EPA/600/R-01/087
October 2001
Cost of Selective Catalytic Reduction (SCR)
Application for NOX Control
on Coal-fired Boilers
Prepared by:
D. Foerter
ESI International, Inc.
1660 L Street, N.W.
Washington, DC 20036
and
W. Jozewicz
ARCADIS Geraghty & Miller, Inc.
P.O. Box 13109
Research Triangle Park, NC 27709
EPA Contract No. 68-C99-201
Work Assignment 1-019
U.S. EPA Project Officer: Ravi K. Srivastava
National Risk Management Research Laboratory
Research Triangle Park, NC 27711
Prepared for:
U.S. Environmental Protection Agency
Office of Research and Development
Washington, DC 20460
-------
11
-------
Abstract
This report provides a methodology for estimating budgetary costs associated with retrofit
applications of selective catalytic reduction (SCR) technology on coal-fired boilers. SCR is a post-
combustion nitrogen oxides (NOX) control technology capable of providing NOX reductions in excess
of 90 percent. With SCR, NOX reductions are achieved by injecting ammonia into the flue gas, which
then passes through layers of catalyst in a reactor. The ammonia and NOX react on the surface of the
catalyst, forming molecular nitrogen and water. In the United States, SCR has been applied mainly to
electrical utility boilers firing coal and natural gas and ranging in capacity from 25 to 800 megawatts
(MW).
The costing methodology presented in this report is applicable to SCR retrofits on coal-fired boilers
ranging in capacity from 100 MW to approximately 850 MW and with design efficiencies greater
than 80 percent and up to 95 percent of NOX removal. The cost equations and variables used in the
methodology are based on information obtained from SCR system suppliers and reflect experience
gained from over 200 SCR applications. It is noted, however, that the budgetary cost estimates for
typical SCR applications that this methodology can provide cannot replace the detailed site-specific
engineering cost studies or cost quotations that are developed by SCR system suppliers.
in
-------
Contents
Abstract ii
List of Tables iv
List of Figures iv
Acknowledgements v
1. Introduction 1
2. Description of the SCR Technology 2
3. Description of the Costing Methodology 4
4. Costing Algorithms 5
Capital Cost 5
Fixed O&M Cost 7
Variable O&M Cost 7
5. Validation of the Costing Methodology 10
6. Conclusions 11
References 12
Appendix. Determination of the Design Margin Accounting for Ammonia Slip 16
iv
-------
List of Tables
Table 1. A summary of equations used in the costing methodology 14
Table 2. Calculation of cost estimates using the costing methodology 15
List of Figures
Figure 1. Process flow schematic for an SCR application 2
-------
Acknowledgements
This work was completed by ARCADIS Geraghty & Miller, Inc. under support provided by U.S.
EPA's Office of Research and Development. ESI International, Inc. was a subcontractor to
ARCADIS for this work. The technical guidance provided by the EPA Work Assignment Manager,
Ravi K. Srivastava, is gratefully acknowledged.
VI
-------
Chapter 1
Introduction
Selective catalytic reduction (SCR) is a post-combustion nitrogen oxides (NOX) control technology
capable of providing NOX reductions in excess of 90 percent.1 With SCR, NOX reductions are
achieved by injecting ammonia (NH3) into the flue gas, which then passes through layers of catalyst
in a reactor. The NH3 and NOX react on the surface of the catalyst, forming molecular nitrogen (N2)
and water.
SCR has been applied to stationary source, fossil-fuel-fired combustion units for NOX emission
control since the early 1970s and is currently being used in Japan, Europe, and the United States.2'3'4 It
has been applied to large utility and industrial boilers, process heaters, and combined cycle gas
turbines. In the United States, SCR has been applied mainly to electrical utility boilers firing coal and
natural gas ranging in size from 25 to 800 megawatts (MW).
This report provides a methodology for estimating budgetary costs associated with typical retrofit
applications of selective catalytic reduction (SCR) technology on coal-fired boilers. The cost
equations and variables used in the methodology are based on information obtained from SCR system
suppliers and reflect experience gained from over 200 SCR applications. It is noted, however, that the
budgetary cost estimates for typical SCR applications that this methodology can provide cannot
replace the detailed site-specific engineering cost studies or cost quotations that are developed by
SCR system suppliers.
The cost estimates developed using the costing methodology have been considered in the context of
estimates made using other approaches 5'6'7'8 as well as estimates from reported case studies.9'10'11 In
general, there tends to be good agreement between predictions of this methodology and reported
estimates of capital and operation and maintenance (O&M) costs.
-------
Chapter 2
Description of the SCR Technology
In the SCR process, NH3 is injected into the flue gas within a temperature range of about 315 to 400
°C (600 to 750 °F), upstream of a catalyst. Subsequently, as the flue gas contacts the SCR catalyst,
NOX, which predominantly is NO in combustion devices, is chemically reduced to nitrogen as
follows:
2NO + 2NH3 + !/202 -> 2N2 + 3H2O
(1)
Figure 1 shows the process layout of an SCR system in which the catalyst is located between the
economizer and the air preheater. This process layout, known as hot-side SCR, is most commonly
used in SCR applications.
Equation (1) indicates that, theoretically, 1 mole of NH3 is required to reduce 1 mole of NO. Any
unreacted NH3 released from the SCR system, known as NH3 slip, is undesirable because it may
combine with sulfur dioxide (SO2) and sulfur trioxide (SO3) in the flue gas to generate ammonium
sulfate and bisulfate compounds that may cause fouling of downstream equipment, especially the air
preheater. By maintaining close to the theoretical stoichiometry, the NH3 slip can be kept at
acceptable levels in properly designed modern SCR systems,12 while NOX reductions in excess of 90
percent can be achieved.
SECONDARY
AIR
Figure 1. Process flow schematic for an SCR application.
-------
NOX reduction with NH3 is exothermic, resulting in the release of heat. However, because the NOX
concentration in the flue gas at the inlet of the SCR is typically 0.02 to 0.01 percent by volume, the
amount of heat released is correspondingly small.
With a properly designed and controlled NH3 injection rate and typically 2 to 4 percent excess
oxygen, the NOX reduction reactions achieve completion, provided the reaction temperature is
maintained within the required range.
The catalyst plays a central role in this NOX control technology. Originally, SCR catalysts were made
of precious metals such as platinum. In Japan, researchers started using base metals consisting of
vanadium, titanium, and tungsten in the late 1970s, thereby significantly reducing costs. Further
improvements in catalyst formulations have resulted in decreased unwanted side reactions such as
SO2 to SO3 conversions, increased resistance to flue gas poisons, and increased catalyst activity. As a
consequence, catalyst volumes needed to achieve a given level of NOX reduction have decreased and
the operating life of catalysts has increased.
SCR systems have been operating for many years on fossil-fuel-fired boilers in Japan, Europe, and
the United States and have experienced relatively few operational or maintenance problems. Thus,
SCR is considered to be a NOX control technology that is capable of providing a high degree of NOX
reduction in a reliable manner.
-------
Chapter 3
Description of the Costing Methodology
The methodology is applicable to SCR retrofits on coal-fired boilers with design efficiencies greater
than 80 percent and up to 95 percent of NOX removal. The methodology was not extended to lower
NOX removal efficiencies for several reasons. First, in general, there has been less of a demand for
achieving low NOX removal efficiencies (i.e., below 70-75 percent) with SCR.13'14'15'16 This is
because, typically, SCR systems with higher NOX removal efficiencies are more cost effective.
Second, the procedures for obtaining budgetary cost estimates at lower NOX removal efficiencies are
already available.7 Finally, SCR systems designed for lower NOX removal efficiencies often may be
more applicable to unique applications such as SNCR/SCR hybrid systems;16'17'18 however, such
applications are not in the scope of this work.
The methodology can be applied to coal-fired boilers ranging in capacity from 100 to approximately
850 MW and is applicable to all coal-fired boiler types. The range of inlet NOX (i.e., NOX
concentration at the inlet of the SCR reactor) in this methodology can vary between 0.15 and 2.5
lb/106 Btu of NOX. The costing equations and variables are based on experience gained at over 200
SCR applications in the U.S. and around the world. Data fit constants used in these equations are also
based on industry data. In general, as a budgetary costing procedure not intended to be a substitute for
a more detailed unit specific study, the methodology is thought to slightly overestimate cost in most
cases.
-------
Chapter 4
Costing Algorithms
Costing algorithms for capital, fixed O&M, and variable O&M costs are described in this section.
Capital Cost
The capital cost of a SCR retrofit is estimated in $/kW (January 2000 dollars) using Equation (2)
below:
D = 75 {300,000 [(B/1.5)005 (C/100) 04]/A}035 (2)
where:
D = capital cost ($/kW)
75 = capital cost ($/kW) associated with a typical SCR retrofit on a 300,000 kW coal-fired unit
300,000 = reference to a 300,000 kW baseline unit (basis for the economy-of-scale adjustment in
the equation)
B =NOX (lb/106 Btu) at the inlet of SCR reactor; range of approximately 0.15 - 2.5 lb/106 Btu
1.5 = reference NOX concentration in lb/106 Btu at SCR inlet
0.05 = exponent for inlet NOXconcentration 13
C = NOX removal efficiency (percent); range of greater than 80 and up to 95 percent
0.4 = exponent for NOX removal efficiency13
A = plant capacity (kW); range of approximately 100,000 - 850,000 kW
0.35 = exponent for an economy-of-scale adjustment factor (scaled from 300,000 kW unit) 16'19
The capital cost estimated by Equation (2) accounts for the costs associated with equipment,
installation, engineering, contingency, spare parts, and commissioning. However, this cost does not
include an allowance for funds during construction (AFDC) because of a relatively short construction
schedule associated with typical SCR installations. Note that capital cost as estimated by Equation (2)
is a function of three primary parameters: plant capacity (kW), initial NOX (lb/106 Btu), and NOX
removal efficiency (percent). Additional comments on these parameters are provided below.
Plant Capacity
The capital cost equation is reliable for plant capacities in the 100 MW to 850 MW range. As noted
above, the capital cost equation was derived based on a 300 MW unit. Capital cost estimates for units
other than 300 MW are obtained by the application of a scaling factor. Applying this equation to
capacities significantly greater than 850 MW could result in economy-of-scale benefits that would not
normally occur.16 However, in situations where SCR retrofits are considered for significantly larger
capacity units (e.g., two 1300 MW boilers at the Gavin and Cumberland facilities), the combined
capacity is often split between multiple air heaters and catalyst reactors. For example, two 1300 MW
capacity units could be treated by utilizing three air heaters and three catalyst reactors, each with
approximately 867 MW capacity.16 Therefore, units significantly larger than 850 MW can generally
be accommodated by this methodology.
-------
Inlet NOX
Equation (2) is reliable for inlet NOX in the range of 0.15 and 2.5 lb/106 Btu. This range would
accommodate most coal-fired boilers including uncontrolled cyclone units that operate with inlet NOX
in the range of approximately 0.9 to 2.5 lb/106 Btu.
NOxRemoval Efficiency
The capital cost equation is reliable for NOX reduction efficiencies greater than 80 and up to 95
percent. Note that these efficiencies depend on catalyst specifications including formulation and size.
These specifications, as well as the associated reactor size, will vary from manufacturer to
manufacturer and are proprietary.
Design and Installation
The costing methodology presented in this report is for a typical (or average) SCR retrofit and,
therefore, assumes a generic SCR system design and installation. With this constraint, the
methodology is not suitable for estimating detailed site-specific costs of an SCR application. The
following discussion briefly explains the major elements that can influence the cost of an SCR retrofit
and how they were accommodated in the methodology.
The major design elements that can influence the cost of an SCR retrofit on a coal-fired boiler are:
reagent storage, vaporizer, dilution fan/chamber, piping, and injection grid;
ductwork for flue gas, and economizer bypass;
SCR reactor bypass;
air heater modifications;
soot blowers and modification to ash handling system;
fans and electrical supply and equipment;
SCR reactor(s) and catalyst;
structural steel;
foundations;
enclosures; and
instrumentation and controls.
The extent of incorporation of the above elements in an SCR system design is dependent on the plant
layout. For example, a relatively constrained plant layout may involve a more difficult SCR system
design compared to a relatively unconstrained layout and, therefore, may require more ductwork and
air heater modifications. In contrast, a relatively unconstrained layout may not need air heater
modifications and more than the typical amount of ductwork. To account for varying levels of design
complexity, a degree of difficulty is associated with an SCR retrofit application.13'14'16'21 In this
context, the average degree of difficulty is assigned to a retrofit where the SCR installation is
relatively simple (i.e., the facility has adequate space for the SCR system). Equation (2) assumes this
average degree of difficulty.
Although the capital cost equation applies only to SCR retrofits on coal-fired boilers, it can be used to
develop estimates for new coal-fired unit SCR applications. Generally, these latter applications will
experience less design and installation constraints since the SCR is part of the overall plant design
and construction. Therefore, in general, the capital cost of a retrofit application would be 20-50
percent higher compared to that of a new application.9
-------
Fixed Operation and Maintenance (O&M) Cost
The fixed O&M cost is a function of the capital cost estimate and is estimated using Equation (3)
below:
E = D * A * C (3)
where:
E = fixed O&M cost ($/yr)
D = capital cost ($/kW) from Equation (2)
A = plant capacity (kW); range of approximately 100,000 - 850,000 kW
C = a constant; 0.0066 yr"1
The fixed O&M cost is the sum of the annual maintenance material and labor cost, and is estimated to
be 0.66 percent16 of the capital cost. It should be noted that while assumptions of as much as 2
percent have been used in a number of other estimation methods, information from recent case studies
have shown this value to generate fixed cost estimates that are generally too high. For example, the
assumption of 2 percent along with 80 $/kW for a 300 MW application would result in a fixed O&M
cost of 480,000 $/yr, a value much higher than approximately $100,000 associated with boiler
capacities of 300 to 350 MW.9 However, it should be noted that fixed O&M cost tends to be a small
portion of the annual cost associated with application of any NOX control technology. It is also
important to note that Equation (3) does not include the cost associated with catalyst replacement. In
this methodology, catalyst replacement cost is included in the variable O&M cost estimate.
Variable O&M Cost
The variable O&M cost, F, is estimated in $/yr using Equation (4):
F = G [225 * (0.37B * H * C/100 * 8760/2000) * 1.005 * 1.05 + 0.025 * D * A *
{(B/1.5)005 (C/100) °4} + 1.45 * A] (4)
As expressed above, Equation (4) addresses the costs associated with reagent usage, catalyst
replacement, and energy consumption. These costs are calculated as follows.
NH3 use cost ($/yr) = G * 225 * (0.37B * H * C/100 * 8760/2000) * 1.005 * 1.05 (4a)
where:
G = annual capacity factor (expressed as a fraction)
225 = anhydrous NH3 cost in $/ton
0.37 = molecular weight of NH3/molecular weight of NOX measured as NO2 = 17.03/46.01
B = inlet NOX (lb/106 Btu); range of 0.15 - 2.5 lb/106 Btu
H = heat input (106 Btu/hr)
C = NOX removal efficiency (percent); range of greater than 80 and up to 95 percent
8760 = hr/year
2000 = Ib to ton conversion
1.005 = design margin that accounts for NH3 slip (see the appendix)
1.05 = design margin that accounts for small amount of NO2 in flue gas (SCR chemistry requires
2 moles of NH3 per mole of NO2 instead of 1 mole of NH3 per 1 mole of NO)
-------
Annual catalyst replacement cost ($/yr) = G * 0.025 * D * A * [(B/l .5)° °5 (C/100) °4] (4b)
where:
G = annual capacity factor (expressed as a fraction)
0.025 = catalyst deactivation factor for coal-fired units (yr"1).13'14'16
D = capital cost ($/kW)
A = plant capacity (kW); range of approximately 100,000 - 850,000 kW
B = inlet NOX (lb/106Btu); range of 0.15 - 2.5 lb/106Btu
1.5 = reference NOX concentration in lb/106 Btu at SCR inlet
0.05 = exponent for inlet NOX
C = NOX removal efficiency (percent); range of greater than 80 and up to 95 percent
0.4 = exponent for percent NOX removal efficiency13
Note that, based on information from industry, 0.025 was selected as the default factor associated
with catalyst replacement in Equation (4b).16 However, it is important to note that this factor is likely
to remain debatable among catalyst, as well as system, suppliers. The factor of 0.025 assumes that
one catalyst layer would be replaced approximately every 15,000 and 20,000 hours of operation.
Other values of this factor have been suggested,13'16 including values as high as 0.075, and as low as
0.01. The lower the factor, the slower the catalyst deactivation being assumed. Given the importance
of catalyst cost, changing factors over a broad range can result in a significantly increased (or
decreased) estimate for variable O&M cost. Selection of the default factor of 0.025 is consistent with
the overall approach of this methodology of developing conservative cost estimates for typical or
average SCR retrofit applications. This factor is dependent upon the number of hours a unit is
assumed to operate per year. The factor of 0.025 assumes full-year operation of the catalyst (i.e., no
bypass). If the catalyst is used only during an ozone season (typically a 5-month period), the number
of useful years for the catalyst could increase, warranting a lower value for the factor. This variable
can also change as a function of fuel quality (e.g., use of high/low sulfur or more/less arsenic content
of coal). Generally, the more "difficult" is the fuel for SCR application (e.g., Powder River Basin coal
with potential for catalyst masking due to high ash alkalinity), the higher is the factor. In such a
context, a higher factor (possibly approaching 0.075) could be deemed reasonable in Equation
(4b).13'14'16
In Equation (4b), the average catalyst life is assumed to be approximately 15,000 to 20,000
hours.6'16 In this methodology, a "three plus one" layer catalyst arrangement is assumed (three catalyst
support layers plus one "spare/dummy" layer). This arrangement includes a spare catalyst layer and
has an advantage such that only one catalyst layer is replaced at the end of 15,000-20,000 hours.
There is also an underlying assumption that the catalyst cost6'16 is $8000/m3, which may be more
representative of the current catalyst market, rather than the historical market (e.g., some studies have
noted catalyst cost at existing retrofits closer to $11,000 to $14,000 /m3).9
Energy requirement cost ($/yr)1? = G * 1.45 * A (4c)
where:
G = annual capacity factor (expressed as a fraction)
1.45 = a constant ($/kW-yr) = 0.03 (cost of energy in $/kWh) * 8760 (h/yr) * 0.0055 (fraction cost
of auxiliary power/unit of generation)
A = plant capacity (kW); range of approximately 100,000 - 850,000 kW
-------
Note that power cost assumes that typical coal-fired SCR applications use 0.5 to 0.6 percent of the
total auxiliary power cost. Therefore, 0.55 percent is used as a default.16 Additionally, an assumption
of 6 to 7 inches (15.24 to 17.78 cm) of water pressure drop through the SCR system is implicit in
Equation (4c).16
Table 1 presents a summary of equations presented in this section and illustrates their use through
an example. Further, Table 2 presents cost estimates calculated with the methodology for SCR
retrofits on a range of plant capacities with varying NOX reduction requirements.
-------
Chapter 5
Validation of the Costing Methodology
The costing methodology was derived from information provided by SCR system suppliers and is
deemed valid for typical coal-fired SCR retrofit applications for NOX removal efficiencies greater
than 80 and up to 95 percent. However, it is important to note that this methodology is intended and
valid for developing budgetary cost estimates and assumes typical installations. Therefore, the
methodology should not be expected to account for reported costs of each site-specific SCR retrofit
application.
In a 1998 study9 conducted for Northeast States for Coordinated Air Use Management (NESCAUM)
and Mid-Atlantic Regional Air Management Association (MARAMA), the capital cost for SCR
retrofit on dry-bottom wall- and tangentially fired boilers to achieve 85 percent NOX reduction was
estimated to be between 70 and 90 $/kW. Similarly, the study estimated that capital cost of SCR
retrofits to achieve 90 percent NOX reduction from wet-bottom boilers would also range between 70
and 90 $/kW. Note that these estimates were based on 330 MW units. In comparison, the costing
methodology results in an estimate of approximately 70 $/kW for similar units.
Recent literature reflects a range of 55 to 140 $/kW as being typical of site-specific retrofit SCR
capital costs for all types of utility boilers.8 By comparison, the costing methodology estimates a
capital cost range of approximately 50 to 110 $/kW to achieve 85 to 95 percent NOX removal
efficiency.
The NESCAUM/MARAMA report (in Appendix D) estimated a combined fixed and variable O&M
cost of reducing NOX by 70 to 80 percent for a 330 MW dry bottom boiler at approximately $1.1
million/yr and approximately $2.8 million/yr for a similar-sized wet bottom boiler.9 The costing
methodology estimates the combined fixed and variable O&M costs for all boiler types to be about
$1.0 million/yr to $1.7 million/yr for 85 percent NOX removal and initial NOX ranging between 0.45
and 1.5 lb/106Btu. In another economic analysis by Gaikwad and Boward,6 for SCR retrofits on 300
and 500 MW boilers to reduce NOX by 80-85 percent, combined O&M costs were estimated to be
between $1.6 million/yr and $3.2 million/yr.
Based on above comparisons, the capital and O&M cost estimates derived from applying this
methodology are deemed reasonably accurate and fall well within the ranges reported elsewhere.9'16
Actual cost reported from site-specific SCR retrofit applications, as well as individual facility
engineering studies, would more accurately reflect the circumstances of individual facilities. As a
result, it is also reasonable to expect that reported actual costs may not always conform to budgetary
cost estimates that have been designed around typical installations. While the methodology has been
shown to consistently estimate costs that fit well with other reported actual and estimated costs,
occasionally it can be expected that there will be data with higher or lower costs.
10
-------
Chapter 6
Conclusions
This report provides a costing methodology for estimating budgetary costs associated with retrofit
applications of SCR technology on coal-fired boilers. The methodology is applicable to SCR retrofits
on coal-fired boilers ranging in capacity from 100 to approximately 850 MW and with design
efficiencies greater than 80 and up to 95 percent of NOX removal.
The cost equations and variables used in the methodology are based on information obtained from
SCR system suppliers and reflect experience gained from over 200 SCR applications in the U.S. and
abroad. Data fit constants used in the cost equations are also based on industry data.
The capital and O&M cost estimates derived from applying this methodology fall well within the
ranges reported elsewhere. For example, recent literature reflects a range of 55 to 140 $/kW as being
characteristic of retrofit SCR system capital cost for all types of utility boilers. By comparison, the
costing methodology estimates a capital cost range of approximately 50 to 110 $/kW to achieve 85 to
95 percent NOX removal efficiency.
The costing methodology presented in this report can be used to estimate budgetary costs of SCR
application over the spectrum of boiler sizes and types. However, the estimates that this methodology
can provide cannot replace the detailed site-specific engineering cost studies or cost quotations.
Actual cost reported from site-specific SCR retrofit applications, as well as individual facility
engineering studies would more accurately reflect specific circumstances of an individual installation.
In general, the methodology is thought to slightly overestimate cost in most cases.
11
-------
References
1. Srivastava, R., Johnson, A., Geyer, M., and Quarles, P., "Application of Selective Catalytic
Reduction Technology on Coal-fired Electric Utility Boilers," In Proceedings: EPRI-DOE-
EPA Combined Utility Air Pollutant Control Symposium: The Mega Symposium: Opening
Plenary Session and NOX, EPRI, Palo Alto, CA, U.S. Department of Energy, Pittsburgh, PA,
and U.S. Environmental Protection Agency, Air Pollution Prevention and Control Division,
Research Triangle Park, NC, EPRI Report TR-108683-V1, August 1997.
2. Khan, S., Shroff, G., Tarpara, J., and Srivastava, R., "SCR Applications: Addressing Coal
Characteristic Concerns," In Proceedings: EPRI-DOE-EPA Combined Utility Air Pollutant
Control Symposium: The Mega Symposium: Opening Plenary Session andNOx, EPRI, Palo
Alto, CA, U.S. Department of Energy, Pittsburgh, PA, and U.S. Environmental Protection
Agency, Air Pollution Prevention and Control Division, Research Triangle Park, NC, EPRI
Report TR-108683-V1, August 1997.
3. Hartenstein, H., Gutberlet, H., and Licata, A., "Utility Experience with SCR in Germany,"
Presented at the 1999 Conference on Selective Catalytic and Non-Catalytic Reduction for
NOX Control, Pittsburgh, PA, May 20-21, 1999.
4. Buschmann, J. and Larsson, A., "Worldwide Experience With Selective Catalytic
Reduction," Presented at the Institute of Clean Air Companies Forum '98, Durham, NC,
March 18-20, 1998.
5. Cox, J., Healy, E., Maxwell, J. Hinton, W., Stallings, J., and Baldwin, A., "Economic
Analysis of Selective Catalytic Reduction Applied to Coal-Fired Boilers," In Proceedings:
EPRI-DOE-EPA Combined Utility Air Pollutant Control Symposium: The Mega
Symposium: Opening Plenary Session and NOX, EPRI, Palo Alto, CA, U.S. Department of
Energy, Pittsburgh, PA, and U.S. Environmental Protection Agency, Air Pollution Prevention
and Control Division, Research Triangle Park, NC, EPRI Report TR-108683-V1, August
1997.
6. Gaikwad, R. and Boward, W., "Economic Evaluation of Controls for PM 2.s - A Case Study,"
In Proceedings: EPRI-DOE-EPA Combined Utility Air Pollution Control Symposium: The
Mega Symposium: NOX and Multi-Pollutant Controls, EPRI, Palo Alto, CA, U.S.
Department of Energy, Pittsburgh, PA, and U.S. Environmental Protection Agency, Air
Pollution Prevention and Control Division, Research Triangle Park, NC, EPRI Report TR-
113187-V2, August 1999.
7. "Analyzing Electric Power Generation Under the CAAA," Office of Air and Radiation, U.S.
Environmental Protection Agency, Washington, DC, March 1998. Available at the web site
www. epa.gov/capi/ipm/update .htm.
8. Schimmoller, B., "SCR Dominates NOX Compliance Plans," Power Engineering, Vol. 104,
No. 7, July 2000, pp. 45-48.
9. Staudt, J., "Status Report on NOX Control Technologies and Cost Effectiveness for Utility
Boilers," Northeast States for Coordinated Air Use Management (NESCAUM) and Mid-
Atlantic Regional Air Management Association (MARAMA), June 1998.
10. Cochran, J., Scarlett, D., and Johnson, R., "SCR for a 460 MW Coal Fueled Unit: Stanton
Unit 2 Design, Startup, and Operation," In Proceedings: EPRI-DOE-EPA Combined Utility
Air Pollutant Control Symposium: The Mega Symposium: Opening Plenary Session and
12
-------
NOX, EPRI, Palo Alto, CA, U.S. Department of Energy, Pittsburgh, PA, and U.S.
Environmental Protection Agency, Air Pollution Prevention and Control Division, Research
Triangle Park, NC, EPRI Report TR-108683-V1, August 1997.
11. Wagner, P., Bullock, D., Sigling, R., and Johnson, R., "Selective Catalytic Reduction:
Successful Commercial Performance on Two U.S. Coal-Fired Boilers," In Proceedings EPRI-
DOE-EPA Combined Utility Air Pollutant Control Symposium: The Mega Symposium:
Opening Plenary Session and NOX, EPRI, Palo Alto, CA, U.S. Department of Energy,
Pittsburgh, PA, and U.S. Environmental Protection Agency, Air Pollution Prevention and
Control Division, Research Triangle Park, NC, EPRI Report TR-108683-V1, August 1997.
12. Johnson, S., Zammit, K., and Engelmeyer, A., "Improved SCR Control to Reduce Ammonia
Slip," EPRI-DOE-EPA Combined Utility Air Pollution Control Symposium: The Mega
Symposium: NOX and Multi-Pollutant Controls, EPRI, Palo Alto, CA, U.S. Department of
Energy, Pittsburgh, PA, and U.S. Environmental Protection Agency, Air Pollution Prevention
and Control Division, Research Triangle Park, NC, EPRI Report TR-113187-V2, August
1999.
13. Personal Communication with J. Buschmann, ALSTOM Power, July 2000.
14. Personal Communication with M. Cohen, ALSTOM Power, July 2000.
15. Personal Communication with W. DePriest, Sargent & Lundy, June 2000.
16. Personal Communication with R. Gaikwad, Sargent and Lundy, July 2000.
17. Cichanowicz, J. and Broske, D., "An Assessment of European Experience with Selective
Catalytic Reduction in Germany and Denmark," In Proceedings: EPRI-DOE-EPA Combined
Utility Air Pollution Control Symposium: The Mega Symposium: NOX and Multi-Pollutant
Controls, EPRI, Palo Alto, CA, U.S. Department of Energy, Pittsburgh, PA, and U.S.
Environmental Protection Agency, Air Pollution Prevention and Control Division, Research
Triangle Park, NC, EPRI Report TR-113187-V2, August 1999.
18. Urbas, J. and Boyle, J., "Design and Optimization of SNCR/SCR Hybrid on a Group 1 Boiler
in the Ozone Transport Region," Presented at the Institute of Clean Air Companies Forum
'98, Durham, NC, March 18-20, 1998.
19. Electric Power Research Institute, Palo Alto, CA, "Technical Assessment Guide," EPRI TR-
102276s Vol. IRev. 7, 1993.
20. Cichanowicz, J., "SCR for Coal-Fired Boilers: A Survey of Recent Utility Cost Estimates," In
Proceedings: EPRI-DOE-EPA Combined Utility Air Pollution Control Symposium, The
Mega Symposium: Opening Plenary Session and NOX, EPRI, Palo Alto, CA, U.S.
Department of Energy, Pittsburgh, PA, and U.S. Environmental Protection Agency, Air
Pollution Prevention and Control Division, Research Triangle Park, NC, EPRI Report TR-
108683-V1, August 1997.
21. Emmel, T. and Maibodi, M., "Retrofit Costs for SO2 and NOX Control Options at 200 Coal-
fired Plants, Volume 1. Introduction and Methodology," EPA-600/7-90-021a (NTIS PB91-
133322), U.S. Environmental Protection Agency, Air and Energy Engineering Research
Laboratory, Research Triangle Park, NC, 1990.
13
-------
Table 1. A summary of equations used in the costing methodology.
Cost Basis: January 2000 dollars
Capital Cost
Capital Cost Equation:
Plant Capacity, kW
Initial NOX, lb/106 Btu
NOX Reduction Efficiency, %
Capital Cost, $/kW
D =
A
B
C
D
For example, if A=600000 kW, B=0.45 Ib/MBtu, C=85%
D, S/kW = 56.32
75 * (300,000 ((B/1.5)A.05*(C/100)A0.4)/A)A0.35
Fixed O&M Cost
Fixed O&M Cost Equation:
Plant Capacity, kW
Capital Cost, $/kW
Fixed O&M Cost, $/yr
For example, if A = 600000 kW, D = 56.32
E, $/yr = 223,027.20
E =
A
D
E
D * A * 0.0066
Variable O&M Cost
J
Variable O&M Cost Equation:
Plant Capacity, kW
Initial NOx, lb/106 Btu
NOX Reduction Efficiency, %
Capital Cost, $/kW
Annual Capacity Factor
Heat Input to Boiler, 106 Btu/hr
Variable O&M Cost, $/yr
F =
A
B
C
D
G
H
F
G*(225*(0.37*B*H*C/100*8760/2000)H.005H.05 +
0.025*D*A*((B/1.5)A.05*(C/100)A.4) + 1.45 *A)
For example, if A=600000 kW, B=0.45 lb/10 Btu, C=85%, D=$56.32/kW, G=0.65, H=6000 106Btu/hr
F, $/yr = 1,623,993.96
14
-------
Table 2. Calculation of Cost Estimates using the costing methodology.
(Costs are based on January 2000 dollars)
EXAMPLE 1
NOK Reduction Efficiency, %
Plant Capacity, kW
Boiler Heat Input, 10fiBtu/hr
Initial NOK, lb/10sBtu
Capital Cost, S/kW
Fixed O&M, S/yr
Variable Cost, S/yr
85
100,000
1000
0.45
105.44
69,593.06
341,097.81
85
200,000
2000
0.45
82.73
109,203
617,061.77
85
300,000
3000
0.45
71.78
142,132.94
878,513.39
85
400,000
4000
0.45
64.91
171,358.20
1,131,917.58
85
500,000
5000
0.45
60.03
198,105.45
1,379,938.58
85
600,000
6000
0.45
56.32
223,030.48
1,624,001.09
85
700,000
7000
0.45
53.36
246,535.60
1,864,979.01
85
800,000
8000
0.45
50.93
268,889.85
2,103,456.82
EXAMPLE 2
NOK Reduction Efficiency, %
Plant Capacity, kW
Boiler Heat Input, 10sBtu/hr
Initial NO,, lb/106Btu
Capital Cost, #kW
Fixed O&M, S/yr
Variable Cost, S/yr
90
100,000
1000
0.45
106.29
70,152.19
351,464.30
90
200,000
2000
0.45
83.39
110,081
635,753.01
90
300,000
3000
0.45
72.36
143,274.87
905,074.46
90
400,000
4000
0.45
65.43
172,734.^4
1,166,096.22
90
500,000
5000
0.45
60.51
199,697.08
1,421,566.04
90
600,000
6000
0.45
56.77
224,822.37
1,672,953.29
90
700,000
7000
0.45
53.79
248,516.33
1,921,159.25
90
800,000
8000
0.45
271.C
2,166,
51.34
50.18
786.74
EXAMPLE 3
NO* Reduction Efficiency, %
Plant Capacity, kW
Boiler Heat Input, 106Btu/hr
Initial NO.,, Ib/lo'Btu
Capital Cost, S/kW
Variable Cost, S/yr
95
100,000
1000
0.45
107.10
361,711.13
95
200,000
2000
0.45
84.03
654,256.49
95
300,000
3000
0.45
72.91
931,391.16
95
400,000
4000
0.45
65.93
1,199,980.24
95
500,000
5000
0.45
60.97
1,462,852.91
95
600,000
6000
0.45
57.20
1,721,522.04
95
700,000
7000
0.45
54.20
1,976,915.63
95
800,000
8000
0.45
51.73
2,229,654.37
NOK Reduction Efficiency, %
Plant Capacity, kW
Boiler Heat Input, 10fiBtu/hr
Initial NOK, lb/10sBtu
Capital Cost, S/kW
Fixed O&M, S/yr
Variable Cost, S/yr
85
100,000
1000
1.5
107.69
71,074.90
577,118.23
85
200,000
2000
1.5
84.49
111,528
1,083,588.27
85
300,000
3000
1.5
73.31
145,159.38
1,574,317.33
85
400,000
4000
1.5
66.29
175,006.94
2,056,317.65
85
500,000
5000
1.5
61.31
202,323.71
2,532,479.03
85
600,000
6000
1.5
57.52
227,779.48
3,004,346.78
85
700,000
7000
1.5
54.50
251,785.09
3,472,868.80
85
800,000
8000
1.5
52.01
274,615.33
3,938,679.05
15
-------
Appendix
Determination of the Design Margin Accounting for Ammonia Slip
NOX (ppm) = NOX (Ib NO2/106Btu)* 1/46.0l(lb-mole NO2/lb NO2)*9780 (106Btu/dscf)*
22.4 (normal L/g-mole flue gas)*530/430 (standard L/normal L)*
103 (cm3/L)*l/(12*2.54)3 (ft3/cm3)*
453.6 (g-mole flue gas/lb-mole flue gas)*(20.9-3)/(20.9)*106 (Al)
where 9780 (106Btu/dscf) is the F-factor for bituminous coal and the factor (20.9-3)720.9
accounts for the fact that NOX (ppm) is generally measured at 3% oxygen (O2) in flue gas.
Simplifying Eqn. (Al),
NOX (ppm) = 786.88 NOX (Ib NO2/106Btu) (A2)
As seen in Eqn. (1), 1 mole of NH3 is needed for each mole of NOX reduced. Consider an
NH3 slip of x (ppm). Then a mass balance of NH3 yields:
(NH3), (ppm) = (N0x)r (ppm) + x (ppm) (A3)
where (NH3); (ppm) relates to the amount of NH3 injected into flue gas to provide a NOX
reduction of (NOx)r (ppm) and resulting in a slip of x (ppm).
Rearranging Eqn. (A3),
(NH3)i (ppm)/(NOx)r (ppm) = 1 + x (ppm)/(NOx)r (ppm) (A4)
Substituing Eqn. (A2) in (A4),
(NH3)i (ppm)/(NOx)r (ppm) = 1 + x (ppm)/[786.88*(NOx)r (lb/106 Btu)] (A5)
(NOx)r (lb/106 Btu) = B (lb/106 Btu)*C/100 (A6)
where
B = NOX (lb/106 Btu) at the inlet of SCR reactor
C = NOx removal efficiency (percent)
Substituting Eqn. (A6) in (A5),
(NH3)i (ppm)/(NOx)r (ppm) = 1 + x (ppm)/[786.88*B (lb/106 Btu)*C/100] (A7)
16
-------
In general, B can range from 0.3 (lb/106Btu) for a coal-fired boiler equipped with state-of-
the-art combustion controls to 2.5 (lb/106Btu) for an uncontrolled cyclone-fired boiler.
Assume x = 2 ppm and C = 85 percent. Then using these values in Eqn. (A7),
1.001 < (NH3)i (ppm)/(NOx)r (ppm) < 1.009
Thus, on average, a design margin of 1.005 would account for NHa slip.
17
------- |