CUECost WORKBOOK DEVELOPMENT
DOCUMENTATION
Version 5.0
William H. Yelverton
U.S. Environmental Protection Agency
Office of Research and Development
Air Pollution Prevention and Control Division
Research Triangle Park, NC 27711
September 2009
EPA/600/R-09/131
-------
COAL UTILITY ENVIRONMENTAL COST
(CUECost) WORKBOOK DEVELOPMENT
DOCUMENTATION
Version 5.0
William H. Yelverton
U.S. Environmental Protection Agency
Office of Research and Development
Air Pollution Prevention and Control Division
Research Triangle Park, NC 27711
Prepared by:
ARCADIS
4915 Prospectus Drive, Suite F
Durham, NC 27713
ARCADIS EPA Contract No. EP-C-04-023
EPA/600/R-09/131
September 2009
-------
Abstract
This document serves as a user's manual for the Coal Utility Environmental Cost (CUECost)
workbook and documents its development and the validity of methods used to estimate
installed capital and annualized costs. The CUECost workbook produces rough-order-of-
magnitude (ROM) cost estimates (+/-30% accuracy) of the installed capital and annualized
operating costs for air pollution control (ARC) systems installed on coal-fired power plants to
control emissions of sulfur dioxide (SO2), nitrogen oxides (NOX), particulate matter (PM),
mercury (Hg), and carbon dioxide (CO2). In general, system performance is an input
requirement for the workbook user. The workbook was designed to calculate estimates of an
integrated ARC system or individual component costs for various ARC technologies used in
the utility industry. Twelve technologies are currently in the workbook: flue gas
desulfurization (FGD)—limestone with forced oxidation (LSFO) and with dibasic acid and
lime spray drying (LSD); particulate matter removal—electrostatic precipitator (ESP) and
fabric filter (FF); NOX control—selective catalytic reduction (SCR), selective non-catalytic
reduction (SNCR), natural gas reburning, and low-NOx burner (LNB); mercury control-
powdered activated carbon (PAC) injection; and CO2 control— monoethanolamine (MEA)
process, chilled ammonia process (CAP) and sorbent injection (SI). It is expected that this
manual will be useful to a broad audience, including: (1) individuals responsible for
developing and implementing SO2, NOX, PM, Hg, and CO2 control strategies at sources, (2)
state authorities implementing pollution control programs, and (3) the interested public at
large. Moreover, persons engaged in research and development efforts aimed at improving
cost-effectiveness of air pollution control technology applicable to coal-fired plants may also
benefit from this manual.
Note:
The original model was delivered by Raytheon Engineers & Constructors, Inc. for Eastern
Research Group, Inc. under EPA Contract No. 68-D7-0001. Subsequent revision was
completed by Andover Technology Partners for ARCADIS under EPA Contract No. EP-C-04-
023. This version 5.0 was revised and accompanying documentation prepared by ARCADIS
under EPA Contract No. EP-C-04-023.
-------
Table of Contents
Table of Contents
Section Page
Abstract i
Acronyms vi i
Introduction and Summary 1
Overview 1
Background 1
Workbook Description 2
CUECost WORKBOOK development and documentation document Contents 4
Project Approach 5
NOX Control Estimate 5
Particulate Matter Control Estimates 6
SO2 Control Estimates 7
CO2 Control Estimates 7
Mercury Control Estimates 7
Default Plant Criteria 8
Results 8
Getting Started 10
Hardware and Software requirements / Internet Access 10
Getting Started 10
Workbook Layout and Methodology 12
Workbook Layout 12
Methodology 14
Input and Output Options 20
Input Data 20
Output Options 21
Worksheet Validation 22
FGD Worksheets - LSFO and LSD Technologies 22
Particulate Matter Control Worksheet 23
NOX Control Worksheet 25
Mercury Control Worksheet 27
Carbon Dioxide Control Worksheet 28
Validation Summary 29
References 30
Appendix A Terminology Definitions, Abbreviations, Acronyms, and Range Names 33
A. 1 Definition of Terms 33
-------
Table of Contents
Appendix B Technology Descriptions/Criteria 36
B.I Limestone Forced Oxidation Design Criteria 36
B. 2 Lime Spray Dryer Design Criteria 39
B.3 Particulate Matter Control Design Criteria 41
B.4 NOX Control Technology Criteria 42
B.4.1 Selective Catalytic Reduction Design Criteria 43
B.4.2 Selective Non-catalytic Reduction Design Criteria 45
B.4.3 Natural Gas Reburning Design Criteria 47
B.4.4 Low-NOx Burner Technology Design Criteria 49
B.5 Hg Control Technology Criteria 50
B.5.1 Mercury Removal Models 50
B.5.2 Mercury Removal by Existing Equipment, fasting equipment 51
B.5.3 Mercury Reduction by PAC injection, fPAc injection 54
B.5.4 PAC Injection Models Developed from Full-Scale Data 56
B.5.5 Mercury Speciation with SCR 62
B.5.6 Conclusions 69
B.6 CO2 Control Design Criteria 69
References 71
Appendix C Design/Economic Criteria 73
C.I General Plant Design Criteria 73
C.2 Economic Criteria 76
Appendix D Cost Algorithm Development/Validation/Sources 77
D.I FGD Cost Algorithm Development 77
D.2 Selective Catalytic Reduction 78
D.2.1 Performance Parameters 78
D.2.2 Capital Costs 79
D.2.3 Operating and Maintenance Costs 81
D.2.4 CUECost Validation 83
D.3 Selective Noncatalytic Reduction 86
D.3.1 Performance Parameters 86
D.3.2 Capital Costs 87
D.3.3 Operating and Maintenance Costs 89
D.3.4 CUECost Validation 91
D.4 Natural Gas Reburning 91
D.4.1 Performance Parameters 91
D.4.2 Capital Costs 94
D.4.3 Operating and Maintenance Costs 95
D.4.4 CUECost Validation 96
D.5 Low-NOx Burner Technology 98
D.5.1 Capital Costs 98
D.5.2 Operating and Maintenance Costs 99
D.5.3 CUECost Validation 100
-------
Table of Contents
D.6 Hg Control Technology 100
D.7 CO2 MEA Control System Cost Algorithm Development 104
D.7.1 Capital Cost 104
D.7.2 Operating and Maintenance Costs 105
D.8 CO2 Cap Control System Cost Algorithm Development 110
D.8.1 Capital Cost 110
D.9 CO2 SI Control System Cost Algorithm Development 112
D.9.1 Preconditioning 113
D.9.2 Absorber 115
D.9.3 Blower/ID Fan 119
D.9.4 Regenerator 120
References 123
Appendix E INPUT WORKSHEET SCREENS 125
E.I Getting Started 125
E.2 Inputs 126
E.2.1 Economic Inputs 126
E.2.2 Power Generation Technique Choices 127
E.2.3 APC Technology Choices 127
E.2.5 Particulate Control Inputs 129
E.2.6 SO2 Control Inputs 129
E.2.7 Mercury Control Inputs 130
E.2.8 CO2 Control Inputs 131
Appendix F Programs for Economic Parameters 133
IV
-------
Table of Contents
List of Figures
Page
Figure 1. CUECost Workbook Map 3
Figure 2. CUECost Logic Diagram 15
Figure B-l. Salem Harbor Mercury Removal without PAC Injection (Durham et al., 2001) 54
Figure B-2a. Gaston Testing 57
Figure B-2b. Gaston Testing 58
Figure B-3. Deviation of the Gaston PAC Algorithm 59
Figure B-4. PPPP Testing 60
Figure B-5. Deviation from the PPPP PAC Algorithm 60
Figure B-6. Brayton Point Testing 61
Figure B-7. Deviation from the Brayton Point PAC Algorithm 62
Figure B-8. Mercury Oxidation without a Catalyst as a Function of Residence Time, Gas
Temperature, and HCI Content (Hocquel et al., 2002) 64
Figure B-9. Mercury Oxidation across SCR Catalysts and without SCR Catalyst (Hocquel et
al., 2002) 64
Figure B-10. Oxidation of Mercury across C-l SCR Catalyst in PRB-derived Flue Gas
(Richardson et al., 2002) 65
Figure B-ll. Effect of Flue Gas Exposure Time on C-l SCR Catalyst Oxidation of Elemental
Mercury: 700 °F and Space Velocity of 1,450 h"1 (Richardson et al., 2002) 65
Figure D-l. PAC, Bituminous FF 101
Figure D-2. PAC, Bituminous ESP 101
Figure D-3. PAC, Subbituminous FF 102
Figure D-4. PAC, Subbituminous ESP 102
Figure D-5. BPAC 102
Figure D-6. Cost of Mercury Reduction, LS Bituminous Coal and ESP 103
List of Tables Page
Table 1. British to Metric Conversion Factors 9
Table 2. Total Capital Requirement Calculation Method 17
Table 3. Annualized Cost Calculation Method 19
Table 4. CUECost-FGD Cost Comparison to FGDCOST by EPRI for Phase 1 Acid Rain
Installations 23
Table 5. Comparison of CUECost ESP Sizing Estimates with Raytheon Model 25
Table 6. Percent Difference between CUECost and Acid Rain Division Studies (Khan and
Srivastava, 2004) for Retrofit Cases 27
Table 7. Estimated Costs of ACI Control Systems according to CUECost 28
Table 8. Estimated Costs of CO2 Control Technologies with CUECost and IECM model 28
Table B-l. Specific Design Criteria for LSFO 38
Table B-2. Input for LSD and the Default Values for the Inputs 41
Table B-3. Inputs for Particulate Matter Control and Its Default Values 42
Table B-4. Default Input Parameters for SCR 45
-------
Table of Contents
Table B-5. Default Input Parameters for SNCR 47
Table B-6. Default Input Parameters for NGR 49
Table B-7. Default Values for LNBT Input Parameters 50
Table B-8. Predicted Collection of Mercury by ESP according to Eqs. B-19 and B-20 52
Table B-9. Values of Constants Used in the PAC Injection from Eqs. B-21 and B-22 56
Table B-10. Coefficients for Curve Fit Algorithms 58
Table B-ll. Summary of Results from Full-Scale SCR Mercury Oxidation Tests (Bustard et
al., 2001) 67
Table B-12. Default Values for Mercury Control Input Parameters 68
Table C-l. Snapshot for a Specific Plant and Its Default Parameters 74
Table C-2. Coal Analysis Library 75
Table C-3. Economic Inputs 76
Table D-l. Variable and Constant Parameters for Wet FGD Cost Algorithm 77
Table D-2. Parameters for LSD Cost Algorithms 78
Table D-3. Direct Capital Costs for Hot-side SCR (Installed equipment costs) 80
Table D-4. Indirect Capital Costs for Hot-side SCR 81
Table D-5. Operating and Maintenance Cost Equations for SCR ($/year) 82
Table D-6. CUECost with Acid Rain Division Study Design for SCR (1990 dollars) 84
Table D-7. Acid Rain Division Study: SCR Applications 85
Table D-8. Direct Capital Costs For SNCR (Installed Equipment Costs) 88
Table D-9. Indirect Capital Costs for SNCR 89
Table D-10. Annual Operating and Maintenance Costs for SNCR 90
Table D-ll. CUECost with Acid Rain Division Study Cases for SNCR (1990 dollars) 92
Table D-12. Acid Rain Division Study: SNCR Applications (1990 dollars) 93
Table D-13. Direct Capital Costs for NGR (Installed equipment cost) 94
Table D-14. Indirect Capital Costs for NGR 95
Table D-15. Annual Operating and Maintenance Costs and Savings for NGR 95
Table D-16. CUECost with Acid Rain Division Study Cases for NGR (1990 dollars) 97
Table D-17. Acid Rain Division Study: NGR Applications (1990 dollars) 98
Table D-18. Total Capital Costs for LNBT Retrofit 99
Table D-19. Annual Operating and Maintenance Costs for LNBT ($/year) 100
Table D-20. CUECost with Acid Rain Division Study Cases for LNBT (1990 dollars) 100
Table D-21. Constants for Eqs. D-9 and D-10 101
Table D-22. Indirect Capital Costs for CO2 Control 105
VI
-------
Acronyms, Units, and Symbols
ACRONYMS
A/C
ACI
AFDC
ARC
APPCD
ARD
BPAC
CAP
CCF
CE
GEMS
CEPCI
CF
CFR
CMU
COHPAC
CUECost
CV
DBA
DC
DCC
DOE
EPA
EPAC
EPRI
ESP
ESPc
ESPh
FF
FGD
GDP
GHG
HHV
IAPCS
Air to Cloth Ratio
Activated Carbon Injection
Allowance for Funds Used During Construction
Air Pollution Control (equipment)
Air Pollution Prevention and Control Division
Acid Rain Division (of EPA)
Brominated Powdered Activated Carbon
Chilled Ammonia Process
Carrying Charge Factor
Chemical Engineering (Magazine)
Continuous Emissions Monitoring System
Chemical Engineering Plant Cost Index
Capacity Factor
Code of Federal Regulations
Carnegie Mellon University
Compact Hybrid Particle Collector
Coal Utility Environmental Cost (model)
Catalyst Volume
Dibasic Acid
Direct Capital
Direct Contact Cooler
Department of Energy
Environmental Protection Agency
Enhanced Powdered Activated Carbon
Electric Power Research Institute
Electrostatic Precipitator
Electrostatic Precipitator (cold)
Electrostatic Precipitator (hot)
Fabric Filter
Flue Gas Desulfurization
Gross Domestic Product
Greenhouse Gas
Higher Heating Value
Integrated Air Pollution Control System
VII
-------
Acronyms, Units, and Symbols
ICR
ID
IECM
IGCC
IPM
k
L/G
LNB
LNBT
LNCFS
LOI
LR
LS
LSD
LSFO
MEA
MEL
MHI
NEMS
NETL
NG
NGR
NM
NRMRL
O&M
OFA
ORD
PAC
PC
PCI
PJFF
PM
PPPP
PR
PRB
Information Collection Request
Induced Draft (fan)
Integrated Environmental Control Model
Integrated Gasification Combustion Combination
Integrated Planning Model
Exponential constant used to related coal type, removal efficiency and ash resistivity to
ESP size in terms of the Specific Collection Area (SCA)
Liquid-to-Gas Ratio
Low NOX Burner
Low NOX Burner Technology
Low-NOX Concentric Firing Systems
Loss On Ignition
Learning Rate
Low Sulfur (bituminous coal)
Lime Spray Drying (flue gas desulfurization)
Limestone (flue gas desulfurization)with Forced Oxidation
Monoethanolamine
Magnesium Enhanced Lime
Mitsubishi Heavy Industry
National Energy Modeling System
National Energy Technology Laboratory (DOE)
Natural Gas Combined Cycle
Natural Gas Reburning
Not Measured
U.S. EPA National Risk Management Research Laboratory
Operation and Maintenance
Over-fire air (used to complete coal combustion in some LNBT applications)
U.S. EPA Office of Research and Development
Powdered Activated Carbon
Pulverized Coal
Plant Cost Index
Pulse-Jet Fabric Filter
Particulate Matter
Pleasant Prairie Power Plant
Progress Ratio
Powder River Basin (coal)
VIM
-------
Acronyms, Units, and Symbols
PV
RAM
RF
RLCS
ROM
SCA
SCPC
SCR
SDA
SI
SNCR
SV
TAG
TCE
TCR
TEG
TOL
TP
TPC
TPI
TTN
TVA
WACC
Present Value
Random Access Memory
Retrofit Factor
Rubber-Lined Carbon Steel
Rough Order of Magnitude
Specific Collection Area (refers to ESP size in terms of plate area (ft2)/1000 acfm)
Supercritical Pulverized Coal
Selective Catalytic Reduction
Spray Dryer Absorber
Sorbent Injection
Selective Non-catalytic Reduction
Space Velocity
Technical Assessment Guide
Total Cash Expended
Total Capital Requirement
Triethylene Glycol (dehydrator)
Technological Optimism Learning
Tax Paid
Total Plant Cost
Total Plant Investment
Technology Transfer Network
Tennessee Valley Authority
Weighted Average Cost of Capital
IX
-------
Acronyms, Units, and Symbols
Units
acfm
gpm
GW
h
hp
kW
kWh
MB
MMacf
MMBtu
MPa
MW
MWe
MWh
ppm
scfm
Actual Cubic Feet per Minute
Gallons Per Minute
Gigawatt
Hour
Horsepower
Kilowatt
Kilowatt Hour
Megabyte
Millions of Actual Cubic Feet
Millions of British Thermal Units
Megapascal
Megawatt
Megawatt (electric)
Megawatt Hour
parts per million
Standard Cubic Feet Per Minute
See Table 1 in the Introduction for units not listed here
Chemical Symbols
CO
C02
H2O
HCI
HCCV
N2O
NH3
NH4+
NH4HSO4
NO
NO2
NOX
S02
S03
SOX
Carbon Monoxide
Carbon Dioxide
Water
Hydrogen Chloride
Bicarbonate
Nitrous Oxide
Ammonia
Ammonium
Ammonium Bisulfate
Nitrogen Oxide
Nitrogen Dioxide
Nitrogen Oxides
Sulfur Dioxide
Sulfur Trioxide
Sulfur Oxides
-------
Getting Started / Installation Guidelines
OVERVIEW
This document serves as a User's Manual for the CUECost workbook and documents its
development and the validity of the methods used to estimate installed capital and
annualized costs. The CUECost economic analysis workbook produces rough-order-of-
magnitude (ROM) cost estimates (±30% accuracy) of the installed capital and annualized
operating costs for air pollution control (ARC) systems installed on coal-fired power plants.
Costs for utility ARC systems are site-specific. These costs are subject to change with
changes in technology, labor rates, and material costs. The costs estimated by the CUECost
workbook come from a variety of sources. With that understanding, one may assume, but it
is not guaranteed, that CUECost will produce estimates in the range of accuracy of ±30% of
the actual cost, which was the goal of this project.
The CUECost workbook was developed in Microsoft Excel workbook format to provide users
with complete insight into the equipment cost estimating methodology. All assumptions are
readily accessible to the user by reviewing the specific equations and references for each
cell in the worksheets. CUECost is composed of technology-specific worksheets with one
common input worksheet for all technologies. This structure allows the workbook to be
expanded to incorporate other technologies in the future.
The original model (1998) was developed by Raytheon Engineers & Constructors, Inc. for
Eastern Research Group, Inc. under EPA Contract No. 68-D7-0001. Subsequent revision was
completed by Andover Technology Partners for ARCADIS under EPA Contract No. EP-C-04-
023. This version 5.0 was revised by ARCADIS under EPA Contract No. EP-C-04-023.
Background
The Air Pollution Prevention and Control Division (APPCD) of the National Risk Management
Research Laboratory (NRMRL) contracted for development of a cost estimating workbook for
APC systems on coal-fired power plants. This workbook was developed in Excel format to
-------
Getting Started / Installation Guidelines
provide the user with more flexibility in modifying the worksheet and outputs to meet the
user's needs for site-specific applications.
The workbook was designed to calculate estimates of an integrated ARC system or individual
component costs for various ARC technologies currently used in the utility industry to reduce
emissions of sulfur dioxide (SO2), particulate matter (PM), nitrogen oxides (NOX), mercury
(Hg) and (in the future) carbon dioxide (CO2) generated by coal-fired boilers. Technologies
currently included in the workbook are:
Flue Gas Desulfurization (FGD)
Particulate Matter Removal
Nitrogen Oxide Control
Mercury Control
Carbon Dioxide Control
Limestone with Forced Oxidation (LSFO)
Lime Spray Drying (LSD)
Electrostatic Precipitator (ESP)
Fabric Filter (FF)
Selective Catalytic Reduction (SCR)
Selective Non-Catalytic Reduction (SNCR)
Natural Gas Reburning (NGR)
Low NOX Burners (LNB)
Powdered Activated Carbon (PAC) injection
Monoethanoamine (MEA) Process
Chilled Ammonia Process (CAP)
Sorbent Injection (SI)
WORKBOOK DESCRIPTION
A map of the CUECost workbook is shown in Figure 1. This design allows the addition of
future technologies by inserting new worksheets into the workbook. The workbook
calculates both new and retrofit plant costs using a 1.0 factor for a new facility, a 1.3 factor
for a moderately difficult retrofit, and a 1.6 factor for a difficult retrofit. The user is also
given the option to input his own retrofit factor based on plant-specific information.
Equipment sizing and variable operating costs are derived based on the calculated material
balances for specific process criteria, including flue gas flow rate, pollutant removal rate,
chemical consumption rate, waste production rate, etc.
-------
Getting Started / Installation Guidelines
Sheet 1 .0 General
Input-General
•Economic Factor
nput
•General Plant Technical Input
•APC Technology Choices
•NOx Control Inputs
•Particulate Control
•SO2 Control Inputs
•Hg Control Inputs
•CO2 Control Input
nputs
Sheet 4.0 Power Generation
Output-Power Generation
•Sizing
•Engineering Calculations
•Equations
•Levelization
•Normalization
Output-SO2 Control
•Sizing
•Engineering Calculations
•Equations
•Levelization
•Normalization
Sheet 10.0 Levelization Cal.
•Carrying Charges
•Expenses (O&M)
Sheet 2.0 Input Summary
Input-Calculations
•Economic Factor
•General Plant Technical Input
•ARC Technology Choices
•NOx Control Inputs
•Particulate Control Inputs
•SO2 Control Inputs
•Hg Control Inputs
•CO2 Control Input
Sheet 3.0 Output Summary
Summary of Emiss. Gene.
Costs
•SO2 Control Costs
•NOx Control Costs
•PM Control Costs
•Hg Control Costs
•CO2 Control Costs
Total Air Pollution Control Costs
Sheet 5.0 NOx Control
Output-NOx Control
•Sizing
•Engineering Calculations
•Equations
•Levelization
•Normalization
Sheet 8.0 Hg Control
Output-Hg Control
•Sizing
•Engineering Calculations
•Equations
•Levelization
•Normalization
Sheet 1 1 .0 Constant _CC
•Coal Library
•Combustions
•Etc.
Sheet 6.0 PM Control
Output-PM Control
•Sizing
•Engineering Calculations
•Equations
•Levelization
•Normalization
Sheet 9.0 CO2 Control
Output-CO2 Control
•Sizing
•Engineering Calculations
•Equations
•Levelization
•Normalization
Sheet 1 2.0 Future Cost Projections
•Learning Curves
•Etc.
Figure 1.
CUECost Workbook Map
-------
Getting Started / Installation Guidelines
The first sheet of the workbook functions as the menu of all sheets in the workbook. Users
can follow the link, by clicking the icons, to the input or output for a specific air pollutant
control technology. In this version 5.0, a toolbar was developed, including Main Menu, Go
to Top, User Input, Outputs, and Print Buttons.
All inputs are integrated into one worksheet, and outputs for a specific control technology
are listed separately in one worksheet. Economic-related outputs are first listed at the top of
the outputs worksheet, with engineering-related calculations listed at the bottom. Version
5.0 of the CUECost workbook contains calculations of the carrying charges and levelizing
factors for expenses in worksheet 10.0. In calculating the capital carrying charges and
operation and maintenance (O&M) levelized cost, a 30-year plant duration was used. The
calculation can be accessed from the worksheet "1.0 General Input" by clicking the
calculator link.
CUECOST WORKBOOK DEVELOPMENT AND DOCUMENTATION DOCUMENT
CONTENTS
This document consists of the following sections:
Overview of CUECost Workbook states the purpose and content of this document.
Getting Started presents an itemized listing of requirements for the user's computer
system and is followed by a series of installation guidelines for use in installing the CUECost
workbook to the user's hard disk. Instruction is also provided for the first-time user on how
to get started producing a cost estimate using CUECost. These starting instructions include
listings of the input sequence and other preliminary steps for the user to complete prior to
using the CUECost workbook.
Workbook Layout and Methodology presents a detailed description of the contents of
each worksheet and provides a layout diagram. This section provides a technical description
of the workbook and discusses how the worksheets are integrated to minimize user input.
The cost estimating methodology is also described, including a logic diagram to illustrate the
calculation sequence that is used to develop capital and annualized cost estimates.
Input and Output Options provides a description of the input and output options available
to the user for cost estimate development.
Worksheet Validation, the final section of the user's manual, summarizes the validation
procedure that was followed during development and subsequent testing of the CUECost
workbook.
Appendix A provides the definitions of terminology used in the text and worksheets.
-------
Getting Started / Installation Guidelines
Appendix B provides process criteria and technology descriptions of equipment included in
each technology cost estimate.
Appendix C presents tabulations of the primary assumptions that served as the estimate
basis for the default values included in the worksheets, including both plant design and
economic criteria.
Appendix D discusses the data sources for the cost-versus-capacity algorithms. Previous
publications, vendor quotations, and costs from recent ARC installations served as the basis
for all cost-versus-capacity curves used in the worksheets.
Appendix E provides a demonstration of the worksheets to show how the workbook is
used. Pictures of the actual Excel screens are provided for easy reference to the screens
shown when running CUECost.
Appendix F provides programs to calculate carrying charges and levelization of O&M costs.
PROJECT APPROACH
The workbook design allows the user to review all of the assumptions and equations
contained in each worksheet and to adjust any of them to fit the user's particular needs. A
multi-worksheet format was selected to allow the addition of other technologies if future
expansion of the workbook is desired. A separate input worksheet was assembled, along
with technology-specific Excel worksheets that perform equipment sizing and economic
calculations for each ARC system.
NOX Control Estimate
NOX control technology design and cost algorithms are based on research conducted for the
EPA Acid Rain Division (ARD) (now the Clean Air Markets Division), the EPA Office of
Research and Development (ORD), and the U.S. Department of Energy (DOE) National
Energy Technology Laboratory (NETL) (Frey and Rubin, 1994). Design parameter
calculations for SCR, SNCR, and NGR are taken from the Integrated Air Pollution Control
System (IAPCS) model, Version 5.0 (Gundappa et al., 1995).
SCR capital cost components are based on algorithms developed for DOE (Frey and Rubin,
1994) as part of the Integrated Environmental Control Model (IECM).1 For SNCR and NGR
total capital equipment costs, ARD research was used to update IAPCS methodology. The
ARD cost data used to update IAPCS are presented in the following:
1 IECM is a computer-modeling program that performs a systematic cost and performance analysis of emission
control equipment at coal-fired power plants. It is developed for the U.S. Department of Energy by Carnegie Mellon
University and is available at http://www.iecm-online.com (accessed February 13, 2009).
-------
Getting Started / Installation Guidelines
• "Cost Estimates for Selected Applications of NOX Control Technologies on Stationary
Combustion Boilers and Responses to Comments," (EPA, 1998) and
• "Investigation of Performance and Cost of NOX Controls as Applied to Group 2 Boilers,"
(EPA, 1997).
Low NOX burner technology (LNBT) total plant costs are based on algorithms presented in
another ARD report (EPA, 1996). The cost estimates presented in the ARD reports are being
used in the NOx-related rulemaking and have been reviewed by stakeholders associated
with the rulemaking process. O&M cost algorithms for all technologies use IAPCS equations
from IAPCS 5.0 (Gundappa et al., 1995). Operating costs are estimated in the workbook
based on simplified material balances calculated within CUECost based on the inputs
supplied by the user. The ultimate coal analysis, including weight percent sulfur, carbon,
hydrogen, oxygen, nitrogen, moisture and ash, serves as the primary input for the
combustion calculations performed by the worksheet. The resulting gas flow is the basis for
the remaining material balance calculations.
In this manual, the default values for NOX control devices were generally adopted from
Integrated Planning Model (IPM)/IECM models (the IPM model can be found at
http://www.epa.gov/airmarkt/progsregs/epa-ipm/index.html). For NOX control technologies,
CUECost results were compared to cost data reported by the EPA ARD (EPA, 1997; EPA,
1998) for NOX controls applied to utility boilers and Chapter 5 of the IPM manual. The ARD
reports are based on an EPA national database of boilers. Using CUECost's default values for
Retrofit Factor, General Facilities, etc., should produce capital cost results that are the same
as or very close to the results that would be produced by the IPM source algorithms.
Particulate Matter Control Estimates
The particulate matter control technology cost estimates are based on IECM model
constructed by Carnegie Mellon University (CMU) for the DOE (Berkenpas et al., 1999). This
model was constructed based on a combination of theoretical equations for Electrostatic
Precipitator (ESP) sizing. The theoretical equations were modified to incorporate the
empirical data obtained from a series of ESP vendors for installations firing different coals.
This framework taken from the CMU model served as the basis for the CUECost ESP design
portion of the worksheet. The CMU worksheet was based on 150 to 200 actual installations
firing a wider variety of fuels. Operating costs are calculated using the inlet-flow rate-
versus-expected-power-consumption algorithms. Maintenance costs are calculated as a
percentage of the installed equipment cost.
Fabric filter (FF) costs were also based on a set of cost equations developed by Berkenpas
et al. (1999) to relate the FF size [calculated as a function of the volumetric flue gas flow
rate times the air-to-cloth ratio (A/C) selected by the user] and the FF inlet flue gas
volumetric flow rate to determine the expected cost for the installed system (Berkenpas et
al., 1999). Operating costs are calculated using the inlet-flow rate-versus-expected-power-
-------
Getting Started / Installation Guidelines
consumption algorithms. Maintenance costs are calculated as a percentage of the installed
equipment cost.
SO2 Control Estimates
For the FGD technologies, cost-versus-capacity equations were based on the historical
database (Keeth, 1991) of actual equipment costs incurred during Phase 1 of the utility
Clean Air Act compliance programs, budgetary quotations for components as received from
vendors during early 1998, and cost data obtained from industry database programs
(Srivastava, 2000). These parametric equations serve as the basis for the FGD system
capital costs calculated by CUECost. Operating cost equations were formulated based on the
consumption rates estimated in the worksheets by the material balance calculations. A
material balance is developed specifically for each FGD system and provides the chemical
consumption rates, wastes production rates, and flow rates through process equipment that
are used to estimate the system power consumption. Operating labor requirements are
based on a formula that relates plant size to the number of operating staff needed to run
the FGD equipment, and maintenance costs are calculated as a percentage of the installed
costs for the system.
CO2 Control Estimates
The monoethanolamine (MEA) CO2 control technology cost estimates are based mainly on a
report from the U. S. DOE/NETL (2007). The cost of a bare erected plant was estimated
based on 30% of MEA, currently the most practical concentration. For MEA islands, the
same total plant cost (TPC) is assumed to occur for KS-1 and MEA solvents. The compressor
island cost, however, depends upon the compressor stages and power consumption through
a regression of cost and compressor power. In the CUECost design, the details of O&M cost,
which are the major concern, are listed for the users. The absorption island of chilled
ammonia process (CAP) is estimated to be 97% of the bare erected cost of the same size
MEA-type island. Although the pressure of CO2 out of the reflux drum is significantly higher
than from the MEA process, the investment estimation of the compressor island follows the
same algorithm as the MEA process, depending upon only the regressed relation of cost-
power. Regarding the sorbent injection (SI) option, due to the lack of information for a full
scale plant, the bare erected cost for absorption island is estimated with a same size of MEA
island total cost. Detailed values for consumption of power, steam and cooling water are
given in the engineering calculation section.
Mercury Control Estimates
The effects of existing equipment on mercury reduction were isolated from the effects of
powdered activated carbon (PAC) injection on mercury reduction, and new algorithms were
developed for PAC injection. The PAC injection algorithms include algorithms based on the
results of two full-scale demonstrations (Bustard et al., 2001; Durham et al., 2001;
Bustard et al., 2002; Sterns, 2002) as well as algorithms developed from pilot-scale data
(EPA, 2000).
-------
Getting Started / Installation Guidelines
Economic criteria supplied by the user are used by CUECost to calculate the capital and
annualized costs for the selected ARC system. The user has the option to use the default
values provided in the worksheet if some of the input data requested are not readily
available. For the convenience of users, a levelization calculation worksheet is included in
the CUECost workbook. This worksheet provides carrying charge and levelization factor for
expenses (O&M) in terms of current dollars and constant dollars.
DEFAULT PLANT CRITERIA
The CUECost workbook includes default values for all input parameters. These criteria are
specific to a generic 500 MW coal-fired power plant located in Pennsylvania. The specific
design and economic criteria used as defaults are provided in Appendix C for reference. A
coal library is also included in worksheet 11.0 so that the user can select a coal similar to
that actually burned at the plant if an actual ultimate analysis is not readily available. User
has the capability to adjust coal properties as desired to create "user-defined" coal in
worksheet 11.0. The coal information was retrieved from the DOE Coal Sample Bank and
Database at http://datamine.ei.psu.edu/index.php.
RESULTS
The CUECost workbook provides rough-order-of-magnitude (ROM) cost estimates (±30%
accuracy) for a wide variety of ARC technology scenarios. Cost estimates for different
combinations of control technologies can easily be compared in the results summaries
presented in five parallel columns on the worksheets. Examples of the input sheets are
shown in Appendix E.
CUECost is designed to produce ROM estimates for a wide range of plant sizes and coal
types. However, appropriate ranges of plant size and operating conditions have been
established based on the limits to the database used to construct the cost-versus-capacity
algorithms. Range limits are provided in the worksheet for each input supplied by the user.
The major criterion limitation for CUECost is the plant size range. Algorithms are based on
the assumption that equipment options will be installed at a facility ranging from 100 to
2000 MW in net capacity. All other criteria are limited only by their technical validity. The
suggested technical limits for each criterion are provided in the worksheets when applicable.
It is expected that this document will be useful to a broad audience, including: (1)
individuals responsible for developing and implementing SO2, NOX, PM, Hg, and CO2 control
strategies at sources, (2) state authorities implementing pollution control programs, and (3)
interested public at large. Moreover, persons engaged in research and development efforts
aimed at improving cost-effectiveness of air pollution control technology applicable to coal-
fired plants may also benefit from this document.
-------
Getting Started / Installation Guidelines
Note that the cost estimates provided in this study and generated by CUECost are
dependent upon the various underlying assumptions, inclusions, and exclusions utilized in
developing them. Actual project costs will differ and can be significantly affected by factors
such as changes in the external environment, the manner in which the project is
implemented, and other factors which impact the estimate basis or otherwise affect the
project. Estimate accuracy ranges are only projections based upon cost estimating methods
and are not guarantees of actual project costs.
EPA policy is to express all measurements in EPA documents in metric units. Values in this
document are given in British units for the convenience of the engineers and other technical
staff accustomed to using the British system. The following conversion factors presented in
Table 1 can be used to provide metric equivalents.
Table 1.
Abbr.
ac
Btu
°F
ft
ft2
ft3
ft/m
ft3/m
gal
gpm
gr
gr/ft3
hp
in.
Ib
Ib/ft3
Ib/h
mi
psi
rpm
scfm
t/h
British to Metric
British Unit
acre
British thermal unit
deg. Fahrenheit - 32
feet
square feet
cubic feet
feet per minute
cubic feet per minute
gallons (U.S.)
gallons per minute
grains
grains per cubic foot
horsepower
inches
pounds
pounds per cubic foot
pounds per hour
miles
pounds per square inch
revolutions per minute
standard (60 °F) cubic
feet/minute
short tons per hour
Conversion Factors
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Conv. Factor
0.405
0.252
0.5556
0.3048
0.0929
0.02832
0.00508
0.000472
3.785
0.06308
0.0648
2.288
0.746
0.0254
0.4536
16.02
0.126
1609
6895
0.1047
1.6077
0.252
Abbr.
ha
kcal
°C
m
m2
m3
m/s
m3/s
L
L/s
g
g/m3
kW
m
kg
kg/m3
g/s
m
Pa
rad/s
nm3/h
kg/s
Metric Unit
hectare
kilocalories
degrees Centigrade
meters
square meters
cubic meters
meters per second
cubic meters/second
Liters
liters per second
grams
grams per cubic
meter
kilowatts
meters
kilograms
kilograms/cubic meter
grams per second
meters
Pascals (Newton/m2)
radians per second
normal cubic
meters/h
kilograms per second
-------
Getting Started / Installation Guidelines
HARDWARE AND SOFTWARE REQUIREMENTS / INTERNET ACCESS
The CUECost workbook is written in Microsoft Excel 2003 format. The hardware and
software requirements for CUECost and User's Manual are listed below:
Computer Hardware:
Operating System:
Memory Requirements:
Installation Requirements:
Commercial Support
Software Required:
500 MHz Processor, 256 MB RAM and 100 MB hard drive
Internet access required for down-loading the worksheet from web site
Windows 2000 or higher
3 MB on hard drive for download of CUECost workbook
Download from EPA's Technology Transfer Network (TTN) web site:
http: //www. epa. gov/ttn/catc/products .html
Search for CUECost under heading: Software (Executables & Manuals)
Download to hard drive
Microsoft Excel 5.0 or higher for Workbook
Microsoft Word 6.0 or higher or Adobe Acrobat Reader (latest version
available for free at: http://get.adobe.com/reader/) for User's Manual
GETTING STARTED
After accessing the workbook via the EPA web site and storing the files on the user's hard
drive (note that the files may have to be decompressed using "WinZip"), the User's Manual
may be called up in WordPerfect or Acrobat, depending upon the format downloaded, and
then printed out for easy access. Each user should read the user's manual to become
familiar with how CUECost works and where various input and technical data are provided
within the workbook. After reviewing the user's manual, the user should then call up the
workbook as an Excel file and begin review of the worksheets contained therein. The file will
10
-------
Getting Started / Installation Guidelines
be active when called up from the web site (after decompression). The user should go to the
home site (cell Al) on the first sheet of the workbook to begin.
NOTE: The CUECost workbook can be modified bv the user. To ensure its integrity,
a copy of the original worksheet should be saved in a separate file in a new
directory and all other copies saved under different file names.
The default values provided in the worksheet will allow the user to immediately run a test
case and print output sheets to test the existing printer setup routine. Familiarity with Excel
worksheet software is required to modify the workbook to correct printing problems.
The input requirements for the worksheet are itemized in the section titled "Input and
Output Options" of this user's manual. The user should first obtain the necessary input data
for all cases to be evaluated. Up to ten cases can be run simultaneously for direct on-screen
comparison of results. Up to twelve site-specific coal analyses can be added to the ten
columns available in the coal library for use in any series of estimating runs. This file can
then be saved for use in the future. The existing default values can be deleted by entering
values in the library cells, and then saving the new file for future use under a different file
name. The input cells are colored blue for identification by the user.
When running the workbook for the first time, it would be best to save your input data to a
separate file on a regular basis. The worksheet provides the capability to select from a
variety of system options, picking alternate control technologies and combinations of the
component options provided.
11
-------
Workbook Layout and Methodology
WORKBOOK LAYOUT
Figure 1 provides the basic layout of the various sheets currently included in the CUECost
workbook. The following descriptions apply to the individual worksheets.
Worksheet Menu
Upon opening CUECost, the user can enter the menu by clicking the "Menu" tab. The menu
provides easy links to specific control technologies. The user can also use the CUECost
toolbar. The toolbar offers quick access to many functions while working in a particular
worksheet.
This worksheet menu provides the primary user interface and basic instructions on how to
proceed. The interface consists of a series of buttons the user selects based on the
technology cost estimates desired and the part of the workbook to be reviewed at that time.
Note that when the user selects a specific technology for evaluation, the inputs for
preceding technologies must be fulfilled according to the selected air pollutant control
sequence. For example, if both selective catalytic reduction (SCR) and limestone forced
oxidation (LSFO) technologies are selected; inputs for SCR must be completed prior to the
inputs for LSFO.
Worksheet 1.0 = General Input - This worksheet contains three subsections: Economic
Factors, Power Generation Technology and ARC Technology Choices (only for pulverized coal
application). Through this worksheet, the user constructs the basis for all pollutant control
technology-related estimates. The various columns in this worksheet are described below:
• Column B provides a text description of the cells in each row.
• Column C defines the units that should be used for the input to the cells in each row.
• Column D supplies a suggested range of input values based on technical limits or
worksheet validity limitations.
12
-------
Workbook Layout and Methodology
• Column E is a listing of the default values included in the worksheet.
• Columns F through O provide entry points for values specific for up to ten simultaneous
case evaluations.
Worksheet 2.0 = Input Summary - This sheet summarizes economic input, power
generation technology choices, and air pollutant control technology choices specified in the
general input worksheet.
Worksheet 3.0 = Output Summary - This worksheet summarizes all ARC technology outputs
from a specific technology evaluation. A filter function is provided in this worksheet for the
user to select the preferred outputs easily.
Worksheet 4.0 = Power Generation - This worksheet is specifically designed to evaluate the
investment and O&M costs of a specific type of power generation technology such as
subcritical, supercritical, ultra-supercritical, integrated gasification combined cycle (IGCC),
and Oxyfuel.
Worksheet 5.0 = NQy Control - NOX calculations are completed on this worksheet. The
results of the combustion calculations provided in worksheet 1.0 "Constants_CC" are used
to calculate the material balance for the NOX systems. These values are then used to
calculate the expected costs for the various cost areas using algorithms developed for the
CUECost workbook.
Worksheets 6.0 = PMF, 7.0 = SO?, 8.0 = Hq, 9.0 = COZ Control Technologies - These
worksheets perform the same function as Sheet 5.0 for the other ARC technologies.
Sheet 10.0 = Levelization Calculations - This worksheet is specifically designed for the user
to calculate carrying charges and non-carrying expenses.
Worksheet 11.0 = Constants CC - This worksheet contains range name definitions, tables of
constants used by the workbook (such as the molecular weights of compounds), and other
macros used by the CUECost workbook. This worksheet also contains the coal library and
the combustion calculation sequence used for all of the material balances performed in the
other process-specific worksheets.
In general, the methodology employed in the workbook for cost development follows the
format used by the IAPCS model (Gundappa et al., 1995), providing installed capital and
operating costs for the selected technologies. The calculation sequence takes advantage of
the vertical arrangement of the worksheet. A series of tables presents the equations (and all
variables used in these equations) contained in each cell and the units of the calculated
results. Descriptive material is included in the documentation to define the purpose and
method employed within various subsections of the worksheets.
13
-------
Workbook Layout and Methodology
METHODOLOGY
The calculation sequence used in the worksheets to estimate capital and annualized costs is
summarized in the following material. Additional details regarding the specific equations and
interrelationships between sections of the worksheets can be found in this documentation
provided in Appendix D. The worksheet design will accommodate the addition of alternate
ARC technologies by inserting new worksheets for system cost estimation and technical
calculations that will use the common input sections and common economic calculations.
The cost worksheet allows the user to select the technologies of interest and calculates the
associated costs for each control system based on the data that the user enters to define
site-specific conditions. Figure 2 is the logic diagram for the workbook and illustrates how
the capital and annualized costs for ARC equipment are calculated. The methodology and
the calculation sequences used by CUECost are described below in the following material.
Step 1
This step begins with input worksheets and can be split into two sub-steps.
Sub-step 1
The user is first asked to select the "1.0 General Input" worksheet. This worksheet provides
a general description of the power plant and desired combination of ARC technologies.
Following the initial process selection, the user enters the necessary technical parameters
specific to the project. Default values are provided for all inputs. The inputs are separated
into the following distinct sections:
• Economic Factor
Inflation adjustment factors are used for cost adjustment from algorithm development
years to current cost basis year. User can select either gross domestic product (GDP) price
deflator or chemical engineering cost index (Chem Index) for cost adjustment. GDP price
deflator can be obtained from the Bureau of Economic analysis, and the Chem Index can
be obtained from the journal of Chemical Engineering. Carrying charges (current dollars,
constant dollars, first year constant dollars and first year current dollars.) and levelization
factors for expenses (current and constant dollar based on a 30-year lifecycle of plant)
follow the definition in EPRI TAG Technical Assessment Guide (EPRI TR-102276-V1R7
volume 1, 1993). For the convenience of user, carrying charges and levelization factors for
expenses can be calculated by clicking the calculator link. Once the inputs in worksheet
10.0 are given by the user, the outputs are automatically sent back to worksheet 1.0.
• General Plant Technical Inputs (boiler operation, coal analysis, excess air, etc.)
14
-------
Workbook Layout and Methodology
Technical Inputs:
- Plant Description
- Boiler Operation
- Coal Analyses
- Excess Air
APC Process Inputs:
- Process Selection
- Operating Criteria
- Equipment Sparing
Combustio
- Gas Flow
- Gas Com]
- Chemical
i
APC Mate
- Inlet Gas
- Inlet Gas
- Reagent C
- Waste Ge
i
nCalc.:
Rate
)osition
Usage
r
r. Balance:
Tlow Rate
Compos.
onsump.
deration
r
Oper. Parameters:
- Chemical Usage
- Waste Disposal
- By-product Rate
Economic Inputs:
- Inflation Rate
- Escalation Rates
- Fixed Charge Rate
- Consumable Costs
i
Equipmenl
- Installed 3
- Installed (
Major C
i
Utility Con
- Power
- Steam
- Water
r
Cost:
/Subsys.
^ost of
omponents
r
sumption:
Indirect Cost Inputs:
- Engineering %
- General Facilities
- Contingency
- Retrofit Factor
i
Total Ca
r
pital Cost
Variable Operating
i
L
Fixed Cost
- Maintenar
- Operating
$/hr. & # o
i
Factors:
ice %'s
f operators
r
First-year and
Levelized Operating
Cost Results
1
k
Figure 2.
CUECost Logic Diagram
15
-------
Workbook Layout and Methodology
• For the power generation technology selection, the user can select sub-critical, super-
critical, ultra super-critical, IGCC, or oxyfuel. When IGCC or oxyfuel power generation is
selected, there will be no control technologies related to them in this version.
• ARC Technology Choices(NOx/SO2/PM/Hg/CO2)
This section is of importance because it tells the system what plant configuration is
desired. As aforementioned, there will be no technology selected when IGCC or oxyfuel
power generation technology is selected.
Sub-step 2
After users complete common input, the user then can access a specific control technology
as described in the ARC Technology Section for process specific input, fully exploiting the
convenience of the toolbar attached to this workbook. Default values are provided for all
inputs.
Step 2
After the user has entered the technical inputs, the workbook performs the combustion
calculations in the Constants_CC worksheet. The flue gas flow rate and composition are
calculated in this step. The results of these calculations are summarized in the
Constants_CC worksheet.
Step 3
Using the results of the combustion calculation and the APC-specific technical inputs, the
necessary material balance calculations are performed. Reagent consumption and waste
generation are calculated based on the inlet gas flow and composition (see ARC technology
worksheets).
Step 4
Following the calculation of the material balance, the equipment costs associated with the
specific equipment areas (ARC worksheets) are calculated. The largest equipment
components for each area [absorber, induced draft (ID) fan, etc.] are broken out and
estimated separately. All capital costs are installed costs (i.e., they include all costs
associated with the installation of the subsystem or component). These installation
expenditures include the costs for the following:
• Earthwork
• Concrete
• Structural steel
• Piping
• Electrical
16
-------
Workbook Layout and Methodology
• Instrumentation and controls
• Painting
• Insulation
• Buildings and architectural.
Costs for demolition are treated as an input, assuming that the user can provide the
expected costs for any demolition that might be required at a specific site. The items listed
above, when added to the bare equipment cost, are equivalent to "A" in the calculation
sequence for the capital cost shown in Table 2.
Table 2. Total Capital Requirement Calculation Method
Installed Process Capital Cost = A
General Facilities at % of A = B
Engineering and Home Office Fees at % of A = C
Contingency at % of (A + B + C) = D
Total Plant Cost (TPC) = A+B+C+D
Total Cash Expended (TCE) = TPC x Adjustment Factor*
Allowance for Funds During Construction (AFDC) = AFDC % (input) x TPC
Total Plant Investment (TPI) = TCE + AFDC
Preproduction Costs = F
Inventory Capital = G
Total Capital Requirement (TCR) = TPI + F + G
* Adjustment Factor is based on the years of construction, the inflation rate, and the escalation rate.
The factor reduces the cost of the capital investment due to the purchase of components prior to the
completion of the construction period, allowing the TCR to be expressed in a single-year dollar value.
Step 5
Adding the costs listed above to the uninstalled bare equipment costs results in the total
direct field cost for the installed equipment (ARC worksheets). The installed equipment costs
(bare equipment cost multiplied by an installation factor composed of various cost accounts
listed above—earthwork, steel, piping, etc.) for each component include the typical indirect
field costs, such as field staff and legalities, craft fringes and insurance, temporary facilities,
construction equipment and tools, and an allowance for start-up and testing. Allowances for
taxes are also included in the final installed cost for each subsystem. The Total Installed
Cost then serves as the basis for the calculation of the engineering and general facilities
cost components and the contingency cost associated with the project capital cost.
Escalation of the capital cost is then performed using the GDP Index or CE Index (see
Economic Indicators found on the last page of each issue of the Chemical Engineering
magazine) for the year selected by the user as the basis for the cost estimate.
17
-------
Workbook Layout and Methodology
For most equipment areas and components, a cost algorithm is supplied to relate installed
component cost to the component capacity. The worksheet was constructed to allow the
user to generate cost estimates for units ranging from 100 MW to 1000 MW and for facilities
firing almost any coal.
Step 6
In addition to the equipment costs, the ARC worksheets also calculate operating parameters
(chemical usage, waste disposal, byproduct rate, etc.) after the calculation of the material
balances (ARC worksheets). The usage and production rates serve as the basis for the
calculation of the variable operating cost components. The workbook uses the operating
parameters and the calculated utility consumption (electrical energy, steam, water, etc.) to
calculate the variable operating costs. The annualized cost calculation method is
summarized in Table 3.
Step 7
Finally, the total capital and operating costs are used to calculate the levelized constant
dollars and also first-year annualized current dollars. Operating costs belong to the non-
carrying charge category. Operating costs will be levelized at the 30-year level (L30) at
constant dollars. Both of these costs (for both capital and operating components) are
represented in absolute ($/year) and normalized terms (i.e., mills/kWh or $/kW). These
costs are summarized in the summary worksheet for direct comparison of case cost
estimates and printing of output summaries.
18
-------
Workbook Layout and Methodology
Table 3. Annualized Cost Calculation Method
Fixed O&M Costs
Operating Labor
Maintenance Labor/Materials
Administrative/Support Labor
Labor Rate x 8760 x Number of Operators Added
= Maintenance Factor x Installed Capital Cost
0.3 x (Operating Labor + Maintenance Labor)
A
B
C
Variable Operating Cost
Chemicals
Solids Disposal
Water Cost
Power
Steam
= Chemical Cost x Consumption Rate/Year
Waste Disposal Cost x Waste Production Rate/Year
Water Cost x Water Consumption Rate
= Power Cost x Power Consumption Rate
Steam Cost x Steam Consumption Rate
D
E
F
G
H
Carrying Charges
Carrying Charges
= Total Capital Requirement x Carrying Charge Rate
I*
Annualized Constant Cost
First Year Current Cost
(A+B+C+(D+E+F+G+H)xCapacity Factor)xL(SO) +1 =
(A+B+C)+(D+E+F+G+H)xCapacity Factor +1
*When calculating levelized/annualized constant cost, I =total capital requirement (TCR) x constant $
carrying charge rate.
When calculating first year current cost, I =total capital requirement (TCR) x first year current $
carrying charge rate.
L(30), levelization factor for 30 year service life, can be calculated with worksheet 10.0. Results are
automatically sent back to worksheet 1.0.
19
-------
Input and Output Options .
INPUT DATA
The worksheet "1.0 general inputs" starts with default, inputs, represented with "D", in
worksheet 1.0. The user can change the "D" to its defined value for a site-specific estimate.
Input cells are colored blue for highlighting purposes. Each column is specific to an
individual case. Duplicate data for each case can simply be copied over into the remaining
columns rather than entered individually for each case.
The general input worksheet is divided into various sections for clarity. The detailed input
requirements are listed below with a brief description of the content of each:
1. Economic Factors - These economic data define the basis for the cost estimates that are
produced, including the basis year, inflation rates, escalation rates, capital carrying
charges, non-carrying expense (O&M), operating labor rates, chemical costs, and utility
costs. The economic factors apply to all the control technologies.
2. Power Generation Technology - These criteria define the operating conditions at the
facility under investigation. Fuel characteristics, heat rate, location conditions, etc., are
requested in this section. These data are then used as the basis for the combustion
calculations and definition of the plant ambient conditions.
3. Air Pollution Control system definition - This is a section of the utmost importance where
the user can select from among various ARC technology configurations for a specific site.
Each specific site can contain one or all of the ARC subsystems.
3.1 Nitrogen Oxides Control technology - All data required to define the NOX control system
are requested in this section. The user also selects the type of control system that is
desired: selective catalytic reduction (SCR), selective non-catalytic reduction (SNCR),
natural gas reburn (NGR) technology, or low-NOx burner technology (LNBT) [including low
NOX burners (LNB) for pulverized coal boilers, and low-NOx concentric firing systems
(LNCFS) for tangentially fired boilers].
20
-------
Input and Output Options .
3.2 Particulate Matter Control Technology - All data required to define the particulate matter
control system are entered in this section. The user also selects the type of control
system that is desired, ESP or a FF, and what type of fabric filter is selected.
3.3 Sulfur Dioxide Control Technology - This section provides a series of inputs that define
the operating conditions for the scrubber system. In this section the user can define
conditions that are specific to vendor data, or the default values can be used to determine
the generic costs for the FGD system. The option to use a dibasic acid (DBA) additive is
also provided. The DBA acts as a buffer in the SO2 absorption reaction, potentially
reducing the operating costs for the FGD system and improving performance at some
sites.
3.4 Lime Spray Dryer FGD Process Definition - The data inputs on this worksheet are similar
to the inputs for the LSFO worksheet. Once again the process operating conditions are
defined for each case being considered.
3.5 Mercury Control Technology - All data required to define the Hg control system are
requested in this section. The user selects the type of sorbent that is desired: enhanced
PAC (EPAC), PAC, or other. The user also specifies the design and operating conditions of
the pulse-jet fabric filter (PJFF) downstream of PAC if desired.
3.6 Carbon Dioxide Control Technology - All data required to define the CO2 control system
are input in this section. The user selects the type of process: MEA, CAP, or SI.
OUTPUT OPTIONS
The input values are then summarized in the Input Summary worksheet. The Output
Summary worksheet compiles the results generated by the technology-specific worksheets.
The tables are constructed for use in printing the output sheets. A summary table is also
available in the section of Summary of Emissions and Generation. The output summary
table primarily provides the cost estimates generated for all of the control technologies
selected for each case. More detailed breakdowns of each technology cost estimate are also
generated in the technology-specific worksheets to identify the components of the
estimates. To obtain outputs from the workbook, the user can return to the General Input
Sheets, ensure that the workbook has been recalculated by pressing the F9 button, and
then click on the Print buttons provided for printing. The user can also enter any worksheet
of interest and click on the print icon. The workbook is set up to automatically print all of
the cost related material in each worksheet.
The workbook also allows the user to select any specific portion of an individual worksheet
that is of interest and print out that material only. A specific range can be selected and that
section printed using the standard Excel methodology.
21
-------
Worksheet Validation
The CUECost workbook was constructed to allow the user to have the maximum flexibility to
modify it to generate site-specific cost estimates without requiring an extensive amount of
input data. The technology worksheets were developed using different sets of cost and
design data. The basis for each set of parametric design and cost equations is described in
the following section.
FGD WORKSHEETS - LSFO AND LSD TECHNOLOGIES
The equipment design parameters and cost data are based on a combination of vendor
quotes and a historical database of installed power projects (Keeth et al., 1991). Cost-
versus-capacity curves were constructed based on this historical information combined with
vendor quotations from both installed FGD systems and budgetary quotes received
specifically for this CUECost project. Many of the sources of information that were used in
this development of the FGD system costs are not available to the public due to the
proprietary nature of the information and the project-specific sensitivity of the cost data.
This equipment cost database was assembled based on the experience gained at FGD
installations for 10-15 plants ranging in size from 300 to 2000 MW. Equipment cost data is
produced by compiling data taken from these 10 to 15 actual installations, vendor
quotations for construction contracts, and budgetary quotations obtained in 1998
specifically to support this CUECost project. The budgetary quotes for large equipment items
were received from one to six vendors depending on the component. The accuracy of the
CUECost is validated by comparing the results generated by the CUECost model to published
cost data for many of the Phase 1 FGD systems installed in response to acid rain
regulations. The validation of the data used in the development of these algorithms is
described in Appendix D.
The cost estimates from CUECost were compared to the results generated by other models,
including the comparison to the Electric Power Research Institute's (EPRI) FGDCOST model
(Keeth et al., 1991). EPRI's FGDCOST model has been used throughout the utility industry
22
-------
Worksheet Validation
for the last seven years and has demonstrated its ability to estimate site-specific costs well
within ROM accuracy requirements. The CUECost estimates were found to agree well with
the results generated by the FGDCOST model when allowance was made for the changes in
the technology that have occurred since the FGDCOST model was constructed, the
escalation of costs, and the reduced level of design data that is required by the CUECost
workbook.
CUECost was also used to calculate cost estimates for many of the Phase 1 FGD systems.
Actual installed cost data have been published in various sources for these systems. These
data were compared to the estimates generated by the CUECost workbook and CUECost
reproduced these actual costs within an accuracy of ±12%. Table 4 provides the results of
this comparative analysis for previously installed FGD systems.
Table 4. CUECost-FGD Cost Comparison to FGDCOST by EPRI for Phase 1 Acid Rain
Installations
un-
Petersburg
Cumberland
Conemaugh
Ghent
Gibson
Bailly
Milliken
Navajo
"•SS1*
657
2600
1700
511
668
600
316
2250
Sulfur
%
3.50
4.00
2.80
3.50
3.50
4.50
3.20
0.75
%
95
95
95
90
91
95
98
92
vSi
317
200
195
215
247
180
348
236
CUECost,
$/kW
291
187
179
229
218
196
362
213
% Difference
-8.20
-6.50
-8.20
+6.5
-11.70
+8.9
+4.0
-9.75
PARTICULATE MATTER CONTROL WORKSHEET
The particulate matter control sizing equations were based on previously published
correlations developed by CMU (Berkenpas et al., 1999). This development process is
described in Appendix D. The CMU model was constructed using design information supplied
by multiple vendors.
The CMU model used a modified version of the Deutsch-Anderson equation (Edgar, 1983) to
relate removal efficiency to collection area and gas flow rate for various coals as part of the
ESP sizing calculations. The original Deutsch-Anderson equation was found to be inaccurate
for removal efficiencies above 95%. Various empirical models were developed to overcome
this inaccuracy, and the CMU model chose to use the White version (White, 1977) of the
modified Deutsch equation provided below (Eq. 1):
23
-------
Worksheet Validation
h = 1- exp {-A/V H wk}k (Eq. 1)
where
h = collector removal efficiency
A = collector area, ft2
V = volumetric flue gas flow rate, actual cubic feet per minute (acfm)
wk = precipitation rate parameter
k = constant varying with coal type.
The wk and k values used in the CMU ESP sizing equations (Berkenpas, 1999) were
correlated with the calculated total ash resistivity (based on the ash analysis provided by
the user or the default database in worksheet 11.0), and separate k curves were developed
for groups of coals that have similar sulfur content. The modified sizing worksheet provides
the expected specific collection area (SCA) for the ESP, and a new set of cost equations was
developed to relate the ESP size (calculated from the SCA and the ESP inlet flue gas
volumetric flow rate) to the expected cost for the installed system.
The ESP equations provided in the CMU model were reviewed and compared to the expected
ESP sizes in terms of SCA, evaluating the various types of coals listed in Table 5. The
"Raytheon" SCA data provided in Table 5 were calculated using a series of parametric
equations developed by Raytheon Co.2 These equations were derived from SCA data for
utility coal-fired installations over the past 25 years obtained by Raytheon Co. and
incorporated into a proprietary model used for confirmation of vendor data and specification
preparation. As can be seen in Table 5, the CUECost workbook calculates SCA values that
are within ±12% of the values generated by the Raytheon model.
The costs generated by CUECost were compared to the current IAPCS results for the same
plant sizes and coals. The results were within 30% of the IAPCS cost estimating model
(Gundappa et al., 1995) over a range of SCA values from 300 to 600. The FF cost
algorithms (one for pulse jet design and one for reverse gas) were developed from 10 to 12
firm price quotations (obtained during 1992-1997) for each FF design. The coal-fired boilers
ranged in size from 50 to 500 MW.
'- The Raytheon database is proprietary.
24
-------
Worksheet Validation
Table 5.
Comparison of CUECost ESP Sizing Estimates with Raytheon Model
Coal Type
Indiantown
WV-EPRI
Low S Bituminous
Keystone
India
Logan, WV
ND Lignite
UT-EPRI
UT-Alternate
Rosebud, MT
WY-PRB
Test Coal
Pitts 8
Carneys
TX Lignite
OH Alternate
IL #6
Armstrong, PA
Jefferson, OH
Sulfur
Content,
%
1.09
0.66
0.97
1.09
0.5
0.89
0.94
0.53
0.66
0.56
0.37
2
2.13
2
1.16
4.7
3.25
2.6
3.43
Removal
Efficiency, %
99.4
99.2
99.4
99.3
99.9
99.7
99.4
99.5
99.6
99.5
99.3
99.1
99.2
99.1
99.8
99.6
99.5
99.3
99.6
Raytheon
SCA*
385
418
403
393
965
569
376
446
435
482
558
287
272
288
549
247
276
277
321
CUECost SCA*
429
375
424
386
883
502
411
442
482
459
558
283
285
281
549
259
261
274
326
%
Difference
+ 11.43
-10.29
+5.21
-1.78
-8.50
-11.78
+9.31
-0.90
+ 10.80
-4.77
0
-1.39
+4.78
-2.43
0
+4.86
-5.43
-1.08
1.56
SCA = square feet of plate area per 1000 actual cubic feet per minute of flue gas flow
NOX CONTROL WORKSHEET
For NOX control technologies, CUECost results were compared to cost data reported by the
EPA Acid Rain Division for NOX controls applied to utility boilers (Khan and Srivastava,
2004), and Chapter 5 of the IPM manual.3 The Acid Rain Division reports (EPA, 1997; EPA,
1998) are based on an EPA national database of boilers (Khan and Srivastava, 2004). The
1990 Clean Air Act Amendments required the EPA to examine NOX control technology costs,
and the resulting Acid Rain Division studies (EPA, 1997; EPA, 1998) were used and reviewed
during the rule-making process. A comparison was made for four cases with various boiler
types, boiler sizes (100 to 400 MW) and coals burned. The boiler design and operating
parameters for each case were input into CUECost to obtain capital and operating and
maintenance costs. In some cases the capital cost estimating algorithms in these sources
3 available at: http://www.epa.gov/airmarkets/proqsreqs/epa-ipm/index.html and can be downloaded at
http://www.epa.gov/airmarkets/proqsreqs/epa-ipm/docs/Section-5.pdf.
25
-------
Worksheet Validation
already included provisions for retrofit factor, general facilities, engineering, contingency,
and other factors that, using CUECost methodology, are in addition to equipment costs. So
it was necessary to adjust for these additional costs in arriving at the equipment cost
algorithms for CUECost. For this reason, the algorithms programmed into CUECost may be
different from those shown in the IPM report or the Acid Rain Division reports (Khan and
Srivastava, 2004). However, using CUECost's default values for retrofit factor, general
facilities, etc., should produce capital cost results that are the same as or very close to the
results that would be produced by the IPM source algorithms.
Different approaches were taken to verify or validate the costs predicted by CUECost for the
various NOX control technologies. For SCR, SNCR and NGR, design parameters used for the
ARD study cases (EPA, 1997; EPA, 1998) were used to calculate preliminary operating
parameters and costs with CUECost. Algorithms for SCR in CUECost were compared to the
ARD study costs to validate the algorithms. However, the ARD data were incorporated into
the algorithms for SNCR and NGR. As a result, the cost comparisons for these technologies
were conducted to benchmark the algorithms and evaluate how well they track the ARD
data. The percent differences found for the four boiler cases are presented in Table 6.
Differences range in magnitude from 0 to 11% for total plant costs and from 0 to 22% for
operating and maintenance costs.
SNCR capital costs are determined from the IPM and are documented in Chapter 5 of the
IPM manual (http://www.epa.QOv/airmarkets/proqsreQS/epa-ipm/index.html), which can be
downloaded at httD://www.eDa.qov/airmarkets/Droqsreqs/eDa-iDm/docs/Section-5.Ddf.
The algorithms used to estimate costs for LNBT in CUECost were taken from an Acid Rain
Division study (EPA, 1996). The cost data upon which the algorithms were based represent
actual LNBT retrofit cases. The capital cost comparison shows 0% difference, as expected,
because the algorithms are based solely on ARD data. A comparison is not presented for
operating and maintenance costs because these costs are highly boiler specific.
26
-------
Worksheet Validation
Table 6. Percent Difference between CUECost and Acid Rain Division Studies (Khan
and Srivastava, 2004) for Retrofit Cases *
Cyclone
Fired
Midwestern Bituminous
140
Boiler
400
Wet Bottom
Vertical Fired Wall Fired
Eastern Bituminous
Size (MW)
100 259
Total Plant Costs
SCR (50% removal)
SNCR (50% removal)
NGR (35% removal)
4%
8%
-11%
0%
0%
-7%
8% -4%
12% 4%
-12% -12%
O&M Costs
SCR (50% removal)
SNCR (50% removal)
NGR (35% removal)
-12%
8%
-11%
-18%
0%
-7%
-16% -22%
12% 4%
-12% -12%
Note: Percent Difference = (Acid Rain Costs - CUECost Results) x 100 /Acid Rain Costs
MERCURY CONTROL WORKSHEET
Mercury control technologies included in CUECost are co-benefit controls from air pollution
control technology used for other pollutants and sorbent-based mercury-specific controls.
Mercury control technology cost and performance estimates are determined by algorithms
described in U.S. EPA (2003), Staudt, Jozewicz, and Srivastava (2003), Srivastava, Staudt,
and Jozewicz (2004), as well as in Appendix F of this manual.
In addition to the impact of sorbent cost on operating cost of mercury control technologies,
calculations include estimates of the impact of parasitic load and filter replacement (if a
fabric filter is retrofit) and the impact the sorbent may have on fly ash marketability when
the sorbent and fly ash are collected in the same PM control device.
Capital cost estimating methodology was made consistent with other technologies with the
sole exception that we included Process Contingency for Hg Control in addition to the other
cost factors because of the relative newness of Hg control technologies. The user may input
a Process Contingency percentage in the Input worksheet or accept the default Process
Contingency value of 5%.
Detailed review of mercury control technologies, performance, and future improvements can
be found elsewhere (Srivastava et al. 2004; EPA 2003). One of the most promising
technologies to control mercury emissions from coal-fired power plants is SI, especially the
activated carbon injection (ACI). Table 7 provides the estimated capital and O&M costs of an
ACI control system from CUECost.
27
-------
Worksheet Validation
Table 7.
Estimated Costs of ACI Control Systems according to CUECost*
ARC Configuration
ACI +Cold-side ESP
ACI +Cold-side ESP
+ Wet FGD System
Capital Cost
(2005$/kW)
19.41
19.41
O&M Cost
(2005$/MWh)
4.06
4.06
Hg Removed by
Sorbent Injection
(Ib/yr)
240.7
188.9
Control Cost
(2005$/lb Hg
removed)
53,380
68,013
ACI+ Dry Scrubber
+ Fabric Filter
3.17
0.32
290.7
3,844
* Note: 500 MW, Wyoming Powder River Basin (PRB) coal, activated carbon injection, capacity
factor = 65%, 80% Hg removal.
CARBON DIOXIDE CONTROL WORKSHEET
Carbon dioxide (CO2) control technologies included in CUECost are based on MEA solvent,
chilled ammonia and sorbent injection. These technologies, as described in Appendix B.6 of
this manual, generally include absorption/regeneration island and compressor island. The
capital cost of the CO2 control facility in the CUECost is the lump-sum of the individual costs
and given by a regressed equation based on currently available bare erected plant cost from
the DOE report (2007). As there is no SI based CO2 control technology adopted in a power
plant, its capital cost is estimated to a comparable cost for MEA process. For all the control
technologies, the compressor island cost is regressed on the basis of power consumption of
compressors. Gas flow rate and specific variable costs are estimates and determined by
algorithms described in Appendix D. Table 8 summarizes the estimated capital and O&M
costs of CO2-related control technologies from CUECost.
Table 8.
Estimated Costs of CO2 Control Technologies with CUECost and IECM model.
Total Plant Cost (TPC), Million $
Total Capacity Requirement
(TCR), Million $
O&M
Fixed O&M, Million $/yr
Variable O&M, Million $/yr
$ (constant)/ton CO2
MEA
350.8
388.3
12.3
76.4
51
CUECost
CAP
322.1
353.1
11.36
42.6
34
SI
350.8
380.2
12.3
39.0
34
IECM
MEA*
232.2
273.3
7.3
103.7
37
Note: estimates are based on a 580 MW plant, firing Illinois bituminous coal. The plant
capacity factor is 65% and demands a 90% CO2 removal efficiency. Capital cost is calculated
for the base year of 2006.
*The MEA data was calculated with the IECM model developed by CMU. The IECM program can be
downloaded from http://www.iecm-online.com/iecm_dl.html.
28
-------
Worksheet Validation
VALIDATION SUMMARY
Costs for utility ARC systems are site-specific. These costs are subject to change with
changes in technology, labor rates, and material costs. The costs estimated by the CUECost
workbook come from a variety of sources. With that understanding, one may assume, but it
is not guaranteed, that CUECost will produce estimates in the range of accuracy of ±30% of
the actual cost, which was the goal of the CUECost development. The operating cost
estimates are more straightforward than the capital cost estimates, relying more on the
accuracy of the input data supplied by the user. The calculation sequences for these
estimates have been verified on a cell-by-cell basis during the course of the workbook
development. The documentation provided in Appendix F also allows any user to verify a
specific calculation sequence that might be in question at some point in the future. The
economic calculation methods used have been well established for many years throughout
the utility industry, and have been documented in the EPRI Technical Assessment Guide
(Ramachandran, 1989).
29
-------
References
Berkenpas, M. B.; Frey, H.C.; Fry, J. J.; Kalagnanam, J.; Rubin, E. S. "Integrated
Environmental Control Model (Technical Documentation)," Prepared for the Federal Energy
Technology Center (U. S. Department of Energy). May 1999.
Bustard, J., Durham, M., Lindsey, C., Starns, T., Baldrey, K., Martin, C., Schlager, R.,
Sjostrom, S., Slye, R., Renninger, S., Monroe, L, Miller, R., Chang, R., 2001. "Full-Scale
Evaluation of Mercury Control with Sorbent Injection and COHPAC at Alabama Power E.G.,
Gaston", DOE-EPRI-U.S. EPA-A&WMA Power Plant Air Pollutant Control "Mega" Symposium,
August 20-23, 2000, Chicago, IL.
Bustard, J., Durham, M., Lindsey, C., Starns, T., Baldrey, K., Martin, C., Schlager, R.,
Sjostrom, S., Slye, R., Renninger, S., Monroe, L., Miller, R., Ramsey, C., 2002. "Gaston
Demonstrates Substantial Mercury Removal with Sorbent Injection", Power Engineering, vol.
106, no. 11.
DOE/NETL. 2007. Cost and Performance Baseline for Fossil Energy Plants (DOE/NETL-
2007/1281).
Durham, M., Bustard, J., Schlager, R., Martin, C., Johnson, S., Renninger, S., 2001. "Field
Test Program to Develop Comprehensive Design, Operating Cost Data for Mercury Control
Systems on Non-Scrubbed Coal-Fired Boilers", AWMA 94th Annual Conference and
Exhibition, Orlando, FL, June 24-28 2001.
Edgar, T. T., 1983. Coal Processing and Pollution Control. Houston, TX. Gulf Publishing Co.
EPA, 1996. "Cost-effectiveness of Low-NOx Burner Technology Applied to Phase I, Group 1
Boilers," prepared by Acurex Environmental Corporation for EPA Acid Rain Division. This
report is available to the public from EPA's Office of Air and Radiation, Acid Rain Division,
Washington, DC 20460 (ph. 202-564-9085).
30
-------
References
EPA, 1997. "Investigation of Performance and Cost of NOX Controls as Applied to Group 2
Boilers," EPA, Washington, DC. This report is available to the public from EPA's Office of Air
and Radiation, Acid Rain Division, Washington, DC 20460 (ph. 202-564-9085).
EPA, 1998. "Cost Estimates for Selected Applications of NOX Control Technologies on
Stationary Combustion Boilers and Responses to Comments," EPA, Washington, DC. This
report is available to the public from EPA's Office of Air and Radiation, Acid Rain Division,
Washington, DC 20460 (ph. 202-564-9085).
EPA, 2000, Performance and Cost of Mercury Emission control Technology Applications on
Electric Utility Boilers, EPA-600/R-00-083.
EPA, 2003. "Performance and Cost of Mercury and Multipollutant Emission Control
Technology Applications on Electric Utility Boilers," EPA-600/R-03-110, October 2003.
EPRI, 2000, An Assessment of Mercury Emissions from U.S. Coal Fired Power Plants, EPRI,
Palo Alto, CA.
Frey, C.H. and E.S. Rubin, 1994. "Development of the Integrated Environmental Control
Model: Performance Models of Selective Reduction (SCR) NOX Control Systems; Quarterly
Progress Report to Pittsburgh Energy Technology Center, U.S. Department of Energy, from
Center for Energy and Environmental Studies, Carnegie Mellon University," Pittsburgh, PA.
Document number DE-AC22-92PC91346-11
Gundappa, M., L. Gideon, and E. Soderberg, 1995. "Integrated Air Pollution Control System
(IAPCS), version 5.0, Volume 2: Technical Documentation, Final," EPA, Air and Energy
Engineering Research Laboratory, Research Triangle Park, NC, EPA-600/R-95-169b (NTIS
PB96-157391).
Keeth, R. J., Baker, D. L., Tracy, P. E., Ogden, G. E., and Ireland, P. A., 1991. Economic
Evaluation of Flue Gas Desulfurization System. No. GS-7193, Research Project 1601-6.
EPRI. Palo Alto, CA.
Khan, S., and Srivastava, R. "Updating Performance and Cost of NOX Control Technologies
in the Integrated Planning Model", EPA-EPRI-DOE Combined Power Plant Air Pollution
Control Mega Symposium, August 30-September 2, 2004, Washington, DC.
Ramachandran, G., 1989. "TAG™ Technical Assessment Guide," EPRI Report No. P-6587-L,
Volume 1: Rev.6.
Srivastava R. K. "Controlling SO2 emissions: A review of technologies." U.S. Environmental
Protection Agency EPA/600-R-00/093, 1999
Srivastava, R. K., Staudt, J., Jozewicz, W. "Preliminary Estimates of Performance and Cost
of Mercury Emission Control Technology Applications on Electric Utility Boilers: An Update",
31
-------
References
EPA-EPRI-DOE Combined Power Plant Air Pollution Control Mega Symposium, August 30-
September 2, 2004, Washington, DC
Starns, T., Bustard, J., Durham, M., Lindsey, C., Martin, C., Schlager, R., Donnelly, B.,
Sjostrom, S., Harrington, P., Haythornthwaite, S., Johnson, R., Morris, E., Chang, R.,
Renninger, S., 2002, "Full-Scale Test of Mercury Control with Sorbent Injection and an ESP
at Wisconsin Electric's Pleasant Prairie Power Plant", AWMA 95th Annual Conference and
Exhibition, Baltimore, June 23-27 2002.
Staudt, J.E.; Jozewicz, W.; Srivastava, R. "Modeling Mercury Control with Powdered
Activated Carbon", AWMA Paper 03-A-17-AWMA, Presented at the Joint EPRI DOE EPA
Combined Utility Air Pollution Control Symposium, The Mega Symposium, May 19-22, 2003,
Washington, D.C.
White, Harry J., 1977. "Electrostatic Precipitation of Fly Ash," J. Air Pollution. Control Assoc.,
March 1977, Volume 27, No. 3, pp. 206-217.
32
-------
Appendix A
A.1 DEFINITION OF TERMS
Allowance for Funds Used During Construction (AFDC) - Represents the time value of
money during the construction period. AFDC is calculated based on the weighted cost of
capital, compounded on an annual basis throughout the period, and applied to all funds
spent during each year. This cost is added to the Total Cash Expended to obtain TPI. See
Table 2 for the use of the AFDC factor. The AFDC factor is input by the user, and is a
function of the years of construction and the discount rate.
Ammonia Slip = The un-reacted ammonia that exits an SCR or SNCR process, and exits
the stack with the flue gas. Ammonia slip is expressed as a concentration in the exit gas
or as a percentage of the mass of ammonia input to the process.
Battery Limits = The boundary limits within a plant used to define the equipment
components contained in a subsystem.
Capacity Factor (CF) - Equivalent to the ratio of the total energy output over a time
period divided by the total gross energy generating capacity of the unit. Typically the CF
is input as the expected average value over the remaining plant life.
Carrying Charge Factor (CCF) - Amount of revenue per dollar of investment that must be
collected from customers in order to pay the carrying charges on that investment. The
CCF is expressed as a decimal that is multiplied by the original investment to obtain a
carrying charge in terms of dollars. The carrying charge rate can be a present value or
levelized quantity over a specified period of time (up to the book life), or an annual
quantity in a specific year of life. The factor includes the return on debt, return on equity,
income and property taxes, book depreciation, rate of return to shareholders, and
insurance.
33
-------
Appendix A
Constant Dollar - Cost estimate presented in terms of the base year dollars without
including the impact of inflation over the plant life. However, real escalation is included in
the calculation of future year costs. Constant dollar analysis requires the use of a
discount rate that does not include inflation.
Contingency - A capital cost included in the estimate to cover the costs for additional
equipment or other costs that are expected to be incurred during a project after the detailed
design is completed. These are funds that are expected to be spent during implementation
of the final project. The contingency is factored as a percent of process capital plus
engineering, home office and general facilities.
Current Dollar - A cost analysis that includes the effects of inflation and real escalation. The
discount rate used for current dollar analyses is equivalent to the return required to attract
investment capital and is equivalent to the weighted average of the return on equity and
return on debt.
Engineering and Home Office Costs - Derived as a percentage of the total direct capital cost.
This indirect cost includes the costs for an architectural/engineering company and for home
office engineering expenses by the user's company. This value typically ranges from 5 to
20% of the Process Capital, with the percentage varying based on the level of complexity
for equipment installation (e.g., a new plant might have a value of 5 to 10% while a retrofit
might experience engineering costs closer to 15-20%).
General Facilities - Includes costs for items such as roads, office buildings, maintenance
shops, and laboratories. The indirect cost for these facilities typically ranges from 5 to 20%
of the Process Capital.
Heat Rate - Equivalent to the fuel energy content (Btu) required to produce 1 kWh of
electric energy. Fuel energy content is typically based on the higher heating value of the
fuel.
Inflation Rate - Equivalent to the rise in prices caused by an increase in the available
currency and credit without a proportionate increase in availability of goods and services of
equal quality. The inflation rate does not include the effects of real escalation.
Operating Costs - Operating costs for each technology are expressed in terms of both $/kW-
year and mills/kWh. The $/kW-year costs are considered to be an expression of annual
costs and, therefore, include the capacity factor in the calculation. The mills/kWh values are
considered instantaneous values, and, therefore, do not include the capacity factor in their
calculation.
Present Value (PV) - Monetary equivalent to the amount of money at a point in time other
than that at which the amount of money is paid or received.
34
-------
Appendix A
Process Capital - Total installed cost of all process equipment.
Total Capital Requirement (TCPO - Equivalent to the Total Plant Cost, AFDC, plant startup
costs, and inventory capital.
Total Plant Cost (TPC) - Equivalent to the total installed cost for all plant equipment,
including all direct and indirect construction costs, engineering, overheads, fees, and
contingency.
35
-------
Appendix B
B.I LIMESTONE FORCED OXIDATION DESIGN CRITERIA
In a limestone with forced oxidation (LSFO) system, the flue gas is contacted with slurry
containing approximately 15% calcium carbonate and sulfate solids. The aqueous sulfite
formed by SO2 absorption is oxidized to sulfate by forced air injection in the tower
recirculation tank to produce slurry with essentially 100% conversion of calcium sulfite to
sulfate. The series of chemical reactions that occur in an LSFO absorber and reaction tank is
described in Eqs. B-l and B-2:
SO2 Reaction: CaCO3 (s) + SO2 (g) + 1/2 H2O -> CaSO3 • 1/2 H2O + CO2 (Eq. B-l)
Sulfite Oxidation: CaSO3 • 1/2 H2O + 1/2 O2 + 3/2 H2O -> CaSO4 • 2 H2O (Eq. B-2)
The CUECost workbook requires that the user input new values for the slurry recycle rate
(Liquid to Gas Ratio = L/G) whenever the SO2 removal efficiency across the FGD system is
changed versus the current 95% removal rate included as the base case default value.
Typically the increase in removal efficiency above this 95% level will require significant
increases in the recycle rate. A value of 140 gallons/1000 actual cubic feet (L/G) would be
typical for a 97% removal system versus the 125 value for a 95% system. Therefore, the
pump sizes and power consumption required in the FGD system would increase significantly.
Values for the limestone feed rate (stoichiometric feed ratio default = 1.05 moles of CaCO3
per mole of SO2 removed) also remain constant with changes in the removal efficiency, but
can be modified by the user if additional vendor information is available.
The slurry produced by the FGD system can be thickened and pumped directly to a gypsum
stack for final disposal, vacuum filtered or centrifuged for landfill disposal, or washed and
dewatered for commercial wallboard production.
36
-------
Appendix B
The LSFO Process Equipment includes the Reagent Handling and Preparation, SO2 Control
System, and the Byproduct Handling.
Reagent Handling and Preparation includes the following:
• Reagent storage
• Reagent feed
• Ball mill and hydroclones
• DBA acid tank.
SO2 Control System includes:
• SO2 removal system
• Absorber tower
• Spray pumps, spray nozzles, associated piping.
Byproduct Handling includes:
• Waste/byproduct handling system
• Thickener system.
ID Fans and Ductwork are:
• Booster fans needed for the system
• Ductwork between components.
Chimney is:
• Cost of replacement chimney and associated foundations.
Support equipment is:
• Electrical support equipment and modifications not included elsewhere.
An alternative design option is provided in the LSFO system to include the addition of DBA.
This additive helps to buffer the SO2 absorption reaction, increasing the available alkalinity
in the slurry. Addition of DBA allows the system to be designed with lower recycle rates and
potentially a lower limestone feed rate while maintaining the removal efficiency.
37
-------
Appendix B
Specific design criteria for LSFO are shown in Table B-l. The default values provided in the
worksheet are considered typical for operating FGD systems recently installed in the U.S.
Reagent costs are typically based on the costs stated in the journal Chemical Marketing
Reporter.
Table B-l. Specific Design Criteria for LSFO
Description
Year equipment placed in service
SO2 Removal Required
L/G Ratio
Design Scrubber with Dibasic Acid Addition?
(1 = yes, 2 = no)
Adiabatic Saturation Temperature
Reagent Feed Ratio
(Mole CaCO3/ Mole SO2 removed)
Scrubber Slurry Solids Concentration
Reheat Air Temperature
Pressure
Stacking, Landfill, Wallboard
(1 = stacking, 2 = landfill, 3 = wallboard)
Number of Absorbers
(Max. Capacity = 900 MW per absorber)
Absorber Pressure Drop
Reheat Required?
(1 = yes, 2 = no)
Amount of Reheat
Reagent Bulk Storage
Reagent Cost (delivered)
Landfill Disposal Cost
Stacking Disposal Cost
Credit for Gypsum Byproduct
Retrofit Factor
Maintenance Factor (% of TPC)
Contingency (% of Installed Cost)
General Facilities (% of Installed Cost)
Engineering Fees (% of Installed Cost)
Time for Retrofit to use for TCE and AFDC factors
Units
year
%
gal / 1000 acf
Integer
°F
Factor
Wt. %
°F
in. H2O
Integer
Integer
in. H2O
Integer
°F
Days
$/ton
$/ton
$/ton
$/ton
%
%
%
%
years
Range
90-98%
95-160
1 or 2
100-170
1.0-2.0
1,2,3
1-6
1 or 2
0-50
Default
2004
95%
125
1
127
1.05
15%
440
1
1
1
6
1
25
60
$15
$30
$6
$2
1.3
3%
15%
5%
10%
2
38
-------
Appendix B
B.2 LIME SPRAY DRYER DESIGN CRITERIA
In a lime spray dryer (LSD) process the flue gas exiting the air heaters enters a spray dryer
vessel. Within the vessel, an atomized slurry of lime and recycled solids contacts the flue
gas stream. The sulfur oxides in the flue gas react with the lime and fly ash alkali to form
calcium salts.
The chemical reactions associated with the SO2 removal from the flue gas are provided
below (Eqs. B-3 through B-6):
Lime Hydration: CaO + H2O -> Ca(OH)2 (Eq. B-3)
SOX Reaction (1): Ca(OH)2 + SO2 -> CaSO3 • 1/2 H2O + 1/2 H2O (Eq. B-4)
SOX Reaction (2): Ca(OH)2 + SO3 + H2O -> CaSO4 • 2 H2O (Eq. B-5)
Sulfite Oxidation: Ca(OH)2 + SO2 + H2O + 1/2 O2 -> CaSO4 • 2 H2O (Eq. B-6)
The water entering with the slurry vaporizes, lowering the temperature and raising the
moisture content of the scrubbed gas. A particulate matter control device downstream of
the spray dryer removes the dry solids and fly ash that did not fall out in the vessel. A
portion of the collected reaction products and fly ash solids is recycled to the slurry feed
system. The remaining solids are transported to a landfill for disposal.
The CUECost workbook responds to changes in the removal efficiency and any other
parameter by using the input values entered by the user and recalculating the material
balance on that new basis. No other changes in the worksheet are done automatically in
response to changes in parameters. The CUECost workbook does modify the solids recycle
rate as the coal sulfur content is modified. The modification is done with a look-up
tabulation of recycle values associated with various coal sulfur percentages. A look-up table
is embedded in worksheet 11.0 Constants_CC of the CUECost workbook.
The LSD system incorporates five specific equipment areas:
• Reagent handling and preparation
• SO2 control system
• Byproduct handling
• ID fans and ductwork
• Support equipment.
The Reagent Handling and Preparation includes the following:
39
-------
Appendix B
• Lime storage and preparation
• Lime slaker.
SO2 Control System includes:
• SO2 removal system
• Absorber tower
• Spray pumps, spray nozzles, associated piping.
Byproduct Handling includes:
• If LSD system is installed upstream of existing ESP, this includes modifications to existing
ESP due to increased solids handling and gas with more moisture
• Otherwise, SDA and new FF or SDA and new ESP need to be added. Their costs can
further be calculated for ESP and FF calculations in the ESP and FF worksheet.
ID Fans and Ductwork are:
• Booster fans needed for the system
• Ductwork between components.
Support Equipment is:
• Electrical support equipment and modifications not included elsewhere
The annual Maintenance (component of the operating cost), additional General Facilities,
and Engineering factors provided in Table B-2 are multiplied by the installed equipment
capital cost to obtain an estimate of these costs to the utility. The Contingency factor is
applied to the total bottom line cost (Equipment Installed Cost plus Site Facilities and
Engineering) and represents an estimate of the capital that will be expended but not
accounted for in the estimate due to the level of detail included in the system design for this
cost worksheet.
40
-------
Appendix B
Table B-2. Input for LSD and the Default Values for the Inputs
Description
SO2 Removal Required
Is SDA being retrofit upstream of existing ESP?
(0 = no, 1 = yes)
Adiabatic Saturation Temperature
Flue Gas Approach to Saturation
Recycle Slurry Solids Concentration
Number of Absorbers
(Max. Capacity = 300 MW per spray dryer)
Absorber Material
(1 = alloy, 2 = RLCS)
Spray Cooler Pressure Drop
Reagent Bulk Storage (days)
Reagent Cost (delivered)
Dry Waste Disposal Cost
Retrofit Factor
Maintenance Factor (% of TPC)
Contingency (% of Installed Cost)
General Facilities (% of Installed Cost)
Engineering Fees (% of Installed Cost)
Project Duration (years)
Units
%
integer
°F
°F
Wt. %
integer
integer
in. H2O
integer
$/ton
$/ton
%
%
%
%
integer
Range
90-95%
0,1
100-170
10-50
10-50
1-7
1 or 2
Default
90%
0
127
20
35%
1
2
1
30
$65
$30
1.3
2%
15%
5%
10%
2
B.3 PARTICULATE MATTER CONTROL DESIGN CRITERIA
In a particulate control system, the flue gas exiting the air heaters enters an ESP or FF
through the inlet manifold. In an ESP, the particulate matter is electrically charged by the
electric fields generated. This charge helps to move the particles to the collecting plates'
surfaces, and holds them in place until the collected material can be discharged into the
collecting hoppers. ESPs are available in a wide variety of designs and construction
materials; collecting plate design, size and spacing; electrode design; etc. These variations
in design among vendors are not addressed in this worksheet, and are not expected to drive
the final system cost estimates beyond the stated ROM estimate accuracy. The dry fly ash
material is typically transferred to final disposal silos by a pneumatic conveying system.
Within the FF, the particulate matter is collected on filter bags suspended vertically within
the FF vessel. The particulate matter is physically removed from the gas as it passes
through the filter bags, by impacting both the bag fibers and the filter cake that collects on
the surface of the bags. Periodically, individual FF compartments are mechanically cleaned
by reversing the gas flow or using a pulse jet design that uses pressurized air to force the
collected fly ash off the bags and into the collection hoppers. The two design options
41
-------
Appendix B
(reverse gas and pulse jet) are available as options in the worksheet. The air-to-cloth ratio
(square feet of cloth required per 1000 actual cubic feet per minute of flue gas flow)
identifies the size of the FF required, quantifying the amount of cloth area required to treat
a given gas flow rate. Once again, the ash is typically transferred to the waste silo by a
pneumatic conveying system.
The CUECost workbook responds to changes in the removal efficiency and any other
parameter by using the input values entered by the user and recalculating the material
balance on that new basis. No other changes in the worksheet are done automatically in
response to changes in parameters. The model does modify the solids collection rate as the
coal ash content is modified.
Specific design criteria associated with particulate matter control are summarized in Table
B-3 below:
Table B-3. Inputs for Particulate Matter Control and Its Default Values
Description
Units
Value
Particulate Matter Control
Outlet Part. Matter Emission Limit
Particulate Matter Control Process
(1 = Fabric Filter, 2 = ESP)
Ibs/MMBtu
integer
0.03
1
Fabric Filter
Fabric Filter Type
(1 = Reverse Gas, 2 = Pulse Jet)
Gas-to-Cloth Ratio
Bag Life
integer
acfm/ft2
years
2
1.8
5
Electrostatic Precipitator
Specific Collection Area (SCA)
ft2 Collecting Plate/
1000 acfm Gas
Calculated based on ash
composition and
collection efficiency
B.4 NOX CONTROL TECHNOLOGY CRITERIA
Four NOX control technologies are included in CUECost:
• Selective Catalytic Reduction (SCR)
• Non-Selective Catalytic Reduction (SNCR)
Natural Gas Reburning (NGR)
Low NOX Burners (LNB).
42
-------
Appendix B
The process design criteria and assumptions that serve as defaults within the worksheet are
described in the following sections.
B.4.1 Selective Catalytic Reduction Design Criteria
Selective catalytic reduction (SCR) is a post-combustion nitrogen oxides (NOX) reduction
process where NOX in the flue gas is reduced to nitrogen (N2) and water (H2O) using
ammonia (NH3) as a reductant. The reduction occurs in the presence of a catalyst at
reaction temperatures between 600 and 750 °F. SCR systems are typically based on one of
two designs:
• A hot-side, high-dust SCR where the SCR system is located between the economizer and
air preheater
• A cold-side, low-dust SCR where the SCR is typically located downstream of the air heater
and particulate control device
• In a variation of this design, the SCR system can be located further downstream, after the
flue gas desulfurization (FGD) system (often called a tail-end SCR system).
The CUECost algorithms estimate costs for hot-side, high-dust systems, because hot-side
systems have been used on most SCR applications (EPA, 1996).
An SCR system reduces NOX concentrations in the flue gas using ammonia as the reducing
agent in a series of gas-phase reactions in the presence of a catalyst to form nitrogen and
water. The chemical reactions for these reduction reactions are provided below:
4 NH3 + 4 NO + O2 -> 4 N2 + 6 H2O (Eq. B-7)
4 NH3 + 2 NO2 + O2 -> 3 N2 + 6 H2O (Eq. B-8)
Small fractions of the ammonia can also be oxidized to alternate forms of nitrogen oxides:
2 NH3 + 2 O2 -> N2O + 3 H2O (Eq. B-9)
Some of the residual ammonia will also react with trace concentrations of the sulfur oxides
in the flue gas in the reactions shown below.
NH3 + SO2 + 1/2 O2 + H2O -> NH4HSO4 (Eq. B-10)
2 NH3 + SO3 + H2O -> (NH4)2SO4 (Eq. B-ll)
The solids formed in this reaction can contribute to catalyst fouling and contamination of fly
ash.
43
-------
Appendix B
The key operating parameters that affect the performance and, consequently, the capital
and operating cost of SCR systems include the allowable NH3 slip emissions, the space
velocity, the NOX reduction efficiency, and the NH3/NOX molar ratio. For SCR systems, these
parameters are interrelated, and their values depend on the type of SCR application (high-
dust or tail-end) and the desired performance levels. Ammonia slip emissions are controlled
by the SCR system design. Typically SCR catalyst suppliers provide a guarantee of 2 ppm
over the catalyst life. Since the 2 ppm NH3 slip is guaranteed at the end of the catalyst's
life, the initial NH3 slip emissions will be very low (<1 ppm). For this reason, ammonia slip
does not affect the catalyst volume calculations in CUECost.
The space velocity is the primary parameter used to specify catalyst volume. If the user
does not input a value for space velocity, CUECost calculates space velocity based on the
NOX reduction efficiency and the NH3/NOX molar ratio. For SCR, NOX reduction efficiency can
range from approximately 60 to 95%, but systems are typically designed to achieve 70 to
90% removal. The NH3/NOX molar ratio generally ranges from about 0.7 to 1.0. Ammonia
can be injected at a greater than 1:1 stoichiometric ratio to increase NOX reduction
efficiency, but NH3slip would also increase significantly.
CUECost estimates capital costs for reactor housing, initial catalyst, ammonia storage and
injection system, flue gas handling including ductwork and induced draft fan modifications,
air preheater modifications and miscellaneous direct costs, including ash handling and water
treatment additions that typically are modified due to the increased concentrations of
ammonium salts in the collected fly ash.
Operating and maintenance costs include NH3, catalyst replacement and disposal, electricity,
steam, labor and maintenance costs. Annual catalyst replacement costs are based on the
catalyst life. For example, if the catalyst life is 3 years and there are three catalyst sections,
then one-third of the catalyst is replaced each year. The catalyst disposal cost reflects the
cost of disposing of the spent catalyst. A typical value of 48 Ib/ft3 was used for the catalyst
density to calculate the mass of the spent catalyst. Default input values for SCR are
presented in Table B-4. The default inputs were taken from EPA's ARD studies (EPA, 1996)
where available. Unit costs are escalated from 1995 dollars to 2004 dollars using Chemical
Engineering Magazine cost indices.
44
-------
Appendix B
Table B-4. Default Input Parameters for SCR
Description
Inlet NOX level
NH3/NOX Stoichiometric Ratio
NOX Reduction Efficiency
Space Velocity (Calculated if zero)
Time to first catalyst replenishment (years)
Ammonia Cost
Catalyst Cost
Solid Waste Disposal Cost
Maintenance (% of installed cost)
Retrofit Difficulty Factor
Contingency (% of installed cost)
General Facilities (% of installed cost)
Engineering Fees (% of installed cost)
Mercury Oxidation Rate - bituminous coal
(removal if downstream)
Mercury Oxidation Rate - subbituminous coal
Duration of Project (years)
Units
Ib/MMBtu
decimal
decimal
1/h
integer
$/ton
$/m3
$/ton
%
decimal
%
%
%
%
%
integer
Range
0.7-1.0
0.60-0.90
2-5
Default
calculated
0.9
0.90
0
3
400
5000
11.48
0.66%
1.5
15%
5%
10%
90.0%
0.0%
2
6.4.2 Selective Non-catalytic Reduction Design Criteria
The selective non-catalytic reduction (SNCR) process involves injection of a nitrogen-
bearing chemical (usually NH3 or urea) into boiler flue gases within a prescribed
temperature range (typically 1600 to 2000 °F). The NH3 or urea [CO(NH2)2] selectively
reacts with NOX in the flue gas to convert it to N2. For the NH3-based SNCR process, either
aqueous or anhydrous NH3 is injected into the flue gas where the temperature is between
1600 and 1900 °F. Most of the NH3 reacts with NO and oxygen in the gas stream to form N2
and H2O. For the CO(NH2)2-based SNCR process, an aqueous solution of CO(NH2)2 is
injected into the flue gas at one or more locations in the upper furnace and/or convective
pass. The CO(NH2)2 reacts with NOX in the flue gas to form N2, H2O, and carbon dioxide
(CO2). The chemical reactions for this conversion process are not well defined, consisting of
a series of dissociation reactions at the elevated gas temperatures in the boiler gas path.
The following summary equation describes the overall reaction that is occurring, while the
actual reaction mechanism is a long series of dissociation and chemical reactions between
various free radicals.
45
-------
Appendix B
Urea Reaction:
CO(NH2)2 + 2 NO + 1/2 O2 -> 2 N2 + CO2 + 2 H2O (Eq. B-12)
Ammonia Reactions:
4 NH3 + 4 NO + O2 -> 4 N2 + 6 H2O (Eq. B-13)
4 NH3 + 2 NO2 + O2 -> 3 N2 + 6 H2O (Eq. B-14)
CUECost allows the user to select either CO(NH2)2 or NH3 as the SNCR reagent. The user is
asked to specify the NOX reduction efficiency and the stoichiometric molar ratio of reagent
to NOX. SNCR can achieve NOx-reduction efficiencies ranging from 30 to 70%.
Approximately 50% reduction is typical. The SNCR process requires stoichiometric reagent-
to-NOx ratios of greater than 1:1 to achieve significant NOX removal. The ratio can range
from about 0.5 to 2.5, but will typically fall within the range of 1 to 2. The NH3 and
CO(NH2)2 injection rates are then calculated based on the stoichiometric ratio, inlet NOX and
boiler heat input.
For the CO(NH2)2-based SNCR process, the user chooses wall injectors, lances, or both. Wall
injectors are nozzles installed in the upper furnace waterwalls. In-furnace lances protrude
into the upper furnace or convective pass and allow better mixing of the reagent with the
flue gas. In-furnace lances require either an air- or water-cooling circulation system. If the
user enters values for both wall injectors and lances, then costs include both lances and wall
injectors. If wall injectors are to be used alone, then the user enters zero for both the
number of lance levels and the number of lances. Similarly, if lances are to be used alone,
the user enters zero for both the number of injector levels and the number of wall injectors.
CUECost uses input parameters for the number of injectors and lances unless the user
wants these parameters to be calculated from the number of levels. If the user inputs zero
for the number of injectors and also inputs the number of injector levels, CUECost will
calculate the number of injectors. Similarly, if the user inputs zero for the number of lances,
the number of lances will be calculated from the number of lance levels. For the NH3-based
SNCR process, the user can choose either steam or air as the atomizing medium. Based on
the user's choice, an annual operating cost for steam or electricity usage is calculated.
The main equipment areas in the battery limits for SNCR include the reagent receiving area,
storage tanks, and recirculation system; the injection system, including injectors, pumps,
valves, piping, and distribution system; the control system; and air compressors. In
addition, NH3-based SNCR systems use electrically powered vaporizers to vaporize the NH3
prior to injection.
Operating labor costs are based on two person-hours required per 8-hour shift of operation.
The annual cost of the reagent is the major operating cost item for the process and is
calculated as the product of the reagent usage in tons/year and the cost in dollars per ton of
46
-------
Appendix B
pure reagent. Electricity, water, and steam requirements are based on vendor information.
The cost of steam or air for atomization of reagent is included as an operating cost.
Default input values for SNCR are presented in Table B-5. The default inputs were taken
from studies by EPA's Acid Rain Division (EPA, 1996; EPA, 1997; EPA, 1998) where
available. Unit costs are escalated from 1995 dollars to 2004 dollars using Chemical
Engineering Magazine cost indices.
Table B-5. Default Input Parameters for SNCR
Description
Inlet NOX level
Reagent (l:Urea 2:Ammonia)
Number of Injector Levels
Number of Injectors
Number of Lance Levels
Number of Lances
Steam or Air Injection for Ammonia (1: Steam, 2: Air)
NOX Reduction Efficiency
NH3/NOX Stoichiometric Ratio
Urea/NOx Stoichiometric Ratio
Urea Cost
Ammonia Cost
Retrofit Factor
Maintenance (% of installed cost)
Contingency (% of installed cost)
General Facilities (% of installed cost)
Engineering Fees (% of installed cost)
Duration of Project (years)
Units
Ib/MMBtu
integer
integer
integer
integer
integer
integer
fraction
decimal
decimal
$/ton
$/ton
decimal
%
%
%
%
integer
Range
0.30-0.70
0.8-2.0
0.8-2.0
Default
calculated
1
3
18
0
0
2
0.50
1.2
1.2
400
300
1.3
1.5%
15%
5%
10%
1
6.4.3 Natural Gas Reburning Design Criteria
Natural gas reburning (NGR) involves substituting natural gas for a portion of the pulverized
coal supplied to the primary combustion zone and injecting the natural gas downstream of
the primary combustion zone to form a reducing zone in which NOX compounds are reduced
to N2. Combustion air for the reburning fuel (natural gas) is injected further downstream.
Because the main combustion zone of furnaces employing this technology operates in its
normal manner, gas reburning is applicable to a wide range of wall-, tangential-, and
cyclone-fired boilers.
47
-------
Appendix B
Boiler modifications for gas reburning involve installation of additional fuel injectors and
associated piping and control valves. In the burnout zone, key components include overfire
air (OFA) ports, a windbox, ductwork, and control dampers. Installation of the gas injectors
and OFA ports requires waterwall modifications. Adequate residence time must be available
both in the reburn zone and the burnout zone to maximize NOX reduction and to minimize
unburned carbon losses. Consequently, for retrofit applications, adequate space between
the top burner row and the furnace exit must be available for appropriately locating the
reburn fuel injectors and OFA ports.
The fraction of boiler heat input contributed by natural gas combustion (reburn fraction)
depends on the desired NOX removal efficiency. The relationship between the reburn fraction
and NOX reduction efficiency applies for NOX reduction efficiencies from 55 to 65% and
corresponding reburn fractions from 0.08 to 0.20. In CUECost, these are the valid input
ranges for the NOX removal efficiency and reburn fraction. If the user inputs both
parameters within the valid ranges, the input values are used for cost calculations. If only
one parameter is outside the valid range, that parameter is calculated using the other
parameter. If both input values are outside of the valid ranges, a default reburn fraction of
0.15 is used with a corresponding 61% NOX removal efficiency. The installed costs of gas
injectors, OFA ports, and related equipment are included in the NGR cost worksheet. Also
included in the NGR cost worksheet is the cost associated with piping natural gas to the
boiler from the metering station located at the utility plant fence-line.
In general, natural gas reburning reduces the boiler operating costs associated with coal-
and ash-handling process areas, including maintenance, electricity, and ash disposal. Fuel
costs are generally higher, because the price of natural gas is typically higher than the price
of coal. Maintenance costs for operating the NGR system are estimated at 2% of the total
plant cost, plus a maintenance credit for operating the coal handling process at reduced coal
feed rates. Savings from reduced fly ash disposal are estimated only for retrofit
applications. The incremental fuel cost for firing gas is estimated by multiplying the amount
of gas burned by the fuel price difference between gas and coal. Default values for NGR
input parameters are presented in Table B-6. The default inputs were taken from ARD
studies (EPA, 1996) where available. Unit costs are escalated using Chemical Engineering
Magazine cost indices.
48
-------
Appendix B
Table B-6. Default Input Parameters for NGR
Description
Uncontrolled NOX level
NOX Reduction Efficiency
Gas Reburn Fraction
Waste Disposal Cost
Natural Gas Cost
Retrofit Factor
Maintenance (% of installed cost)
Contingency (% of installed cost)
General Facilities (% of installed cost)
Engineering Fees (% of installed cost)
Duration of Project (years)
Units
Ib/MMBtu
fraction
fraction
$/ton
$/MMBtu
%
%
%
%
integer
Range
0.55-0.65
0.08 - 0.20
Default
calculated
0.61
0.15
11.48
9.00
1.30
1.5%
15%
2%
10%
1
6.4.4 Low-NOx Burner Technology Design Criteria
Low-NOx burner technology (LNBT) limits NOX formation by controlling both the
stoichiometric and temperature profiles of the combustion process in each burner flame
envelope. This control is achieved with design features that regulate the aerodynamic
distribution and mixing of the fuel and air, yielding one or more of the following conditions:
• Reduced O2 in the primary combustion zone, which limits fuel NOX formation;
• Reduced flame temperature, which limits thermal NOX formation; and
• Reduced residence time at peak temperature, which limits thermal NOX formation.
Low NOX burner designs for wall-fired boilers can be divided into two general categories:
"delayed combustion" and "internally staged." Delayed combustion LNBT is designed to
decrease flame turbulence (thus delaying fuel/air mixing) in the primary combustion zone,
thereby establishing a fuel-rich condition in the initial stages of combustion. Internally
staged LNBT is designed to create stratified fuel-rich and fuel-lean conditions in or near the
burner. In the fuel-rich regions, combustion occurs under reducing conditions, promoting
the conversion of fuel nitrogen to N2 and inhibiting fuel NOX formation. In the fuel-lean
regions, combustion is completed at lower temperatures, thus inhibiting thermal NOX
formation.
Conventional tangentially-fired boilers consist of corner-mounted vertical burner assemblies
from which fuel and air are injected into the furnace. The fuel and air nozzles are directed
tangent to an imaginary circle in the center of the furnace, generating a rotating fireball in
the center of the boiler. Each corner has its own windbox that supplies primary air through
49
-------
Appendix B
the air compartments located above and below each fuel compartment. For tangentially-
fired boilers, LNBT changes the air flow through the windbox by decreasing the amount of
primary air and directing secondary air away from the fireball and toward the furnace wall.
Default input parameters for LNBT and suggested ranges are presented in Table B-7. The
user selects the boiler type and the retrofit difficulty. CUECost calculates total capital cost as
a function of boiler size. The NOX reduction efficiency input does not affect the capital cost
estimate, but is used to estimate emissions reduction.
Table B-7. Default Values for LNBT Input Parameters
Description
Uncontrolled NOX level
Boiler Type (T:T-fired, W:Wall)
Burner Type
1 = LNBor LNC1,
2 = LNBand OFA or LNC2,
3=LNC3
Retrofit Difficulty Factor
General Facilities
Engineering
Contingency
Duration of Project (years)
Units
Ib/MMBtu
letter
integer
number
percent
percent
percent
integer
Range
Default
calculated
T
1
1.3
5.0%
10.0%
15.0%
1
B.5 Hg CONTROL TECHNOLOGY CRITERIA
Injection of powdered activated carbon (PAC) has been developed and tested at full scale on
coal-fired utility boilers. Test programs have been performed on a utility boiler firing
subbituminous coal with a downstream cold-side ESP, on utility boilers firing bituminous
coal with a downstream cold-side ESP, and firing bituminous coal with a compact hybrid
particle collector (COHPAC) arrangement (upstream hot-side ESP with downstream
baghouse after the air preheater). Performance models were developed.
B.5.1 Mercury Removal Models
EPA's Information Collection Request (ICR)4 showed that mercury released from coal
combustion may be partly removed from the exhaust gases by existing equipment without
additional retrofit technology. The existing equipment may be one or more pieces of
equipment that contribute to mercury removal.
'Available at http://www.epa.gov/icr/
50
-------
Appendix B
If /equipment is equal to the fraction of mercury removed from the boiler gases by a piece of
equipment, then (1 - /"equipment) equals the fraction of mercury remaining in the gases after
that piece of equipment. The fraction of mercury remaining after n pieces of equipment is
equal to
Fraction of mercury remaining after n pieces of equipment =
L\l ~~ 'equipment lj X (. 1 "'equipment 2) X 11 ~~ 'equipment 3) X . . . X ^1 — requjpment nJJ l^q. D-ljJ
Therefore, the total mercury removal fraction = /Votai
/Total = 1 ~ [(1 ~ /equipment!.) X (l~/equipment 2) X (1 ~ /equipments) X ... X (1 - /equipment n)] (Eq. B-16)
If one of the pieces of equipment is PAC injection, then the total mercury removal fraction =
/Total =
1 — [(1 — 'equipment l) X (1-/equipment 2) X (1 — /equipment 3) X ... X (1 — / PAC injection) X ... X (1 — /equipment n)] C^Q1
where
f PAC injection is the fraction of mercury removed by PAC injection.
If PAC injection is simply added to existing equipment and the removal effects of the
existing equipment are combined into one term, then we can represent Eq. B-17 as
/Total = 1 ~ [(1 ~ /existing equipment) x (1 - fpAC injection)] (Eq. B-18)
where
/"existing equipment is the removal fraction of the existing equipment.
In this effort, data from full-scale tests of mercury reduction were used to formulate models
for mercury reduction from existing equipment and from PAC injection. Full-scale data for
mercury removal by existing equipment are available from the ICR data. Full-scale testing
results for mercury reduction from PAC injection are available from the Department of
Energy's field testing programs at Southern Company's Gaston plant, Wisconsin Electric
Power Company's Pleasant Prairie power plant (PPPP) and at PG&E National Generating
Group's Brayton Point and Salem Harbor plants.
B.5.2 Mercury Removal by Existing Equipment, fexisting equipment
Through statistical analysis of the ICR data, EPRI (2000) shows that mercury reduction is a
function of both emission control equipment configuration and a function of chlorine content
of the coal, and in some cases a function of the SO2 emissions level from the boiler. EPRI
(2000) provides algorithms to estimate mercury capture as a function of the plant
configuration, the coal chlorine content, and the SO2 emissions. These algorithms are:
51
-------
Appendix B
Algorithm 1 (cold-side ESP):
Existing equipment = Ci x In [(coal Cl, ppm)/(SO2, in Ib/MMBtu)] + C2
where
Ci and C2 = Algorithm 1 constants
Algorithm 2 (all other categories):
(Eq. B-19)
xisting equipment
= Ci X In (C03l CI, ppm) + C2
(Eq. B-20)
where
Ci and C2 = Algorithm 2 constants
Minimum and maximum allowable values are set for the results of Equations B-19 and B-20.
Values of Ci and C2, minimum and maximum are shown in the left columns in Table B-8 for
hot and cold side ESP operating conditions.
According to Eqs. B-19 and B-20, the predicted mercury reduction efficiencies for conditions
at Gaston (Bustard et al., 2001; Durham et al., 2001; and Bustard et al., 2002), Pleasant
Prairie power plant (PPPP) (Bustard et al., 2001; Durham et al., 2001), Brayton Point
(Durham et al., 2001) and at Salem Harbor (Durham et al., 2001) are presented in Table
B-8.
Table B-8. Predicted Collection of Mercury by ESP according to Eqs. B-19 and B-20
ESPc
ESPh
Chlorine, % by weight in coal
Coal Chlorine, ppm
Flue Gas SO2, Ib/MMBtu
Ci
0.1233
0.0927
C2
-0.3885
-0.4024
Min
0.0%
0.0%
Max
55.0%
27.0%
Gaston
0.03
300
0.650
PPPP
0.0015
15
0.360
Brayton
Point
0.08
800
0.820
Salem
Harbor
0.03
300
0.500
Predicted Mercury Reduction
12.6%
7.1%
46.0%
40.0%
ESPc = cold-side ESP
ESPh = hot-side ESP
Source: EPRI (2000)
Gaston fires bituminous coal and has a hot-side ESP followed by an air preheater and then a
low-pressure pulse-jet FF for a COHPAC arrangement (Bustard et al., 2001; Durham et al.,
2001; and Bustard et al., 2002). EPRI (2000) did not include algorithms for facilities with
this arrangement. One might expect that the mercury reduction without PAC might
52
-------
Appendix B
correspond approximately to the predicted mercury reduction in Table B-8 for a hot-side
ESP (ESPh). Under the conditions at Gaston, predicted mercury reduction equals 12.6%.
However, tests at Gaston showed negligible mercury removal. But considering the range of
variability in the possible results, the difference may be reasonable. However, this example
demonstrates that this algorithm will not give precise values, but reasonable estimates.
At the Pleasant Prairie power plant (PPPP) (Starns et al., 2002), a facility firing PRB coal
with a cold-side ESP (ESPc), the test results showed about 5% actual mercury removal from
existing equipment compared to about 7% as estimated by the algorithm of (EPRI 2000) for
the conditions at PPPP, and shown in Table B-8 (Bustard et al., 2001; Durham et al., 2001).
Therefore, the value estimated by the algorithm is approximately in the same range. The 15
ppm chlorine content of the coal used at PPPP (which is much lower than that of most other
PRB sites) probably contributes to the low removal by the existing equipment. With chlorine
content more typical of a PRB coal, around 100 ppm or more, the algorithm predicts that
mercury would be reduced by a greater amount.
For Brayton Point, a facility firing bituminous coal and equipped with an ESPc, the algorithm
of EPRI (2000) produces an estimated mercury reduction by existing equipment of about
46% (see Table B-8) versus an actual measured removal efficiency of 32% (Durham et al.,
2001). These values, which are in about the same range, further illustrate that the
algorithm of EPRI (2000) is not exact, but approximate, at estimating mercury removal by
existing equipment.
At Salem Harbor, a facility firing bituminous coal and equipped with an ESPc, 87% mercury
reduction from existing equipment was measured (Durham et al., 2001). This measured
value compares to about 40% estimated from the algorithm of EPRI (2000) as shown in
Figure B-l. The significant difference can be explained as follows: First, Salem Harbor
operates with fly ash loss on ignition (LOI) in the range of 25-35%. According to Bustard et
al. (2001), this fly ash loss is approximately equivalent to a carbon loading of 60-84
Ib/MMacf in the exhaust stream. This value is higher than a typical plant's inject rate. So,
the carbon present in the fly ash has likely contributed to a very high intrinsic capture of
mercury. Second, temperature plays a role in intrinsic mercury capture. Because Salem
Harbor has the ability to increase its ESP inlet temperature through operation of steam
heaters, parametric tests of intrinsic mercury removal as a function of temperature could be
performed. Figure B-l shows the results of that testing under various firing conditions and
also with data taken from another test using low sulfur bituminous coal (not the baseline
coal). The trend is quite clear that increasing temperature reduces intrinsic mercury capture
from around 90% down to around 10%. Thus, mercury absorption by fly ash is enhanced
when flue gas is cooled. Cooling the flue gas can enhances mercury uptake by flash.
However, when PAC is injected, its large capacity for mercury absorption allows the sorbent
to be operated at temperatures of 350 °F or higher. As such, spray cooling usually promotes
little or nearly zero mercury absorption by PAC.
53
-------
Appendix B
Because a facility's mercury reduction by existing equipment may be significantly different
from what the algorithm of EPRI (2000) determines, this algorithm should be used with care
and only for making approximate estimates. As the measurements at Salem Harbor clearly
indicate, LOI or other ash qualities and gas temperature can have a very significant impact
on the level of mercury being removed by existing equipment and may be worth including
as parameters in this algorithm at some future date when more information is available.
Therefore, the algorithm of Equation B-20 and EPRI (2000) may provide reasonable
estimates in many cases. But there is a chance that actual mercury capture may differ
significantly from what Equation B-20 predicts. For any specific facility, actual
measurements of mercury removal, if available, should be used.
100!
90
80
70
>
SS 60
40
30
20
10
270
0A
O17-19% LOI (45 Ib/M Macf)
X 20-24% LOI (55 Ib/M Macf)
A 25-29% LOI (68 Ib/M Macf)
• >30%LOI
H30-35% LOI, C1, High Load
A 21-27% LOI, C2, High Load
<>LS bitum coal
290
310 330
Temperature (degrees F)
350
370
Figure B-l. Salem Harbor Mercury Removal without PAC Injection (Durham et al., 2001)
6.5.3 Mercury Reduction by PAC injection, fPAC injection
EPA (2000) has algorithms developed from pilot-scale data for mercury reduction on boilers
equipped with PAC injection. In this work, we have made the following model
improvements:
1. The algorithms of EPA (2000) were developed from pilot-scale tests and characterize
total mercury reduction from both PAC injection and from existing equipment as a
function of PAC injection concentration. When using the algorithms of EPA (2000), it is
necessary to have a different PAC injection algorithm for each type of equipment
configuration, including upstream equipment. These PAC injection algorithms may have
to be updated as new information regarding mercury control from existing equipment
54
-------
Appendix B
becomes available. In the effort described in this paper, the mercury reduction from PAC
injection was isolated from that of the other equipment. Therefore, as we gain more
information on reduction of mercury from equipment other than PAC injection, it should
not be necessary to perform new regressions on the PAC injection models and it will also
be possible to assess the fate of mercury in equipment that is either upstream or
downstream of the PAC injection system.
2. The algorithms of EPA (2000) are of a form where it is possible for Hg removal to
approach 100% by injection of very high concentrations of PAC. As will be shown,
experience at PPPP showed that under some circumstances it is not possible to achieve
such extremely high reduction of mercury emissions with PAC injection. Therefore, the
algorithm for mercury reduction from PAC injection was modified to permit an upper
limit to mercury removal that may be less than 100%.
3. Because the algorithms of EPRI (2000) are based on the full-scale ICR data, it is
desirable to use them to characterize mercury reduction from existing equipment.
However, it is not possible to integrate the algorithms of EPRI (2000) into the approach
used in EPA (2000). By treating the mercury reduction from PAC injection independently
from mercury reduction from other equipment, it is possible to use the algorithms of
EPRI (2000) to characterize mercury reduction from existing equipment.
In the case of PPPP, PAC injection test results demonstrated that mercury reduction
behaved asymptotically with a maximum achievable mercury reduction from PAC that is well
below 100%, regardless of PAC injection rate. For this reason, the equation that is used in
EPA (2000) to characterize the relationship between mercury reduction and PAC injection
% Hg reduction = n. = 100 x ffrom PAC injection = 100-[A/(M+B)/VC] (Eq. B-21)
where
M = the mass injection rate of PAC (in Ib/MMacf) and A, B, C are curve-fit constants
determined with available data.
% Hg reduction = n. = 100 x ffrom PAC injection = 100 x D-[A/(M + B)/VC] (Eq. B-22)
where
D = the asymptotic fraction of mercury reduction that is approached but is not achieved.
Constants A, B, C, and D appearing in Eq. 8 are specified for a given plant configuration and
gas temperature. At this time, these constants can only be developed for full-scale
applications similar to the conditions where full-scale data exists. For some other boiler
configurations there is test data available from pilot-scale (Bustard et al., 2001) tests that
can be used until full-scale data becomes available. For other configurations where neither
full-scale nor pilot-scale data exists, the constants can be developed as data becomes
available from future tests. The constants A, B, C and D used in CUECost are listed in Table
B-9.
55
-------
Appendix B
Table B-9. Values of Constants Used in the PAC Injection Eqs. B-21 and B-22
Coal
Bituminous
Bituminous
Bituminous
Bituminous
Bituminous
Bituminous
Subbituminous
Subbituminous
Subbituminous
Subbituminous
Subbituminous
Subbituminous
FF retrofit
in flight
in flight
in flight
FF
FF
FF
in flight
in flight
in flight
FF
FF
FF
PAC Capacity
EPAC
PAC
other
EPAC
PAC
other
EPAC
PAC
other
EPAC
PAC
other
A
0
-0.6647
0
1.6944
0.8837
3.308
0
-0.4318
B
1.207
2.1232
2.5007
-1.1267
0.4485
0.754
2.5007
1.9551
C
-0.2277
-0.0665
-2.2097
-0.0009
-0.575
-0.5925
-2.2097
-0.8937
D
100%
100%
100%
100%
100%
70%
100%
100%
6.5.4 PAC Injection Models Developed from Full-Scale Data
For the purpose of modeling, we are interested in estimating the necessary PAC injection
rate to achieve a specified level of mercury control. Therefore, we developed algorithms of
PAC injection rate as a function of desired mercury reduction by PAC. So, rather than
plotting mercury reduction versus PAC injection concentration, as is done in Bustard et al.
(2001; 2002) and Starns et al. (2002), we have reversed the axes from those shown in
these references.
In these tests (Bustard et al., 2001; Bustard et al., 2002; Starns, 2002) several different
PAC sorbents were tested. The different PAC sorbents will be designated on the legends of
the figures. In this effort, we did not have specific information regarding the sorbent
properties of the tested sorbents. Therefore, since we did not want to conjecture on the role
of particular sorbent properties on mercury removal performance, we did not evaluate
sorbent choice effects except to determine whether or not sorbent type has an effect on
performance under a particular condition.
Gaston
Figure B-2a shows mercury collection results measured from an on-line mercury analyzer
during testing conducted at Gaston. Data are plotted as PAC injection concentration versus
mercury reduction percent. Data include results obtained with several different sorbent
types (Bustard et al., 2001). Figure B-2a also shows a curve developed in the form of
Equation B-22 to approximately correspond to the results achieved at Gaston. The
coefficients for the algorithm are listed in Table B-10. At Gaston the choice of sorbent
appeared to have little or no impact on performance. At mercury removal rates in the range
56
-------
Appendix B
of 92-96%, mercury reduction is less sensitive to changes in PAC injection rate. Figure B-2b
shows the data for 92-96% mercury reduction in greater detail. The enclosed region on
Figure B-2b includes the estimated 95% confidence range for these mercury reduction
data.5 Figure B-3, a plot of deviation of the predicted and measured PAC injection rate,6
demonstrates this trend in another way. For most mercury reduction levels, the deviation
between model and actual PAC injection rates is only about 10%. For mercury reduction in
excess of 90%, however, the deviation is higher on a percentage basis. While Figure B-3
shows that at high removal rates the deviation between the model and measured value
expressed as a percent of predicted level is -30% to +40%, in fact this range of values only
corresponds to a range of under ± 1 Ib/MMacf. The high percentage of the deviation is due
to the actual values being relatively small at Gaston.
4.5
4.0 -
3.5 -
=. 3.0 -
c
g
"TO 2.^.
§ 2.0 -
o
<-> 1.5 -j
O
tS 1.0
0)
I?
0.5 H
0.0
R-squared of algorithm 68%
95% confidence area for all data
over 90% reduction of mercury.
0 10 20 30 40 50 60 70 80 90 100 110
Percent Hg Removal
Data provided by Jean Bustard, ADA Environmental Services, September 16, 2002
Figure B-2a. Gaston Testing
5 95% confidence range for Hg reduction and for PAC injection concentration are determined by ±2 standard
deviations from the arithmetic mean, with correction for sample size.
6 Calculated as (actual rate - predicted rate) / predicted rate and expressed in percent.
57
-------
Appendix B
Table B-10. Coefficients for Curve Fit Algorithms
Plant
Gaston
PPPP
Brayton
Point
Algorithm
a
b
c
a
b
c
A
53
150
140
145
300
300
300
B
0.1
5
1
3
3
0
1.5
C
2
1
1
1
0.8
0.8
0.8
D
1.00
0.72
0.69
0.705
1.13
1.05
1.09
Sorbent Type
FGD, PAC 20
FGD
(14 micron
fraction)
FGL
Insul
FGD and FGD I
FGL
Insul
Note 1: Algorithms were developed for a specific plant and a specific sorbent.
Note 2: A, B, C, and D are coefficients for curve fit.
0.5 -
0.0
Includes the 95% confidence area for
all data over 90% reduction of
mercury.
90
91
92
98
99 100
93 94 95 96 97
Percent Hg Removal
Data provided by Jean Bustard, ADA Environmental Services, September 16, 2002
Figure B-2b. Gaston Testing
58
-------
Appendix B
50%
40% -
30%
deviation = (actual PAC rate minus predicted PAC rate) divided by predicted PAC rate
2 20%
I 10%
£
Q.
® 0%
0) ;
I -10% -
I -20% -
ra
'>
° -30% -
-40% -
40
50
60
70
80
90
100
Percent Hg Reduction from PAC
Figure B-3. Deviation of the Gaston PAC Algorithm
Pleasant Prairie Power Plant
Figure B-4 shows mercury collection results measured from an on-line mercury analyzer
during testing conducted at PPPP. Data include results with several different sorbent types.7
Figure B-4 also shows a data point for the total mercury removal as measured by the
Ontario Hydro method. The Ontario Hydro method shows a somewhat higher, but
nevertheless a similar, mercury removal as the on-line mercury analyzer used for the
testing. Two curves were developed in the form of Equation 9 to correspond to specific sets
of data and are plotted on Figure B-4. The coefficients of these algorithms (A, B, C, D) are
listed in Table B-10. Unlike the results at Gaston, at PPPP the choice of sorbent has a
significant effect, possibly a result of the fact that at Gaston there is a downstream fabric
filter, which provides improved sorbent-gas contact, while at PPPP all of the mercury
absorption had to occur in the duct. Figure B-5 is a plot of deviation of the predicted and
measured PAC injection rate.8 Had one algorithm been used for all of the sorbents, the
deviations would have been very high in some cases. Nevertheless there is enough scatter
in some of the data that, even with different algorithms for each sorbent, deviation can be
on the order of 40%. Note that the one data point with very high percent deviation (over
70%) was actually at a low removal rate and the absolute difference between the algorithm
results and measured results was quite small. For other plants with conditions similar to
those at PPPP (sub-bituminous coal), some consideration should be made for the sorbent
type.
7 Data provided by Jean Bustard, ADA Environmental Services, September 16, 2002.
8 Calculated as (actual rate - predicted rate) / predicted rate.
59
-------
Appendix B
45
40
35
30
25
20
I
o
o
I
'•B
0)
15 -
10 -
5 -
0
R-squared of algorithms
algorithm b and FGL
99%
algorithm c and Insul (7 micron)
98%
algorithm a and FGD (14 micron) data
85%
0
* FGD
• FGD (14 microns)
A FGL
X Insul (7 micron)
—CD— algorithm a
—O—algorithm b
^^"algorithm c
+ Ontario Hydro
excluded from r-
squared calculation
20 40 60
% Hg removal from PAC injection
Data provided by Jean Bustard, ADA Environmental Services, September 16, 2002
Figure B-4. PPPP Testing
deviation = (actual PAC rate minus predicted PAC rate) divided by predicted PAC rate
o
0.
I
O
1
Q.
•s
60%
40%
20%
0% -
20%
4CIOA -
the absolute difference was under
X
0 35 4
0.5lb/MMacf
XFGD (14 microns), algorithm a
X Insul (7 micron)
i
i i
i i
i
i
v
X
>
A
<
A
A X V
I £—1 1 I /
0 45 50 55
X '
60X
6
X
1 1 1
1 1 1
1 1 1
5 7
Percent Hg Reduction from PAC
Figure B-5. Deviation from the PPPP PAC Algorithm
60
-------
Appendix B
Bravton Point
Figure B-6 shows results of testing at Brayton Point. Data include results with several
different sorbent types. 7 Figure B-6 also shows curves developed in the form of Equation 9
that correspond to specific sorbent types. The coefficients of these algorithms are listed in
Table B-10. Like PPPP and unlike the results at Gaston, at Brayton Point the choice of
sorbent appears to have a significant effect. When considered with the PPPP results, this set
of results provides further evidence that the sorbent choice may have a greater impact
when a downstream fabric filter is not installed. While good correlation is possible for all
data with algorithm c (R2 = 77%), improved correlation was possible by using different
correlations for different sorbents, as demonstrated by the higher correlations of algorithms
a and b with the sorbents indicated on Figure B-6. Figure B-7 shows that the predictive
accuracy of the algorithms across a broad mercury removal range does not change much.
However, Figure B-7 shows that improved accuracy will result if the algorithm is tailored to
the sorbent. For algorithm c, maximum deviation ranges from -60% to +50%. But, by
tailoring the algorithm to the sorbent, as shown for Alt Sorbent 1 with algorithm b and FGD1
with algorithm a, the deviation is reduced sharply.
30
c-
u
ro
25 -
5 20 --
o
R-squared of algorithm
algorithm c and all data
77%
algorithm b and Alt. Sorbent 1 data
86%
algorithm a and FGD, FGD1 & FGD (SOS
off) 85%
~ 15
u
o
O
o
= 5
10
* FGD1
• Alt. Sorbent 1
A FGD (SOS off)
X FGD
-O— algorithm a
-O—algorithm b
^"algorithm c
20 40 60
% Hg removal from PAC injection
80
100
Figure B-6. Brayton Point Testing
61
-------
Appendix B
deviation = (actual PAC rate minus predicted PAC rate) divided by predicted PAC rate
(1)
•s
o
0.
S
1
Q.
•5
C
-------
Appendix B
• The space velocity of the catalyst
• The temperature of the reaction
• The concentration of ammonia
• The age of the catalyst
• The concentration of HCI in the flue gas stream.
Bustard et al. (2001) showed that, in tests on a laboratory combustor, mercury oxidation
without a catalyst was enhanced with higher Cl concentration (higher HCI in the flue gas)
and that oxidation increased with residence time and at lower temperatures, as shown in
Figure B-8. Hocquel et al. (2002) also describe the results of laboratory tests of oxidation of
mercury across SCR catalysts. The results of these tests, shown in Figure B-9,
demonstrated that the catalyst significantly increased the amount of mercury that oxidized
to mercuric chloride.
In Richardson et al. (2002), tests of mercury oxidation by SCR catalyst were conducted
using simulated flue gas and slip-streams from actual units. Results showed similar trends
for both simulated flue gas and slip-streams from actual units with the exception that the
effect of increasing space velocity appeared somewhat more significant with the slip-
streams. Multiple catalyst types were tested with similar results obtained. According to
Bustard et al. (2001), at space velocities in the range of 400 h"1, mercury oxidation was in
the range of about 80% to 90% for fresh catalyst. However, the oxidation rate falls off
quickly with increased space velocity; oxidation might be in the range of 30-80% at a space
velocity of 4000 h"1. The wide range of oxidation performance at a space velocity of 4000 h"1
is the result of the influence of other factors - temperature, ammonia and possibly other
effects.
As shown in Figure B-10, Richardson et al. (2002) showed that oxidation of mercury across
fresh SCR catalyst was highest at temperatures in the range of 550 °F and lowest in the
range of 800 °F, consistent with the fact that oxidation of mercury to mercuric chloride
occurs mostly at lower temperatures.
63
-------
Appendix B
100
1200
2345
Retention time [s]
400
Figure B-8. Mercury Oxidation without a Catalyst as a Function of Residence Time, Gas
Temperature, and HCI Content (Hocquel et al., 2002)
i1 — 35-cells
Is
« 2 8Q plate-type
c o
CD
'S
0
I
(O
60
40
20
22-ceils
without catalyst
f
510 540 570
600 630
Temperature [K]
660
690
720
Figure B-9. Mercury Oxidation across SCR Catalysts and without SCR Catalyst (Hocquel et
al., 2002)
64
-------
Appendix B
*- 7WF; o»mNH3 I
O- TOO'F:
-*— «ST:
«• 7eOf,!)pp«>NH3 (ESP hint)
aOOQ 19000
Sp.ce Veloclly (hr")
Figure B-10. Oxidation of Mercury across C-l SCR Catalyst in PRB-derived Flue Gas
(Richardson et al., 2002)
The presence of ammonia, which is the NOX reducing reagent normally used in SCR
systems, was shown by Richardson et al. (2002) to inhibit the oxidation of elemental
mercury. This effect is most pronounced with catalyst that has been exposed to boiler
exhaust gases for a number of months. As shown in Figure B-ll, mercury oxidation without
ammonia present remained between 80% and 90% after 4200 hours (about six months) of
exposure to boiler gases at a space velocity of 1450 IT1. When exposed to 300 ppm of
ammonia, fresh catalyst continued to oxidize 80-90% of the elemental mercury. However,
after 4200 hours of exposure no oxidation was measured across the catalyst when ammonia
was present.
wo i,0DQ i.soo 2,000 2,500
Tim* (hrtj
3,0DQ 3,500 4,000
Figure B-ll. Effect of Flue Gas Exposure Time on C-l SCR Catalyst Oxidation of Elemental
Mercury: 700 °F and Space Velocity of 1,450 h'1 (Richardson et al., 2002)
65
-------
Appendix B
Oxidation of elemental mercury to mercuric chloride across an SCR catalyst, therefore, may
be a function of: space velocity, temperature, ammonia concentration, and catalyst life.
Other factors, such as fly ash characteristics, are also believed to play a role.
Bustard et al. (2002) describe the results of a program that evaluated mercury oxidation
across full-scale utility boiler SCR systems. A summary of the results of these tests is shown
in the first four entries in Table B-ll. Testing was performed at four coal-fired electric utility
plants having catalyst age ranging from around 2500 hours to about 8000 hours. One plant
fired subbituminous coal and three other plants fired eastern bituminous coal. The test
results showed high levels of mercury oxidation in two of the three plants firing eastern
bituminous coal and insignificant oxidation at the other two plants (one firing bituminous
coal and the other subbituminous). However, for both of the plants where little or no
mercury oxidation was measured (SI and S3) over 85% of the mercury at the particle
control device inlet was already in the non-elemental (oxidized) form. For the one
bituminous coal fired plant with low mercury oxidation (S3), over 50% of the mercury at the
SCR inlet was already in the oxidized form. At the plant firing subbituminous coal (SI),
mercury oxidation was fairly low. But, due to the high carbon in that plant's fly ash, the
elemental mercury was apparently adsorbed onto the ash, resulting in high particulate
mercury levels. Finally, in contrast with the studies of Hocquel et al. (2002) and Richardson
et al. (2002), ammonia appeared to have little or no effect on mercury oxidation on these
actual, full-scale facilities.
Subsequent tests on sister units at those plants and at other plants are shown in the second
four entries in Table B-ll. All of the units fired bituminous coal and showed that mercury
oxidation was generally enhanced to high levels of oxidized mercury at the SCR outlet. In
each case where a scrubber was installed the mercury removal was high. For the unit with
an ESP and no scrubber, mercury removal was not improved by the SCR.
At this point in time, the understanding of the effects of SCR catalyst on mercury oxidation
is fairly limited. Clearly, mercury oxidation is substantial under some conditions, but less
significant under others. However, significant mercury oxidation by SCR catalyst appears to
occur with bituminous coal and oxidation may be less certain with PRB coals. Where
bituminous coal was fired with an SCR and an FGD, high levels of mercury removal
generally occurred.
Default values for mercury control input parameters are shown in Table B-12.
66
-------
Appendix B
Table B-ll. Summary of Results from Full-Scale SCR Mercury Oxidation Tests (Bustard et al., 2001)
Power Plant
S1.650MW
gross
Cyclone, ESP
S2, 1360 MW
gross, Wall,
ESP+FGD
(MEL)
S3, 750 MW
gross,
Tangential, ESP
S4, 704 MW
gross, Cyclone,
scrubber
684 MW gross,
Wall, ESP+FGD
800 MW gross,
Tangential, ESP
1360MW
gross, Wall,
ESP+FGD
(MEL)
Cyclone, Lime
venturi scrubber
Catalyst Vendor,
Type,
SV (h-1)
Cormetech
Honeycomb
1800
Westinghouse
Phtp
nZD
KWH
Honeycomb
-3930
Cormetech
Honeycomb
2275
Halder-Topsoe
"corrugated"
-3750"?
Cormetech
Honeycomb
3800
C&l Ceramics
Plate
2125
Honeycomb
22/5
Catalyst
Age
8000 h
3.5
months
1 ozone
season
1 ozone
2 months
2 seas
2 layers
repl. after
1st season
2 ozone
seasons
2 ozone
seasons
Coal
Type
PRB
OH Bit
PA Bit
blend
KYBit
PAAA/V Bit
KYAA/V Bit
OH Bit
KYBit
Sin
Coal
(%)
0.2
3.9
1.7
2.9
3.6
1
3.9
3.1
Cl in Coal
(ppm)
<60
1640
1150
360
470
1000
520
750 bypass
250 w/SCR
NH3 Slip
(ppm)
2*
0.1
0.8
0.2
0.3
0.1
0.5
0.1
S03
ppm
0.4*
33*
24
16#
10.6
14
30
12
Cl
ppm
1.5*
108*
81 #
19
Not
Measu
red
(NM)
NM
NM
NM
Oxidized mercury
content,
SCR in/out
Unit 2: 8% -18%; net
10%; small increase;
1 OH sample
48% -91%; net 43%;
significant increase;
2 OH samples
No effect of alkali
injection (Unit 1)
55% -65%; net 10%;
small increase; 2 OH
samples
35% -61%; net 26%; for
2nd coal in sister unit; 2
OH samples
9% -80%; net 71%;
significant increase; 2
OH samples
Oxidation to 80+%; Net
+38% increase
Oxidation to 80+%; Net
+21% increase
Oxidation to 80+%; Net
+33% increase
Oxidation to 60+%; Net
+20% increase; "More"
oxidation if 1 outlier data
not used
Oxidized mercury
content, w/o and w/
SCR, PM inlet
5% -8%; net 3%;
small increase;
1 OH sample each
73% -97%; net 24%;
significant increase;
2 OH samples each
77% -67%; net -10%;
possible filter effects
due to reactive ash
Not tested in 2nd
coal/sister unit
56% -87%; net 31%;
significant increase; 2
OH samples each
Oxidation to 95%; Net
+15%
(using data from sister
unit w/o SCR)
Oxidation to 89%' Net
-0%
(using data from sister
unit w/o SCR)
Oxidation to 95+%;
Did not test w/o SCR
Oxidation to 90+%;
Net +39%
Cl in coal changed
between tests
Total Hg Removal
across PM+FGD, w/o
and w/ SCR
60% -65%; net 5%;
small increase- within
experimental error; 1
OH sample each
51% -88%; net 37%;
significant increase;
FGD removed 94% of
oxidized Hg; 2 OH
samples each
16% -13%; net -3%;
within experimental
error; 2 OH samples
each
Not tested in 2nd
coal/sister unit
46% -90%; net 44%;
significant increase; 2
OH samples each
Significant increase to
90+%; net +40%
No effect; actually lower
Hg removal in ESP (-
6% vs 23%)
-85% Hg removal; Did
not test w/o SCR
Significant increase to
90+%; net 47%
Fffort nf MH,
onHg
Oxidation
(SCR in/out)
No effect
Not tested
Small neg.
effect.
Not tested in 2nd
coal/sister unit
Small negative
effect
Not tested
Not tested
Not tested
Not tested
NH3, Cl, S03 - Sampled at SCR outlet unless noted (* - ESP outlet, # - Particulate control inlet)
-67
-------
Appendix B
Table B-12. Default Values for Mercury Control Input Parameters
Description
Units
Range
Default
Sorbent Injection Inputs
Hg GEMS (0=no, l=yes)
Hg Reduction Required from Coal
Sorbent Type, 1 = EPAC, 2=PAC, 3=other
Maximum Temperature before Spray Cooling
Sorbent Recycle Used?
Spray Cooling Desired?
EPAC Cost (delivered cost of brominated PAC)
PAC Cost (delivered)
Other Sorbent Cost (delivered)
Does sorbent adversely impact fly ash sales?
(0=no, l=yes)
Before Sorbent Injection,
Fly Ash Sold (1) or Disposed of (2)
Revenue from Fly Ash Sales
Dry Waste Disposal Cost
Retrofit Factor
Maintenance Factors (% of Installed Cost)
Process Contingency, % of process capital
General Facilities (% of Installed Cost)
Engineering Fees (% of Installed Cost)
Project Contingency
Duration of Project (years)
integer
percent
deg F
yes/no
yes/no
$/ton
$/ton
$/ton
integer
1 or 2
$/ton
$/ton
%
%
%
%
%
integer
0 or 1
up to 325 F
0 or 1
0 to 35
1 to 25
1
80.0%
2
325
no
no
$1,500
$1,000
$1,000
1
2
$6.00
$6
1.30
5%
5%
5%
10%
15%
1
PJFF downstream of PAC Inputs
PJFFto COHPAC (i.e., TOXECON), 0=no, l=yes
Cost of Bags, installed ($/bag)
Estimated Number of Bags/MW
Average bag life
Pressure Drop
Outlet Emissions
Retrofit Difficulty Factor
Process Contingency, % of process capital
General Facilities, % of Process Capital
Engineering, Home Office, etc.
% of Process Capital and General Facilities
Project Contingency, % of Process Capital and
Gen Facilities
Owner's Overhead and costs
Inventory Capital and Prepaid Royalties, etc.
Maintenance, % of process capital and excluding bags
Period of construction, yrs
0 or 1
$/bag
integer
years
iwc
Ib/MMBtu
%
% of process
capital
%
% of process
capital
1
$80
20
5
8
0.012
1.30
5%
5.0%
10.0%
15%
5.0%
1.0%
1%
1
68
-------
Appendix B
6.5.6 Conclusions
Correlations for mercury removal from coal-fired power plants have been developed in the
CUECost model, incorporating information on mercury removal from existing equipment that
was developed from the ICR data in EPRI (2000). CUECost also incorporates mercury
removal from injection of PAC, as developed from full-scale demonstrations of PAC injection
where data are available. Algorithms developed with CUECost should be continuously
updated and modified as more information becomes available on experience with mercury
removal.
The following summarize some important findings that influence modeling mercury removal:
• The CUECost workbook that permits isolation of the effects of different air pollution control
equipment on the fate of mercury will facilitate modeling combined effects with PAC
injection over a wide range of boiler configurations and scenarios without the need for new
regressions of PAC injection test data. Impact of a specific piece of equipment can be
estimated with models best suited for that equipment.
• PAC injection followed by a fabric filter results in much lower injection concentrations
being necessary for a given level of mercury reduction than for PAC injection followed by a
cold-side ESP. Thus, economic modeling may show that in some cases the additional
capital cost of a fabric filter may be justified by reduced operating costs associated with
PAC consumption.
• Sorbent selection appears to have little effect on performance when PAC injection is
followed by a fabric filter. But sorbent choice appears to have a significant effect when
PAC injection is followed by an ESP.
• As demonstrated by the Salem Harbor test results, LOI and temperature can have a
significant effect on the mercury removal by existing equipment. For this reason, the
correlations of EPRI (2000), which do not include these effects, do not always provide an
accurate indication of mercury removal by existing equipment.
• In some cases PAC injection without a downstream fabric filter may not be able to achieve
very high mercury removal rates of 90% or more, regardless of PAC injection
concentration.
B.6 CO2 CONTROL DESIGN CRITERIA
In a monoethanolamine (MEA)-based CO2 control system, a continuous scrubbing system is
used to separate CO2 from the flue gas. The system consists of an MEA island and a
compressor island. The temperature of flue gas coming out the wet scrubber system is often
higher than the temperature required by the MEA process. Therefore, in order for CO2 to be
efficiently scrubbed and to reduce solvent losses, the flue gas must be cooled down below
50 °C. As SO2 reacts with MEA, the concentration of SO2 prior to the absorber should be low
69
-------
Appendix B
(<10 ppm) to reduce the losses and degradation of MEA by SO2. NaOH scrubbing is often
required before the absorber.
Flue gas then flows through the absorber where CO2 binds to MEA. The CO2-rich solution
leaves the absorber and passes through the heat exchanger and finally enters the
regenerator where CO2 is released from MEA by external heat from steam supply or natural
gas burning. The hot CO2 lean solvent then flows back to the heat exchangers where it is
cooled, and then is sent back to the absorber. To supplement the MEA losses, fresh MEA is
added. Eqs. B-23 and B-24 show this cycle.
CO2 absorption:
2R-NH2 + CO2 (g) -> R-NH3++R-NH-COO" (Eq. B-23)
MEA regeneration:
R-NH-COO" + R-NH3+ (heat) -> CO2+2R-NH2 (Eq. B-24)
The regeneration of MEA consumes a great deal of energy when the MEA concentration is
low in the solvent. Inhibitors are therefore added to the solvent to increase the MEA
concentration. In the worksheet, a typical MEA concentration is 30% with the addition of
inhibitors.
In sorbent injection (SI) to capture CO2, a continuous scrubbing system is used to separate
CO2 from the flue gas. The system consists of a sorbent absorption and regeneration island
and a compressor island. The flue gas coming out of the wet scrubber system is cooled to
relatively low temperatures (30 to 35 °C) for the easy capture of CO2 by the sorbent. When
SO2 reacts with sorbent to degrade the sorbent, the concentration of SO2 prior to the
absorber should be lowered, in general <10 ppm, to minimize sorbent consumption. As
such, additional scrubbing is required before the absorber. Flue gas then flows through the
absorber where CO2 binds to sorbents. The CO2-rich sorbent leaves the absorber and passes
through the heat exchanger in the regenerator where CO2 is released from sorbent under
the assistance of external heat.
In the CAP, CO2 is absorbed in an ammoniated solution at 32 °F. Cooling the flue gas to
such a low temperature is a necessary step within the process. As the result of flue gas
cooling, moisture in the flue gas is also condensed, leading to less actual flue gas flow rate
through the booster fan. In the absorption process, the formation of aqueous ammonium
carbonate [(NH4)2CO3] with the precipitation of ammonium bicarbonate [(NH4)HCO3] solids
at low temperatures optimizes the energy demand, improves CO2 removal efficiency, and
reduces ammonia slip. The formation of ammonium bicarbonate solids is a reversible
reaction. With heat in the regenerator, the ammonium bicarbonate solids are dissolved with
eventual evolution of ammonia, water and CO2 gases. The CO2 stream leaves the
regeneration vessel from the CAP at a higher pressure than the other two CO2 processes
70
-------
Appendix B
(MEA and SI) which results in fewer stages of downstream CO2 compression. The ammonia
and water reaction products are stripped and condensed from the resulting gas stream for
reuse as reagent and flue gas wash solvent (Sherrick, B. 2008).
Gas exiting the regenerator must be compressed and dehydrated to accommodate transport
and disposal. Moist CO2 from the CO2 regenerator's reflux drum enters the compressor at
21 °C (69 °F) and nominally 160 kPa (23 psi). CO2 is compressed in a six-stage integrally
geared compressor. Intercoolers between stages cool the gas using chilled water from the
plants' cooling tower. After exiting the compressor, and presumably a final heat exchanger,
the CO2 is dried to < 20 ppm water in a triethylene glycol (TEG) dehydrator. Dry gas exiting
the dehydrator is at 15.27 MPa (2215 psi) and 51 °C (124 °F) (DOE 2007).
REFERENCES
Bustard, J., Durham, M., Lindsey, C., Starns, T., Baldrey, K., Martin, C., Schlager, R.,
Sjostrom, S., Slye, R., Renninger, S., Monroe, L, Miller, R., Chang, R., 2001. "Full-Scale
Evaluation of Mercury Control with Sorbent Injection and COHPAC at Alabama Power E.G.,
Gaston", DOE-EPRI-U.S. EPA-A&WMA Power Plant Air Pollutant Control "Mega" Symposium,
August 20-23, 2000, Chicago, IL.
Bustard, J., Durham, M., Lindsey, C., Starns, T., Baldrey, K., Martin, C., Schlager, R.,
Sjostrom, S., Slye, R., Renninger, S., Monroe, L., Miller, R., Ramsey, C., 2002. "Gaston
Demonstrates Substantial Mercury Removal with Sorbent Injection", Power Engineering, vol.
106, no. 11.
DOE, 2003, National Energy Technology Laboratory Mercury Field Evaluation - PG&E NEG
Salem Harbor Station - Unit 1, Project No. 00-7002-76-10, Field Evaluation Summary
Report, January 2003.
DOE/NETL. 2007. Cost and Performance Baseline for Fossil Energy Plants ( DOE/NETL-
2007/1281).
Durham, M., Bustard, J., Schlager, R., Martin, C., Johnson, S., Renninger, S., 2001. "Field
Test Program to Develop Comprehensive Design, Operating Cost Data for Mercury Control
Systems on Non-Scrubbed Coal-Fired Boilers", AWMA 94th Annual Conference and
Exhibition, Orlando, FL, June 24-28 2001.
EPA, 1996, "Cost-effectiveness of Low-NOx Burner Technology Applied to Phase I, Group 1
Boilers," prepared by Acurex Environmental Corporation for EPA Acid Rain Division. This
report is available to the public from EPA's Office of Air and Radiation, Acid Rain Division,
Washington, DC 20460 (202-564-9085).
EPA, 2000, Performance and Cost of Mercury Emission control Technology Applications on
Electric Utility Boilers, EPA-600/R-00-083.
71
-------
Appendix B
EPRI, 2000, An Assessment of Mercury Emissions from U.S. Coal Fired Power Plants, EPRI,
Palo Alto, CA.
Gundappa, M., L. Gideon, and E. Soderberg, 1995, "Integrated Air Pollution Control System
(IAPCS), version 5.0, Volume2: Technical Documentation, Final" EPA, Air and Energy
Engineering Research Laboratory, Research Triangle Park, NC, EPA-600/R-95-169b (NTIS
PB96-157391).
Hocquel, M., Unterberger, S., Hein, K., Bock, J., 2002, "Behavior of Mercury in Different Gas
Cleaning Stages", International Conference on Air Quality, September 9-12, 2002, Crystal
City, VA.
Richardson, C., Machalek, T., Miller, S., Dene, C., and Chang, R., 2002, "Effect of NOX
Control Processes on Mercury Speciation in Utility Flue Gas", International Conference on Air
Quality, September 9-12, 2002, Crystal City, VA.
Sherrick, B.; Hammond, M.; Spitznogle, G.; Murashin, D.; Black, S.; Cage, M.; CCS with
Alstom's Chilled Ammonia Process at AEP's Mountaineer Plant, Present in the Power Plant
Mega Symposium. Baltimore, MD. 2008.
Starns, T., Bustard, J., Durham, M., Lindsey, C., Martin, C., Schlager, R., Donnelly, B.,
Sjostrom, S., Harrington, P., Haythornthwaite, S., Johnson, R., Morris, E., Chang, R.,
Renninger, S., 2002, "Full-Scale Test of Mercury Control with Sorbent Injection and an ESP
at Wisconsin Electric's Pleasant Prairie Power Plant", AWMA 95th Annual Conference and
Exhibition, Baltimore, June 23-27 2002.
72
-------
Appendix C
C.I GENERAL PLANT DESIGN CRITERIA
The plant design and operating default values provided below were taken from the
criteria established by EPA's Integrated Air Pollution Control System (IAPCS) model
(Gundappa et al., 1995) and were generally replaced with IPM/IECM values (Table C-l).
The user can override any default value as long as the value input is within the range of
the parameter stated on the worksheet. Table C-2 lists the coal analysis embedded in
Sheet 11.0 Constant_CC (Coal Analysis Library). More information for coal analysis can
be found from DOE coal bank and database (http://datamine.ei.psu.edu/index.php).
73
-------
Appendix C
Table C-l. Snapshot for a Specific Plant and Its Default Parameters
Plant Information
Cost Basis -Year
(For Power Generation Estimation only)
Location - State
Power Generation Technologies
General Plant Factors
Gross Plant output
Net Plant Output
Plant Heat Rate
Plant Capacity Factor
Coal Type
Price of Coal
Other Operating Information
Percent Excess Air in Boiler
Uncontrolled NOX from Boiler
Air Heater Inleakage
Air Heater Outlet Gas Temperature
Inlet Air Temperature
Ambient Absolute Pressure
Pressure After Air Heater
Moisture in Air
Ash Split:
Fly Ash
Conversion of SO2 to SO3
Units
MW
MW
Btu/kWh
%
$/MMBtu
%
%
°F
°F
in. Hg
in. H2O
Ib/lb dry air
%
Range
All States
500-800
500-750
40-90%
Default
2005
PA
1
580
500
10,500
65%
5
2.05
120%
algorithm
12%
300
80
29.4
-12
0.013
80%
1.0%
74
-------
Appendix C
Table C-2. Coal Analysis Library
Go Back to Input Sheet
COAL ANALYSIS LIBRARY
Index Number
Coal Name
Coal Cost
$/MMBtu
PROXIMATE ANALYSIS (ASTM, as rec'd)
Moisture - Enter below in Ultimate Analysis
Volatile Matter
Fixed Carbon
Ash - Enter below in Ulti
COAL ULTIMATE ANALYSIS (AS
Moisture
Carbon
Hydrogen
Nitrogen
Chlorine
Sulfur
Ash
Oxygen
TOTAL
Mercury
Modified Mott Spooner HHV (B
COAL ASH ANALYSIS (ASTM, a
SiO2
AI2O3
TiO2
Fe2O3
CaO
MgO
Na2O
K20
P2O5
SO3
Other Unaccounted for
TOTAL
wt%
wt%
mate Analysis
aTM, as rec'd)
wt%
wt%
wt%
wt%
wt%
wt%
wt%
wt%
wt%
mg/kg
Btu/lb
s rec'd)
wt%
wt%
wt%
wt%
wt%
wt%
wt%
wt%
wt%
wt%
wt%
wt%
Coal 1, Wyoming
PRB: 8,227 Btu,
0.37% S, 5.32%
ash
1
Wyoming PRB
1.50
31.39
33.05
100.00
30.24
48.18
3.31
0.70
0.003
0.37
5.32
11.87
99.99
0.10
8,227
35.51
17.11
1.26
6.07
26.67
5.30
1.68
2.87
0.97
1.56
1.00
100.00
PRB
Coal 2,
Armstrong,
PA: 13,100
Btu, 2.6% S,
9.1% ash
2
Armstrong, PA
1.50
36.20
48.70
100.00
6.00
71.55
4.88
1.40
0.000
2.60
9.10
4.47
100.00
0.10
13,100
46.92
21.00
2.40
20.20
3.25
2.65
0.90
0.30
0.00
1.38
1.00
100.00
Bituminous
Coal 3,
Jefferson,
OH:
11,922
Btu, 3.43%
S, 13% ash
3
efferson, Ol
1.50
37.20
44.80
100.00
5.00
65.72
4.53
1.21
0.100
3.43
13.00
7.01
100.00
0.10
11,922
51.35
30.00
1.80
9.00
4.50
2.00
0.40
0.20
0.16
0.59
0.00
100.00
Bituminous
Coal 4, Logan,
WV: 12,058
Btu, 0.89% S,
16.6% ash
4
Logan, WV
1.50
35.40
43.00
100.00
5.00
65.99
4.75
0.70
0.100
0.89
16.60
5.97
100.00
0.10
12,058
50.68
29.00
1.70
9.00
5.50
1.00
0.40
0.90
0.60
1.22
0.00
100.00
Bituminous
Coal 5, No. 6
Illinois:
10,100 Btu,
4% S, 16%
ash
5
No. 6 Illinois
1.50
33.00
39.00
100.00
12.00
55.35
4.00
1.08
0.100
4.00
16.00
7.47
100.00
0.10
10,100
50.82
19.06
0.83
20.00
3.43
3.07
0.60
0.37
0.17
1.22
0.43
100.00
Bituminous
Coal 6,
Rosebud, MT:
8,789 Btu,
0.56% S,
8.15% ash
6
Rosebud, MT
1.50
36.40
30.30
100.05
25.20
51.52
3.29
0.69
0.100
0.56
8.15
10.49
100.00
0.10
8,789
27.00
19.00
1.08
9.00
18.50
2.40
2.80
0.45
0.42
18.85
0.50
100.00
Subbituminous
Coal 7, Lignite,
ND: 7,500 Btu,
0.94% S, 5.9%
ash
7
Lignite, ND
1.50
42.00
20.10
100.00
32.00
45.06
2.80
1.50
0.100
0.94
5.90
11.70
100.00
0.10
7,500
29.80
10.00
0.40
9.00
21.40
10.50
4.40
0.49
0.00
14.01
0.00
100.00
Lignite
Coal 8, DOE HS:
12,676 Btu, 3%
S, 9% ash
8
DOE HS
1.50
40.40
47.50
100.00
3.10
69.82
5.00
1.26
0.120
3.00
9.00
8.70
100.00
0.10
12,676
29.00
17.00
0.74
36.00
6.50
0.83
0.20
1.20
0.22
7.30
1.01
100.00
Bituminous
Coal 9, DOE
LS: 14,175
Btu, 0.6% S,
3.8% ash
9
DOE LS
1.50
44.00
50.00
100.00
2.20
78.48
5.50
1.30
0.120
0.60
3.80
8.00
100.00
0.10
14,175
51.00
30.00
1.50
5.60
4.20
0.76
1.40
0.40
1.80
2.60
0.74
100.00
Bituminous
Coal 10, DOE
PRB: 8,304
Btu, 0.48% S,
6.4% ash
10
DOE PRB
1.50
30.79
32.41
100.00
30.40
47.85
3.40
0.62
0.003
0.48
6.40
10.82
99.97
0.07
8,304
31.60
15.30
1.10
4.60
22.80
4.70
1.30
0.40
0.80
16.60
0.80
100.00
PRB
Coal 11, K
Fuel: 11,718
Btu, 0.38% S,
6.42% ash
11
KFuel
1.50
40.20
45.50
99.62
7.50
66.70
4.80
1.00
0.030
0.38
6.42
13.20
100.03
0.04
11,718
28.40
17.30
1.60
6.00
23.50
4.00
1.40
0.27
2.43
13.63
1.47
100.00
PRB
Coal 12, Med
S: 11,570
Btu, 1.5% S,
8.15% ash
12
Med S
1.50
36.19
43.80
100.00
11.86
65.12
4.22
1.33
0.380
1.50
8.15
7.44
100.00
0.10
11,570
51.35
30.00
1.80
9.00
4.50
2.00
0.40
0.20
0.16
0.59
0.00
100.00
Bituminous
75
-------
Appendix C
C.2 ECONOMIC CRITERIA
Economic inputs for CUECost workbook calculations are shown in Table C-3.
Table C-3. Economic Inputs
Description
Cost Basis, Year Dollars
Service Life (Levelization Period, years)
Sales Tax Rate
Escalation/Inflation Adjustment (GDP or Chem Index*)
Units
date
integer
%
Range
Cor D
Economic Factors During Construction Period
Construction Labor Rate
Prime Contractor's Markup
Current Inflation Rate
Current Escalation Rates
After Tax Discount Rate (Current $'s)
Capital Carrying Charges
First-year Carrying Charge (Current $'s)
Levelized Carrying Charge (Constant $'s)
Non-Carrying Expense (O&M)
Levelizing Factor (L30) (Constant $'s)
Variable Cost Factors
Operating Labor Rate (include benefit)
Power Cost
Steam Cost
Demineralized Water
Makeup Water
$/h
%
%
%
%
%
%
$/h
Mills/kWh
$/1000 Ibs
$/lb
$/1000 Ib
Calculator
Default
2008
30
6%
GDP
$35
3%
2%
3%
9%
16%
8%
1.48
$25.0
60
3.5
$0.0030
$0.05
* Chem Index =Chemical Engineering Magazine - Plant Index updated in each issue. This is the user
input value for the year selected. The model divides the input value by the January 1998 index value
to determine the escalation factor that is needed.
76
-------
Appendix D
D.I FGD COST ALGORITHM DEVELOPMENT
The cost algorithms associated with the flue gas desulfurization processes were developed
based on historical data and new equipment quotations received by Raytheon during 1998
for some of the major equipment items. Algorithm development began with derivations from
Raytheon's in-house historical database. These data sets were then modified by adding the
additional data points from the new budgetary quotations, and then deriving new equations
to represent the costs for equipment areas and for specific large pieces of equipment.
Performance data were sent to multiple vendors for one or two of the major equipment
components identified in each cost area. These vendor contacts included a minimum of four
vendors in each case. Responses to cost data requests were received from a minimum of
one and normally three or more of the vendors solicited. Where vendor responses were
limited due to refusals or delayed responses, additional data sources were obtained from
recent projects to add to the data base of cost information for specific components. The cost
data requests were made over the expected range of component sizes that could be used in
the CUECost estimating workbook. LSFO capital cost algorithms are shown in Table D-l.
Table D-l. Variable and Constant Parameters for Wet FGD Cost Algorithm
LSFO Process
Equipment
ID Fans and
Ductwork
Chimney
Support Equipment
x =
MW
Chimney acfm
Chimney acfm
MW
Equation
(x x 1000 x A x
x^B)/1.3
(A x x + B)/1.3
(A x x + B)/1.3
A=
4456.5
1.6225
3.4736
B =
-0.6442
3,000,000
5,000,000
=(0.0003 x x^3-1.0667 x x^2 + 1993.8 x
x+1177674) x 1.22
77
-------
Appendix D
LSFO Process Equipment includes the Reagent Handling and Preparation, SO2 Control
System, and the Byproduct Handling. The capital costs for these equipments are described
in Appendix B.I.
LSD capital cost algorithms are shown in Table D-2.
Table D-2. Parameters for LSD Cost Algorithms
Reagent Handling and
Preparation
SO2 Control System
Byproduct Handling
Flue Gas Handling and
ID Fans
Chimney Modification
Support Equipment
X=
pph lime
Inlet gas acfm
pph byproduct
(if upstream of
existing ESP)
If new ESP or FF
acfm
KW
MW
Equation
(Ax+B)/1.3
(Ax+B)/1.3
(Ax+B)/1.3
A=
136.84
9.262
31.124
B=
3,000,000
5,000,000
2,000,000
Determined by ESP and FF Worksheets
(Ax+B)/1.3
(Ax+B)/1.3
2.9232
3.4
3.00E+06
0
(-1.211 x x^2+2704.2 x x +1354716.2) x 1.22
The capital costs for LSD Process Equipments are described in Appendix B.2.
D.2 SELECTIVE CATALYTIC REDUCTION
D.2.1 Performance Parameters
The key operating parameters that affect the performance and, consequently, the capital
and operating costs of SCR systems include the allowable NH3 slip emissions, the space
velocity, the NOX reduction efficiency, and the NH3/NOX molar ratio. For SCR systems
these parameters are interrelated, and their values depend on the type of SCR
application (high-dust or tail-end) and the desired performance levels. Ammonia slip
emissions are controlled by the SCR system design. Typically SCR catalyst suppliers
provide a guarantee of 2 ppm over the catalyst life. Since the 2 ppm NH3 slip is
guaranteed at the end of the catalyst's life, the initial NH3 slip emissions will be very low
(<1 ppm). Ammonia slip is not taken into consideration in the catalyst volume
determination. The space velocity is the primary parameter used to specify catalyst
volume. If the user does not input a value for space velocity, CUECost calculates it based
on the NOX reduction efficiency and the NH3/NOX molar ratio (molecular weight of NOX =
molecular weight of NO2):
78
-------
Appendix D
Space Velocity
SV = 6131.06 / 3 x (n)-°-241 x (NH3:NOX ratio)'2'306 (Eq. D-l)
where
SV = space velocity, 1/h
n = NOX reduction efficiency, fraction
NH3:NOX ratio = stoichiometric ratio of NH3 to NOX.
The NOX reduction efficiency (n) and molar ratio of NH3 to NOX (NH3/NOX ratio) are user-
specific input values. The gross catalyst volume and NH3 injection rate are determined from
the following equations taken from IAPCS sources (Gundappa et al., 1995):
Ammonia Injection Rate
NH3 = 3.702 x 10"4 x NH3:NOX ratio x BSIZE x HTR x NOX (Eq. D-2)
where
NH3 = ammonia injection rate, Ib/h
BSIZE = boiler size, MW
HTR = net heat rate, Btu/kWh
NOX = inlet NOX emissions, Ib/MMBtu
CV = gross catalyst volume, ft3
Q = flue gas volume flow rate, SCFH.
Gross Catalyst Volume
CV = Q/SV (Eq. D-3)
where
SV= space velocity.
D.2.2 Capital Costs
CUECost estimates capital costs for reactor housing, initial catalyst, ammonia storage and
injection system, flue gas handling including ductwork and induced draft fan modifications,
air preheater modifications and miscellaneous direct costs, including ash handling and water
treatment additions. CUECost equations for SCR direct capital costs are shown below.
For all items except flue gas handling, cost algorithms are based on regression models
developed for the Integrated Environmental Control Model (IECM) (Frey and Rubin, 1994).
The IECM regression models were developed from cost data for 12 coal-fired power plants
(Robie and Ireland, 1991). The flue gas handling cost algorithm is taken from the
Integrated Air Pollution Control System (IAPCS) model, version 5.0 (Gundappa et al.,
1995). Costs derived from the IAPCS equations for flue gas handling were found to be on
79
-------
Appendix D
the same order of magnitude as costs reported by the Acid Rain Division study (EPA, 1997;
EPA, 1998). IECM equations (Frey and Rubin, 1994) were used for the other direct capital
cost items because they are based on more current cost data than IAPCS (Gundappa et al.,
1995). Installation costs for items such as structural supports, foundations, concrete,
earthwork are accounted for in the cost data used to develop the IECM and IAPCS equations
and, therefore, are not a separate item in CUECost. Plant cost indices from Chemical
Engineering Magazine are included in the equations to update direct capital costs. Direct
capital costs for hot-side SCR are shown in Table D-3.
Table D-3. Direct Capital Costs for Hot-side SCR (Installed equipment costs)
Reactor Housing
DC r= 18.65 x Nr,totx (CV/ Nr,tot)^0.489 x 1000 x RF x PCI/ 357.3
Ammonia Storage and Injection System
DC NH3 = 50.8 x (NH3)/V0.482 x 1000 x RF x PCI / 357.3
Flue Gas Handling: Ductwork and Fans
DC fgh = 143.66 x[Gfg x (750+460) / (70+460)^0.694 x RF x PCI / 314.0
Air Preheater Modifications
DC aph,mod= 1370 x Nt,aph x (UAt,aph/ 4.4 / 10^67 Nt,aph)^0.8 x 1000 x RF x PCI / 357.3
Miscellaneous Direct Costs
DC misc = [100 + 300 x (BSIZE/ 550)^0.6] x 1000 x RF x PCI / 357.3
where
Gfg = flue gas volumetric flow rate for SCR ductwork, scfm
Nr,tot = Number of SCR reactors
Nt,aPh = total number of air preheaters
RF = retrofit factor
PCI = chemical engineering plant cost index from Chemical Engineering Magazine
= 388 for 1998 dollars, 314.0 for 1982 dollars and 357.3 for IECM base year dollars
UAt,aph = product of universal heat transfer coefficient and heat exchanger surface area
= Qaph , Btu/°R
q aph = heat transfer = Flue gas scfm x 60 x 7.9 x (Tfiue aas. out -1 fiue aas. m)
0.7302 x 530
dTLM, aph = log-mean temperature difference
= LL flue aas. in^ Ta|r. out ) " (Tf|ue gas, out ^_L air. inJ
flue gas, in ~ ' air, out ) /(.' flue gas, out ~ ' air, inJJ
80
-------
Appendix D
The flue gas inlet temperature (Tf|Ue gas, m) and the outlet temperature (Tflue gas, out) are
assumed to be the respective typical values of 725 and 600 °F.
Capital costs for instruments and controls, sales tax and freight are calculated from
percentages of the equipment cost subtotal. The equipment cost subtotal is the sum of the
equations listed above. For instruments and controls and freight, the respective default
percentages are 2% and 5%. The sales tax rate is a user input value. The total direct cost is
determined by applying the retrofit factor to the capital equipment cost subtotal, which is
the sum of the equipment costs listed above as well as instruments and controls, sales tax
and freight. The retrofit factor is a user input value that ranges from one for new
applications to three for the most difficult retrofit cases. Equations for indirect capital costs
are given in Table D-4.
Table D-4. Indirect Capital Costs for Hot-side SCR
General Facilities = Total Direct Cost with Retrofit x General Facilities (% of installed
cost)
Engineering fees = Total Direct Cost with Retrofit x Engineering Fees (% of installed cost)
Contingency = Total Direct Cost with Retrofit x Contingency (% of installed cost)
Total Plant Investment = Sum of Total Direct Cost with Retrofit, General Facilities,
Engineering fees, Contingency taking into account allowance for funds during construction
Preproduction = Total Plant Investment x 0.02 + One month fixed operating costs +
One month variable operating costs (at full capacity)
Initial Ammonia (60 days) = NH3 x 24 x CF x 60 x UCNH3/ 2000
Initial Catalyst = CV x UCCAT
where
CF = capacity factor, fraction
UCNH3 = ammonia cost rate, $/ton
UCcAT = unit cost of catalyst, $/ft3
CV = gross catalyst volume, ft3
NH3= Ammonia injection rate, Ib/h.
D.2.3 Operating and Maintenance Costs
Operating and maintenance costs include NH3, catalyst replacement and disposal, electricity,
steam, labor and maintenance costs. The CUECost operating and maintenance cost
equations presented below are based on IAPCS equations (Gundappa et al., 1995). IAPCS
equations were selected instead of IECM equations (Frey and Rubin, 1994) for operating
and maintenance costs because the level of detail required for IAPCS input parameters was
closer to that of other CUECost inputs. Additionally, the parameters affecting operation and
maintenance costs are not likely to have changed significantly since the IAPCS equations
were developed. With the exception of catalyst replacement costs, the equations from
81
-------
Appendix D
IAPCS were derived from data reported by TVA for the high-dust system (Maxwell and
Humphries, 1985). Annual catalyst replacement costs are based on the catalyst life. For
example, if the catalyst life is 3 years, then one-third of the catalyst is replaced each year.
The catalyst disposal cost reflects the cost of disposing of the spent catalyst. Catalyst
disposal is typically included in the purchase cost of the catalyst. As a result, the
recommended default for this line item is zero. However, an equation is included to allow
the user to estimate a disposal cost, if applicable. A typical value of 48 Ib/cubic foot was
used for the catalyst density to calculate the mass of the spent catalyst. Operation and
maintenance cost equations for SCR are shown Table D-5.
Table D-5. Operating and Maintenance Cost Equations for SCR ($/year)
Ammonia Cost = (8,760/2,000) x (NH3 x CF x UCNH3)
Catalyst Replacement Cost = CV/N x UCCAT
Catalyst Disposal Cost = 48 x Catalyst Replacement Cost x
2,000 x UCCAT
Electricity = (-545,133 + S.SOlxG) x (CF / 0.628) x UCELEc
Steam = (-14.91 + 33.29 x NH3 xCF) x UCSTEAM
Operating Labor = (1,341 + 5.363 x BSIZE) x UCOL
Maintenance Costs = Maintenance (%) x TPC
where
BSIZE= boiler size, MWe
CF= capacity factor, fraction
CV= gross catalyst volume, ft3
G = flue gas flow rate, acfm
N = overall catalyst life, years
Maintenance (%) = annual maintenance cost as a percent of total plant cost
TPC = total direct and indirect capital costs, $
UCCAT= catalyst cost, $/ft3
UCELEc = electricity rate, $/kWh
UCoL = operating labor wage, $/person-h
UCNH3= ammonia cost rate, $/ton
= steam rate, $/MMBtu
= solid waste disposal rate, $/ton
NH3= Ammonia injection rate, Ib/h.
82
-------
Appendix D
D.2.4 CUECost Validation
Total plant costs and operating and maintenance costs estimated by CUECost algorithms
were compared to current cost data developed and validated by EPA's ARD. Cost and design
information for four applications of SCR on various boiler types, boiler sizes and coals was
taken from a 1996 Acid Rain Division (ARD) study (EPA, 1996) (Tables D-6 and D-7). The
design information for these SCR applications was used to evaluate equations from
CUECost. Total plant capital costs include the reactor housing, initial catalyst, ammonia
storage and injection system, flue gas handling including ductwork and induced draft fan
modifications, air preheater modifications and miscellaneous direct costs, including ash
handling and water treatment additions. Other direct capital costs for taxes, freight,
instruments and controls and initial inventory are included in the comparison of direct
capital costs. The total plant cost includes direct costs listed above as well as indirect capital
costs for engineering, general facilities and contingencies. Chemical engineering plant cost
indices from Chemical Engineering Magazine were used to normalize costs in consistent year
dollars.
The percent difference between ARD study costs and the CUECost estimates for total plant
costs ranged from -4% to +8% for the cases evaluated. Operation and maintenance costs
estimated by CUECost are 23 to 31% lower than those estimated by the ARD study. The
largest difference appears to be the catalyst replacement cost.
83
-------
Appendix D
Table D-6. CUECost with Acid Rain Division Study Design for SCR (1990 dollars)*
Selective Catalytic Reduction
Cyclone-Fired
Midwestern
Wet-Bottom
Vertical-
Fired
Wall-
Fired
Eastern Bituminous
Boiler Size (MW)
150
400
100
259
CUECost with Acid Rain Division Design Parameters
Input Parameters Taken from Acid Rain Division Study
NOX Reduction Efficiency fraction
NH3/NOX Molar Ratio fraction
Inlet NOX Ibs/MMBtu
Design Parameters Calculated by CUECost
Ammonia Injection Rate Ib/hr
Gross Catalyst Volume ft3
Flue Gas at Air Heater Outlet SCFM
0.50
0.50
1.4
340
1,385
273,571
Capital Costs Using Acid Rain Division Design Parameters ($
Reactor Housing and
Ammonia Handling and
Flue Gas Hand 1 ing :Ductwork and Fans
Air Preheater Modifications
Misc. Other Direct Capital Costs
Initial Catalyst
Total Capital Equipment Cost
Freight, Sales Tax and Inst. & Controls
Total Plant Cost (TPC)
TPC ($/kW)
% Difference from Acid Rain Division Study
1,188
1,097
2,238
481
309
485
5,798
691
8,590
57.3
4%
0.50
0.50
1.3
884
3,883
766,250
1000)
1,967
1,739
4,574
1,096
453
1,359
11,188
1,278
16,353
40.9
0%
0.50
0.50
0.95
155
935
182,280
981
752
1,689
348
270
327
4,367
525
6,489
64.9
8%
0.50
0.51
0.92
399
2,485
482,464
1,582
1,185
3,318
757
379
870
8,090
939
11,884
45.9
-4%
O&M Costs usina Acid Rain Division Desian Parameters f$1000/vear)
Ammonia
Catalyst Replacement
Catalyst Disposal
Electricity
High-dust SCR Steam
Maintenance
O&M Total
% Difference from Acid Rain Division Study
157
162 '
0.10
112
34
122
586
-23%
407
453
0.28
366
88
225
1,539
-24%
72
109
0.07
66
15
92
354
-31%.
184
290
0.18
220
40
165
899
-30%
Source: EPA, 1997; EPA, 1998
84
-------
Appendix D
Table D-7. Acid Rain Division Study: SCR Applications*
Selective Catalytic Reduction
Cyclone-Fired
Midwestern Bituminous
Wet-Bottom
Vertical- Wall-Fired
Fired
Eastern Bituminous
Boiler Size (MW)
150 400
100
259
Acid Rain Division Costs and Design Parameters
Design Parameters from Acid Rain Division
NOX Reduction Efficiency fraction
NH3/NOX Molar Ratio fraction
Inlet NOX Ibs/MMBtu
Ammonia Injection Rate Ib/hr
Gross Catalyst Volume ft3
Flue Gas at Air Heater Outlet SCFM
Acid Rain Division Capital Costs {$ 1000)
SCR Reactors/Ammonia Storage
Piping/Ductwork
Electrical/PLC
Draft Fans
Platform/Insulation/Enclosure
Air Preheater Modifications
Total Capital Equipment Cost
Total Plant Cost (TPC)
TPC ($/kW)
Acid Rain Division O&M Costs ($ 1000/year)
Power Consumption
Ammonia Consumption
Catalyst Consumption
General Maintenance
0.50 0.50
0.50 0.50
1.4 1.3
339 882
3,690 10,020
292,924 821,164
3,180 7,040
945 1,600
450 720
1,065 1,760
180 440
285 520
6,105 12,080
8,242 16,308
55.05.0 40.8
56 200
156 408
430 1,168
123 246
0.50
0.50
0.95
155
2,571
191,279
2,150
860
460
650
100
250
4,470
6,035
60.4
55
72
300
89
0.50
0.51
0.92
398
6,675
498,215
4,921
1,528
803
1,166
285
466
9,169
12,378
47.8
140
184
779
183
O&M Total
* Source: EPA, 1997; EPA, 1998
764
2,023
516
1,286
85
-------
Appendix D
D.3 SELECTIVE NONCATALYTIC REDUCTION
D.3.1 Performance Parameters
The CUECost workbook allows the user to select either urea [CO(NH2)2] or ammonia (NH3)
as the SNCR reagent. The user is asked to specify the NOX reduction efficiency and the
stoichiometric ratio of reagent to NOX (molecular weight of NOX = molecular weight of NO2).
The NH3 and CO(NH2)2 injection rates in pounds of pure reagent per hour are then
calculated based on the stoichiometric ratio, inlet NOX and boiler heat input:
Urea Injection Rate
Urea = 6.5 x 10"4 x UREA:NOX ratio x BSIZE x HTR x NOX (Eq. D-4)
Ammonia Injection Rate
NH3 = 3.702 x 10"4 x NH3:NOX ratio x BSIZE x HTR x NOX (Eq. D-5)
where
Urea = CO(NH2)2 injection rate, Ib/h
NH3 = NH3 injection rate, Ib/h
BSIZE = boiler size, MWe
HTR = net heat rate, Btu/kWh
NH3:NOX ratio = stoichiometric ratio of NH3to NOX
NOX = inlet NOX emissions, Ib/MMBtu
UREA:NOX ratio = normalized stoichiometric ratio of CO(NH2)2 to NOX (i.e., moles of reagent
nitrogen to moles of uncontrolled NOX).
For the CO(NH2)2-based SNCR process, the user may select to use wall injectors, lances, or
both. Wall injectors are nozzles installed in the upper furnace waterwalls. In-furnace lances
protrude into the upper furnace or convective pass and allow better mixing of the reagent
with the flue gas. In-furnace lances require either an air- or water-cooling circulation
system. Additionally, since the location of the temperature window changes with load,
multiple levels of injectors and/or lances will be required for effective NOX reduction over
the operating load range of the boiler. If the user specifies a number of injector lance levels,
but inputs zero for the number of injectors or lances, CUECost calculates the number of
injectors or lances using the equations below:
NI = (8.6 + 0.03 x BSIZE - 0.013 x Red) x NIL (Eq. D-6)
NL = (2 + 0.013 x BSIZE) x NLL (Eq. D-7)
where
NI = number of wall injectors
Red = NOX reduction efficiency, %
86
-------
Appendix D
NIL = number of injector levels
ML = number of lances
BSIZE = boiler size, MW
NLL = number of lance levels.
If the user enters values for both wall injectors and lances, then costs include both lances
and wall injectors. If wall injectors are to be used alone, then the user enters zero for both
the number of lance levels and the number of lances. Similarly, if lances are to be used
alone, the user enters zero for both the number of injector levels and wall injectors. For the
NH3-based SNCR process, the user can choose either steam or air as the atomizing medium.
Based on the user's choice, an annual operating cost for steam and/or electricity usage is
calculated.
D.3.2 Capital Costs
The main equipment areas in the battery limits for SNCR include the reagent receiving area,
storage tanks, and recirculation system; the injection system, including injectors, pumps,
valves, piping, and distribution modules; the control system; and air compressors. In
addition, NH3-based SNCR systems use vaporizers to vaporize the NH3 prior to injection.
The capital costs are estimated using modified equations from IAPCS v.5.0 (Gundappa,
1995). The IAPCS equations were modified to incorporate the extensive current cost data
developed and validated by EPA's ARD. IAPCS is a computer model developed for the EPA
NRMRL-RTP (formerly the Air and Energy Engineering Research Laboratory) to estimate
costs and performance for emission control systems applied to coal-fired utility boilers.
IAPCS was developed in the 1980s and has been updated over the years. Documentation for
the latest revision to IAPCS (Gundappa, 1995), completed in 1995, presents equations in
1982 dollars, with adjustments made using cost indices to normalize costs to other-year
dollars.
Cost and design information was available in a 1996 ARD study (EPA, 1996) for six
applications of urea-based (50% solution) SNCRs on various boiler types and sizes. The
design information for these cases was input to the IAPCS model, and the capital cost
estimates from IAPCS were compared to the ARD study estimates (EPA, 1996). The ratio of
the ARD study costs to costs calculated using IAPCS equations was determined for each
case. The ratios were then averaged, and the resulting average ratio was incorporated into
each IAPCS capital cost equation. The ratios were determined for Total Direct Capital Cost.
Itemization of equipment in major equipment areas varied between IAPCS and the ARD
study so that unique ratios could not be established for each equipment area. As a result,
the same ratio was added to each equipment cost equation. This approach was applied for
both urea- and ammonia-based SNCR, because the capital costs do not vary significantly
between the two processes (EPA, 1996). The algorithms for SNCR direct capital costs are
presented below. Plant cost indices from Chemical Engineering Magazine are included in the
equations to update direct capital costs. Direct capital costs for SNCR are shown in Table
D-8.
87
-------
Appendix D
Table D-8. Direct Capital Costs For SNCR (Installed Equipment Costs)
Urea-Based SNCR Process
Urea Storage & Handling = 38,143 x (Urea/8.7)0-417 x 0.915 x PCI/ 357.6
Urea Injection = (117,809 + 10,477 x NI + 53,111 x NL) x 0.915 x PCI/ 357.6
Misc. = (96,082 +106 x BSIZE + 898 x NI + 2,433 x NL) x 0.915 x PCI / 357.6
Air Heater Modifications = 11.2 x (acfm)0-772 x 0.915 x PCI / 357.6
Ammonia-Based SNCR Process
Ammonia Storage = 63,822 x (BSIZE)0-6 x 0.655 x PCI / 357.6
Handling, Injection, Controls
Air Heater Modifications = 11.2 x (acfm)0-772 x 0.655 x PCI / 357.6
where
Urea = urea injection rate, Ib/h
NH3 = ammonia injection rate, Ib/h
NI = number of wall injectors
NL = number of lances
acfm = flue gas volumetric flow rate at air heater inlet, ft3/min.
PCI = chemical engineering plant cost index from Chemical Engineering Magazine
= 388 for 1998 dollars and 357.6 for 1990 dollars.
Capital costs for instruments and controls, sales tax and freight are assumed to be included
in the algorithms listed above because they are updated with ARD costs that include these
items. The total direct cost with retrofit is determined by applying the retrofit factor to the
capital equipment cost subtotal, which is the sum of the equipment costs listed above. The
retrofit factor is a user input value that ranges from one for new applications to three for
the most difficult retrofit cases. Equations for indirect capital costs are given in Table D-9.
88
-------
Appendix D
Table D-9. Indirect Capital Costs for SNCR
General Facilities = Total Direct Cost with Retrofit x General Facilities (% of installed
cost)
Engineering fees = Total Direct Cost with Retrofit x Engineering Fees (% of installed cost)
Contingency = Total Direct Cost with Retrofit x Contingency (% of installed cost)
Total Plant Investment = Sum of Total Direct Cost with Retrofit, General Facilities,
Engineering fees, Contingency taking into account allowance for funds during construction
Preoroduction = Total Plant Investment x 0.02+ One Month Fixed Operating Costs +
One Month Variable Operating Costs (at full capacity)
Initial Ammonia (60 days) = NH3 x 24 x CF x 60 x UCNH3 /2000
Initial Urea (60 davs) = NH3 x 24 x CF x 60 x UCUREA /2000
where
CF = capacity factor, fraction
UCNH3 = ammonia cost rate, $/ton
= CO(NH2)2 cost rate, $/ton.
D.3.3 Operating and Maintenance Costs
The operating and maintenance cost equations for SNCR, taken from IAPCS v.5.0
(Gundappa, 1995), are shown below. Equations for the urea- and ammonia-based processes
are shown separately in the table. As in IAPCS, the operating labor costs are based on
2 person-hours required per 8-hour shift of operation. The default for maintenance labor
and materials costs is 4% of the total direct and indirect capital cost. The annual cost of the
reagent is the major operating cost item for the process and is calculated as the product of
the reagent usage in tons/year and the cost in dollars per ton of pure reagent. Electricity,
water, and steam requirements are based on vendor information. The increase in the
energy requirement for steam or air atomization is included in the operating cost
algorithms. Annual operating and maintenance costs for SNCR are shown in Table D-10.
89
-------
Appendix D
Table D-10. Annual Operating and Maintenance Costs for SNCR
Urea-Based SNCR Process ($/vear)
Operating and Supervisory Labor = 0.25 x 8,760 x UC0L
Maintenance Labor and Materials = Maintenance (%) x TPC
Reagent Requirement = Urea x 8760 x CF/2,000 x UCUREA
Electricity Requirement = (5.97 + 0.29 x NI + 0.87 x NL) x 8760x CF x UCELEC
Water Requirement = (1.0 x NI + 2.5 x NL) x 60 x 8760 x CF/1,000 x UCH2o
Ammonia-Based SNCR Process ($/vear)
Operating and Supervisory Labor Requirement = 0.25 x 8,760 x UC0L
Maintenance Labor and Materials Cost = Maintenance (%) x TPC
Reagent Requirement = NH3 x 8760 x CF/2,000 x UCNH3
Steam Requirement (for steam atomization) = BSIZE x 99.2 x 8,760 x CF/1,000 x UCSTEAM
Electricity Requirement (for steam atomization) = BSIZE x 0.12 x 8,760 x CF x UCELEC
Electricity Requirement (for air atomization) = BSIZE x 4.23 x 8,760 x CF x UCELEC
where
TPC = total direct and indirect capital costs, $ (see Table 3-1)
UCELEc = electricity rate, $/kWh
UCH2o = unit cost water, $/l,000 gallon
UCNH3 = NH3 cost rate, $/ton
= steam rate, $/MMBtu
= CO(NH2)2 cost rate, $/ton.
90
-------
Appendix D
D.3.4 CUECost Validation
To determine how successfully the IAPCS algorithms were modified using the ARD data,
CUECost was run using the design information upon which the ARD cases were based. Total
plant costs and operating and maintenance costs estimated using CUECost were compared
to the costs developed by ARD (EPA, 1996). Results from this comparison are presented in
Tables D-ll and D-12.
Total plant costs presented below include reagent storage and handling, injection system,
air heater modifications, and miscellaneous direct capital costs. Total plant costs also
include indirect capital costs such as engineering, general facilities and contingencies.
Chemical Engineering Magazine plant cost indices were used to report costs in consistent
year dollars.
The percent difference between ARD study costs and the CUECost estimates for total plant
costs ranged from -15% to +7% for the cases evaluated. Operation and maintenance costs
estimated by CUECost are 0 to 12% greater than those estimated by the ARD study (EPA,
1996).
D.4 NATURAL GAS REBURNING
D.4.1 Performance Parameters
The fraction of boiler heat input contributed by natural gas (reburn fraction) depends on the
desired NOX removal efficiency. The relationship between the reburn fraction and NOX
reduction efficiency, taken from IAPCS v.5.0, is based on vendor information and review of
NGR performance data:
RBFRAC = (NOxEFF - 0.48)/0.86 (Eq. D-8)
where
RBFRAC = boiler heat input contributed by natural gas (fraction)
NOX EFF = NOX reduction efficiency (fraction).
The relationship applies for NOX reduction efficiencies from 55 to 65% and yields reburn
fractions from 0.08 to 0.20. In CUECost, these are the valid input ranges for the NOX
removal efficiency and reburn fraction. If the user inputs both parameters within the valid
ranges, the input values are used for cost calculations. If only one parameter is outside of
the valid range, that parameter is calculated using the other parameter. If both input values
are outside of the valid ranges, a default reburn fraction of 0.15 is used with a
corresponding 61% NOX removal efficiency.
91
-------
Appendix D
Table D-ll. CUECost with Acid Rain Division Study Cases for SNCR (1990 dollars)*
Cyclone-!
Midwest
Bitumin
Selective Noncatalytic Reduction 150
Ired Wet-Bottom
Vertical- Wall-Fired
Fired
ern Eastern
ous Bituminous
Boiler Size (MW)
400 100 259
CUECost with Acid Rain Division Design Parameters
Default Input Parameters
Number of Injectors integer 1 8
Number of Lances integer 0
Urea/NOX Stoichiometric Ratio fraction 0.90
36 18 36
000
0.90 0.90 0.90
Desisn Parameters calculated by CUECost
Urea Injection Rate Ib/hr 2, 1 39
Air Heater Inlet ACFM ACFM 611,455 1
Capital Costs
Urea Storage
using Acid Rain Division Design Parameters ($ 1000)
& Handling 451
Urea Injection 364
Controls/Miscellaneous 152
Air Heater Modifications 39 1
Total Capital
Equipment Cost 1 ,358
Total Plant Cost (TPC) 1,833
TPC($/kW) 12.2
% Difference from Acid Rain Division Cost Study -7
5,297 973 2,439
712,635 407,633 1,078,935
658 324 476
589 364 589
203 146 185
865 286 605
2,314 1,120 1,855
3,124 1,513 2,505
7.81 15.1 9.67
7 -15 6
O&M Costs using Acid Rain Division Design Parameters ($1000/vear)
Operating and Supervisory Labor 46
Maintenance
Reagent
Electricity
Water
O&M Total
% Difference
Labor and Materials 27
1,102
3
2
1,181
from Acid Rain Division Cost Study 8
46 46 46
47 23 38
2,730 501 1,257
535
525
2,832 575 1,350
0 12 4
Source: EPA, 1996
92
-------
Appendix D
Table D-12. Acid Rain Division Study: SNCR Applications (1990 dollars)*
Selective Noncatalytic Reduction
Cyclone-Fired
Midwestern
Bituminous
Wet-I
Vertical-
Fired
ottom
Wall-Fired
Eastern Bituminous
Boiler Size (MW)
150
400
100
259
Acid Rain Division Costs and Design Parameters
Design Parameters from Acid Rain Division
Number of Injectors
Number of Lances
Urea/NOX Stoichiometric
Economizer Outlet
Ratio
integer 1 8
integer 0
fraction 0.90
ACFM 648,029
36 18
0 0
0.90 0.90
1,812,657 416,969
36
0
0.90
1,085,858
Acid Rain Division Capital Costs f$ 10001
Tanks, Pumps & Injectors
Pipes/Valves/Heat Tracing
Electrica!/PLC
Platform/Insulation/Enclosure
Total Capital Equipment Cost
Total Plant Cost (TPC)
TPC ($/kW)
Acid Rain Division Q&M Costs ($ 1 OOP/year)
Coal Consumption
Power consumption
Ash Disposal
General Maintenance
Urea Consumption
Water Consumption
O&M Total
* Source: EPA, 1996
615
510
180
135
1,000
680
160
280
480
530
180
90
673
725
155
155
1,440
1,980
13.2
1,094
2,120
2,920
7.3
2,824
1,280
1,770
17.7
512
1,709
2,357
9.1
74
19
3
31
961
7
198
59
7
48
2,494
18
36
7
1
27
437
3
97
31
3
37
1,119
7
1,295
93
-------
Appendix D
D.4.2 Capital Costs
Direct capital cost equations for NGR are presented below. The first equation includes the
installed costs of gas injectors, OFA ports, and related equipment. This equation was
developed by modifying the IAPCS equation for the same equipment area [Cost =
6,644,400 x (BSIZE/500)0'214] to reflect recent cost estimates from an ARD study (EPA,
1996). The ARD study estimated NGR costs for four different boiler sizes. To bring the
IAPCS model up to date, the constant in the equation (6,644,400) was replaced with a
variable. Then the equation was set equal to each of the ARD cost cases, and the equation
was solved to determine a new constant. The results showed that the new "constant" varied
linearly with boiler size. Therefore, the constant in the IAPCS equation was replaced with an
expression that is a function of boiler size (BSIZE x 3238 + 1504675).
The second equation shown includes the costs associated with piping natural gas to the
boiler from the metering station located at the utility plant fence line. The equation was
derived by fitting an exponential curve to ARD costs for natural gas piping. Plant cost indices
from Chemical Engineering Magazine are included in the equations to update direct capital
costs. Direct capital costs for NGR are shown in Table D-13.
Table D-13. Direct Capital Costs for NGR (Installed equipment cost)
Fuel injectors, overfire air ports, associated piping, valves, windbox, and control dampers
" '17e PCI
500 ) 357.6
PCI
Gas pipeline from fence line to boiler = 372xexp (2.64xlO"3x BSIZE)>
357.6
where
BSIZE = Boiler capacity (MW)
PCI = chemical engineering plant cost index from Chemical Engineering Magazine
= 388 for 1998 dollars and 357.6 for 1990 dollars.
Capital costs for instruments and controls, sales tax and freight are assumed to be included
in the algorithms listed above because they are updated with ARD costs that include these
items. The total direct cost with retrofit is determined by applying the retrofit factor to the
capital equipment cost subtotal, which is the sum of the equipment costs listed above. The
retrofit factor is a user input value that ranges from 1 for new applications to 3 for the most
difficult retrofit cases. Equations for indirect capital costs are given in Table D-14.
94
-------
Appendix D
Table D-14. Indirect Capital Costs for NGR
General Facilities = Total Direct Cost with Retrofit x General Facilities (% of installed
cost)
Engineering fees = Total Direct Cost with Retrofit x Engineering Fees (% of installed cost)
Contingency = Total Direct Cost with Retrofit x Contingency (% of installed cost)
Total Plant Investment = Sum of Total Direct Cost with Retrofit, General Facilities,
Engineering fees, Contingency taking into account allowance for funds during construction
Preproduction = Total Plant Investment x 0.02+ One Month Fixed Operating Costs + One
Month Variable Operating Costs (at full capacity)
D.4.3 Operating and Maintenance Costs
In general, natural gas reburning reduces the boiler operating costs associated with coal-
and ash-handling process areas, including maintenance, electricity, and ash disposal. Fuel
costs are generally higher, because the price of natural gas is typically higher than the price
of coal per unit of energy. The equations used by CUECost and taken from IAPCS for
estimating operating costs and savings are given below. The electricity requirement for coal
and ash handling processes decreases in proportion to the amount of reburn fuel used. The
default for maintenance costs for operating the NGR system is 1.5% of the total plant cost.
The empirical equation for estimating waste disposal savings includes a reduction of bottom
and fly ash as a result of firing gas. As in IAPCS, savings from reduced fly ash disposal are
estimated only for retrofit applications. The incremental fuel cost for firing gas is estimated
by multiplying the amount of gas burned by the fuel price difference between gas and coal.
Annual operating and maintenance costs and savings for NGR are shown in Table D-15.
Table D-15. Annual Operating and Maintenance Costs and Savings for NGR
Electrical Consumption Savings ($/year)
ELEC = 9.51 x 107 x Qin x CF x RBFRAC/HHV x UCELEc
Maintenance Cost ($/vear)
MAINT = Maintenance (%) x TPC - 1387.5 x RBFRAC x (BSIZE/500)0'6
Waste Disposal Savings ($/vear)
WASTE = [BA x RBFRAC + (NR - 1) x 4.336 x RBFRAC x PPHPRT x CF] x UCWASTE
Natural Gas Consumption Cost ($/vear)
GAS = Qin x RBFRAC x 8,760 x CF x (UCGAS - UCCOAL)
95
-------
Appendix D
where
Qin = boiler heat input, MMBtu/h
CF = capacity factor, dimensionless
HHV = higher heating value of coal, Btu/lb
UCELEc = electricity rate, $/kWh
TPC = total plant capital costs, $
BA = bottom ash rate, tons/year estimated from:
BA = BAF x ASH x, 500/HHV x Qin x 8,760 x CF/2,000
where
BAF = bottom ash factor, dimensionless
ASH = percent ash in coal, wt.%
NR = retrofit status, 1 for new "grass root" installation (retrofit factor = 1) and 2 for retrofit
application (retrofit factor > 1)
PPHPRT = fly ash rate, Ib/h
UCWASTE = waste disposal rate, $/ton
UCGAS = gas rate, $/MMBtu
UCCOAL = cost for coal, $/MMBtu.
D.4.4 CUECost Validation
Total plant costs and operating and maintenance costs estimated by CUECost algorithms
were compared to current cost data developed and validated by EPA's ARD (See Tables
D-16 and D-17). Four applications of NGR for various boiler types, boiler sizes and coals
were evaluated with CUECost. The design information provided by ARD for the four NGR
applications was used to evaluate the direct capital cost equations from CUECost.
Total plant costs presented below include the fuel injectors, overfire air ports, associated
piping, valves, windbox, and control dampers and the gas pipeline from the fence line to
boiler. The total plant costs include direct costs listed above as well as indirect capital costs
for engineering, general facilities and contingencies. Chemical Engineering Magazine plant
cost indices were used to report costs in consistent year dollars.
The percent difference between ARD study costs and the CUECost estimates for total plant
costs ranged from 0 to 11% for the cases evaluated. Operation and maintenance costs
estimated by CUECost are 7 to 12% lower than those estimated by the ARD study.
96
-------
Appendix D
Table D-16. CUECost with Acid Rain Division Study Cases for NGR (1990 dollars)*
Natural Gas Reburning
Cyclone-Fired
Midwestern Bituminous
Wet-B
Vertical-
Fired
ottom
Wall-Fired
Eastern Bituminous
Boiler Size (MW)
150 400
100 259
CUECost with Acid Rain Division Design Parameters
Design Parametersfrom AcidRain Division
Gas Reburn Fraction
Capital Costs using Acid Rain Division Design Parameters f$ 1000)
Gas Pipeline from Fenceline to Boiler
Fuel Injectors, Overfire Air Ports and Associated
Piping, Valves, Windbox and Control Dampers
Total Capital Equipment Cost
Total Plant Cost (TPC)
TPC($/kW)
% Difference from Acid Rain Division Cost Study
2,000
2,720
3,590
23,9
11
Q&M_Costs using Acid Rain Division Design Parameters ($1000/vear)
Electrical Consumption Savings
Maintenance
Waste Disposal Savings
Natural Gas Consumption
O&M
% Difference from Acid Rain Division Cost Study
Source: EPA, 1996
3,470
4,863
6,419
16.1
1,684
2,315
3,056
30.6
0
2,646
3,606
4,760
18.4
(54)
54
(43)
1,467
1,423
-11
(152)
96
(122)
4.110
3,933
-7
(34)
46
(23)
866
855
-12
(89)
71
(61)
2.290
2,212
-12
97
-------
Appendix D
Table D-17. Acid Rain Division Study: NGR Applications (1990 dollars)*
Natural Gas Reburning
Cyclone-Fired
Midwestern Bituminous
CHAPTER 2
Vertical-
Fired
WET-
Wall-Fired
Eastern Bituminous
Boiler Size (MW)
150 400
100
259
Acid Rain Division Costs and Design Parameters
Design Parameters from Acid Rain Division
Gas Return Fraction 0.16 0.16 0.16 0.16
Acid Rain Division Capital Costs ($1000)
Fuel Piping System 510 1040 500 803
Burners/OFA 585 1840 540 1191
Electrical/BMS Modifications 735 1000 750 907
Windbox/Duct/Modifications 165 120 60 104
Platform/Insulation/Demolition 405 520 410 466
Total Capital Equipment Cost 2,400 4,520 2,260 3,471
Total Plant Cost (TPC) 3,225 6,080 3,050 4,662
TPC($/kW) 21.5 15.2 30.5 18.0
Acid Rain Division O&M Costs ($1 OOP/year)
Coal Consumption
Ash Disposal
General Maintenance
Natural Gas Consumption
O&M Total 1,607 4,236 969 2,510
* Source: EPA, 1996
(1,630)
(50)
50
3,239
(4,564)
(141)
93
8.848
(1,201)
(27)
47
2.150
(3,184)
(71)
71
5,694
D.5 LOW-NOx BURNER TECHNOLOGY
D.5.1 Capital Costs
CUECost estimates capital costs for retrofitting tangentially-fired and wall-fired boilers with
LNBT. The cost algorithms are based on a study of LNBT by ARD (EPA, 1996). The study
obtained information from 56 boilers--35 wall-fired and 21 tangentially-fired. The
information provided for these retrofit cases was used to develop empirical equations that
estimate total capital cost for LNBT retrofits as a function of boiler size. CUECost only
addresses retrofit installations because most new boilers include LNBT in their base design.
The "bottom-line" costs include direct capital costs and indirect costs such as engineering,
general facilities, and contingencies. The scope of direct costs collected for the ARD study
includes (1) for the burner portion: burners or air and coal nozzles, burner throat and
98
-------
Appendix D
waterwall modifications and windbox modifications; (2) for applicable combustion air
staging: waterwall modifications or panels, windbox modifications, and ductwork; and (3)
scope adders or supplemental equipment such as replacement or additional fans, dampers,
or igniters necessary for the LNBT. The scope of installed LNBT retrofit capital costs includes
materials, construction and installation labor, engineering, and overhead costs (40 CFR, Part
76, Appendix B).
The ARD study found that capital costs vary greatly depending on the scope of the retrofit
and the degree of modification necessary. As a result, the cost data were statistically
separated into subsets of high and low cost cases for each boiler type. Cost equations were
then developed by ARD for the high and low cost subsets, as well as for the entire set of
cost data. The CUECost user selects from any of the three ARD cost equations based on the
estimated retrofitting difficulty: high, average or low. The equations are given in 1995
dollars and include the user input Chemical Engineering Magazine plant cost index (PCI) to
escalate to the desired cost year. Total capital costs for LNBT retrofit are shown in Table D-
18.
Table D-18. Total Capital Costs for LNBT Retrofit
Tangential-fired Boilers
High Cost: 57.04 x (300/BSIZE)/V0.679 x 1000 x BSIZE x PCI / 357.6
Average Cost: 21.20 x (300/BSIZE)^0.35 x 1000 x BSIZE x PCI/ 357.6
Low Cost: 11.71 x 1000 x BSIZE x PCI / 357.6
Wall-fired Boilers
High Cost: 27.72 x (300/BSIZE)/V0.573 x 1000 x BSIZE x PCI / 357.6
Average Cost: 15.37 x (300/BSIZE)/V0.35 x 1000 x BSIZE x PCI/ 357.6
Low Cost: 6.53 x (300/BSIZE)^0.857 x 1000 x BSIZE x PCI/ 357.6
where
BSIZE = boiler size, MW
PCI = Chemical Engineering Plant Cost Index for desired cost basis year.
A cost comparison between CUECost and IAPCS cost algorithms was not possible because
design and economic parameters were not given in the ARD study of NGR technology.
D.5.2 Operating and Maintenance Costs
The only direct operating costs associated with LNBT are for maintenance labor and
materials. No energy penalty is assumed to be incurred with this technology. Costs for the
controls, administration and support labor, including overhead, are 30% of the maintenance
labor costs. Annual operating and maintenance costs for LNBT are shown in Table D-19.
99
-------
Appendix D
Table D-19. Annual Operating and Maintenance Costs for LNBT ($/year)
Maintenance Labor = TPC ($) x Maintenance Labor (0.8%)
Maintenance Materials = TPC ($) x Maintenance Materials (1.2%)
Administration/Overhead = Maintenance Labor ($/year) x 30%
where
Maintenance Labor = Annual maintenance labor cost, $/year
Maintenance Materials = Annual maintenance materials cost, $/year
Administration/Overhead = Annual costs, $/year
TPC = Total Plant Costs ($).
D.5.3 CUECost Validation
Total plant costs estimated by CUECost for the four boiler sizes examined for the other NOX
technologies are shown in Table D-20. The CUECost algorithm for total plant cost is identical
to the cost function presented by the ARD study of LNBT (EPA, 1996). A comparison is not
presented for operating and maintenance costs because these costs are highly boiler-
specific.
Table D-20. CUECost with Acid Rain Division Study Cases for LNBT (1990 dollars)*
Low NO., Burner Technology
CUECost Total Plant Cost ($ 1000)
Wail-Fired
T-Fired
150
Boiler Si
400
ze (MW)
100
259
Average Case
2,938
4,053
5,559
7,668
2,258
3,114
4,191
5,781
% Difference from Acid Rain Division Study
Wall-Fired
T-Fired
Source: EPA, 1996
0
0
0
0
0
0
0
0
D.6 HG CONTROL TECHNOLOGY
The algorithm of PAC control cost has the form
x = MIN(X,D)
y = Logio(Injection Rate) = Ax2+Bx + C
(Eq. D-9)
(Eq. D-10)
100
-------
Appendix D
where X is the mercury reduction fraction desired and the injection rate is expressed in
Ib/MMacf. A, B, and C are provided in the table below. D is used to specify the maximum
fraction of mercury that can be removed, essentially an upper limit. In CUECost, D is
actually multiplied by 0.99 so that the maximum removal that can be calculated equals 99%
of D. Calculation results are shown in Figures D-l through D-5.
Constants for Eqs. D-9 and D-10 are shown in Table D-21.
Table D-21. Constants for Eqs. D-9 and D-10
PAC, Bituminous FF
PAC, Bituminous ESP
PAC, Subbituminous FF
PAC, Subbituminous ESP
Treated PAC, Subbituminous FF
Treated PAC, Subbituminous ESP
Treated PAC, Bituminous ESP
A
1.6944
-0.6647
-0.4318
3.308
0.0
0.8837
0.0
B
-1.1267
2.1232
1.9551
0.754
2.5007
0.4485
1.207
C
-0.0009
-0.0665
-0.8937
-0.5925
-2.2097
-0.575
-0.2277
D
1.0
1.0
1.0
0.7
1.0
1.0
1.0
Log Injection Rate vs Reduction
Bituminous FF
60% 65% 70% 75% 80% 85% 90% 95% 100%
reduction
Figure D-l. PAC, Bituminous FF
y = 1.6944X2 - 1.1267x - 0.0009
R2 = 0.8409
x < 100%
Log Injection Rate vs Reduction
Bituminous ESP
10% 20% 30% 40% 50% 60% 70% 80% 90% 100
reduction "
Figure D-2. PAC, Bituminous ESP
y = -0.6647X2 + 2.1232x - 0.0665
R2 = 0.8797
x < 100%
101
-------
Appendix D
Log Injection Rate vs. Reduction
Subbituminous FF
0.8
0.6 -
0.4-
0.2 -
0 -
-0.2 -
-0.4 -
-0.6 -
-0.8
-/-=-
:_-Q.4318x + 1.9551X - 0.8937
R2 = 0.9_55 _!_
0%
10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Reduction
Figure D-3. PAC, Subbituminous FF
y = -0.4318x2 + 1.9551x - 0.8937
R2 = 0.955
x < 100%
Log Injection Rate vs Reduction
Subbituminous ESP
1.6
Iog10 (injection)
-best fit
• U./b4x - U.5925
10% 20% 30% 40% 50% 60% 70% 80%
redn
Figure D-4. PAC, Subbituminous ESP
y = 3.308x2 + 0.754x - 0.5925
R2 = 0.7856
x < 70%
0.8 T—
0.6
0.4
0.2
£ 0
ro
2 -0.2 ]
-0.4
Log Injection Rate vs Reduction
BPAC
BPAC In Flight, Parametric
in-flight best fit
BPAC FF
-FF best fit
y = 2.5007X - 2.2
R2 = 0.6062
097
0% 10% 20% 30%
% 50% 60% 70% 80% 90% 100%
reduction
Figure D-5. BPAC
Treated PAC, Subbituminous FF
y = 2.5007X - 2.2097
R2= 0.6062
x < 100%
Treated PAC, Subbituminous ESP
y = 0.8837X2 + 0.4485x - 0.575
R2 = 0.8497
x < 100%
102
-------
Appendix D
The algorithm of PAC control cost was incorporated into CUECost and a lookup table, located
in the Constants_CC worksheet, and developed to ease users' selection. To use the lookup
table in the CUECost workbook, a three-digit index key is constructed by summing the
following digits:
• 100 or 200 for bituminous or subbituminous coal, respectively
• 10 or 20 for in-flight or filter capture, respectively
• 1, 2, or 3 for enhanced PAC (denote EPAC, such as brominated PAC), standard PAC,
or other sorbent, respectively.
For example, a PRB coal fired boiler with a cold-side ESP and using enhanced PAC would
have an index key of 211. If the same boiler were retrofit with a PJFF for a TOXECON
arrangement and standard PAC were used, the index key would then be 222.
For the purpose of CUECost, subbituminous and lignite coals are treated the same way. For
all practical purposes, the two categories are bituminous and low rank. In reality some
bituminous coals with very low chlorine levels may behave more like low rank coals and
some low rank coals with unusually high chlorine may behave more like bituminous coals.
This issue will be addressed in the future.
Sample Calculations
To see how the new algorithms worked, some calculations of control cost were made. Figure
D-6 shows comparison calculations for cost of controlling mercury for various situations on a
500 MW low sulfur bituminous coal-fired boiler equipped with an ESPc as a function of total
mercury removal.
Cost of Mercury Reduction
LS Bituminous Coal and ESP
3.00
2.50
2.00 |
= 1.50
0.00
•LS Bituminous Coal with
CS-ESP and EPAC
•LS Bituminous Coal with
CS-ESP and PAC
Percent Total Reduction
Figure D-6. Cost of Mercury Reduction, LS Bituminous Coal and ESP
103
-------
Appendix D
In all calculations, addition of sorbent is assumed to cause fly ash in contact with sorbent to
be disposed of at a differential cost of $30/ton. Also, the cost of sorbent is assumed to be
$1000/ton for standard PAC and $1500/ton for EPAC.
As shown in Figure D-6, EPAC incurs the lowest cost while maintaining the same level of Hg
removal efficiency. For this reason, a TOXECON retrofit with PAC is not cost effective
compared to EPAC. However, if fly ash is currently land filled, the differential disposal cost is
negligible and an estimated 0.38 mills/kWh could be deducted from the cost of controlling
mercury with sorbent injection upstream of an ESP (Staudt et al., 2003).
D.7 CO2 MEA CONTROL SYSTEM COST ALGORITHM DEVELOPMENT
The cost algorithms associated with the CO2 MEA control processes were developed based
on DOE/NETL dataset in 2007 (DOE 2007). Algorithm development began with derivations
from DOE/NETL database by running a series of data regressions and identified suitable
equations. These datasets were then utilized to predict the cost by assuming the MEA
concentration was at 30%. The derived regression equations represent a typical MEA
operating plant for equipment areas and for specific O&M costs.
D.7.1 Capital Cost
The MEA island contains a pretreatment unit, a CO2 absorber, and a CO2 stripper. Costing
has been based on the most recent DOE/NETL (2007) cost analysis of MEA CO2 capture.
MEA mainly reacts with two moles of amine and one mole of CO2. Mitsubishi Heavy Industry
(MHI) has developed a new solvent (named KS-1) primarily reacting with one mole of amine
and one mole of CO2. Little information has been found for the specific cost of the KS-1
based system. The capital costs found in MHI presentations did not provide sufficient detail
to determine a comparable basis for Bare Erected Costs. Further, these cost estimates
aggregated the recovery island and the compression island costs. As the KS-1 contains the
same processes as the MEA, bare erected costs for the KS-1 island are therefore calculated
in the same manner as the MEA island bare erected costs. As a regression of DOE dataset,
the model MEA island cost will be:
Y=69,412,748 x X0'5741 (Eq. D-ll)
where
Y = bare erected cost, 2007 $
X = CO2 capture, metric ton/h.
Gas exiting the CO2 stripper must be compressed and dehydrated to accommodate
transport and disposal. In DOE's model (DOE/NETL, 2007), moist CO2 from the CO2
stripper's reflux drum enters the compressor at 21 °C (69 °F) and nominally 160 kPa (23
psi). CO2 is compressed in a 6-stage integrally geared compressor. Intercoolers between
stages cool the gas using chilled water from the plant's cooling tower. After exiting the
104
-------
Appendix D
compressor and presumably a final heat exchanger, the CO2 is dried to <20 ppmv water in
a TEG dehydrator. Dry gas exiting the dehydrator is at 15.27 MPa (2215 psi) and 51 °C
(124 °F). Regression of the two coal-fired plants and one natural gas combustion plant
cases presented results in a power law model for capital costs scaled to the power used for
compression raised to 0.5429 power. This regression is based on a very limited data set.
The uncertainty of the capital estimate increases as conditions deviate from those used in
model development. As the result, the model compressor island cost will be
Y=103,045 x X °-5429 (Eq. D-12)
where
Y = bare erected cost, 2007 $
X = compressor power, kW.
Indirect capital costs for CO2 control are shown in Table D-22.
Table D-22. Indirect Capital Costs for CO2 Control
General Facilities = Total Direct Cost with Retrofit x General Facilities (% of installed
cost)
Engineering fees = Total Direct Cost with Retrofit x Engineering Fees (% of installed cost)
Contingency = Total Direct Cost with Retrofit x Contingency (% of installed cost)
Total Plant Investment = Sum of Total Direct Cost with Retrofit, General Facilities,
Engineering fees, Contingency taking into account allowance for funds during construction
Preoroduction = Total Plant Investment x 0.02 + One month fixed operating costs +
One month variable operating costs (at full capacity)
Inventory = 0.5% Total Plant Cost (TPC)
D.7.2 Operating and Maintenance Costs
Steam
Steam is used in the reboiler of the CO2 stripper to reverse the CO2 reactions that took place
in the CO2 absorber. In addition to the heat required for CO2 regeneration, some steam is
used evaporating water in the stripper. In the recent DOE analysis (DOE/NETL, 2007), 1529
Btu/lb CO2 were required to regenerate CO2 in most cases for coal combustion while 1590
Btu/lb CO2 were required for natural gas combined cycle (NGCC) (DOE, 2007). The major
portion of this difference arises from the lower concentration of CO2 in the NGCC gas. Rao
(2002) reported a range of steam use of 3800-4000 kJ/kg CO2 (1636-1723 Btu/lb CO2).
Steam use at the KS-1 installation at a Malaysia urea plant was 3270 kJ/kg CO2 (1409
Btu/lb CO2) with a feed gas containing 8% CO2 on a dry basis. Data presented by DOE
(2007) and Rao (2002) suggest that the steam requirement decreases with increasing CO2
concentration. Steam consumption in the reboiler is estimated in this model based on a
power law curve fit re-created from MHI's presentation of steam use. Assuming the CO2
105
-------
Appendix D
concentration is on a dry volumetric basis, this model predicts 3140 kJ/kg CO2 steam
consumption for an 8% CO2 flue gas as documented for the Malaysia facility, a 4% error.
Assuming there is no significant difference for different MEA processes, the steam use for
regenerating MEA solvent in the worksheet will be regressed as:
Y = 4109.2 x X'0-13 (Eq. D-13)
where
Y = energy demand, kJ/kg CO2
X = CO2 concentration, %.
Cooling Water Makeup
Cooling water will be used to remove heat from the direct contact cooler (DCC) during
pretreatment, remove heat generated in the absorber, condense steam in the reflux drum
of the CO2 stripper, remove heat from the lean solvent returned from the CO2 stripper, and
remove heat generated by the compressor. As a budgetary estimation of cooling water
makeup, we simplify the total use of cooling water makeup as:
Cooling water makeup = Loss due to DCC + Lump-sum Loss from MEA island + Loss from
compressor island.
Enthalpy of the flue gas entering the direct contact cooler is calculated based on mass flows
and temperatures exiting from the previous unit operation, for example, wet scrubbers.
Enthalpy of the gas flow exiting the direct contact cooler is based on the mass flow exiting
the direct contact cooler, assuming the exiting gas is saturated with water. Heat loss from
the direct contact cooler is then calculated as the difference between the above two
enthalpies. Although the pretreatment may also involve SO2 polishing, this heat duty is
expected to be inconsequential in comparison with the heat duty of condensing water vapor
from the flue gas.
The MEA island cooling water requirement is estimated based on the steam requirement for
the CO2 stripper reboiler. The heat supplied to the reboiler is sufficient to reverse the CO2
absorption, evaporate water and increase the enthalpy of the stripper effluent. The heat of
reaction is removed in heat exchangers associated with the absorber. Steam is condensed in
the stripper reflux drum and returned as reflux. Enthalpy of the stripper effluent in excess of
the heat transferred to the stripper influent must be removed in a heat exchanger
associated with the absorber; the lean solvent from the stripper is cooled to a lower
temperature than the rich solvent effluent from the absorber. For simplicity, the heat input
from steam will be equal to the heat rejected through cooling water evaporation.
Intercoolers are heat exchangers located between compressor stages with an intention to
reduce the temperature of the gas, and, in turn, to protect the compressor from heat
damage and reduce the power requirements. Chilled water is required for this purpose. The
heat duty is assumed to be a fraction of power used by the compressors, as shown in Eq.
106
-------
Appendix D
(D-14). The fraction is equal to the overall compressor efficiency which is equal to the
isentropic efficiency of the compressor multiplied by the efficiency of the drive. The energy
losses from the drives are assumed to flow to the surrounding environment, not to the CO2
compression. In general, the isentropic efficiency is assumed to be 84% and the drive
efficiency of the electric motor is 95%. Consequently, the overall efficiency is 80%.
8W (Eq. D-14)
.
0.84
The makeup cooling water flow rate (gpm) is equal to evaporation rate of water
(approximately 2 gpm per 1 million Btu/h of heat) multiplied by an appropriate correction
factor: l/(cycle of concentration-1). The cycles of concentration = chlorides in tower water/
chlorides in makeup water.
Power
The power used in the MEA Island is primarily consumed by an induced draft fan after the
direct contact cooler. Pumps used to recirculate condensate and the MEA solvent represent
the remainder of the MEA island power demand.
Power used in the induced draft fan is estimated based on the average volumetric flow
entering and exiting the fan and the pressure differential across the fan. The recent DOE
model (DOE, 2007) indicates a pressure differential of 0.014 MPa (2 psi) across this fan to
overcome the pressure drop in the absorber. Gas is assumed to enter the fan saturated with
water at 32 °C from the direct contact absorber. The recent DOE model (DOE, 2007)
indicates a temperature rise across the fan of 17 °C for PC cases; an outlet temperature of
49 °C will be used in all cases. The pressure difference across the absorber with the MHI
design using structured packing is substantially less than the power required with a
randomly packed column. MHI claimed the pressure differential is 1/7 that of conventional
MEA technology. Assuming isentropic compression and a k of 1.4, a temperature rise of 2
°C is estimated across the fan using MHI's design. The fan inlet temperature is therefore
assumed to be 47 °C. The flue gas entering the fan is expected to be at nearly atmospheric
pressure. An overall efficiency of 80% will be used to calculate the expected power
requirement of the fan. Consequently, the power required by fan will be:
Power (HP) = Gas flow (acfm) x AP(psi)/229/efficiency (Eq. D-15)
At this stage of estimation, the power for all the remaining pumps is estimated at 0.006
kWh/kg CO2 removed, the average MEA island power use for PC units in the 2000 and 2004
analysis. Since the DOE 2007 analysis reflects similar steam requirements for PC units and
NG-fired units, the loading of the MEA and the parasitic power is assumed to be similar for
PC and NG-fired units. No new power requirements could be assessed from the DOE 2007
analysis since the fan power is added to the rest of the MEA system power requirements.
Though a higher recirculation rate in the DCC is anticipated for the NG-fired units, this
power consumption was disregarded at this level of analysis.
_ 107
-------
Appendix D
The larger share of the power consumed in the compressor island is associated with
compression of the CO2. At the study level of estimation, power required to run the TEG
dehydrator is ignored. Power consumed by the intercoolers is contained in the power
required for compression as pressure loss across the intercooler. The efficiency of the drive
is included in the overall efficiency of the compressor.
Each stage of the compressor is nearly isentropic as there is limited surface area within the
stage to remove heat in large compressors. Heat is, therefore, largely removed between
stages by heat exchangers. Power consumption for compression is assumed to be isentropic
in each stage with an efficiency factor applied to correct for non-ideal behavior of the
compressor. Due to high pressures involved, the power estimate must account for the non-
ideal behavior of the gas as well. The deviation from ideal gas behavior is corrected with a
compressibility factor (Z). The estimation procedure used for Z is shown in Appendix F. An
overall efficiency of 0.80 has been used consistent with the DOE (2007) analysis. The
overall compression work is then calculated for each stage of compression as follows (Ulrich,
1984).
_ ~ k~l
RTls7 ( P ~\ k
(Eq. D-16)
m k-l
where
PI= inlet pressure of the compressor
P2=outlet pressure of the compressor
TI= inlet gas temperature, K
R=188.9 J/(kg x K) for CO2,
k=1.28 for CO2. Ratio of constant pressure to constant volume heat capacity
m= flow rate of gas, kg/s.
n=0.8, overall efficiency
Z=compressibility factor (See appendix F for calculation).
The power requirement for compression appears to decrease with each additional stage of
compression. This work advantage is offset somewhat by the cost of the interstage coolers
and the pressure drop between each of the stages. Because of the pressure drop across the
interstage cooler, the pressure of the gas exiting a stage is slightly higher than the pressure
of the gas entering the successive stage. At this level of estimation, a constant pressure
drop of 0.01 MPa (1.5 psi) per cooler is assumed.
Generally, the power required and costs of a compressor are minimized when the same
amount of compression work is accomplished in each stage. Since compressibility can vary
from nearly 1 to 0.5 in these compressors, the compression ratio in installed equipment is
likely to be different in each stage. This effect is most dramatic in the supercritical region
where compressibility will be managed by controlling the stage inlet temperature. For
computational simplicity, the pressure ratio of the final stage of compression is estimated to
108
-------
Appendix D
be 30% higher than the other stages if the final pressure is supercritical. The pressure ratio
for each of the preceding stages is evenly distributed prior to considering the pressure drop
of the interstage cooler.
(Eq-D-17)
The overall compressor efficiency includes the isentropic efficiency of the compressor and
the efficiency of the drive. The work lost to the drive efficiency is assumed lost to the
environment and not transferred to the CO2. The isentropic efficiency is assumed to be 84%
while the overall efficiency is assumed to be 80%; the resulting drive efficiency of the
electric motor is 95%.
(Eq. D-18)
0.84
During estimation, the average Z is 0.8 for the first n-1 stage compressors, and equal P2/Pi
is also assumed for the first n-1 stage compressors.
Cost of MEA
MHI estimated MEA consumption at 0.45 kg MEA/metric ton CO2 for its CO2 capture
technology. For the other MEA based processes, Rao (2002) reported a range of 0.5-3.1
kg/metric ton CO2 and a typical value of 1.5 kg MEA/metric ton CO2. For estimation
purposes, the consumption rate of MEA employed in the worksheet will be at the typical
value of 1.5 kg MEA/metric ton CO2 for MEA process. Inhibitors are added to the absorber to
prevent corrosion. The cost of inhibitors is estimated at 20% cost of MEA.
Cost of NaOH
Sodium hydroxide is used to bring down the SO2 concentration in the influent gas to less
than 10 ppm and to regenerate the MEA from the sulfate salts. The consumption of NaOH
for SO2 removal is based on the removal of SO2 from the influent gas. For the consumption
of NaOH in the reclaimer, no data on sodium hydroxide use were found in MHI papers or
presentations for KS-1. Rao (2002) reported a typical value of 0.13 kg NaOH/metric ton
CO2. This value will be used in the worksheet.
Cost of Activated Carbon
Activated carbon is used to remove high molecular weight products. Rao (2002) reported a
typical value of 0.075 kg carbon/metric ton CO2. MHI claims a consumption of 0.06 kg
carbon/metric ton CO2. As the difference is not significant and the impact on the total cost is
minor, a typical value of 0.075 kg carbon/metric ton CO2 will be used in the worksheet.
109
-------
Appendix D
D.8 CO2 CAP CONTROL SYSTEM COST ALGORITHM DEVELOPMENT
The algorithm developed for CO2 CAP control was based on the system description and data
sets in the DOE/NETL report (DOE 2007). The algorithm for capital cost is developed
through the comparisons with the MEA process. O&M cost is estimated based upon the
system description by Sherrick et al. (2008).
D.8.1 Capital Cost
As with the MEA process, the CO2 CAP control system contains a pretreatment unit, a CO2
absorber, and a CO2 stripper in the absorption island. The only cost information found to
date for the CAP is based on analysis presented at Lyon, France in 2007 (EPRI, 2007). This
analysis aggregates the total plant cost for a 3891 MMBtu/h supercritical pulverized coal
(SCPC) plant with and without CAP and with MEA CO2 capture.
For modeling purposes, the bare erected cost of the CAP separation island will be
approximated as a fixed fraction of the modeled cost for a comparable MEA island. The total
cost of the CAP CO2 capture, CAP separation island and compression island is estimated
based on the difference in TPC of the SCPC with CAP and the SCPC without CAP to be $120
million. Though this approach includes plant effects other than the CO2 capture system costs
such as steam takeoffs, turbines, condenser, and chilled water systems, this shortcut is
considered expedient in lieu of replicating the entire plant economic analysis. To return
these costs to the bare erected cost basis, a constant escalation factor, 23% in aggregate,
is applied. The estimated bare erected cost of the CAP CO2 capture is $97.6 million. The
bare erected cost of the MEA system analyzed in the DOE/NETL report (2007) is $111.8
million.
The capital costs associated with a CAP and MEA CO2 capture are expected to be distributed
differently between the separation island and the compression island. The compression
island for the CAP is expected to be significantly cheaper than the compression island for
MEA due to the high pressure, >400 psi, output of the CAP regenerator; the MEA
regenerator is evaluated at 27.2 psia. The cost of the CAP separation island and the MEA
separation island is estimated by subtracting the estimated cost of the respective
compression islands from their respective bare erected costs of the CO2 capture. The power
requirement of the CAP compression island is estimated assuming 400 psia inlet pressure,
1217 psia outlet pressure, and 69 °F (21 °C) inlet temperature using a single stage
compressor. Using the correlation developed from the DOE/NETL report (2007), the 6277
KW compressor estimated for the CAP CO2 capture would have a bare erected cost of $11.9
million in 2007, January 2000 bare erected costs are estimated at $9.1 million. Using the
correlation developed in the DOE/NETL report (2007), the 29730 kW compressor specified
for the MEA CO2 capture would have a bare erected cost of $27.6 million in 2007. January
2000 bare erected costs are estimated at $21.2 million.
The estimated bare erected cost of the CAP separation island is $88.5 million. The estimated
bare erected cost for the MEA separation island in the 2000 Parsons study is $90.6 million.
110
-------
Appendix D
The CAP separation island bare erected costs are estimated at 97.7% of the comparable
cost of an MEA separation island.
Electricity
The main power consumers in the CAP separation island are expected to be the blower
upstream of the absorber, recirculation pumps, and chiller. The Alstom analysis provides
parasitic power for the overall study plant without providing details on power requirements
for compressor, blower, pumps, or chillers. The relative power consumption of each function
is expected to change with inlet gas composition and temperature. Since detailed
information is not available, process conditions will be assumed to fix power consumption.
Power used in the induced draft fan is estimated based on the average volumetric flow
entering and exiting the fan and the pressure differential across the fan. The recent DOE
model (2007) indicates a pressure differential of 0.014 MPa (2 psi) across this fan to
overcome the pressure drop in the absorber. Gas is assumed to enter the fan saturated with
water at 32 °C from the direct contact absorber. The recent DOE model indicates a
temperature rise across the fan of 17 °C for PC cases; an outlet temperature of 49 °C will
be used in all cases. An 80% overall efficiency is assumed for the fan.
The power for pumping is assumed to be 0.006 kWh/kg CO2 removed. This value was
derived from the DOE/NETL report (2007) for MEA CO2 capture.
The chiller is used to cool the flue gas exiting the ID fan prior to the absorber and to remove
the heat of reaction. The flue gas is assumed to exit the fan at 49 °C (saturated at 32 °C)
and will be cooled to saturation at 2 °C with the chiller. The heat load at the absorber will be
approximated using the steam heating duty of the regenerator, 267 Btu/lb CO2. Power use
by the chiller will be approximated as 1/4 total cooling duty of the chiller. The chiller is a
mechanical chiller for removing heat.
Steam
Steam is used in the CAP CO2 capture to reverse ammonium bicarbonate back to NH3 and
CO2 (NH3HCO3+heat=NH3+CO2+H2O). The Alstom analysis (DOE/NETL, 2007) uses 179,500
Ib/h low pressure steam in the recovery of 710,423 Ib/h of CO2, 0.253 Ib steam/lb CO2.
Assuming 1058 Btu/lb steam can be utilized, 267 Btu/lb CO2 is required. The heat duty is
very close to the heat of reaction suggesting minimal reflux in the regenerator and
extraordinarily efficient heat transfer in the cross flow heat exchanger between the absorber
and regenerator.
Cooling Water
The cooling water duty is obtained by subtracting the sum of the other energy flows out of
the CAP island from the energy flows into the CAP island; the energy flows out of the CAP
island must balance the energy flows into the CAP island.
Ill
-------
Appendix D
Specific enthalpy of the flue gas entering the CAP island will be calculated based on
temperature and composition information resulting from operation of the prior unit. Much of
the water in the flue gas will be condensed in pretreatment contributing enthalpy to a CAP
water balance.
Specific enthalpy of the flue gas exiting the CAP island will be calculated based on the mass
balance composition assuming 90% CO2 removal, 100% SO2 removal, and water saturation;
no significant removal of other gases is anticipated in the absorber. Though the exhaust
temperature from the CAP island is not known, Alstom includes a second direct contact heat
exchanger to recover some of the heat removed in preconditioning (DOE/NETL, 2007). The
temperature of the gas exiting the CAP island is assumed to be the wet bulb temperature of
the flue gas entering the CAP island.
CO2 exiting the stripper reflux drum and leaving the CAP island is estimated to be pure CO2
saturated with water. The temperature of the CO2 exiting the stripper reflux drum in the
MHI design was not known but is assumed to be 21 °C as found in the MEA island analysis.
Steam is primarily used in the reboiler to regenerate solvent and produce concentrated CO2.
Steam use is estimated by the net heat required for regeneration, based on the amount of
CO2 recovered, 267 Btu/lb CO2.
The heat balance in the water streams is difficult to estimate with certainty because the
amount of fresh makeup water added to scrub the absorber outlet gas is not known for the
MHI design. It is not clear whether the scrubbing water is derived from the direct contact
cooler or fresh makeup water. For estimation purposes, the water used for scrubbing the
absorber outlet gas is assumed to be derived entirely from the direct contact cooler. The net
amount of water condensed is therefore the difference in the water in the flue gas entering
the CAP island and the water in the gases leaving the CAP island in the stack gas and CO2
gas streams. The specific enthalpy of the net condensed water is estimated at 5 °C warmer
than the wet bulb temperature of the flue gas entering the CAP island.
Work is transferred to the flue gas and working fluids through the action of blowers and
pumps. All this work is assumed to be powered by electricity. The electric motors driving
this equipment are assumed to be 95% efficient; 5% of the electric power used is assumed
lost to the ambient environment and does not contribute to the energy balance around the
CAP island.
D.9 CO2 SI CONTROL SYSTEM COST ALGORITHM DEVELOPMENT
Sorbent-based CO2 capture can be developed in a variety of configurations to conform to
sorbent properties and market constraints. For this estimate, sorbent-based CO2 capture is
assumed to utilize an internally cooled moving bed reactor for CO2 sorption. Sorbent
regeneration is assumed to require indirect steam in a separate moving bed reactor. Parallel
to MEA costs, sorbent-based CO2 capture costs will be estimated with two islands: a sorbent
island and a compressor island. Sorbent island costs will require inputs specific to the
112
-------
Appendix D
sorbent system. As there is no base plant for comparison, the capital cost for the sorbent
island will be estimated based on major components and modified to a comparable cost by
MEA island. The costs for compressor island will be modeled in the same way as described
in the MEA process.
The sorbent island consists of three major subsystems: preconditioning, sorption, and
regeneration.
D.9.1 Preconditioning
Preconditioning is assumed to be required for most sorbent applications to prevent
condensation on the sorbent. If the absorption temperature is less than 5 °C above the flue
gas supply temperature, a DCC is assumed to be required to cool the gas and remove
moisture. For absorption temperatures greater than 5 °C above the island inlet
temperature, it is assumed that preconditioning is not required.
During pretreatment, water is circulated through a direct contact cooler resulting in
condensation of water from the flue gas. The condensed water is recycled through a heat
exchanger to reduce the water temperature and is then sprayed back into the direct contact
cooler. A slip stream of condensed water is purged from the direct contact cooler prior to
the heat exchanger. The heat duty of the heat exchanger is estimated as the difference
between the enthalpy of the flue gas entering the direct contact cooler and the enthalpy of
the gas exiting the direct contact cooler plus the enthalpy difference of the moisture
condensed in the direct contact cooler. Enthalpy of the flue gas entering the direct contact
cooler is calculated based on mass flows and temperatures exiting previous unit operation.
Enthalpy of the gas flow exiting the direct contact cooler is based on the mass flow exiting
the direct contact cooler assumed to be saturated with water at 5 °C less than the absorber
temperature; a 35 °C absorber temperature would require a 30 °C DCC exhaust
temperature. Enthalpy of the moisture condensed from the flue gas is based on the mass
flow of moisture condensed and a temperature 5 °C lower than the wet bulb temperature of
the flue gas entering the DCC. For applications after a wet FGD, the gas entering the DCC is
essentially saturated and the inlet temperature is equal to the wet bulb temperature.
Though the pretreatment may also involve SO2 polishing, this heat duty is expected to be
inconsequential in comparison with the heat duty of condensing water vapor from the flue
gas.
Direct Contact Cooler
Water recirculating within the DCC is assumed to be cooled in a counter-current shell and
tube heat exchanger. Cooling water is assumed to be available at 16 °C and is discharged
from the heat exchanger at 27 °C. Condensed water enters the DCC 5 °C cooler than the
inlet flue gas wet bulb temperature (T-5) and discharges 10 °C cooler than the absorber
temperature (Ta-10). Using an overall heat transfer coefficient of 1200 J/(m2 s °C), the
surface are of the heat exchanger can be estimated as:
113
-------
Appendix D _
Q life -10-16)/(7;. - 5 - 27)1
= -!=—*-^ - ^-^ - ^ (Eq. D-19)
1200 (ra -10-ie)- (T; -5-27)
Q = heat duty in J/s
Ta = absorber temperature in °C
Tt = inlet flue gas wet bulb temperature in °C
The maximum surface area in a single unit is assumed to be 1000 m2. The number of heat
exchangers is estimated by dividing the total surface area by 1000 m2 and rounding up to
the next largest integer. The bare module cost of these heat exchangers can then be
estimated using a power law:
/- , , \O.S6
Cosf(2002$) = n • 80,000 — ^- (Eq. D-20)
V '
where
A= surface area, m2
n=number of exchangers.
DCC Recirculation Pump
The amount of water recirculating through the heat exchangers and to the DCC is calculated
based on the heat duty of the DCC and the temperature change of the water across the
DCC.
(Eq.D-21)
4,187,000
where
Ta= temperature gas at the inlet of absorber, C
Ti=temperature of gas at the inlet of the direct contact cooler, C
Q= heat duty removed by the direct contact cooler, J/Q.
At this level of estimate, a single centrifugal pump is assumed to be associated with each
heat exchanger.
/• . \ 0.40
(2002$) = 71 • 19200*^ (Eq. D-22)
Power consumption will depend greatly on the pressure drop through the nozzles and the
type of pump selected. For costing purposes, power consumption will assume an 85% pump
efficiency and a 95% drive efficiency and a 350 kPa (51 psi) pressure drop.
114
-------
Appendix D
P(kW) = n. * (Eq.D-23)
V ! 0.85.0.95
where
P=power consumption, kW.
D.9.2 Absorber
Though the absorber in a dry sorbent system may be engineered in a variety of
configurations, at this time a counter-current moving bed is assumed. Hot sorbent from the
regenerator is assumed to be added to the top of the moving bed at the regeneration
temperature and loaded sorbent removed from the bottom of the moving bed at the
sorption temperature. Conditioned flue gas is assumed to enter the bottom of the reactor
heated slightly above DCC exhaust due to the compression of a blower and to exhaust the
absorber at the regeneration temperature. The absorber is assumed to be cooled with non-
contact cooling water to cool the sorbent from regeneration temperature to absorption
temperature and maintain the sorbent at absorption temperature during carbon capture.
Absorber Feed Conveyer
Due to the anticipated conveying capacity, 3-belt conveyors are assumed to collect, raise,
and distribute the absorber feed for each conveying system. Each conveyer system is
assumed to be limited to 0.66 m3/s and a 20° incline. Estimating the cost of the absorber
feed conveyer requires estimation of the volume of sorbent to be fed and the conveying
distance and height. The estimated volume of sorbent to be fed depends on estimated CO2
loading and sorbent bulk density which demands the input by the user.
= CO2 •L/p (Eq. D-24)
where
V=estimated volume of sorbent to be fed, m3/s
CO2 = removal rate (kg CO2/s)
L = sorbent loading (kg CO2/kg sorbent)
p = bulk density (kg sorbent/m3).
Given the user supplied height [m], the length of the lifting conveyor is:
D,= 7 r (Eq. D-25)
' sin(20°)
where
AZ=supplied height, m
D|=length of the lifting conveyor, m.
115
-------
Appendix D
For estimation purposes, the collection and distribution conveyors are assumed to run the
length of the absorber. The length of the absorber is estimated with the user supplied face
velocity of the inlet flue gas and an assumed width of 6.1 m (20 ft).
Dc=Dd=
(Eq. D-26)
where
Dc = length of collection, m
Dd= length of distribution, m
Va = gas volumetric flow into absorber (m3/s)
n = number of conveyor systems
F = face velocity of gas (m/s).
Cost of the entire conveyor system, excluding drive, is estimated as:
(2002$) = n • [25000 + 2200 • (Dc + D, +Dd)]
(Eq. D-27)
Conveyor Drives
Each conveyor is assumed to be driven by an electric motor. The power for the conveyor is
estimated with the length, lift, and loading of the conveyor; conveyor speed is assumed
constant for this estimate. A constant 80% efficiency is assumed for each drive. The power
required for collection and distribution is assumed equal.
P - P -
rc — rd —
3.91 + 0.07245 • D + 0.0295 • 0.4
91.42
• V • p
\ '
(Eq. D-28)
where
PC = power for the collection conveyor, kW
Pd = power for the distribution conveyor, kW.
Power required for lifting
3.91 + 0.07245 •£>, +0.0295* 0.4
91.45
' AZ*F*p
(Eq. D-29)
where
PI = power of lifting conveyor, kW.
The total cost for drives for conveyers is estimated as:
(Eq. D-30)
116
-------
Appendix D
Absorber
At this stage of estimation, the moving bed costs will be based on the cost of heat
exchangers plus the cost of a shell. The cost of the shell will be estimated based on volume.
fV A7^°'55
(2002$) = 11300* -2-^ (Eq. D-31)
\ F )
where
Va = gas volumetric flow into absorber (m3/s)
AZ = height of the absorber (m)
F = face velocity of gas (m/s).
Absorber costs are assumed to be driven by the heat removal requirement. Heat is removed
with non-contact cooling water. The heat duty of the absorber is approximated by the heat
of adsorption and the sensible heat of the sorbent less the sensible heat of the flue gas.
Q = Qa+Qs-Qf (Eq. D-32)
where
Q = total heat, kJ
Qs = adsorption heat, kJ
Qs = sensible heat of sorbent, kJ
Qt=sensible heat of the flue gas, kJ.
The heat of sorption is estimated base on the required CO2 removal and the user supplied
specific heat of sorption.
Qa=CO2»AH (Eq. D-33)
where
Qs = adsorption heat, kJ
CO2 = CO2 removal (kg/s)
AH = specific heat of sorption (J/kg CO2).
The sensible heat of the sorbent is estimated using the user supplied sorbent heat capacity
Q,=Cp*V*p*(T,-Ta) (Eq. D-34)
where
Qs= sensible heat of sorbent, kJ
Cp=specific heat capacity, kJ/kg
\/=volume of sorbent per hour
117
-------
Appendix D
Tr= Temperature at the outlet of regenerator, °C
ra=Temperature at the outlet of the absorber, °C.
Flue gas enters the absorber at a temperature close to the absorption temperature and exits
the absorber at regeneration temperature; there is heat removal associated with the
exhausting flue gas. For estimating purposes, the sensible heat of the exhaust gas rising
from the absorber inlet temperature to the regeneration temperature will be used to
estimate this heat removal. The sensible heat should be summed for each component of the
flue gas exhausting the absorber. The heat associated with SOX, NOX, and CO is expected to
be minimal and/or unaffected by CO2. Water condensation is likely in many sorption
schemes. The condensed water is then evaporated during sorbent regeneration, which is, in
turn, condensed in the reflux to produce nearly pure CO2. The heat required and released
from the absorber and regenerator can be canceled out, leaving only heat removals across
only the reflux.
(Eq.D-35)
where
Qf= sensible heat of the flue gas, kJ
Cp = the specific heat capacity of flue gas component., kJ/kg. °C
m = the mass of the gas component, kg
Tr.= temperature of the flue gas out of the regenerator, °C
TJ= temperature of flue gas after the direct contact cooler, °C.
The heat exchanger surface areas are estimated from the heat removal requirement and
the temperature driving force. The overall heat transfer coefficient is likely a function of the
sorbent and the gas velocity. The user-specified heat transfer coefficient is assumed
constant across the absorber for this estimate; a default of 250 J/(m2s °C) will be assumed.
Potential for condensation on heat exchange tubes will be ignored for this estimate.
-27)] (& -Qf) ln[(rr -27)/fc -16)]
\ EU . L^~OO )
— 7 \ 7 \ 7 \
U (27-16) U (rr-27)-(rfl-16)
where
Qa.= adsorption heat, kJ
U = heat transfer coefficient, default at 250 J/(m2 s °C)
Qs = sensible heat of sorbent, kJ
Qf = sensible heat of the flue gas, kJ
Tr = temperature of the flue gas out of the regenerator, °C
Ta = temperature of the flue gas out of the absorber, °C.
For costing purposes, the heat exchange is assumed to be performed in U-tube heat
exchangers with a maximum surface area of 1000 m2.
_ 118
-------
Appendix D
(2002$) = H • 360
'74
(Eq. D-37)
where
n = number of exchangers
A = total surface area of exchangers, m2
(2002$) = purchase price of the material in 2002 dollars.
D.9.3 Blower/ID Fan
At this stage of costing, a blower is specified based on the anticipated pressure drop across
the absorber bed without additional consideration of bypassing or system failure. The
pressure drop across the absorber, AP, will be estimated employing user provided head loss
(Pa/m) across the sorbent and the user provided sorbent height (m) used in the absorber
pricing. The blower is assumed to be adiabatic. At this stage, a constant heat capacity ratio
of 1.4 will be used. For cases where the direct contact cooler is used to cool the gas to 5 °C
below the absorber temperature, the fluid power is estimated:
Wf =
P+AP
,0.286
(Eq. D-38)
where
Wf = work performed by the fan, kW
m = flow rate of the flue gas, kg
R = gas constant, 8.314, JKernel"1
PI = flue gas pressure before the absorber, pa/m
AP = head loss across the absorber, pa/m.
Purchase price of the blower is then estimated as:
(2002$) =3170 9Wf
0.60
(Eq. D-39)
The enthalpy (AH) increase of the gas is estimated using a constant efficiency of 80%.
W^
0.80
The power consumption (P) will include drive efficiency, assumed to be 95%
(Eq. D-40)
119
-------
Appendix D
A TJ
P = - (Eq. D-41)
0.95
D.9.4 Regenerator
Carbon dioxide will be recovered in a regenerator, analogous to a stripper for the MEA
process. The gas recovered from the regenerator is expected to consist of nominally the CO2
and water removed by the absorber without significant contamination. At this point, the
regenerator is expected to rely on temperature swing with a regeneration temperature
significantly higher than the sorption temperature. The regenerator, expected to be large
and capable of accepting solid feed, is expected to operate at near atmospheric pressure.
The loaded sorbent is expected to enter the top of the regenerator at the absorber
temperature and be heated to near the regeneration temperature. For this estimate,
exhaust gases are assumed to be withdrawn from below the top of the sorbent bed at
regenerator temperature to avoid potential condensation issues. Lean sorbents are removed
from the bottom of the regenerator at regeneration temperatures.
Regenerator Feed Conveyer
At this stage, the regenerator is assumed to be the same size and shape as the absorber.
For cost estimates, the conveyors are assumed to cost the same as the absorber feed
conveyors.
Conveyor Drives
The cost of conveyor drives is estimated in the same manner as the absorber conveyor
drives. The sizes of the conveyor drives are expected to be slightly larger than estimated
due to the mass of CO2 and water absorbed on the sorbent. The volume of the sorbent is
assumed to remain unchanged while the density increases. Therefore the product of
volumetric flow and density, the mass flow, is equal to the mass flow of absorbent feed plus
the CO2 and water absorbed.
Regenerator
The regenerator, comprised of shell and heat exchanger, is priced in the same way as the
absorber. The shell price is estimated to be the same as the absorber shell since they are
assumed to be the same dimensions.
(2002$) = 11300* -2-^ - (Eq. D-42)
I F )
where
Va = gas volumetric flow into regenerator, m3/s
AZ = height of the regenerator, m
F = face velocity of gas, m/s.
The heat exchanger required to reverse the heat of absorption is calculated as
120
-------
Appendix D
Qr=Qa=CO2*AH (Eq. D-43)
where
CO2 = CO2 removal, kg/s
AH = specific heat of sorption, J/kg CO2
Qr=heat required for sorbent regeneration, kJ
Qa=heat released for sorbent absorption, kJ.
The heat exchangers are required to (1) heat the sorbent and absorbed CO2 and water (as
gases) to regeneration temperatures and (2) to reverse the adsorption process.
(Eq.D-44)
where
Cp = specific heat capacity, kJ/kg. °C
V = gas volumetric flow into regenerator, m3/s
p = density of the gas, kg/m3
m = mass of the sorbent
Tr= temperature of gas at outlet of the regenerator, °C
Ta= temperature of gas at outlet of the absorber, °C.
At this stage of estimation, heat is assumed to be provided from saturated condensing
steam at 150 °C. The overall heat transfer coefficient is assumed to be equal to the overall
heat transfer coefficient used in the absorber.
A a , a*in[(i5o-ra)/(i5o-rr)]
A= - 7 - r-H -- 7 - r - (Eq. D-45)
tf .(150-7;) U*(T,-Ta)
where
A = surface area of the exchangers, m2
Tr = temperature of gas at outlet of the regenerator, °C
Ta = temperature of gas at outlet of the absorber, °C
Qr= heat required for sorbent regeneration, kJ
Qa = heat released for sorbent absorption, kJ
U = heat transfer coefficient, default at 250 J/(m2 s °C).
For costing purposes, the heat exchange is assumed to be performed in U-tube heat
exchangers with a maximum surface area of 1000 m2.
(/i Y'74
(2002$) = n• 360• — (Eq. D-46)
121
-------
Appendix D
Gas Cooler
A cooler/partial condenser will bring the CO2 and steam down to a low temperature ahead of
compression. For this estimate, the gas exit temperature is assumed to be 21 °C. The
amount of heat removed is equal to the sensible heat of cooling the CO2 and water vapor
plus the latent heat of condensing the water, 2.541 106 J/kg.
Q = 2.541. 106 .mwater + ^(c p\ •mi*(T,-2l) (Eq. D-47)
mi = mass of the sorbent
Tr = temperature of gas at outlet of the regenerator, °C
Q = heat removed by the exchanger, kJ/s
Cooling water is assumed to enter the gas cooler at 16 °C and exit at 27 °C. For this
estimate, an overall heat transfer coefficient of 250 J/(m2s°C) is assumed. The heat
exchange surface area is estimated assuming countercurrent flow
C/«(rr-32)
where
A = surface area of the exchangers, m2
Tr = temperature of gas at outlet of the regenerator, °C
Q = heat removed by the exchanger, kJ/s
U = heat transfer coefficient, default at 250 J/(m2 s °C).
The maximum surface area in a single unit is assumed to be 1000 m2. The number of heat
exchangers is estimated by dividing the total surface area of a unit by 1000 m2 and
rounding up to the next largest integer. The bare module cost of these heat exchangers can
then be estimated using a power law:
/- ,, \0.86
Corf(2002$) = n» 80,000 — ^- (Eq. D-49)
v '
122
-------
Appendix D
REFERENCES
DOE/NETL. 2007. Cost and Performance Baseline for Fossil Energy Plants (DOE/NETL-
2007/1281).
EPA, 1996. "Cost-effectiveness of Low-NOx Burner Technology Applied to Phase I, Group 1
Boilers," prepared by Acurex Environmental Corporation for EPA Acid Rain Division. This
report is available to the public from EPA's Office of Air and Radiation, Acid Rain Division,
Washington, DC 20460 (202-564-9085).
EPRI, 2000. "Evaluation of Innovative Fossil Fuel Power Plants with CO2 Removal,"
Document #1000316. Co-sponsored by Dep't. of Energy (Office of Fossil Energy / NETL)
Interim Report, December 2000. Requests for copies of this report should be directed to the
EPRI Distribution Center, 207 Coggins Drive, P.O. Box 23205, Pleasant Hill, CA 94523,
(800) 313-3774.
EPRI, 2007. "Chilled Ammonia Process Update," May 24, 2007. Lyon, France.
http://www.CO2captureandstorage.info/docs/capture/10th%20cap%20network%20web%20
files/K%20-%20Rhudy%20-%20Chilled%20Ammonia%20as%20solvent.pdf
Frey, C.H. and E.S. Rubin, 1994, "Development of the Integrated Environmental Control
Model: Performance Models of Selective Reduction (SCR) NOX Control Systems; Quarterly
Progress Report to Pittsburgh Energy Technology Center, U.S. Department of Energy, from
Center for Energy and Environmental Studies, Carnegie Mellon University," Pittsburgh, PA.
Gundappa, M., L. Gideon, and E. Soderberg, 1995, "Integrated Air Pollution Control System
(IAPCS), version 5.0, Volume 2: Technical Documentation, Final," EPA, Air and Energy
Engineering Research Laboratory, Research Triangle Park, NC, EPA-600/R-95-169b (NTIS
PB96-157391).
Maxwell, J. D. and L. R. Humphries, 1985, "Economics of Nitrogen Oxides, Sulfur Oxides,
and Ash Control Systems for Coal-Fired Utility Power Plants," Environmental Protection
Agency, Air and Energy Engineering Research Laboratory, Research Triangle Park, NC,
EPA-600/7-85-006 (NTIS PB85-243103).
Merrow, E. W., L. McDonnell, and R. Y. Arguden, 1988. Understanding the Outcomes of
Megaprojects: a Quantitative Analysis of Very Large Civilian Projects. Santa Monica, CA:
RAND Corp.
Presto, A. A., and E. J. Granite, 2006. Survey of Catalysts for Oxidation of Mercury in Flue
Gas. Environmental Science & Technology 40 (18):5601-5609.
Rao, Anand. 2002. A Technical, Economic and Environmental Assessment of Amine-Based
CO2 Capture Technology for Power Plant Greenhouse Gas Control. DOE Contract No.: DE-
FC26-OONT40935
123
-------
Appendix D
Robie, C.P. and P.A. Ireland, 1991, "Technical Feasibility and Cost of Selective Catalytic
Reduction (SCR) NOX Control," GS-7266, Prepared by United and Engineers and
Constructors, Inc. for the Electric Power Research Institute, Palo Alto, CA.
Rubin, E. S., M. Antes, S. Yeh, and M. Berkenpas, 2006. Estimating the Future Trends in the
Cost of CO2 Capture Technologies. Report No. 2006/6. Cheltenham, UK: IEA Greenhouse
Gas R&D Programme (IEA GHG).
Sherrick, B.; Hammond, M.; Spitznogle, G.; Murashin, D.; Black, S.; Cage, M., 2008. CCS
with Alstom's Chilled Ammonia Process at AEP's Mountaineer Plant. Presented in the Power
Plant Mega Symposium. Baltimore, MD. 2008
Staudt, J.E.; Jozewicz, W.; Srivastava, R., 2003. Modeling Mercury Control with Powdered
Activated Carbon, AWMA Paper 03-A-17-AWMA. Presented at the Joint EPRI DOE EPA
Combined Utility Air Pollution Control Symposium, The Mega Symposium, May 19-22, 2003,
Washington, D.C.
Taylor, M., E. S. Rubin, and D.A Hounshell, 2003. The effect of government actions on
technological innovation for SO2 control. Environmental Science & Technology 37 (20):4527
- 4534.
Ulrich, Gael D., 1984. A guide to Chemical Engineering Process Design and Economics, John
Wiley and Sons, Inc., New York, NY.
Wright, T. P. 1936. Factors affecting the cost of airplanes, Journal of Aeronautical Sciences,
3 (2):122-128.
124
-------
Appendix E
E.I GETTING STARTED
After downloading the workbook to the hard drive, the first thing to do is to create a copy of
the workbook and save it under a different name. Once the workbook has been saved to the
hard drive, it can be opened using Microsoft Excel 5.0 or a newer version of Excel.
The workbook will originally open to the "1.0 General Input Sheet". This is the worksheet
where all of the necessary inputs are entered. However, a main menu is created for the user
where all of the sheets are linked with buttons. The screen the user will encounter is:
Main Menu
Air Pollution Control Technology,
1.0 Economic INPUT
2,0General PlantlNPUT
3.0 Air Pollution Control Technologies!
10.0 Leuelization CALCULATIONS
4.0 Power Generation CALCULATIONS
Go To Main Menu
INPUT Selections Print Functions
Kyi Irptjf output * FrsntButftrK*
Go to TOP OUTPUT Se|ect!ons
3.1 NOX Control INPUT j
3.2PM Control INPUT!
3.3 SO2 Control INPUT |
3.4 Hg Control INPUT;
3.5 CO2 Control INPUT!
5.0 NOX Control CALCULATIONS
i.O PM Control CALCULATIONS
7.0 SO2 Control CALCULATIONS
8.0 Hg Control CALCULATIONS
9,8 CO2 Control CALCULATIONS
125
-------
Appendix E
E.2 INPUTS
As the user proceeds down following the menu, (s)he will encounter the following
worksheets:
• Power Generation Technique Choices
• Air Pollution Control (APC) Technology Choices
• General Plant Technical Inputs
• Economic Inputs
• Limestone Forced Oxidation (LSFO) Inputs
• Lime Spray Dryer (LSD) Inputs
• Particulate Control Inputs
• NOX Control Inputs
• Hg Control Technology Inputs
• CO2 Control Amine Technology Input
E.2.1 Economic Inputs
This is the area of the worksheet where the economic factors are input. These factors are
used in developing the capital and O&M costs for the control technologies.
Item/Description Units
Economic Factor
Cost Basis -Year Dollars year-
Service Life (Levelization Period) years
Sales Tax %
Escalation/Inflation Adjustment (GDP or Chern Index) ^
Construction Labor Rate S/h
Prime Contractor's Markup %
Inflation Rate %
Escalation Rate %
Capital Carrying Charges
First-year Carrying Charge (Current S's) %
Levelized Carrying Charge (Constant S's) %
Non-Carrying Expense (OS.M)
Levelizing Factor (L30) (Constant S's}
Variable Cost Factors
Operating Labor Rate (include benefit) S/h
Power Cost Mills/kWh
Steam Cost s/1000 Ibs
Demineralized Water s/!b
Makeup Water s/1000 Ib
Range
Default
Calculator
2006
30
6%
GDP
S35
3%
2%
3%
9%
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
S25.0
60
3.5
SO.0030
SO.05
User needs to click the icon of calculator, a quick worksheet pops up to facilitate calculating
carrying charges and levelization factors.
126
-------
Appendix E
INPUT
Factors during Operation (for Carrying Charge and levelization)
Return of Debt
Ratio of Debt
Return of Equity
Ratio of Equity
Property Taxes and Insurance
Income Tax
Investment Tax Credit
Book life of the Piant
Average (price) Inflation rate (Long-Term)
Depreciation method
(1= 30 Yr Straight Line; 2=30 Yr Straight Une;3=2Q Yr ARCS Schedule)
Consumables (O&.M) Escalation
3
3
50%
10%
30
3%
50%
10%
50%
50%
10%
30
OUTPUT
Carrying Charges
First-year Carrying Charge (current S's)7"
Levefszed Carrying Charge (current S's)
First-year Carrying Charge (constant S's)
Leveiized Carrying Charge (constant S's)*
Non-Carrying Expense
Levelizing Factor (L30) (Current S's) ( for O&M)
Levelizing Factor (L3Q) (Constant S's) ( for O&M)*
15.9% 15.9% IS.9% 15.9% 15.9%
11.4% 11.4% 11.4% 11.4% 11.4%
11.7% 11.7% 11.7% 11.7% 11.7%
8.3% 8.3% 8.3% 8.3% 8.3%
2.08
1.49
2.08
1.49
2.08
1.49
2.08
1.49
2.08
1.49
15.9%
11.4%
11.7%
8.3%
2.08
1.49
15.9% 15.9% 15.9%
11.4% 11.4% 11.4%
11.7% 11.7% 11.7%
8.3% 8.3% 8.3%
2.08
1.49
2.08
1.49
2.08
1.49
£.2.2 Power Generation Technique Choices
This is the area of the worksheet where the user can choose what power generation
technique will be evaluated. The following screen shows how this area looks and what
options are available.
t->Zitfj ^rKt Tec lire \-t;,;
ItemfDescription
Plant Information
Cost Basis -Year (For Power Generation Estimation only)
Location - State
Power Generation Technologies
General Plant Factors
Gross Plant output
Net PI ant Output
Plant Heat Rate
Plant Capacity Factor
Coal Type
Price of Coal
Other Operating Information
Percent Excess Air in Boiler
Uncontrolled NOx from Boiler
Air Heater Inleakage
Air Heater Outlet Gas Temperature
Inlet Air Temperature
Ambient Absolute Pressure
Pressure After Air Heater
Moisture in Air
Ash Split:
Fly Ash
Seismic Zone
Conversion of SO2 to SO3
Units
Range
500-800
500-750
MW
MW
BtufkWh
K 40-90%
Goto Coal Data
*F
T
in. Hg
in. HjO
Ibrflb dry air
integer
Default
2005
PA
1
580
500
10,500
65%
5
2.05
120%
algorithm
12%
300
Case 2
Case 3
1-5
80%
1
1.0%
D
D
3
|l=Suberitical; 2=Sup<
580
D
D
D
2
D
D
2
srCritical;
580
D
D
D
2
D
D
2
3=Ultra-SuperCritic
580
D
D
D
2
See Coal Tupes
D
D
, " D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
£.2.3 APC Technology Choices
This is the area of the worksheet where the user can choose what control technologies are
needed. The following screen shows how this area looks and what options are available.
127
-------
Appendix E
APC Technology Choices
NOx Control
Combustion Control (0=no additional control, 1 = combustion control)
Post Combustion Control (0=none. 1 = SCR, 2 = SNCR, 3 « NGR)
Particular Control (0= none. 1 = Fabric Filter, 2 = ESPc. 3=ESPh]
SO2Control
(0=none, 1 = LSFQ, 2 = LSD]
Additional Mercury Control with Sorbent Injection?
(0=no, l=yes)
If Sorbent Injection, add downstream PJFF?
(0=no, 1=PJFF)
CO2 Control (0-none, 1-CAP, 2-MEA, 3-Sorbent)
integer
integer
integer
integer
integer
integer
Integer
0,1 1
0, 1, 2, 3 1
0, 1, 2, 3 2
0, 1, 2 1
Oorl 1
Oor1 1
0, 1, 2, 3 1
1111
1111
1111
1111
1111
1111
1311
E-2.4 NOX Control Inputs
Data necessary for sizing and costing the NOX control processes are input in the worksheet
below. This information is used with the combustion calculations to size one of the four
processes.
3.1 NOX Control Technoloy
I temf Descri pti on
Combustion Technology Selected?
Uncontrolled NQx level
Boiler Type
Burner Type
Retrofit Difficulty Factor
General Facilities
Engineering
Contingency
Duration of Project
SCR Technology Selected?
inlet NOx level
Nr-yNQxStoichiometric Ratio
NOK Reduction Efficiency
Space Velocity (Calculated if zero)
Time to First Catalyst Replenishment
Ammonia Cost
Catalyst Cost
Solid Waste Disposal Cost
Retrofit Difficulty Factor
General Facilities (% of Installed Cost)
Engineering Fees (% of Installed Cost)
Contingency (% of Installed Cost)
Duration of Project
Maintenance (% of Installed Cost)
Co-benefit Application
Mercury Oxidation Rate - bituminous coal
Mercury Oxidation Rate - subbiluminous coal
SNCR Technology Selected?
Inlet NOx level
Reagent
Range
IWMMBtu
W:Wall
2=LNB and OF A
number
percent
percent
percent
years
IWMMBtu
NH^QK
Fraction
years
$Kon
0.740
0.60-0.90
2-5
IWMMBtu
integer Urea 2:Arnrnoni
Default
calculated
T
1
1.3
5.0%
10.0%
15.0%
1
calculated
0.9
0.90
0
3
400
5000
11.48
1.5
5%
10%
15%
2
0.66%
90.0%
0.0%
calculated
1
Casel
Selected
D
D
D
D
D
D
D
D
Selected
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
Ci
Not Selected
D
D
Case 2
Selected
D
D
D
D
D
D
D
D
Selected
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
Not Selected
D
D
Case 3
Selected
D
D
D
D
D
D
D
D
Selected
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
Not Selected
D
D
128
-------
Appendix E
£.2.5 Particulate Control Inputs
Data necessary for sizing and costing the participate control equipment are input into the
worksheet below. This information is used with the combustion calculations to size either an
ESP or FF.
3.2 PM Control Technology
ItemlDescription
Outlet Particulate Emission Limit
Fabric Filler Selected
Pressure Drop
Type (1 = Reverse Gas, 2 = Pulse Jet)
Gas-to-Cloth Ratio
Bag Material (RGFF fiberglass only)
(1 = Fiberglass, 2 = Nomex, 3 = Ryton)
Bag Diameter
Bag Length
Bag Reach
Compartments out of Service
Bag Life
Retrofit Factor
Contingency (% of installed cost)
General Facilities (% of installed costj
Engineering Fees (% of installed cost)
Project Duration
Maintenance (% of installed cost)
ESP Selected
Strength of the electric field in the ESP = E
Plate Spacing
Plate Height
Pressure Drop
Retrofit Factor
Contingency (% of Installed Cost]
General Facilities (% of Installed Costj
Engineering Fees (% of Installed costj
F'roiect Duration
Units
Ibs/MMBtu
in. H2D
Integer
ACFMft!
Integer
inches
feet
Range
fl.
in. HjD
Default
0.03
6
2
3.5
2
6
20
3
10%
5
1.3
15%
10.0
12
36
3
1.3
15%
5%
10%
2
Selected
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
Mot Selected
D
D
D
D
D
D
D
D
D
Case 2
D
Selected
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
Not Selected
D
D
D
D
D
D
D
D
D
CaseS
D
Selected
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
Not Selected
D
[i
D
D
D
D
D
D
D
£.2.6 SO2 Control Inputs
Data necessary for sizing and costing an SO2 control system are input into the worksheet
below. This information is used with the combustion calculations to design the system.
129
-------
Appendix E
3.3 SO2 Control Technology
I temlDescri pli on
Lime Spray Dryer Selected?
SQz Removal Required
is SDA being retrofit upstream of existing ESP? [0 = no, 1 = yes)
Adiabatic Saturation Temperature
Flue Gas Approach to Saturation
Recycle Slurry Solids Concentration
Number of Absorbers
(Max. Capacity = 300 MW per spray dryer)
Absorber Material
(1 = Alloy,2 = RLCS)
Spray Cooler Pressure Drop
Reagent Bulk Storage
Reagent Cost (delivered)
Dry Waste Disposal Cost
Retrofit Factor
Contingency (% of installed Cost)
General Facilities (% of installed Cost)
Engineering Fees (% of Installed Cost)
Project Duration
Maintenance Factor (% of TPC)
LSFO Selected?
Year Equipment Placed in Service
SOS Removal Required
UG Ratio
Design Scrubber with Dibasic Acid Addition?
(1 = yes, 2 = no)
Adiabatic Saturation Temperature
Reagent Feed Ratio
integer
*F
T
Wt. %
integer
integer
in. H£l
days
IKon
Range
70-95%
0,1
100-170
10-50
10-50
1-7
1or2
Default
90%
0
127
20
35%
1
1
30
$65
$30
1.3
15%
5%
10%
Casel
Not Selected
D
D
D
D
D
D
Case 2 Case 3
Not Selected Not Selected
D D
D D
D D
D D
D D
D D
Selected
year
%
gal I 1000 acf
integer
T
Factor
90-98%
95-160
1or2
100-170
1.0-2.0
2004
95%
125
1
127
1.05
D
D
D
D
D
D
Selected
D
D
D
D
D
D
Selected
D
D
D
D
D
D
E.2.7 Mercury Control Inputs
This is where the data necessary for sizing and costing the mercury control processes are
input. This information is used with the combustion calculations to size powdered activated
carbon (PAC) and pulse-jet fabric filter (PJFF) processes.
130
-------
Appendix E
3.4 Mercury Control Technology
ltem4Description Units
Sorbent Injection Technology Selected?
HgCEMS lo-no, 1-yes) integer
Hg Reduction Required from Coal percent
Sorbent Type, 1-EPAC, 2-PAC, 3=olher
Maximum Temperature before Spray Cooling T
Sorbent Recycle Used? Yes/No
Spray Cooiing Desired? Yes/No
EPAC Cost (Delivered Cost of Treated PAC) »on
PAC Cost (Delivered) Won
Other Sorbent Cost (Del i vered] Van
Before Sorbent Injection, Fly Ash Sold (1) or Disposed of (2] 1 or 2
Does Sorbent Adversely Impact Fly Ash Sales? (0=no, 1=yes) integer
Revenue from Fly Ash Sales &Son
Dry Waste Disposal Cost Won
Retrofit Factor
Process Contingency, % of Process Capital %
General Facilities [% of Installed Cost) %
Engineering Fees (% of Installed Cost) %
Project Contingency %
Duration of Project years
Maintenance Factors (% of Installed Cost] %
PJFF Downstream of PAC Selected?
PJFF to COHPAC (ie, TOXECON), 0=no, 1=yes 0 or 1
Cost of Bags, I nstal I ed (Ubag) Jfcag
Estimated SElagsMW integer
Average Bag Life years
Pressure Drop in. H2O
Outlet Emissions Ib/MMBtu
Retrofit Difficulty Factor
Process Contingency, % of Process Capital %
Range
Oor1
M25
Oor1
Oto35
1to25
Default
1
0.8
2
325
No
No
1500
1000
1000
1
0
6
6
1.3
5%
5%
10%
15%
1
5%
1
80
20
5
a.o
0.012
1.3
5%
Casel
Selected
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
Selected
D
D
D
D
D
D
D
D
Case 2
Selected
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
Selected
D
D
D
D
D
D
D
D
Case 3
Selected
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
Selected
D
D
D
D
D
D
D
D
£-2.8 CO2 Control Inputs
This is where the data necessary for sizing and costing the CO2 control processes are input.
This information is used with the combustion calculations to size amine-based CO2 control
processes.
131
-------
Appendix E
3.5 Carbon Dioxide Control Technology
i ' ' f i1' ^ ^ i 11 r t ' i* } ^ *t IK , o ' f lin k * •(
1 temfDescri pti an
Chilled Ammonia Process Selected?
Efficiency of COj removal
Flue gas temperature out of direct contact cooler
Rue gas temperature entering Hie absorber
Flue gas temp exiting the absorber
CO; temperature exiting the stripper reflux
Cycle of Concentration for Cooling Water
Reagent of Ammonia
Concentration of Ammonia
Price of Ammonia (28%)
Ammonia slip to flue gas
Regenerator Pressure
Compressor
CO; Product Pressure
CO2 compressor stage
Number of operator
Retrofit Factor
Maintenance Factor (% of TPC)
Contingency (% of installed Cost)
General Facilities (% of Installed Cost)
Engineering Fees (% of Installed Cost)
Time for Retrofit
MEA Process Selected?
Efficiency of CO; removal
KS-lorOtherMEA
Cycle of Concentration for Cooling Water
Reagen Price
Price of ME A
Price of NaOH ( 20% solution)
i' 1 llf ' *v , ' l' 4
Units
%
T
T
T
T
%
Won
ppm
Psi
psi
%
%
%
%
years
%
1=KS-1, 2= MEA
Bon
Won
iin k * ^ ' t . i
Range
MO
35
30-32
32-50
65-72
2-10
28-30%
100-200
2-10
300-600
500-2500
3
6-8
90%
2-10
l\t I L *l V ^
Default
90%
35
32
35
70
5
28%
150
5
400
2200
3
8
to
3%
15%
10%
7%
2
90%
1
5
2142
413
i 1 I Jl 1 >'t r •}
Casel
Selected
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
Not Selected
D
1
D
D
D
1, t 1 ilr 1 i ^
Case 2
Not Selected
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
Not Selected
D
2
D
D
D
t ' »«• i *
Case3
Selected
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
Not Selected
D
1
D
D
D
132
-------
Appendix F
The programs included in this appendix are for calculations of economic parameters,
including the TEC, TPI, current $ carrying charge, constant $ carrying charge, first year
current $ carrying charges, first year constant $ carrying charges, and levelization factors.
Comments were added in each program for clarity.
'This module contains functions to calculate the Economic parameters
'This function calculates TCE value
'Arguments: inflation rate, float; escalation, float; period, float
Function TCE (inflation, escalation, period)
EA = (1 + inflation) x (1 + escalation) - 1
TCE = PV(EA, period, -1) x (1 + EA) / period
End Function
'This function calculates TPI value, the parameters
'Inside the parenthesis already show what you should input
Function TPI(inflation, escalation, interest, period)
EA = (1 + inflation) x (1 + escalation) - 1
Z = (1 + interest)/ (1 + EA)
TPI = (Z A period - 1)
TPI = TPI / period
TPI = TPI / (Z - 1)
End Function
'This function calculates Current Carrying Charge
'Arguments: rd (cost of debt), float; wd(ratio of debt), float; re(cost of equity), float;
'we (ratio of equity), float;PTI (property tax and insurance), float; T ( tax), float;
'ITC (investment tax credit), float; BL(book life), float; depreMethod (depreciation method"),
integer
133
-------
Appendix F
Function CurrentCC(ByVal rd, ByVal wd, ByVal re, ByVal we, ByVal PTI, ByVal T, ByVal ITC,
ByVal BL, ByVal depreMethod)
rd = rd
wd = wd
re = re
we = we
PTI = PTI
T = T
ITC = ITC
Dim Arcs(21) As Double
'Define depreciation rate for each year
Dim Depr(31) As Double
'Loop variable
Dim i As Integer
'Calculate the weighted average cost of capital (WACC)
Dim EFI As Double
'Straight line depreciation
Dim sltxdp As Double
sltxdp = 1 / BL
'Cumulative CC x V in the EPRI formula
'Detail of the original formula can be seen in TAG-Technical Assessment Guide, Vol. 3,
'Fundamentals and Methods, Supply-1986, EPRI P-4463-SR
Dim SP As Double
SP = 0
V current value factor in the EPRI formula
Dim V As Double
'Deferred income tax
Dim DT As Double
'tax paid
Dim TP As Double
'tax depreciation rate
Dim TD As Double
'Carrying charge
Dim cc As Double
'An annuity factor in EPRI formula
Dim A As Double
'temp variable of rd
Dim rdTemp As Double
'temp variable of re
Dim reTemp As Double
'Book depreciation rate net of ITC
Dim BD As Double
BD = (1 - ITC)/ BL
134
-------
Appendix F
'Remaining book value
Dim SB As Double
SB = 1 - ITC
'Assign value to ACRS schedule
Arcs(O) = 0
Arcs(l) = 7.5
Arcs(2) = 6.9
Arcs(3) = 6.4
Arcs(4) = 5.9
Arcs(5) =5.5
Arcs(6) = 5.1
Arcs(7) = 4.7
Arcs(8) = 4.5
Arcs(9) = 4.5
Arcs(lO) = 4.5
Arcs(ll) = 4.5
Arcs(12) = 4.5
Arcs(13) = 4.5
Arcs(14) = 4.5
Arcs(15) = 4.5
Arcs(16) = 4.4
Arcs(17) = 4.4
Arcs(18) = 4.4
Arcs(lO) = 4.4
Arcs(20) = 4.4
'Assign value 0 to initialize the array
A = 0#
'Compute return rate used to determine carrying charges
EFI = (wd x rd + we x re)
For i = LBound(Depr) To UBound(Depr)
Depr(i) = 0
Next i
'Calculate tax depreciation with 3 cases, we assume that the book life is greater than the
depreciation life
'Calculate straight line tax depreciation
If (depreMethod = 1) Then
'For straight line tax depreciation over book life
For i = 1 To BL
Depr(i) = 1 / BL
Next i
'For straight line tax depreciation over ACRS
Elself (depreMethod = 2) Then
For i = 1 To 20
Depr(i) = 1/20
135
-------
Appendix F
Next i
'ForACRS depreciation schedule
Else
For i = 1 To 20
Depr(i) = Arcs(i) / 100
Next i
End If
'Start loop over the book life
For i = 1 To BL
'Present value factor
V = (1 + EFI) ^ (-i)
If (EFI <> 0) Then
A= (1-V)/(EFI)
Else
A = i
End If
'tax depreciation rate
TD = Depr(i)
'return on equity
reTemp = SB x re x we
'return on debt
rdTemp = SB x rd x wd
'deferred income tax
DT = (TD - sltxdp) x T
'tax paid.
TP = T / (1 - T) x (BD - TD + DT + reTemp)
' year by year carrying charges
cc = BD + DT + rdTemp + reTemp + TP + PTI
'cum. present value of carrying charge
SP = SP + cc x V
'Depreciation book value, net def. tax
SB = SB - BD - DT
Next i
'earn/Charges = SP / A
CurrentCC = SP / A
End Function
'This function calculates first year Current Carrying Charge
'Arguments: rd (cost of debt), float; wd(ratio of debt), float; re(cost of equity), float;
'we (ratio of equity), float;PTI (property tax and insurance), float; T ( tax), float;
'ITC (investment tax credit), float; BL(book life), float; depreMethod (depreciation method),
integer
136
-------
Appendix F
Function firstYearCurrentCC(ByVal rd, ByVal wd, ByVal re, ByVal we, ByVal PTI, ByVal T,
ByVal ITC, ByVal BL, ByVal depreMethod)
rd = rd
wd = wd
re = re
we = we
PTI = PTI
T = T
ITC = ITC
Dim Arcs(21) As Double
'Define depreciation rate for each year
Dim Depr(31) As Double
'loop variable
Dim i As Integer
'Calculate the weighted average cost of capital (WACC)
Dim EFI As Double
'straight line depreciation
Dim sltxdp As Double
sltxdp = 1 / BL
'cumulative CC x V in the EPRI formula
'Detail of the original formula can be seen in TAG-Technical Assessment Guide, Vol. 3,
'Fundamentals and Methods, Supply-1986, EPRI P-4463-SR
Dim SP As Double
SP = 0
'V current value factor in the EPRI formula
Dim V As Double
'deferred income tax
Dim DT As Double
'tax pa id.
Dim TP As Double
'tax depreciation rate
Dim TD As Double
'Carrying charge
Dim cc As Double
'A annuity factor in EPRI formula
Dim A As Double
'temp variable of rd
Dim rdTemp As Double
'temp variable of re
Dim reTemp As Double
'book depreciation rate net of ITC
Dim BD As Double
137
-------
Appendix F
BD = (1 - ITC)/ BL
'remaining book value
Dim SB As Double
SB = 1 - ITC
'assign value to ACRS schedule
Arcs(O) = 0
Arcs(l) = 7.5
Arcs(2) = 6.9
Arcs(3) = 6.4
Arcs(4) = 5.9
Arcs(5) =5.5
Arcs(6) = 5.1
Arcs(7) = 4.7
Arcs(8) = 4.5
Arcs(9) = 4.5
Arcs(lO) = 4.5
Arcs(ll) = 4.5
Arcs(12) = 4.5
Arcs(13) = 4.5
Arcs(14) = 4.5
Arcs(15) = 4.5
Arcs(16) = 4.4
Arcs(17) = 4.4
Arcs(18) = 4.4
Arcs(lO) = 4.4
Arcs(20) = 4.4
'assign value 0 to initialize the array
A = 0#
'compute return rate used to determine carrying charges
EFI = (wd x rd + we x re)
For i = LBound(Depr) To UBound(Depr)
Depr(i) = 0
Next i
'calculate tax depreciation with 3 cases, we assume that the book life is greater than the
depreciation life
'calculate straight line tax depreciation
If (depreMethod = 1) Then
'for straight line tax depreciation over booklife
For i = 1 To BL
Depr(i) = 1 / BL
Next i
'for straight line tax depreciation over ACRS
Elself (depreMethod = 2) Then
For i = 1 To 20
138
-------
Appendix F
Depr(i) = 1/20
Next i
'for ACRS depreciation schedule
Else
For i = 1 To 20
Depr(i) = Arcs(i) / 100
Next i
End If
For i = 1 To 1
'Prsent value factor
V = (1 + EFI) ^ (-i)
If (EFI <> 0) Then
A= (1-V)/(EFI)
Else
A = i
End If
'tax depreciation rate
TD = Depr(i)
'Return on equity
reTemp = SB x re x we
'Return on debt
rdTemp = SB x rd x wd
'Deferred income tax
DT = (TD - sltxdp) x T
'tax paid.
TP = T / (1 - T) x (BD - TD + DT + reTemp)
' Year by year carrying charges
cc = BD + DT + rdTemp + reTemp + TP + PTI
'cum. present value of carrying charge
SP = SP + cc x V
'Depreciation book value, net def. tax
SB = SB - BD - DT
Next i
'carryCharges = SP / A
firstYearCurrentCC = SP/ A
End Function
139
-------
Appendix F
'This function calculates constant Carrying Charge
'Arguments: rd (cost of debt), float; wd(ratio of debt), float; re(cost of equity), float;
'we (ratio of equity), float;PTI (property tax and insurance), float; T ( tax), float;
'ITC (investment tax credit), float; BL(book life), float; inflation, float; depreMethod
(depreciation method), integer
Function ConstantCC(ByVal rd, ByVal wd, ByVal re, ByVal we, ByVal PTI, ByVal T, ByVal
ITC, ByVal BL, ByVal inflation, ByVal depreMethod)
rd = rd
wd = wd
inflation = inflation
re = re
we = we
PTI = PTI
T = T
ITC = ITC
Dim Arcs(21) As Double
'Define depreciation rate for each year
Dim Depr(31) As Double
'loop variable
Dim i As Integer
'Calculate the weighted average cost of capital (WACC)
Dim EFI As Double
'Straight line depreciation
Dim sltxdp As Double
sltxdp = 1 / BL
'Cumulative CC x V in the EPRI formula
'Detail of the original formula can be seen in TAG-Technical Assessment Guide, Vol. 3,
'Fundamentals and Methods, Supply-1986, EPRI P-4463-SR
Dim SP As Double
SP = 0
'V current value factor in the EPRI formula
Dim V As Double
'Deferred income tax
Dim DT As Double
'tax paid.
Dim TP As Double
'tax depreciation rate
Dim TD As Double
'Carrying charge
Dim cc As Double
'An annuity factor in EPRI formula
140
-------
Appendix F
Dim A As Double
'temp variable of rd
Dim rdTemp As Double
'temp variable of re
Dim reTemp As Double
'book depreciation rate net oflTC
Dim BD As Double
BD = (1 - ITC)/ BL
'Remaining book value
Dim SB As Double
SB = 1 - ITC
'Assign value to ACRS schedule
Arcs(O) = 0
Arcs(l) = 7.5
Arcs(2) = 6.9
Arcs(3) = 6.4
Arcs(4) = 5.9
Arcs(5) =5.5
Arcs(6) =5.1
Arcs(7) = 4.7
Arcs(8) = 4.5
Arcs(9) = 4.5
Arcs(lO) = 4.5
Arcs(ll) = 4.5
Arcs(12) = 4.5
Arcs(13) = 4.5
Arcs(14) = 4.5
Arcs(15) = 4.5
Arcs(16) = 4.4
Arcs(17) = 4.4
Arcs(18) = 4.4
Arcs(lO) = 4.4
Arcs(20) = 4.4
'Assign value 0 to annuity
A = 0#
'Compute return rate used to determine carrying charges
'Calculate rd and re without inflation
rd = (1 + rd) / (1 + inflation) - 1
re = (1 + re) / (1 + inflation) - 1
EFI = wd x rd + we x re
'Initialize depreciation value in the depreciation array.
For i = LBound(Depr) To UBound(Depr)
Depr(i) = 0
141
-------
Appendix F
Next i
'Calculate tax depreciation with 3 cases, we assume that the book life is greater than the
depreciation life
'Calculate straight line tax depreciation
If (depreMethod = 1) Then
'For straight line tax depreciation over book life
For i = 1 To BL
Depr(i) = 1 / BL
Next i
'For straight line tax depreciation overACRS
Elself (depreMethod = 2) Then
For i = 1 To 20
Depr(i) = 1/20
Next i
'forACRS depreciation schedu\e
Else
For i = 1 To 20
Depr(i) = Arcs(i) / 100
Next i
End If
'start the loop over the book life
For i = 1 To BL
'Present value factor
V = (1 + EFI) ^ (-i)
If (EFI <> 0) Then
A= (1-V)/(EFI)
Else
A = i
End If
'Tax depreciation rate
TD = Depr(i)
'return on equity
reTemp = SB x re x we
'return on debt
rdTemp = SB x rd x wd
'Deferred income tax
DT = (TD - sltxdp) x T
'tax paid.
TP = T / (1 - T) x (BD - TD + DT + reTemp)
'Year by year carrying charges
cc = BD + DT + rdTemp + reTemp + TP + PTI
'Cum. presents value of carrying charge
SP = SP + cc x V
'Depreciation book value, net def. tax
142
-------
Appendix F
SB = SB - BD - DT
Next i
'carryCharges = SP / A
ConstantCC = SP / A
End Function
'This function calculates first year Current Carrying Charge
'Arguments: rd (cost of debt), float; wd(ratio of debt), float; re(cost of equity), float;
'we (ratio of equity), float;PTI (property tax and insurance), float; T ( tax), float;
'ITC (investment tax credit), float; BL(book life), float; inflation, float; depreMethod
(depreciation method), integer
Function firstYearConstantCC(ByVal rd, ByVal wd, ByVal re, ByVal we, ByVal PTI, ByVal T,
ByVal ITC, ByVal BL, ByVal inflation, ByVal depreMethod)
rd = rd
wd = wd
inflation = inflation
re = re
we = we
PTI = PTI
T = T
ITC = ITC
Dim Arcs(21) As Double
'Define depreciation rate for each year
Dim Depr(31) As Double
'Loop variable
Dim i As Integer
'Calculate the weighted average cost of capital (WACC)
Dim EFI As Double
'Straight line depreciation
Dim sltxdp As Double
sltxdp = 1 / BL
'Cumulative CC x V in the EPRI formula
'Detail of the original formula can be seen in TAG-Technical Assessment Guide, Vol. 3,
'Fundamentals and Methods, Supply-1986, EPRI P-4463-SR
Dim SP As Double
SP = 0
'V current value factor in the EPRI formula
Dim V As Double
'Deferred income tax
Dim DT As Double
'tax paid.
Dim TP As Double
'tax depreciation rate
143
-------
Appendix F
Dim TD As Double
'Carrying charge
Dim cc As Double
'An annuity factor in EPRI formula
Dim A As Double
'temp variable of rd
Dim rdTemp As Double
'temp variable of re
Dim reTemp As Double
'book depreciation rate net oflTC
Dim BD As Double
BD = (1 - ITC)/ BL
'Remaining book value
Dim SB As Double
SB = 1 - ITC
'Assign value to ACRS schedule
Arcs(O) = 0
Arcs(l) = 7.5
Arcs(2) = 6.9
Arcs(3) = 6.4
Arcs(4) = 5.9
Arcs(5) =5.5
Arcs(6) = 5.1
Arcs(7) = 4.7
Arcs(8) = 4.5
Arcs(9) = 4.5
Arcs(lO) = 4.5
Arcs(ll) = 4.5
Arcs(12) = 4.5
Arcs(13) = 4.5
Arcs(14) = 4.5
Arcs(15) = 4.5
Arcs(16) = 4.4
Arcs(17) = 4.4
Arcs(18) = 4.4
Arcs(lO) = 4.4
Arcs(20) = 4.4
'Assign value 0 to annuity
A = 0#
'Compute return rate used to determine carrying charges
'Calculate rd and re without inflation
rd = (1 + rd) / (1 + inflation) - 1
re = (1 + re) / (1 + inflation) - 1
144
-------
Appendix F
EFI = wd x rd + we x re
'Initialize depreciation value in the depreciation array.
For i = LBound(Depr) To UBound(Depr)
Depr(i) = 0
Next i
'Calculate tax depreciation with 3 cases, we assume that the book life is greater than the
depreciation life
'Calculate straight line tax depreciation
If (depreMethod = 1) Then
'For straight line tax depreciation over book life
For i = 1 To BL
Depr(i) = 1 / BL
Next i
'For straight line tax depreciation overACRS
Elself (depreMethod = 2) Then
For i = 1 To 20
Depr(i) = 1/20
Next i
'ForACRS depreciation schedule
Else
For i = 1 To 20
Depr(i) = Arcs(i) / 100
Next i
End If
'Start the loop over the book life
For i = 1 To 1
'present value factor
V = (1 + EFI) ^ (-i)
If (EFI <> 0) Then
A= (1-V)/(EFI)
Else
A = i
End If
'Tax depreciation rate
TD = Depr(i)
'Return on equity
reTemp = SB x re x we
'Return on debt
rdTemp = SB x rd x wd
'Deferred income tax
DT = (TD - sltxdp) x T
'tax paid.
TP = T / (1 - T) x (BD - TD + DT + reTemp)
' Year by year carrying charges
145
-------
Appendix F
cc = BD + DT + rdTemp + reTemp + TP + PTI
'Cum. present value of carrying charge
SP = SP + cc x V
'Depreciation book value, net def. tax
SB = SB - BD - DT
Next i
'carry Charges = SP / A
firstYearConstantCC = SP / A
End Function
'This function calculate current levelization for O&M cost
'Arguments: rd (cost of debt), float; wd(ratio of debt), float; re(cost of equity), float;
'we (ratio of equity), float; BL(book life), float; inflation, float; escalation, float
Function currentl_L(rd, wd, re, we, BL, inflation, escalation)
inflation = inflation
escalation = escalation
rd = rd
wd = wd
re = re
we = we
'Define discount (weighted average cost of capital)
Dim discount
discount = wd x rd + we x re
Dim EA
EA = (1 + inflation) x (1 + escalation) - 1
Dim k
k= (1 + EA)/ (1 + discount)
Dim An
An = ((1 + discount) ^ BL - 1) / (discount x (1 + discount) ^ BL)
Dim Ln
Ln = (k x (1 - k ^ BL)) / (An x (1 - k))
currentLL = Ln
End Function
Function constantLL(rd, wd, re, we, BL, inflation, escalation)
inflation = inflation
escalation = escalation
rd = rd
wd = wd
re = re
we = we
146
-------
Appendix F
rd = (1 + rd)/ (1 + inflation) - 1
re = (1 + re) / (1 + inflation) - 1
Dim discount
discount = wd x rd + we x re
Dim EA
EA = (1 + escalation) - 1
Dim k
k = (1 + EA) / (1 + discount)
Dim An
An = ((1 + discount) ^ BL - 1) / (discount x (1 + discount) ^ BL)
Dim Ln
Ln = (k x (1 - k ^ BL)) / (An x (1 - k))
constantLL = Ln
'constantLL = re
End Function
The function below is a function to calculate the compressibility of compressors across the
compressor island in CO2 compression
'Define a function
Function CompressionPower(initialPressure, finalPressure, numOfStage, CompressionTemp)
Dim pressureRatio As Double
Dim pressureOfEachStage(lO) As Double
Dim Z(10) As Double
Dim i As Integer
Dim averageZ
'initalize pressure
pressureOfEachStage(O) = initialPressure
'initalize Z
If pressureOfEachStage(O) > 800 Then
Z(0) = 0.5
Else
Z(0) = 1-0.4/ 800 x pressureOfEachStage(O)
End If
pressureRatio = ((finalPressure + 1.5 x numOfStage) / initialPressure) ^ (1# /
numOfStage)
'calculate Z factors
CompressionPower = 0#
For i = 1 To numOfStage
147
-------
Appendix F
pressureOfEachStage(i) = pressureOfEachStage(i - 1) x pressureRatio
'calculate outlet Z
Z(i) = 1-0.5/ 2000 x pressureOfEachStage(i)
If Z(i) < 0.5 Then
Z(i) = 0.5
End If
'calculate inlet Z
If pressureOfEachStage(i - 1) > 800 Then
Z(i - 1) = 0.5
Else
Z(i - 1) = 1 - 0.4 / 800 x pressureOfEachStage(i - 1)
End If
'Calculate average Z
averageZ = (Z(i) + Z(i - 1)) / 2
CompressionPower = CompressionPower + 188.9 x (CompressionTemp + 273.15) x
1.28 x averageZ / (1.28 - 1) x (pressureRatio ^ (0.28 / 1.28) - 1) / 0.8
Next
End Function
148
------- |