNCR should explain whether or not in-furnace lances are included.
Table 1-5 should have a note that the reduction percentages assume some
baseline NOx levels.
Table 2-1 should show a more realistic carrying charge factor, higher
anhydrous ammonia costs, higher ash disposal costs, and SCR disposal
costs.
-------
SEP-26-19S6 15:18 FROM EPA fiCID Rfllh D!U.
TO
913018694078 P. 02
BOUSKi-S. BD.C1UM .
WARSAW'. POLAND
HOWKONO
NEW YORK. icv> YORK
ATLAM1A. SCOQGIA
O. VIRONIA . •
HIINTON & WILLIAMS
180OK STSBBT. N.W.
WASHINGTON. D.C. 2OOOO-UO6
TCLCPHOME (2O2J OSS-IEOO
FACSIMILE-1202177e-S3Ot
August 7, 1996
MCLEAN. VTCMA
NOOTOUC. VIRGN'A
CMABiOTTC. NORTH
HNOXVILLC TEKKESS^E
DIRECT DIAL; (202)778-2J4<
Dwight Alpern. ..
Bnvironmentai Protection Agency
501 3rd Street, N.W.
Washington/DC 20001
Re: Request for Meeting Between UARG and EPA Consultants
Dear .Dwight: • . • •'
Thank you for your letter dated June 26, 1996 regarding the
request by UARG''s consultancs to meeting with EPA's consultants
to discuss technical issues on NOX control technologies. These
meetings were and are needed to resolve questions raised by the
Agency's reports and analyses in the § 407 NOX rulemaking, the
Ozone Transport Assessment Group (OTAG) discussions, and other
regulatory forums. As you knew, UARG wanted to settle the
outstanding issues in these proceedings as expeditiously as
possible, and it is unfortunate that the Acid Rain Division did
not allow its consultants: to meet 'with UARG's consultants
regarding technical issues related to the § 407 ruiemaking in
May. We also understand that where technical issues relate to
matters pther than the § 407 NOX rulemaking, EPA. will agree to a
meeting on technical issues only if:
1. Our clients are not represented at the meeting, by
counsel; -
2. UARG's consultants are limited to raising questions
regarding the need for additional information
-------
SEP-26-1996 15:19 FROM EPfi ftCID RflIN D!U.
TO
913018694B78 P.23
,j,
HTTNTON & WILLIAMS
We find this, a very curiojiis approach to developing a
technically sound basis for NOX control initiatives. As you '!
know, CAA 5 .307(d)(3) requires disclosure in connection with j
proposed Agency actions of "the factual data" on which a; propose^
action is based, as well as "the method used" in obtaining and j
analyzing 'the data. ' Given that meaningful comment is not |
possible unless there is full disclosure of'the facts and. i
methodology on which the Agency .relies, UARG and others have
traditionally worked closely with EPA .technical staff to develop
and to understand the basis for proposed regulatory actions.
This has often involved meetings that include EPA and UARG
technical consultants, in order that the technical experts on
each side have the benefit of each others' expertise and
professional judgment. For these 'reasons, we are confused as to
why EPA would want to close these lines of communication. This
does not appear to be a step designed, to foster either good
science or sound policy. ..'••. '
The enclosed comments by Ed Cichanowicz are specifically
directed to Bechtel's draft "Cost Estimates for Selected
Applications of NOX Control Technologies on Stationary Combustion
Boilers," which apparently is intended for use in OTAG.. .Many of
the data, and methodology issues raised in the Cichano'wicz
comments directly relate to the proposed § 407 NQX rule, which
uses a companion Bechtel report for its basis. These questions
go both to the need for complete disclosure of data and
methodology, and to the technical merit of specific analytical
approaches and assumptions. Among these questions, which are-
detailed in the Cichanowicz report, are the following:
• how is the remaining plant life determined;
what: are the specific boilers in Bechtel's "in-house
data base" whose design details are assumed to be
representative of typical boilers in each category.; .
what do the layout drawings look like of the "similar
boiler installations'1 that.are assumed by Beehtel to
represent this nation's entire boiler population;
what space velocities are assumed for each fuel in SCR
applications;
what limits .are specified for the conversion of S02 to
. SQj-in SCR applications.;
what assunptions are made for boiler economizer by-
passes in the design of SCR applications;
what proportion of the boiler population has unusual
'site features .that would force significant equipment
; relocation in SCR applications; • .
-------
SEP-26-1996 15=20 FROM EPfl ftCID RftIN DIU.
HUNTON
TO
913018694078 P.04
what -SNCR design concepts are used to account:'for'deep
cycling, .of. oil and gas-fired units,-
what is the.distribution of residence time in .the
boiler population and how does this affect the.
feasibility and cost of reburn applications; and • •
how are the results in Table 1-2 employed to create
Table 1-3? .
Resolution of these issues is critical to sound cos;:
estimates for QTAG and a sound .§407 rule. The fact that there
remain outstanding technical questions in the § 407 rulemaking
illustrates the lack of meaningful opportunity for the regulated
industry to continent on EPA1 s proposed rule.. We therefore ask
that EPA reconsider the restrictions that it has placed on the
exchange of technic.al information between the Acid Rain Division
consultants and UARG consultants.
Sincerely,
F, William BrowneII
Craig. S. Harrison
Enclosure
cc: Air docket A-95-28
-------
SEP-26-199S 15:21 FROM EPA flClD RfllN D!U.
TO
913018694078 P. 05
Summary Of Comments . For
the Draft Report Prepared fir U.S. EPA
by Bechtel/Cadmus
"Cost Estimates For Selected Applications Of
NOx Control Technologies
On Stationary Combustion Boilers"
Comments Prepared for
The Utility Air Regulatory Group
. Cichanowic2
. July 31, 1996
-------
SEP-26-1996 15=21 FROM EPfl ACID RflIN DIU.
TO
313018394078 P.06
Summary Of Comments For
the Draft Report:Prepared for U.S. EPA
by Bechtel/Cadmxs
Cost Estimates For Selected Applications Of
NOx Control Technologies
On Stationary Combustion Boilers
Comments Prewired by
J.E. Cidvanowio;
1.0 OVERVIEW
The subject report, entitled "Cost Estimates For Selected Applications Of NOx
Control Technologies On Stationary Combustion Boilers" (March 1996);
prepared for EPA by Bechtel Power Corporation (under subcontract to
Cadmus) addresses the cost of broadly applying advanced and presently
evolving NOx control options to the national boiler population. This report
is distinguished from previous NOx control technology assessments as it
addresses the specific task of bringing the entire inventory of coal-, oil-, and
natural-gas fired boilers into compliance with an extremely stringent NOx
level of 0.15 Ibs/MBtu. This analysis presumes all coal-, oil-, and natural-gas
fired boiler have already successfully applied combustion controls for Title IV
or RACT purposes. The analysis is based on background information and a
database of control technology developed by Bechtel for use in a previous
evaluation for EPA through Cadmus, described in the August 1995 report
"Investigation Of Performance and Cost of NOx Controls as Applied to Group
2 Boilers7', hereafter referred to as the Bechtel/Cadmus Group 2 Report
The key assumption of this analysis is the use of a power-law scaling
relationship to project capital cost over a wide range of generating capacity,
and process conditions. This critical assumption has been addressed in earlier
supplemental comments prepared for UARG, regarding the proposed Group
2 boiler NOx limits1. To reiterate, these cost evaluations employ a simple
power-law relationship which can introduce significant error if die range in
generating capacity over which cost is projected is too large. Generally, the
range of extrapolation should be within a factor of two so as to NOT require
changes in process design; otherwise an inappropriate design is considered as
the cost basis.
1 See Section 3.1 of "Supplemental Comments for Group 2 Boiler NOx Emission Limits,,
addressing Cost Evaluation. Methodology. And Technical Applicability of Selected NOx
Control Options", prepared by 7-E. Qchanowicz for UARG, June, 1996 (Attachment 1).
-------
SEP-26-1996 15:22 FROM EPR fiCID RRIN DIU.
TO
913018694078 P. 07
Comments; Cssl Estimates For Selected Applications Of
NOx Control Technologies On Stationary Boilers
Draft Report Prepared March 19S6-
Within the present study, I noted trends in capital cost estimates that are
counterintuitive and thus suggest errors in extrapolating cost information
from the Bechtel database. Specifically, the results suggest SCR capital cost for
cyclone-, cell-, and conventional wall- and tangential-fired boilers are similar
($~69/kW), despite the significantly higher, boiler NOx production rates from
the former two categories. (Under the premises of this study, cyclone and cell-
fired boilers are assumed to produce 220% and 140% of the NOx from
pulverized coal-fired boilers). This anomalous trend is most notable at 200
MW, but persists at higher generating capacity.
Similarly, the capital cost for rebum and SNCR were projected with a power-
law relationship of the same form, from generating capacities of 200 MW to
900-1000 MW. Capital cost estimates developed for deploying these
technologies at 200 MW appear consistent with utility industry experience
and recent process design information, these capital cost estimates are: for
SNCR, $16-18/kW on coal and $9-11/kW on oil/gas applications; for reburn
$19-22/kW on oil/gas. (Rebum was not judged capable of providing the
requisite NOx reduction capability on coal). However, projecting these capital
costs to large generating capacity (960 MW for oil/gas and 1030 MW for coal)
produces extremely low values. Accordingly, I suggest the power-law
relationship oyercredits economies of scale inherent, and process designs
developed for large capacity process conditions must be more complex, and
thus costly.
Generally, the Bechtel/Cadmus analysis accurately represents NOx reduction
capabilities of the candidate technologies. For coal, I concur that the only NOx
control technology capable of consistently meeting a 015 Ib/MBtu limit from
the boiler baseline NOx production rates cited is SCR; rebum and SNCR
cannot provide the minimum of 67% NOx control capability consistently
over the entire boiler population. For oil- .and natural gas-fired applications,
up to three of these technologies may be capable of meeting the 0.15 Ib/MBtu
limit from the assumed boiler NOx production rates (030 and 0.25 Ibs/MBtu,
respectively): SCR, SNCR, and reburn; However, RACT implementation
between various states' is not consistent, and the national boiler population
may contain a significant number of units that produce NOx in excess of the
assumed rates. Also, reburn may be limited to 35% NOx reduction for those
cases where LNB and OFA have already been deployed to achieve the 030 and
0.25 Ibs/MBtu limit SNCR may be limited to 35% NOx reduction on sulfur
bearing fuels due to formation of ammonium sulfates/bisulfates on
downstream equipment. Subsequently, I recommend that SNCR and reburn
are not capable of meeting the 0.15 Ib/MBtu limit for all oil- and gas-fired
units. .. . "••''.• • ••"•••'
Finally, additional information is requested from EPA and Bechtel/Cadmus
defining key assumptions and premises of the analysis. Examples of these
-2-
-------
SEP-26-1996 15:24 FROM EPR RCID RftIN DIU.
TQ
913018694078 P.08
Comments: Cost Estimates For Selectfd Applications Of
NOx Control Technologies On Stationary Boilers
Draft Report Prepared March 1996 -
are: basis for 20 year remaining life given the diversity of boilers in the
national population" specification of space velocity for SCR, design concepts
assumed to provide load-following capability.for SNCR, residence time
assumed for the boiler population for reburn, and specifics of calculating
levelized cost and cost per ton from information presented in summary
tables.
10 GENERAL METHODOLOGY AND ECONOMIC PREMISES
This section addresses the methodology and basic assumptions inherent to
the analysis. Most of these issues have been treated in earlier UARG
comments and thus will not be addressed in detail; rafter the previous
comments will be referenced as appropriate. :
Assumption Of RAGT-Equivalent .Boiler NOx Production Rates. The subject
Bechtel/Cadmus analysis assumes the entire boiler population has deployed
combustion NOx controls, contributing to a reduction in cost for
pe-'combustion or other advanced technology. This assumption is valid for
most cases; however two issues must be addressed;
• Oil-fired and gas-fired boilers. Most of the boiler inventory, particularly
those units located in the northeast and near ozone non-attainment areas
will probably emit at the 030 and 025 Ibs/MBtu rate assumed for mis
study. However, a significant inventory of units in the southeast may not
be in regions of non-attainment; these units will likely produce NOx
much higher than the assumed levels. Recommendation: Recognize and
account for an approximately 20% of ihe-pil/gas-fired boiler inventory
that will not be operating at RACT NOx limits.
• Cyclone/cell-fired boilers. Title JV Group 2 boiler NOx production
emission rates of 0.94 and 0.68 Ibs/MBtu are proposed for cyclone-fired and
cell-fired boilers, respectively. However, the boiler baseline NOx
production rates rot the cost evaluation are assumed to be 1.17 and 1.0
Ibs/MBtu, respectively. These higher boiler NOx production rates increase
capital cost, and also lower cost per ton of NOx removed.
Recommendation: Assume cyclone- and cellared boilers (and other
Group 2 boilers as appropriate) emit NOx at the Title TV proposed limit.
Absence of Detailed
T,ist&. UARG has previously cited the
importance in developing specific equipment lists as a necessary prerequisite
to developing meaningful cost information. An industry position paper
addressing SCR cost (initially issued by UARG and EPRI in 1993, and revised
-3-
-------
SEP-26-1996 15=25 FROM EPP flCID RflIN DIU.
TO
913018694078 P.09
Comments: Cost Estimates For Selected Applications Of
NOx Control Technologies On Stationary Boilers
• Draft Report Prepared March 1996
in 1994 for UARG and the National Mining Association2) stressed the need to
develop layout drawings defining the location of equipment as necessary for a
realistic cost estimate. EPA recognized this requirement, and directed Bechtel
to develop specific equipment lists (although layout drawings were not
prepared) for the "reference" cases developed for each boiler type in the
Bechtel/Cadmus Group 2 Report.
However, the scope and ambition of the subject Bechtel/Cadmus analysis - to
project advanced NOx control technology cost for .the national utility boiler
population - eliminates the possibility of preparing detailed equipment lists
for the candidate plants. Not all regulators support this position, as the Air
Director for New Hampshire proposed such a unit-by-unit detailed
assessment in OTAG as feasible3': Despite the claims of this one state
regulator, developing equipment lists for the majority of generating units in
the national boiler population (or even within the subset of OTAG) is not
feasible or possible under almost any conditions. Recommendation: EPA
should recognize the uncertainty regarding any results from such an
ambitious study, and assign an appropriate margin for error.
A minimum requirement for the ambitious objectives of the
Bechtel/Cadmus study is to (a) develop specific process designs
-------
SEP-26-19S6 15:26 FROM EPfi flCID RftIN DIU.
TO
913318594878 P. 10
Comments: Cost Estimates for Selected -Applisiations Of
MOx Control Technologies On Stationary Bailers
Draft Report Prepared. March 1996.
The subject Report does not define the. "study boilers" that serve as the basis
for the analysis. Recommendation: EPA should further define the boilers
that served as "baseline" for the analysis, disclosing site features, balance-of-
plant equipment, and the anticipated process impacts. - .
As the subject Bechtel/Cadmus Report is silent'on this issue, and derives data
heavily from the Bechtel/Cadmus Group 2 Report, I examined background
information presented in the latter report for insight. As noted in Appendix
B, page B2-1, "...the design details established for each boiler are representative
of typical boilers in the corresponding category". Although it is encouraging
that Bechtel Power has used a database with "typical" boilers, it must be
recognized that by definition "atypical" boilers are excluded from this cost
evaluation. .
Recommendation: EPA should recognize a certain fraction of boilers feature
site conditions presenting a greater challenge than assumed fcr this analysis,
and thus will incur higher cost. Accordingly, some boilers may require a
premium for further "Scope Adders" beyond that assumed by the Bechtel
database. . . . .
Selected Use Of EPRI TAG-Derived Cost Methodology. The subject report,
similar to the Bechtel/Cadmus Group 2 Report, claims that EPRI Technical
Assessment Guide (TAG) methodology is fully adopted as the basis for
estimating cost.
As noted in earlier UARG comments, this statement is not completely true.
Most significantly, the cost for financing capital equipment and construction
during the construction period - referred to as Allowance For Funds Used
During Construction (AFDC) • was ignored. As stated in earlier comments,
the construction period for SCR is anticipated to be one year - and thus
ignoring AFDC is inappropriate. This leads to underestimating capital costs
by nominally 5%.
In addition, the remaining plant lifetime and subsequent capital recovery
factor may not be appropriate. EPA assumes a 20 year remaining lifetime, and
assigns a capital recovery factor from the 1993 TAG of 0.127. (This capital
recovery factor is appropriate for a 20 year recovery period, and consistent
with TAG recommendations). However, the 20 year recovery period,
although appropriate for newer units, may be optimistic for many older units,
particularly Group 2 boilers. Recommendation: As the Bechtel/Cadmus
study is intended to address the national boiler population, an aggregate
representation of 17-18 years - and a corresponding capital recovery factor of
0.14 * may be more appropriate. •'.-.: .
-------
SEP-26-1996 15:27 FROM EPft flCID RfllN DIU.
TO
913018S94078 P. 11
Comments: Cost Estimates For Selected Applications Of
NOx Control Technologies On Stationary. Boilers
Draft Report Prepared Marsh 1996
3.0. COAL-HRED APPLICATIONS :
3.1. SCR CapitalCost Estimates (Table 1*3)
The capital cost analysis for SCR applied to coal-firing is predicated on a
scaling relationship (a) derived from one base case plant desigp developed for
.200 MW capacity, and (b) assuming the identical pjocess design developed for
200 MW applies to large capacity units (500 MW and greater). .As stated in
supplemental comments regarding the Bechtel/Cadmus Group 2 (see
footnote 1), the use of scaling relationships over a wide generating capacity
range violates the basic premises under which these relationships are
developed/arid can invalidate the results, .
Specifically, a recent article by Remer5 contains a final section entitled
"Limitations And Potential Errors" which discusses the use of cost vs. capacity
relationships, and proper scaling factor (referred to as 'R' value): .
"The use of 'the cost-capacity equation and the R factors presented in this
article can simplify the complex task of estimating equipment costs for air
pollution control. Users of these factors must be careful, however, not to
extrapolate outside the range for which the R value is determined. Jhe
cost found using this method are ball park estimates; when more exact
costs are required, actual vendor quotes should be sought."
This reference clearly cautions regarding the use of the power law '
relationship outside of the range for which the bask process design was
developed.
Due to either misuse of the scaling relationship and/or other assumptions in
this analysis, capital cost estimates developed for SCR applied to coal-fired
plants are not internally consistent or are they logical with basic design
trends. Specifically, first consider mat SCR capital cost reported for wall-fired
boilers ($69.38/kW) exceeds that projected for tangential-fired boilers
($66.82/kW). Presumably, this trend is due to higher boiler NOx production
for wall-fired boilers (050 vs. 0.45 Ibs/MBtu) requiring greater catalyst quantity
to meet the 0.15 Ibs/MBtu limit. Consistent with this logic, capital cost-for
cyclone and cell-fired units should significantly exceed capital cost for both
wall- and tangential-fired boilers/ regardless of capacity, as NOx production
rates from cyclone and cell-fired boilers (assumed by the Bechtel/Cadmus
5 A recent article in Chemical Engineering entitled "Air Pollution Control: Estimate The Cost
Of Scale-Up" (November, 1994, by Reiner et at.) addressed the concerns for employing the
conventional scaling relationship (Attachment 3).
-------
SEP-26-199S 15:25 FROM EPS flCID RfllN DIU.
TG
513018594078 P. 12
Comments: Cost .Estimates For Selected Applications Of
' • tVO* Control Technologic On Stationary Boilers
Draft Report Prepared Match 1996 •
analysis to be 1.17 and 1.00 ib/MBtu) will require SCR reactors with significant
catalyst quantity. - •
Contrary to this well-recognized design trend, the Bechtel/Cadmus analysis
reports SCR capital cost for cyclone, and cell-fired boilers, to be approximately
equivalent to wall-fired boilers at 200 MW, despite the significant difference
in boiler NOx production. Even at the largest boiler capacity evaluated
(1030 MW) there is a negligible difference in capital cost between ceE-fired and
wall/tangential-fired boilers, and only a 12% premium for a cyclone
application.
This observation suggests a flaw in the evaluation methodology. The only
conditions under which SCR capital cost for cyclone and cell-fired boilers
could, be equivalent to wall- and tangential-fired boilers are either/or (a)
ability to relax the residual ammonia limit for cyclone and cell-fired units, or
(b) reduction in cost of the non-catalyst (e.g. ancillary) components. Neither
items (a) or (b) appear likely. Accordingly, SCR cost is probably
underpredicted by the power-law scaling relationship for cyclone boilers.
Recommendation: EPA should recognize the potential for errors in capital
cost, due to the selection of the "reference" site, and extrapolation from the
Bechtel database over generating capacity and process conditions. These
results further support UARC-suggested capital estimates (-r$18/kW higher).
A second issue to be addressed is the ability of SCR to provide in excess of 80%
NOx reduction for coal-fired applications. Regarding high NOx reduction,
several factors prevent achieving greater than 80% NOx reduction without
incurring significant capital cost penalties: these are maintaining strict t<3
ppm) limits of residual N7H3 in flue gas, achieving uniform NH3/NO mixing,
managing maldistribution in flue gas velocity due to flow path, and relatively
low boiler NOx production rates. The potential barriers these issues present
to achieving a NOx limit of C.15 Ib/MBtu have been described in previous
OTAG deliberations6. Recommendation: Except for the case of exclusive
natural gas firing, assign SCR a NOx reduction capability of 80%.
Finally, the assumption that no unusual site features exist that complicate
retrofit of equipment has been discussed in Section 2 of this document
Recommendation: EPA should recognize that some units mil require
additional equipment for retrofit, and assign a. cost accordingly.
6 See letter from Hunton t William's Craig Harrison to Brock Nicholson, dated June 17,1996
(Attachment 4). . ..'.'• '•;''.
r7-
-------
SEP-26-1996 15!30 FROM EPfl ftCID RfllN DIY.
TO
913018594078 P.13
Comments: Cost Estimates for Selected Applications Of I
NOx Control Technologies On Stationary Boilers \
. ' Draft Xepori Prepared March 199$ I
' , ' ' ' •' • ' '.'••.. \
3.2. SNCR Capital Estimates (Table 1-4)
For 200 MW capacity, the-capital cost estimated for SNCR ($16-18/kW) is
similar to tiiat reported for most commercial installations (e,g.
PSE&G/Mercer, Atlantic Electric/Englund, New England Power/Salem
Harbor). The relatively low capital cost estimate developed for SNCR applied
to 1030 MW f$6-7/kW) could be a consequence of inappropriately employing
the power kw scaling relationship. Specifically, the significant increase in
physical distance across the boiler over which reagent must be mixed could
require increasing the number of SNCR injectors, or deploying high
momentum lances in lieu of low energy wall injectors. In addition, the
greater mixing distance probably requires an increase in the sophistication
and complexity of the reagent process control system. Any of these
modifications will increase the capita] cost and are not accounted for in the
scaling relationship derived by Bechtel/Cadmus. Recommendation: EPA
should recognize the increase in complexity of SNCR technology .with greater
generating capacity will negate any economies of scale, and employ the .capital
requirement developed at 200 MW for all capacities.
4.0. OIL/NATURAL GAS FIRED APPLICATIONS
4.1. SCR ' . ' : ••• '. V
For both oil and natural gas-firing, SCR capital cost estimates (Table 1-3) for
200 MW appear consistent with industry experience at Southern California
Edison, San Diego Gas & Electric, and the Los Angeles Department of Water &
Power. There.is no experience with SCR; capital cost at 930 MW by which to
compare Bechtel/Cadmus estimates. Similar to the discussion for coal-firing,
the use of the power-law extrapolation without considering the need for a
change in process design can underpredict capital cost estimates.
As stated in the 1995 SCR cost white paper, the importance of exclusive firing
a boiler with natural gas in contributing to low SCR capital cost cannot be
discounted. The referenced document describes how exclusive use of natural
gas - relegating fuel oil backup operation 1-2 weeks per year at most - auows
the use of extremely small pitch, high vanadium content SCR catalyst.
Consequently, proportionally small quantities of catalyst are required, with
space velocities for these applications exceeding 25,000 1/h (compared to 3500-
5000 for coal-fired SCR). Clearly, the five-fold reduction in catalyst quantity
compared to coal-fired application is key to minimizing SCR capital cost for
such applications. Recommendation: EPA should recognize SCR cost
projected is most applicable to natural gas*fired applications,; EPA should
include a cost premium for applications to sulfur-containing fuel oil.
-------
SEP-26-19S6 15:31 FROM EPfl fiCIB RAIN DIU. TO 913018SS4078 P.14
Comments: Co&t Estimates For Selected Applications Of
Control Technologies On Stationary Boilers.
Draft Report Prepared Maircfc 1996 •
4.2. SNCR, Rebum
Projected capital cost for SNCR applied to oil-fired and gas-Hied boilers (Table
1-3) at 200 MW capacity ($9-11/kW) is reasonably consistent with industry
experience {e.g. ULCo/Pdrt Jefferson Station). The capital cost projected lor
SNCR at 930 MW ($4/kW) appears artificially low; it is .possible scaling-issues
are not properly treated. As stated for coal-fired SNCR applications, achieving
the targeted NOx reduction may require increasing tire number of reagent
injectors, or employing the high momentum lance-type injectors.
For rebum, there is no domestic commercial experience with oil and natural
gas-firing at any capacity by which to judge cost (It will be assumed that oil-
fired stations, are equipped with natural gas on-site, and will not require
additional investment for natural gas- access). Reburn capital cost for 200 MW
capacity ($19-22/kW) for oil and gas is probably accurate, due to the relatively
compact and simple furnace arrangement compared to coal-firing. However,
estimates of rebum capital cost at 930 MW ($lM3/kW) could be
underpredicted as the scaling relationship does not recognize the need to
increase the complexity of the system to accommodate greater mixing
distances for rebum fuel .
-Recommendation: For both SNCR and reburn, the potential for requiring
increased complexity for either reagent or reburn fuel injectors with higher
capacity may negate any economies of scale. Thus, capital requirement
developed at 200 MW should be applied to all capacities.
Also; as noted in the following section/ NOx reduction capability for rebum
on oil- and gas-fired boilers to meet the proposed 0.15 Ibs/MBtu limit may not
be adequate, due to prior deployment of aggressive combustion controls such
as low NOx burners and overfire air. Further process enhancements to
increase NOx reduction may be necessary, elevating cost.
5.0 NOx CONTROL PERFORMANCE
NOx control performance for candidate technologies as assumed in the.
Bechtel/Cadmus Report, and summarized in Table 1-5, appear reasonable,
with several exceptions. The following further clarifications are offered:
5.1. Coal Rebum .'',.•
The maximum NOx reduction for coal reburn on coal-fired boilers (50%) is
probably achievable for boilers approximating 200 MW generating capacity (as
opposed to 930 MW capacity), featuring sufficient furnace "height" or
residence time, and firing relatively high volatility coals.
-------
SEP-26-1996 15:32 FROM EPfl flCID RftIN DIU.
TG
913018694078 P. 15
Comments: Cost -Estimates For Selected Applications Of
. NOz Control Technologies On Stationary Boilers
Draft Report Prepared 'March 1996
Conversely, boilers of larger capacity (e.g. significantly greater than 200 MW),
with restricted furnace "height" or residence time, and firing relatively Jpw
volatility coal will probably be limited to the 35% reduction efficiency.
The significant impact of boiler design criteria on NOx removal capability for
coal-reburn was evidenced by the analysis conducted by Babcock & Wilcdx
(B&W) on residence time requirements for cyclone boilers; results were
highlighted in a letter from B&W to the Acid Rain Division's L. Kertcher7.
Specifically, this analysis showed that no more than 35% NOx reduction can
be anticipated for a significant portion of the cyclone boiler inventory. A
similar trend may characterize pulverized coal-fired boilers, with a significant
fraction of the inventory not providing the necessary residence time for
reaction.
Accordingly, the assumption that coal rebum NOx control cannot provide
sufficient NOx control capability for coal-fired boilers (minimum of 67%) to
meet the targeted 0.15 Ibs/MBtu level is consistent with the best available
experience and process design information.
5.2. Natural Gas Reburn
The maximum NOx reduction (60%) is potentially achievable for boilers
approximating the 200 MW capacity size (as opposed to 930 MW capacity), and
with sufficient furnace "height" or residence time.
One critical unresolved issue regarding natural gas rebum performance is the
NOx reduction capability subsequent to application of combustion controls
such as LNB and overfire .air technology. As discussed in the draft topical
report addressing NOx control technology for the OTAG Control Options
Workgroup8, most demonstrations concerning natural gas reburn addressed
uncontrolled boilers that had not previously installed LNB or OFA
technology. As addressed on page 24 of the report referenced in footnote 8,
results from the sole demonstration of natural gas rebum on a coal-fired
boiler suggest LNB contributes to rebum NOx reduction. These results
suggest NOx reduction with gas reburn applied to boilers mat have already
retrofit LNB and OFA may be limited to 35%.
In addition to concerns regarding NOx reduction subsequent to RACT
controls, boilers of larger capacity (e.g. significantly greater than 200 MW), and
7 Letter from JAt Piepho (Babcock t Wilcox) to LF. Kertcher (EPA Acid Rain Division),
October 27,1995 (Docket item A-95-28, Attachment 5). . .
8 "Electric Utility Nitrogen Oxides Reduction Options For Application By The Ozone
Transport Assessment Group", Prepared for the OTAG Control Technologies & Options
Workgroup, January 1996 {Attachment 6).
-10-
-------
SEP-2S-1996 15=33 FROM EPfi ftCID RAIN DIO.
TO
913018694078 P. 16
Comments: Cost Estimates For Selected Applications Of
NOx Control Technologies On Stationary Boilers
•'.''•• Draft Report Prepared'March 1996
with restricted furnace "height" or residence time will probably "be limited to
the 40% reduction efficiency.
5.2:1. Coal-Firing. Accordingly, the assumption that natural gas rebum NOx
control' cannot provide sufficient NOx control capability (minimum of ;67%)
on coal to meet the targeted 0.15 Ibs/MBtu level is consistent with th^ best
available experience and process design information.
5.2.2. Oil/Gas Firing. NOx reduction of 40-50% is required of natural gas
reburn to meet a 0.15 Ib/MBtu emission limit. For the reasons delineated
earlier, mis level of reduction may be beyond the capability of reburn on
oil/gas.
5.3. Selective Catalytic Reduction
The maximum NOx reduction cited for SCR (90%) is achievable only for
boilers firing exclusively natural gas, with essentially no significant oil backup
capability. As discussed previously, this is due to die absence of ash and
sulfur, which allow the use of small pitch, high activity catalysts that provide
significant surface area for NOx reduction. This nigh NOx reduction is not
considered feasible for coal and oil-fired applications, due to limits on
residual ammonia. . . .
Conversely, the more common 80% NOx reduction is typical for coal-fired
applications, where residual ammonia must be maintained to 5 ppm or
below. The letter referenced in footnote 6 cites key reason why. NOx
reduction for large, coal-fired boilers is nominally limited to 80%.
5.4. Selective Non-Catalytic Reduction
The maximum NOx reduction (50%) is probably achievable for boilers
approximating the 200 MW capacity size (as opposed to 930 MW),.firing
natural gas, where the absence of sulfur allows relatively high (>5 ppm)
residual NH3. This high degree of NOx reduction is not considered feasible
for coal and oil-fired applications, which are probably limited to 25-30% NOx
reduction, unless unusual circumstances allow an increase in residual
f-g- from 5-10 ppm to 10-20 ppm). The reasons fox limiting NOx
reduction to 25-30% are addressed in Section 5.3.5 of the document referenced
in footnote 8.
Accordingly/ the assumption that SNCR cannot provide sufficient NOx
control capability with coal to meet me'targeted 0.15 Ibs/MBtu level is
consistent with the best available experience and process design information.
For oil/gas firing, SNCR does provide sufficient NOx control capability .
(40-50%) to meet a 0.15 Jb'/MBtu emission limit.
-------
SEP-26-1996 15:35 FROM EPfi ftCID RflIN DIU.
TO
913018694073 P. 17
Comments: Cost Estimates For Selected Applications Of
NOx Control Technologies On Stationary Bjilers
Draft Report Prepared Mtirsh 1996
6.0. ADDITIONAL JNFORMAT1ON REQLORED
•' . , ' .
The Bechtel/Cadmus report did not disclose many specific details of thb
analysis. The following discussion identifies basic assumptions and
additional information necessary for a complete review of this report:
6.1. Economic Assumptions .
The most significant economic assumption is remaining plant life, and the
implications "for capital recovery charge. The assumed capital recovery charge
is particularly important in this analysis as capital-intensive SCR is the sole
feasible technical option for coal-firing. Bechtel/Cadmus is requested to
justify the 20 year period .as an appropriate remaining lifetime.
As an alternative, several categories of plant age could be determining from
FERC-derived information, and remaining lifetime selected accordingly.
Although a significant fraction of units would be characterized by the 20 year .
remaining lifetime, a large fraction would also be represented by 15 and 10
year lifetimes. These units would be forced to incur a higher capital recovery
factor.
6.2. Technical Assumptions
Site Features Of Reference Plants
The subject Bechtel/Cadmus report is silent on the specific source(s) used to
construct the Bechtel database of reference plants; accordingly the statement
from the Bechtel/Cadmus Group 2 report is assumed relevant. This
statement (Appendix B, page B2-1) states "The design basis for each boiler
plant was developed from the Bechtel in-house database for similar operating
boiler installations. The design details established for each boiler are
entative of typical boUers in the corresponding category".
EPA/Bechtel is required to share the specifics of the "similar boiler
installations'' used as input to the database, including layout drawings of the
proposed typical plants. -
6.2.2. SCR.
The following additional information is requested:
velocitv
tiantitv of catalvst
tired for NOx removal to iheet
0.15 fes/MBtul. Three separate SCR application cases should be addressed:
coal-fired, fueloil-fired, and natural-gas fired. -•
-12-
-------
5EP-2S-199S 15=36 FROM EPfi flCID RflIN DIU.
TO
913018694078 P.18
Comments: Cast Estimates For Selected Applications Of.
NOx Control Technologies On Stationary Boilers
, Draft Report Prepared March 1996
* SQ2 Oxidation. The report should identify the specified limit (if any) on
the conversion of SO2 to SOS. .
• Boiler Economizer Bypass. It is not clear if a boiler economizer bypass is
included in the design to allow SCR operation at low load. (Presumably,
SCR would be widely deployed under this scenario, and thus load-
following units would require a boiler economizer bypass).
• Sj^e Physical Constraints. The report should Specify the fraction of the
boiler population that must address unusual site features, that force
significant equipment location. Under the present analysis, the fraction
assumed appears to be zero, but this is not explicitly stated.
6.23. SNCR .
SNCR NOx control performance for load-following applications is strongly
dependent upon the ability to inject reagent effectively into flue gas in the.
correct temperature window. The proposed broad application of SNCR over
the oil/gas boiler population suggests many candidate units would operate in
load-following or "deep" cycling mode. What SNCR design concepts are
assumed to vary reagent injection, such as the use of a multi-level injector
assembly, and are these fully accounted for in the capital cost estimate? -Also,
what methods, if any, are assumed available to mitigate the impact of residual
ammonia on fly ash or downstream equipment?
6.2.4. Rebum
The Bechtel/Cadmus evaluation assumes both coal and natural gas rebum
are applicable to most boilers. As indicated by the document referenced in
footnote no. 6, a certain fraction of the boiler population may feature a
residence time distribution that eliminates reburn as a feasible technology.
What assumptions has Bechte! Power made regarding residence time
distribution for the boiler population? The text suggests, but does not state,
the entire population is assumed to offer adequate residence time for rebum.
This assumption is important/ as one vendor of reburn technology (Energy
arid Environmental Research Corporation) claims that NOx reduction is
dependent on residence time available, and depending on NOx reduction
desired, any boiler can deploy rebum. Thus, residence time assumptions are
necessary to define me NOx reduction potential.
6.3. Economic Results
It is not clear how the results in Table 1-2 are employed to create Table t3.
Bechtel is requested to define the methodology showing how the resulfe in
Table 1-2 are used to construct a mills AWh cost, which is then translated into
a tevelized cost per ton.
-13-
TOTflL P.18
------- |