EPA
TVA
United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park, NC 27711
EPA-600/7-8i-12
July 1981
Tennessee Valley
Authority
Office of Power
Energy Demonstrations
and Technology
Muscle Shoals. AL 35660
TVA/OP/EDT-81/2
Evaluation of the Advanced
Low-NOx Burner, Exxon,
and Hitachi Zosen DeNOx
Interagency
Energy/Environment
R&D Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wide range of energy-related environ-
mental issues.
EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect
the views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/7-81-120
TVA/OP/EDT-81-28
JULY 1981
Evaluation of the Advanced
Low-NOx Burner, Exxon,
and Hitachi Zosen DeNOx Processes
By
J.O. Maxwell and L.R. Humphries
Tennessee Valley Authority
Office of Power
Division of Energy Demonstrations and Technology
Muscle Shoals, AL 35660
EPA Interagency Agreement No. 79-D-X0511
Program Element No. INE829
EPA Project Officer: J. David Mobley
Industrial Environmental Research Laboratory
Office of Environmental Engineering and Technology
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
fl.S, KbivlroTraontal Protection
"' " '"''
-------
LEGAL NOTICE
This report was prepared by the Tennessee Valley Authority and has been
reviewed by the U.S. Environmental Protection Agency and approved for
publication. Neither TVA, EPA, nor any person acting on their behalf:
1. makes any warranty or representation, express or implied, with
respect to the accuracy, completeness, or usefulness of the
information contained in this report, or that the use of any
information, apparatus, method, or process disclosed in this
report may not infringe privately owned rights; or
2. assumes any liabilities with respect to the use of, or for
damages resulting from the use of, any information, apparatus,
method, or process disclosed in this report.
This report does not necessarily reflect the views and policies of TVA or
EPA.
ii
-------
ABSTRACT
A technical discussion and a preliminary economic evaluation are made for
three nitrogen oxide (NOX) emission control methods at 50% NOX reduction
and three NOX control methods at 90% NOX reduction. The base-case power
plant is a new 500-MW coal-fired unit emitting 0.6 Ib N02/MBtu in the flue
gas. The three 50% NOX reduction processes are the EPA-sponsored advanced
low-NOx burner (ALNB), the Exxon Thermal DeNOx process, and the Hitachi
Zosen process, which have capital investments of $4.8/kW, $19.7/kW, and
$31.4/kW, respectively, and levelized annual revenue requirements of 0.20,
1.9, and 4.7 mills/kWh respectively. For 90% NOX reduction, the ALNB
process is combined with the Hitachi Zosen process, the Exxon process is
combined with the Hitachi Zosen process, and the Hitachi Zosen process is used
alone. Capital investment and levelized annual revenue requirements for these
three processes are $51.8/kW and 6.7 mills/kWh for the ALNB/Hitachi Zosen
process, $64.2/kW and 8.2 mills/kWh for the Exxon/Hitachi Zosen process, and
$50.9/kW and 7.9 mills/kWh for the Hitachi Zosen process. The ALNB, a
combustion modification, is the least expensive NOX control method. As
would be expected, the costs for obtaining high levels of NOX reduction
(90%) are significantly greater than for more moderate levels (50%).
ill
-------
CONTENTS
Abstract iii
Figures vii
Tables ix
Abbreviations and Conversion Factors ... x
Executive Summary xiii
Introduction 1
Background ...... .......... 3
NOZ Control Regulations 3
NOX Formation Chemistry 3
Status of Technology 4
Combustion Modification 4
Flue Gas Treatment 8
Status of Processes Evaluated 12
Advanced Low-NOx Burner 12
Process Description 12
Technical Considerations 14
Development Status 19
Exxon Process 22
Process Description 22
Technical Considerations 29
Development Status 30
Hitachi Zosen Process 31
Process Description 31
Technical Considerations 33
Development Status 40
Premises 45
Design Premises 45
Plant Size and Fuel 45
Flue Gas Composition 46
NOX Control System 46
Raw Materials 48
Economic Premises 48
Capital Investment Estimates ........ 49
Annual Revenue Requirements 51
Systems Estimated 53
Advanced Low-NOx Burner 53
Exxon Process ..... 57
v
-------
NH3 Storage and Injection 57
Air Preheater Section 57
Hitachi Zosen Process 61
Process Description (90% NOX reduction) 62
Process Description (50% NOX reduction) 67
Results and Comparison 72
Capital Investment 72
Fifty Percent NOX Reduction 72
Ninety Percent NOX Reduction 73
Overall Capital Investment Comparison 74
Annual Revenue Requirements ... 76
Fifty Percent NOX Reduction 77
Ninety Percent NOX Reduction 77
Overall Annual Revenue Requirements Comparison 78
Overall Capital Investment and Annual Revenue Requirements
Comparison ....... 81
Energy Consumption ... 81
Conclusions 86
References 88
Appendix A 91
Appendix B 109
vi
-------
FIGURES
Number Page
S-l Assumed NOX reduction for the six cases studied xv
S-2 Cost for reduction of a pound of NOX based on levelized
annual revenue requirements . xxii
S-3 Capital investment comparison and accuracy range (based
on a -20% to +40% range for Exxon and Hitachi Zosen
processes and a -20% to +100% range for the ALNB) xxiv
S-4 Levelized annual revenue requirements and accuracy range
(based on a -20% to +40% range for Exxon and Hitachi Zosen
processes and a -20% to +100% range for the ALNB) xxv
1 Baseline NOX emissions - coal-fired utility boilers 5
2 Pulverized-coal-fired boiler burner zone heat release
rates 6
3 Major stages of the ALNB development 13
4 Alternatives for injection of ALNB tertiary air 14
5 Effects of ALNB zone stoichiometry, single burner 15
6 Effects of ALNB primary swirl, single burner 16
7 Effects of ALNB secondary swirl, single burner 17
8 Effects of ALNB zone stoichiometry, four burners fired at
12.5 x 106 Btu/hr each 18
9 Effects of ALNB tertiary ports out of service, four burners
fired at 12.5 x 106 Btu/hr each 20
10 Effects of ALNB coal type, four burners fired at 12.5 x 106
Btu/hr each 21
11 Variables evaluated in the single ALNB test program 23
12 Effect of temperature on NO reduction for various levels
of NH3 injection with the Exxon process 25
13 Effect of temperature on NO reduction for various fuel
sources with the Exxon process 26
14 Comparison of NO reductions at the optimum temperature
condition with the Exxon process 27
15 Comparison of the NH3 emissions for all fuels tested
at the peak NO reduction temperature with the Exxon process . 28
16 NOz reduction versus reaction temperature for the Hitachi
Zosen process 34
17 Configuration of Hitachi Zosen NOXNON 500 and 600 series
catalyst 35
18 Relationship between exhaust NH3, NOZ reduction, and molar
ratio for the Hitachi Zosen process 36
19 Conditions for the formation of ammonium snlfate/
bisulfate 38
vli
-------
20 Pressure drop versus operating time for the Hitachi Zosen
process ..... ...... 39
21 NOX reduction versus catalyst age for the Hitachi Zosen
process 37
22 Influence of NOX concentration on NOX reduction with the
Hitachi Zosen process .... 41
23 Influence of 02 concentration on NOX reduction for the
Hitachi Zosen process 42
24 Assumed NOX reduction for the six cases studied 54
25 Advanced low-NOx burner flow diagram 55
26 Exxon process flow diagram . 58
27 Hitachi Zosen process (90% NOX reduction) flow diagram ... 63
28 Hitachi Zosen process (50% NOX reduction) flow diagram ... 68
29 Cost for reduction of a pound of NOX based on levelized
annual revenue requirements 82
30 Capital investment comparison and accuracy range (based on
a -20% to +40% range for Exxon and Hitachi Zosen processes
and a -20% to a +100% range for the ALNB) 83
31 Levelized annual revenue requirements and accuracy range
(based on a -20% to +40% range for Exxon and Hitachi Zosen
processes and a -20% to +100% range for the ALNB) 84
viii
-------
TABLES
Number Page
S-l Summary of Capital Investments ziz
S-2 Summary of Annual Revenue Requirements zx
S-3 The Effect of Catalyst Life on Annual Revenue Requirements . . zzi
S-4 Energy Requirements zzvi
1 NOZ Emissions Standards and Projected Research Objectives for
Large Fossil-Fuel-Fired Boilers ....... 3
2 Summary of Commercial Applications of Ezzon Thermal DeNOx
Process 32
3 Influence of S02 and H20 Concentrations 40
4 Hitachi Zosen Pilot-Plant Experience 43
5 Commercial Plants Using the Hitachi Zosen Process 44
6 Coal Plant Base Mechanical Equipment 45
7 Base Case Coal Composition and Input Flow Rate 46
8 Flue Gas Composition and Flow Rate at the Economizer Outlet . 47
9 Levelized Annual Capital Charges for Regulated Utility
Financing 49
10 Cost Indezes and Projections 50
11 Cost Factors 52
12 Advanced Low-N0x Burner Material Balance ........... 56
13 Exxon Material Balance 59
14 Exxon Equipment List 60
15 Hitachi Zosen (90% NOX Reduction) Material Balance 64
16 Hitachi Zosen (90% NOZ Reduction) Equipment List 65
17 Hitachi Zosen (50% NOX Reduction) Material Balance ...... 69
18 Hitachi Zosen (50% NOX Reduction) Equipment List 70
19 Summary of Capital Investments 73
20 Contribution of Direct Investment, Royalties, and Catalyst to
Capital Investment 75
21 Summary of Annual Revenue Requirements ..... 76
22 Contribution of Raw Materials and Utilities to Annual Revenue
Requirements ..... 79
23 The Effect of Catalyst Life on Annual Revenue Requirements . . 80
24 Comparison of Energy Requirements . 85
ix
-------
ABBREVIATIONS AND CONVERSION FACTORS
ABBREVIATIONS
ac acre
aft3/min actual cubic feet per
minute
ALNB advanced low-NOz burner
bbl barrel
Btu British thermal unit
°F degrees Fahrenheit
dia diameter
F6D flue gas desulfurization
F6T flue gas treatment
ft feet
ft2 square feet
ft3 cubic feet
gal gallon
gpm gallons per minute
gr grain
hp horsepower
hr hour
in. inch
k thousand
kW kilowatt (electric)
kWh kilowatthour
Ib pound
L/G liquid to gas ratio in
gallons per thousand
actual cubic feet of
gas at outlet condi-
tions
M million
mi mile
mo month
MW megawatt (electric)
MWt megawatt (thermal)
ppm parts per million
psig pounds per square inch
(gauge)
rpm revolutions per minute
RR first-year annual revenue
requirement
SCR selective catalytic
reduction
sec second
sft3/min standard cubic feet per
minute (60°F)
Sp swirl angle, primary air
SNR selective noncatalytic
reduction
SRg stoichiometric ratio, total
air minus tertiary air
SRp stoichiometric ratio,
primary air
SRj stoichiometric ratio,
theoretical
Sjj swirl angle, secondary air
SS stainless steel
TCI total capital investment
yr year
-------
CONVERSION FACTORS
EPA policy is to express all measurements in Agency documents in metric units. Values in this
report are given in British units for the convenience of engineers and other scientists accustomed to
using the British systems. The following conversion factors may be used to provide metric equivalents.
To convert British
Multiply by
To obtain Metric
ac acre 0.405
bbl barrels of oila 158.97
Btu British thermal unit 0.252
°F degrees Fahrenheit minus 32 0.5556
ft feet 30.48
ft2 square feet 0.0929
ft3 cubic feet 0.02832
ft/min feet per minute 0.508
ft3/min cubic feet per minute 0.000472
gal gallons (U.S.) 3.785
gpm gallons per minute 0.06308
gr grains 0.0648
gr/ft3 grains per cubic foot 2.288
hp horsepower 0.746
in. inches 2.54
lb pounds 0.4536
lb/ft3 pounds per cubic foot 16.02
Ib/hr pounds per hour 0.126
psi pounds per square inch 6895
mi miles 1609
rpm revolutions per minute 0.1047
sft3/min standard cubic feet per 1.6077
minute (60°F)
hectare
liters
kilocalories
degrees Celsius
centimeters
square meters
cubic meters
centimeters per second
cubic meters per second
liters
liters per second
grams
grams per cubic meter
kilowatts
centimeters
kilograms
kilograms per cubic meter
grams per second
pascals (newton per square meter)
meters
radians per second
normal cubic meters per
hour (0°C)
ha
L
kcal
°C
cm
m2
m3
cm/s
m3/s
L
L/s
kW
cm
kg »
kg/m3
g/s
Pa (N/m2)
m
rad/s
m3/h (normal)
ton
ton, long
ton/hr
a.
b.
tons (short)6
tons (long)b
tons per hour
Forty-two U.S. gallons per
All tons, including tons of
barrel of oil
sulfur, are
0.9072
1.016
0.252
•
expressed
metric tons
metric tons
kilograms per
in short
tons
second
tonne
tonne
kg/s
in this report.
-------
ACKNOWLEDGEMENTS
The authors wish to acknowledge David G. Lachapelle, Program Manager of
Utility/Large Industrial Boiler Technology Development of Industrial Process
Combustion Equipment Technology Development for EPA, and G. Blair Martin,
Program Manager of Advanced Concepts Technology Development and of Fundamental
Combustion Research for EPA, for their assistance in the planning and
execution of this project.
xii
-------
EVALUATION OF THE ADVANCED LOW-NOX BURNER, EXXON,
AND HITACHI ZOSEN DENOX PROCESSES
EXECUTIVE SUMMARY
INTRODUCTION
The new source performance standards (NSPS) for steam electric generating
boilers promulgated after the Clean Air Act Amendments of 1977 require the
installation of the best available control technology (BACT) for nitrogen
oxides (NOX) on new or modified facilities, including electric power
generating units. Therefore, as NOX control technology is further
developed, more stringent standards may be required. The 1977 Clean Air Act
Amendments also require the promulgation of a short-term nitrogen dioxide
(N02> ambient air quality standard (three hours or less averaging time)
unless it can be demonstrated that the standard is unnecessary for public
health protection. This short-term ambient standard may require additional
NOX emission control for stationary sources. Prevention of significant
deterioration (PSD) regulations for N02 are also required by the 1977 Clean
Air Act Amendments and these requirements may also lead to more stringent
NOX emission control for stationary sources.
There are two basic types of NOX control technology currently under
development: combustion modifications and flue gas treatment (FGT).
Combustion modifications primarily include the use of low excess air, staged
combustion with either overfire air or burners out of service, flue gas
recirculation, burner design and operating modifications, or combinations of
the above. Combustion modification techniques have received the most
development emphasis in the United States. However, FGT may be needed to
achieve the NOX emission reductions which may be required in the future.
FGT can be divided into two general categories: dry and wet. The
majority of dry processes involve a gas-phase reaction with a reducing agent,
NH3, which is added to the flue gas. If the NH3 is injected into the
superheater region of the boiler where temperatures are high (1,740°F), a
catalyst is not necessary. These processes are known as selective
noncatalytic reduction (SNR) processes. If the flue gas is treated after the
boiler economizer, a catalyst is necessary to produce the needed reaction
rate. These processes are typically described as selective catalytic
reduction (SCR) processes. Development of wet NOX FGT processes has
practically ceased because of the complexity and unfavorable economics of
these processes compared with dry processes.
xiii
-------
The major purposes of this study are to provide current technical
information on the various selected NOX control methods and to compare the
economics of these NOX control methods using common design and economic
premises. The NOZ control techniques selected for evaluation are combustion
modification, SNR, and SCR processes. Wet processes are not evaluated. The
combustion modification technique evaluated is the advanced low-NOx burner
(ALNB) being developed under the U.S. Environmental Protection Agency (EPA)
sponsorship by the Energy and Environmental Research Corporation (EERC); a 50%
NOX reduction capability is assumed for the ALNB in this study. The Exxon
Thermal DeNOx process, an SNR process, is also evaluated; moderate NOX
reduction (30% to 60%) is attainable by this process. The SCR process
evaluated in this study is the Hitachi Zosen process. This type of process
has the capability for high levels of NOX reduction (90%).
Two levels of NOX reduction from NSPS level are examined. Assuming a
baseline emission for the boiler of 0.6 Ib NC>2/MBtu (450 ppm at 3% 02)
costs are determined for a moderate 50% NOX reduction to 0.3 Ib N(>2/MBtu
(225 ppm at 3% 02) for each of the three processes although 50% NOX
reduction is not typical for the Hitachi Zosen process. The alternative
control level, 90% NOX reduction to 0.06 Ib N02/MBtu (45 ppm at 3% C^),
involves the following three situations:
1. Moderate NOX reduction achieved by the ALNB (0.6 to 0.3 Ib
N(>2/MBtu) and the remaining NOX reduction (80%) achieved by the
Hitachi Zosen process (0.3 to 0.06 Ib N02/MBtu).
2. Moderate NOX reduction achieved by the Exxon process (0.6 to 0.3 Ib
N(>2/MBtu) and the remaining NOX reduction (80%) achieved by the
Hitachi Zosen process (0.3 to 0.06 Ib N02/MBtu).
3. All of the NOX reduction (90%) achieved by the Hitachi Zosen process
(0.6 to 0.06 Ib N02/MBtu).
The levels of NOX reduction assumed for each process and process
combination evaluated in this report are graphically illustrated in
Figure S-l.
DESIGN AND ECONOMIC PREMISES
A specific set of design and economic premises is used to compare the
process economics of the different types of NOX control technology on a
consistent basis. The basic premises used in this report were established by
TVA for comparative economic evaluations of power plant stack gas emission
control processes.
Design Premises
The power plant assumed as a basis for this study is a new 500-MW,
horizontally opposed, balanced-draft boiler burning pulverized coal and
situated in a north-central location (Illinois, Indiana, Ohio, Michigan, or
xiv
-------
CN|
O
S3
53
O
M
C/2
CO
M
S
w
0.7 -.
0.6
0.5
0.4
0.3
0.2,
0.1
0.06,
0
Single Processes
Combination Processes
ALNB Exxon Hitachi
Zosen
ALNB/Hitachi Exxon/ Hitachi Zosen
Zosen Hif-arhi 7ncp>n
3 4
CASES STUDIED
Figure S-l. Assumed NC> reduction for the six cases studied.
A.
XV
-------
Wisconsin). The fuel is a bituminous coal with a heating value of 11,700
Btu/lb as fired and containing 3.5% sulfur and 15.1% ash. The boiler heat
rate is 9,500 Btu/kWh. On-stream time for the boiler is 5,500 hr/yr. Raw
materials are assumed to be received by either rail or truck. Thirty-day
storage facilities at full load (500 MW) are provided.
The estimated costs for the selected NOX control methods reported in
this study are differential costs for new boilers, i.e., only additional
expenses above present new boiler costs are included. Costs are not included
for present combustion modification techniques being applied to new boilers,
which are reportedly capable of reducing NOX emissions to the NSPS of 0.6 Ib
N(>2/MBtu for bituminous coal. In the case of the ALNB, only the
differential costs over those for regular burners are included. Also,
separate induced draft (ID) fans are not included. Where applicable, a larger
boiler ID fan is used to compensate for the higher pressure drops of the
combined boiler-NOx control process and the increased costs are assigned to
the NOX control process.
Economic Premises
Capital investment estimates are based on projected mid-1982 construction
costs. The revenue requirements are based on projected 1984 costs. Delivered
raw material costs and labor rates are based on a north-central location.
Capital investment consists of direct investment, indirect investment,
and other capital investments. The direct investment is based on equipment
costs. Other installation costs (such as piping, electrical, instrumentation,
etc.) are factored from the equipment costs. Indirect investment (engineering
design and supervision, construction expense, etc.) is based on the direct
investment. Other capital investments, such as allowance for startup and
modification and interest during construction, are based on the total direct
and indirect investment. Other capital investments also include, when
applicable, land, working capital, royalties, and initial catalyst charges.
Two types of annual revenue requirements are projected—first year and
levelized. Both are based on 5,500 hr/yr of operation at full load (about a
63% capacity factor) and both use a levelized capital charge. Levelized
annual revenue requirements differ from first-year annual revenue requirements
in that they take into consideration the time value of money over the life of
the NOX reduction unit. They are calculated using a 10%/yr discount factor,
a 6%/yr inflation factor, and a 30-year economic life.
PROCESS BACKGROUND AND SYSTEMS ESTIMATED
Advanced Low-NOy Burner
An ALNB for both utility and industrial applications is being developed
under EPA sponsorship by EERC to minimize NOX formation during the
combustion of coal. The primary objectives are: (1) to provide an initial
fuel-rich, i.e., oxygen-deficient, zone which maximizes the conversion of
xvi
-------
organic nitrogen coumpounds to N2, and (2) to blanket the fuel-rich zone
with an oxidizing atmosphere to maximize burnout and to minimize the potential
for corrosion in the lower furnace section of the boiler. The research goal
is to attain an NOX emission level equivalent to 0.2 Ib N02/MBtu. For
this study, the NOX emission level assumed for the ALNB is 0.3 Ib
N02/MBtu.
The utility burner design consists of a central burner, similar to the
Babcock & Wilcox (B&W) dual register burner, plus four tertiary air ports
located about one throat diameter from the central burner. The air admitted
through the central burner is divided into three streams referred to as
primary and inner and outer secondary air. Primary air that carries the
entrained pulverized coal is about 25% of stoichiometric requirements.
Secondary air is injected from an annulus around the primary air port and is
about 45% of stoichiometric requirements. Swirl vanes in the inner secondary
annulus impart swirl to control mixing of the primary and inner secondary
streams and to control flame shape. The remaining secondary air is admitted
through the outer secondary annulus. About 50% of the stoichiometric air
quantity is admitted through the tertiary air ports. Thus, with a total air
supply of 120% of stoichiometric requirements, the central burner operates at
about 70% stoichiometry to minimize oxygen availability and conversion of fuel
nitrogen to NOX, while the remaining stoichiometric air supplied through the
tertiary air ports completes combustion and maintains an overall oxidizing
atmosphere.
Exxon Process
The Exxon process, developed by Exxon Research and Engineering, reduces
NOX in flue gas by dry SNR. NH3 with air (94-98 volume % air) is injected
into the cavity of the secondary superheater tube bank region where the
temperature is high enough for NOX and NHg to react to form N2 and
H20. For NOX reduction of 40% to 60%, the optimum temperature range is
1,650°F to 1,830°F, and the NH3:NOX molar ratio varies from 1:1 to
2:1. Injecting H2 with the NR$ reduces the temperature level at which
reaction rates are adequate for efficient NOX reduction. At H2:NH3
molar ratios of about 2:1 NOX reduction can occur rather rapidly at about
1,290°F. NH3 is injected through an insulated piping grid which covers
the entire cross-sectional area of the flue gas flow path.
Since NOX reduction is sensitive to temperature, steps must be taken to
maintain NOX reduction efficiency with varying boiler loads. There are many
alternative techniques presently used, i.e., the use of one or multiple grids
to inject only the NH3, or one or multiple grids with both NH3 and H2
injection. In this study a two grid system injecting only NH3 is used. An
NH3:NOX molar ratio of 1.5:1 is assumed to achieve 50% NOX reduction.
The major equipment for this process includes only NHg storage tanks,
NH3 injection grids, and compressors for the air used as the NH3
carrier. Also included in this study under equipment costs are costs for air
heater design and operating modifications that may be necessary when using
this process with a coal-fired boiler. (These modifications result from the
xvii
-------
NH4HS04 and (NH4)2S04 deposited in the air heater by reaction of
NH3 with 863 present in coal-fired flue gas.)
Hitachi Zosen Process
The Hitachi Zosen process is a dry SCR process in which NE^ is injected
into the flue gas and reacts with NOX selectively, in the presence of a
catalyst, to form N2 and I^O. The catalyst allows the reaction to proceed
rapidly at temperatures in the 600°F to 750°F range. The corrugated
catalyst units, made with a proprietary catalyst on a steel support, permit
treatment of flue gas with high particulate loadings. Therefore, based on the
temperature range and the lack of required particulate removal, flue gas from
a coal-fired boiler may be fed directly from the boiler economizer to an NOX
reduction reactor. With NH3:NOX molar ratios of 1:1 or greater, 90% NOX
control is reportedly achievable. This process requires only NH3 storage
and injection facilities and a catalytic reactor. In this study an economizer
bypass is also included to maintain acceptable reactor temperatures when the
boiler load is low. The temperatures during low boiler loads may not be high
enough to obtain adequate NOX control and may also be low enough to allow
NfyHSCfy and (1^4)2804 formation in the catalytic reactor. Also, air
heater design and operating modifications are included to minimize the
potentially adverse effects of these salts, as is done for the Exxon process.
For the case in which overall NOX reduction is 90%, two reactor trains
are used, each treating 50% of the flue gas. An NH3:NOX molar ratio of
1:1 is used. For an overall 50% NOX reduction only one reactor train is
used to treat 56% of the flue gas. Again, an NH3:NOX molar ratio of 1:1
is used. A third case is for an overall reduction of 80%, but from a lower
initial NOX level (from 0.3 to 0.06 Ib N02/MBtu). This case is used in
combination with either the ALNB or Exxon process to achieve the same overall
degree of NOX reduction as with the Hitachi Zosen 90% NOX reduction
case. Two reactor trains similar in size to the first case are used, but with
less catalyst. An NH3:NOX molar ratio of 0.9 is used. A catalyst life of
one year is assumed for all three cases.
ECONOMIC RESULTS AND COMPARISON
The process economics consist of capital investment, first-year revenue
requirements, and levelized annual revenue requirements. Because of the
different sources of data, the simplifying assumptions made, and the necessity
of projecting costs into the future, these estimates are considered to be
accurate to an overall variation of -20% to +40% for the Exxon and Hitachi
Zosen processes. For the ALNB the estimates are considered to be accurate to
an overall variation of -20% to +100%. This larger range is a result of the
less advanced development status of the ALNB and the relative lack of detailed
design, equipment needs, and costs in comparison with the NOX F6T
processes. The findings of this report apply only to a new installation.
xviii
-------
Results - Capital Investment
The capital investment results for the 50% and 90% NOX reduction cases
are shown in Table S-l. The ALNB has the lowest capital investment of the
three 50% NOZ reduction processes studied. Because of the much greater
amounts of process equipment, the investments for the Exxon and Hitachi Zosen
processes are significantly higher. A royalty fee is not charged for the ALNB
technology while royalty fees are $1.5M and $0..5M for the Exxon and Hitachi
Zosen processes, respectively. The cost of the initial catalyst charge
($3.7M) represents a large portion of the Hitachi Zosen capital investment.
TABLE S-l. SUMMARY OF CAPITAL INVESTMENTS
Capital
investment,
mid-19825
Process M$ $/kW
50% NOX reduction
ALNB 2.4 4.8
Exxon 9.9 19.7
Hitachi Zosen 15.7 31.4
90% NOX reduction
ALNB/Hitachi Zosen 25.9 51.8
Exxon/Hitachi Zosen 32.1 64.2
Hitachi Zosen 25.5 50.9
For 90% NOX reduction, the Hitachi Zosen and ALNB/Hitachi Zosen
processes have similar capital investments, while the investment for the
Exxon/Hitachi Zosen process is higher. Royalty fees are $0.5M for the
ALNB/Hitachi Zosen and the Hitachi Zosen processes, and $2.0M for the
Exxon/Hitachi Zosen process. The initial catalyst charge for the ALNB/Hitachi
Zosen and the Exxon/Hitachi Zosen processes costs $5.0M for each process while
the Hitachi Zosen process has a cost for the initial catalyst charge of
$6.5M. In contrast to the other processes, the royalty fee for the Exxon
process is a major cost item. It is equal to 15% of the capital investment.
Although both the Exxon process and the Hitachi Zosen process use
the equipment costs associated with NH3 are much larger for the Exxon
process. The investment for NH3 storage and injection with the process is
four times larger than that for the Hitachi Zosen 90% NOX reduction
process. This is a result of a larger NHg consumption and subsequent larger
storage needs, a more intricate and expensive injection grid, and a more
expensive air carrier system for the Exxon process.
xix
-------
Results - Annual Revenue Requirements
The first-year and levelized annual revenue requirements for both the 50%
and 90% NOX reduction cases are shown in Table S-2. As with the capital
investment, for 50% NOX reduction the ALNB has significantly lower revenue
requirements than the Exxon or Hitachi Zosen processes. The only substantial
revenue requirement item for the ALNB is the levelized capital charges, while
the Exxon and Hitachi Zosen processes not only have higher capital charges,
but also substantial raw material, utility, and maintenance requirements. The
annual catalyst replacement cost is a large portion (33%) of the Hitachi Zosen
annual revenue requirement.
TABLE S-2. SUMMARY OF ANNUAL REVENUE REQUIREMENTS
Annual revenue requirements
(1984$)
First year Levelized
Process M$ Mills/kWh M$ Mills/kWh
50% NOX reduction
ALNB
Exxon
Hitachi Zosen
90% NOX reduction
ALNB/Hitachi Zosen
Exxon/Hitachi Zosen
Hitachi Zosen
0.45
3.4
8.0
11.5
14.2
13.3
0.17
1.2
2.9
4.2
5.2
4.9
0.54
5.1
13.0
18.4
22.6
21.9
0.20
1.9
4.7
6.7
8.2
7.9
For 90% NOX reduction, although the Hitachi Zosen and ALNB/Hitachi
Zosen processes have similar capital investments, the Hitachi Zosen process
levelized annual revenue requirements are $3.5M higher than those for the
ALNB/Hitachi Zosen process due to greater catalyst requirements of the Hitachi
Zosen process. The revenue requirements for the Exxon/Hitachi Zosen process
are the highest of the three because of higher conversion cost and levelized
capital charges. Annual catalyst replacement is required for all three
processes. The annual catalyst replacement cost contributes appreciably to
the levelized annual revenue requirements: 32% for the ALNB/Hitachi Zosen
process, 26% for the Exxon/Hitachi Zosen process, and 34% for the Hitachi
Zosen process.
Since a one-year catalyst life is guaranteed it was used as a basis for
the cost estimate however, actual catalyst life could be longer. Should a
two-year life be obtainable for the Hitachi Zosen process a significant
savings can be realized in the annual revenue requirements. Table S-3 shows
that a two-year catalyst life will reduce the levelized annual revenue
xx
-------
requirements of the Hitachi Zosen (50% NOX reduction), ALNB/Hitachi Zosen,
and Hitachi Zosen (90% NOX reduction) processes by 30% and the Exxon/Hitachi
Zosen process by 24%. Even though this is a significant reduction for the
processes requiring catalyst, it is not sufficient to change the cost
relationship of the processes. For 50% NOX reduction the ALNB is still
lowest followed by the Exxon and Hitachi Zosen processes and for 90% NOX
reduction the ALNB/Hitachi Zosen remains the lowest followed by the Hitachi
Zosen and Exxon/Hitachi Zosen processes.
TABLE S-3. THE EFFECT OF CATALYST LIFE ON
ANNUAL REVENUE REQUIREMENTS
Process
Annual
Hitachi Zosen catalyst
replacement cost. M$
1-year 2-year
catalyst catalyst
life life
Levelized
annual revenue
requirements. M$
1-year 2-year
catalyst catalyst
life life
50% reduction
Hitachi Zosen
90% reduction
ALNB/Hitachi Zosen
Exxon/Hitachi Zosen
Hitachi Zosen
4.3
5.8
5.8
7.5
2.1
2.9
2.9
3.7
13.0
18.4
22.6
21.9
9.0
12.9
17.1
14.8
On the basis of cost per pound of
removed, as shown in Figure S-2,
50% NOX reduction is substantially less expensive than 90% NOX reduction,
with the exception of the Hitachi Zosen 50% reduction process. The large cost
difference between the Hitachi Zosen 50% reduction case and the two other 50%
reduction cases occurs because the ALNB and Exxon processes do not require the
expensive annual catalyst replacement.
The Hitachi Zosen 50% NOX reduction case has a slightly higher cost per
pound of N(>2 removed than the Hitachi Zosen 90% reduction case for two
reasons. First, there is some economy of scale in the capital investment for
the 90% reduction case compared with the 50% reduction case. Therefore, the
level ized capital charges and maintenance costs, which are factored from the
capital investment, are a smaller portion of the annual revenue requirements
for the 90% reduction case. Also, certain items are of equal cost at both the
90% and 50% reduction levels, such as operating and analysis labor, resulting
in a smaller cost per pound of N02 removed at the 90% reduction level.
The N% cost for the Exxon process is over twice that of the Hitachi
Zosen process (50% NOX reduction case) and even 1-1/2 times higher than that
xxi
-------
X
H-
C/l
w
w
u
§
eu
ALNB
Exxon
Hitachi Zosen 50%
ALNB/Hitachi Zosen
Exxon/Hitachi Zosen
Hitachi Zosen 90%
i i
I I • I
50% NOX Reduction
90% NOV Reduction
0 0,2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
$/LB OF N02 REDUCED
Figure S-2. Cost for reduction of a pound of NOX based on levelized annual revenue
requirements.
-------
for the Hitachi Zosen 90% NOX redaction process because of the higher
NH3:NOX molar ratios required. The NH3 costs are about 15% of the Exxon
levelized annual revenue requirements while they are only 2% of the levelized
annual revenue requirements for either case of the Hitachi Zosen process.
Overall Capital Investment and Annual Revenue Requirements Comparison
A comparison of the capital investment and levelized annual revenue
requirements for each of the six NOX control processes is shown in
Figures S-3 and S-4 respectively. Also included in the figures are the
accuracy ranges on the capital investment and levelized annual revenue
requirements.
For 50% NOX reduction the ALNB has the lowest capital investment and
levelized annual revenue requirements. The Exxon process has the second
lowest and Hitachi Zosen has the highest capital investment and levelized
annual revenue requirements.
The capital investment of the ALNB/Hitachi Zosen and the Hitachi Zosen
processes are almost equal for 90% NOX reduction, but the levelized annual
revenue requirements are lower for the ALNB/Hitachi Zosen process. In
comparison with the above two processes, the Exxon/Hitachi Zosen process
capital investment is substantially higher; however, the Exxon/Hitachi Zosen
levelized revenue requirement is comparable.
Capital investment and levelized annual revenue requirements are
significantly higher for 90% NOX reduction than for 50% NOX reduction.
ENERGY CONSUMPTION RESULTS AND COMPARISON
Energy consumption for all the 50% and 90% NOX reduction cases studied
is less than 1% of the boiler capacity, as shown in Table S-4. Energy
requirements for the three 50% reduction cases range from none for the ALNB to
0.4% of the boiler capacity for the Exxon process. The range for 90%
reduction is from 0.4% of the boiler capacity for the ALNB/Hitachi Zosen and
Hitachi Zosen processes to 0.7% of the boiler capacity for the Exxon/Hitachi
Zosen process. The NOX control alternatives containing the Exxon process
are the highest energy consumers at both the 50% and 90% NOX reduction
levels.
CONCLUSIONS
The economic conclusions of this study are based on NOX control
technology at various early stages of development applied to a new power
plant. Further development and retrofit applications could greatly alter both
the absolute and relative costs of the processes. To develop accurate and
timely economics in this rapidly evolving field, continued monitoring of
developments in NOX control technology is necessary.
xxiii
-------
co
CO
w
u
o
Pi
p-l
ALNB
Exxon
Hitachi Zosen 50%
ALNB/Hitachi Zosen
Exxon/Hitachi Zosen
Hitachi Zosen 90%
50% NOV Reduction
A.
90% NOX Reduction
I
I
I
I
10 20 30 40 50 60 70 80 90
CAPITAL INVESTMENT, S/kW
100
110 120
Figure S-3. Capital investment comparison and accuracy range (based on a -20% to +40% range for
Exxon and Hitachi Zosen processes and a -20% to +100% range for the. ALNB).
-------
w
u
8
P-!
ALNB
Exxon
Hitachi Zosen 50%
ALNB/Hitachi Zosen
Exxon/Hitachi Zosen
Hitachi Zosen 90%
50% NOX Reduction
D
90% NOX Reduction
1 2 3 4 5 6 7 8 9 10 11 12
LEVELIZED ANNUAL REVENUE REQUIREMENTS, mills/kWh
Figure S-4. Levelized annual revenue requirements and accuracy range (based on a -20% to +40%
range for Exxon and Hitachi Zosen processes and a -20% to +100% range for the ALNB),
-------
TABLE S-4. ENERGY REQUIREMENTS
Total equivalent
Steam, Electricity, energy consumption,
Process MBtu/hr MBtu/hr % of boiler capacity*
50% reduction
ALNB
Exxon
Hitachi Zosen
90% reduction
ALNB/Hitachi Zosen
Exxon/Hitachi Zosen
Hitachi Zosen
0.0
5.7
6.0
8.0
11.7
10.9
0.0
11.5
7.7
10.2
21.7
10.3
0.0
0.4
0.3
0.4
0.7
0.4
a. Does not include energy requirement represented by raw materials.
Based on a 500-MW boiler, a gross heat rate of 9,500 Btu/kWh for
generation of electricity, and a boiler efficiency of 90% for
generation of steam.
For moderate NOX reduction of 50%, the ALNB is by far the most
economical alternative, even if its costs were to increase several times
relative to the other processes. The ALNB NOX reduction efficiency and its
effects on boiler efficiency and operation remain to be fully demonstrated in
utility applications, however.
The Hitachi Zosen process has a higher capital investment than the Exxon
process at the 50% reduction level because of the initial catalyst charge,
reactor, additional ductwork, and additional fan capacity required. It also
has higher revenue requirements, primarily because of annual catalyst
replacement costs, although the NHg requirements are much less than those of
the Exxon process. Changes in catalyst cost or NH3 consumption or cost
would appreciably affect the cost relationship of these processes.
The royalties for the Exxon process are a significant portion of the
capital investment.
For NOZ reductions of 90% the combination of the ALNB/Hitachi Zosen
process is, overall, the most cost effective alternative. Although the
capital investment for the ALNB/Hitachi Zosen process is slightly higher than
the capital investment for the Hitachi Zosen process, the annual revenue
requirements are substantially lower. The magnitude of the difference in
annual revenue requirements is large enough to overcome the slightly higher
capital investment and make the ALNB/Hitachi Zosen process the most
economically attractive.
xxvi
-------
Except for the ALNB the energy requirements to achieve 50% and 90%
reduction are greater than those for particulate removal. However, energy
requirements for NOX reduction are still modest, much less than 1% of the
boiler output, in comparison with the energy needed for removal of S02 from
flue gas.
Catalyst cost is a very important economic factor for an SCR-type
process. With the Hitachi Zosen process, the catalyst cost may represent as
much as 25% of the capital investment and 35% of the levelized annual revenue
requirement. Catalyst life is also a very important economic factor. A
two-year catalyst life will reduce the levelized annual revenue requirements
of the Hitachi Zosen process by about 30%.
Since with current technology it is necessary to use a process such as
the SCR-type system to achieve low emission levels (0.06 Ib N02/MBtu), the
costs for obtaining these low levels versus more moderate emission levels (0.3
Ib N02/MBtu) are substantially greater. Achieving low emission levels, as
compared with moderate levels, may result in as much as a tenfold increase in
capital investment and a thirtyfold increase in annual revenue requirements.
xxvii
-------
EVALUATION OF THE ADVANCED LOW-NOX BURNER, EXXON,
AND HITACHI ZOSEN DENOX PROCESSES
INTRODUCTION
Manmade nitrogen oxide (NOX) emissions are classified, depending upon
their source, as stationary or mobile. Stationary sources are responsible for
about 60% of the total U.S. NOX emissions. Fuel combustion, especially in
industrial and utility boilers, produces most of the stationary source
NOX. In fact, industrial and utility boilers firing gas, oil, and coal were
responsible for approximately 50% of the manmade NOX produced in the United
States in 1978, with coal-fired utility boilers alone accounting for 22%
(27). This and the trend toward increased reliance on coal as the major fuel
for electrical energy generation have resulted in greater attention on NOX
control for stationary sources, especially in the utility industry.
There are two basic types of NOX control technology under development
for coal-firing applications: combustion modifications and flue gas treatment
(FGT). Combustion modifications primarily include the use of either low
excess air, staged combustion with overfired air or burners out of service,
flue gas recirculation, burner design and operating modifications, or
combinations of the above. Combustion modification techniques have received
the most development emphasis in the United States. However, to achieve the
NOX emission reductions that may be required in the future, FGT may be
needed.
The majority of FGT processes are dry processes involving a gas-phase
reaction with a reducing agent, usually NH3, that is added to the flue gas.
If the NH3 is injected into the cavities of the secondary superheater zone
of the boiler where the temperature is high (1,740°F), a catalyst is not
necessary. These processes are known as selective noncatalytic reduction
(SNR) processes. If the flue gas is treated after the boiler economizer,
where the flue gas temperature is low (730°F), a catalyst is necessary to
produce the needed reaction rate. These processes are typically described as
selective catalytic reduction (SCR) processes. Presently, development of wet-
scrubbing NOX FGT processes has practically ceased because of the complexity
and unfavorable economics of these processes in comparison with the dry
processes.
The major purposes of this study are to provide current technical
information on the various NOX control methods being developed and to
-------
compare the economics of selected types of NOZ control methods using a set
of consistent design and economic premises. The NOZ control techniques
selected for evaluation are combustion modifications and SNR and SCR
processes. The combustion modification technique evaluated is the advanced
low-NOx burner (ALNB) design being developed under the U.S. Environmental
Protection Agency (EPA) sponsorship by the Energy and Environmental Research
Corporation (EERC). The Exxon Thermal DeNOx process, an SNR process, is
also evaluated. Although the maximum NOZ reduction is lover than for SCR,
moderate NOZ reduction is achievable by this process. The SCR process
evaluated in this study is the Hitachi Zosen process. This type of process
has the capability of high (90%) levels of NOZ reduction.
Two levels of NOZ reduction are examined. Costs are determined to
achieve a moderate NOZ reduction to 0.3 Ib N02/MBtu (225 ppm at 3% 02),
for each of the three processes. In addition, costs are determined for a
greater reduction in NOZ to 0.06 Ib NC^/MBtu (45 ppm at 3% 02). This
degree of NOZ reduction requires either use of the Hitachi Zosen process or
combination of the Hitachi Zosen process with one of the other two processes.
The processes evaluated in this study are based on technology provided by
the process vendor through mid-1980. Since this time Hitachi Zosen and Exxon
have recommended process changes. Although the new technology may slightly
affect the cost of each process as presented in this study, the overall
comparability between processes should not be affected.
-------
BACKGROUND
NOX CONTROL REGULATIONS
Following enactment of the Clean Air Act of 1970, EPA promulgated new
source performance standards (NSPS) for control of NOX emissions from fossil-
fuel-fired steam electric generating plants in December 1971 (25). Following
the 1977 Clean Air Act Amendments, the NSPS were revised by EPA and
promulgated in June 1979 (26). The latest standards for large boilers O250
MBtu/hr) are shown in Table 1. The Clean Air Act Amendments of 1977 require
(1) the promulgation of a short-term N02 ambient air quality standard (three
hours or less averaging time), and (2) the prevention of significant
deterioration (PSD) regulations for NOX. These standards may result in
additional NOX emission control on electric power generating facilities.
TABLE 1. NOX EMISSIONS STANDARDS AND
PROJECTED RESEARCH OBJECTIVES FOR LARGE
FOSSIL-FUEL-FIRED BOILERS
June 1979 NSPS
Lb NOx/MBtu input NOX
to boiler* ppmb
Gaseous fuel 0.2 150
Liquid fuel 0.3 225
Solid fuel 0.5 (subbituminous) 375
0.6 (bituminous) 450
a. Expressed as
b. Calculated at 3% excess Oo» dry basis.
NOX FORMATION CHEMISTRY
For most combustion processes, particularly fossil-fuel combustion units,
the only significant quantities of NOX present in the flue gases are nitric
oxide (NO) and nitrogen dioxide (N02) with NO usually representing 90% to
-------
95% of the total NOX from the combustion unit (7) . Two separate mechanisms
contribute to the formation of NOX. One source, thermal NOX, results from
the oxidation of molecular nitrogen present in the combustion air, while the
second source, fuel NOX, results from the oxidation of nitrogen compounds
released from coal (10). For a large coal-fired boiler, the fuel NOX
contribution may range from 30% to 80% of the total NOX (15).
Thermal NOX emission levels, as the name implies, are primarily
dependent on the peak flame temperature generated by the burner and the
residence time at that temperature. Therefore, a decrease in the peak flame
temperature will decrease the thermal NOX emissions. Principle reactions
for the oxidation of molecular N2 are as follows:
N2 + 0 -> NO + N (1)
N + 02 -*• NO + 0 (2)
N + OH + NO + H (3)
Unlike thermal NOX, temperature has little effect on fuel NOX. Fuel
NOX is dependent on the availability of ©2 in the flame. There are two
competing reactions for the nitrogen containing compounds in the volatile
flame region:
I + R -» NO + ... (4)
I + NO + N2 + ... (5)
Where I is a nitrogen containing intermediate and R is a hydrocarbon.
Under fuel-rich conditions (high concentration of fuel in air) reaction (5) is
dominant, resulting in lower NOX concentrations; however, under fuel-lean
conditions (low concentration of fuel in air) reaction (4) is dominant,
resulting in higher NOX concentrations.
NOX formation is dependent on the combustion method, which varies with
boiler design. The typical baseline NOX emissions from different boiler
types are shown in Figure 1.
STATUS OF TECHNOLOGY
Combustion modifications and F6T are the two basic types of NOX control
technology being developed for use with coal-fired boilers.
Combustion Modification
Combustion modification includes methods to inhibit the thermal and fuel
NOX formation. One combustion modification technique for reducing NOX
emissions is to increase the burner zone surface area. This was not developed
as a control technique but as a method for reducing slagging in boilers. By
increasing the burner zone surface area, the burner zone heat release rate
-------
1600
1400
1200
e 1000
o-
800
01
CHJ
O
600
400
200
I I
Wet bottom
•'/"/•'Cyclone:'; :."/'.
•VYx-'V.*.:' •.'.'•*•"'•;:'•''•'•
Wall
0 200
Dry botto
Horizontally opposed
I I
I
400 600
MW
800 1000
Figure 1. Baseline NOX emissions - coal-fired utility
boilers (23).
-------
(defined as the heat input divided by the burner zone surface area) was
decreased. NOX emissions decreased as a result because gas temperatures
were lowered by the increased heat removal capacity. The larger boiler also
proved valuable as lower turbulence burners, such as the low-NOx burner,
were developed because the increased size allows longer residence times to
complete carbon burnout. Commercial acceptance for reduced burner zone heat
release rates came about in the 1970's. Figure 2 shows burner zone heat
release rates for pulverized-coal-fired boilers ordered before and after
1970.
w
W M-l
H 4-i
w cs
o 3
N O
w
g
800
600
500
400
1966
1970
ORDER YEAR
1974
Figure 2. Pulverized-coal-fired boiler burner zone heat release
rates (11).
Another combustion modification method is flue gas recirculation. Flue
gas is extracted at the economizer outlet and returned to the furnace. The
cooled flue gas lowers the overall temperature of the gas inside the furnace
and reduces the oxygen concentration. Flue gas recirculation works well for
oil and gas but has not been proven effective for coal.
Overfire air, unlike the above two techniques, which primarily lower
burner zone temperatures, is designed to limit oxygen availability at the
flame. Air ports are installed above the burner zone to inject a portion of
the combustion air. The burners are thus fired more fuel rich than normal.
Fuel-rich conditions decrease fuel NOX formation. (This method maintains
-------
the burner zone in a reducing atmosphere with an oxidizing atmosphere above
the flame region. This reducing atmosphere is a more corrosive environment
for the boiler walls.)
A technique similar to overfire air is burners out of service. Instead
of placing air ports in the boiler wall, individual burners or rows of burners
admit only air. The remaining burners make up the difference by firing
additional coal if pulverizer capacity is available. As in overfire air, the
active burners operate at more fuel-rich conditions than normal. This
technique was developed primarily for retrofit applications.
Another technique to minimize oxygen availability is low excess air. In
this technique, sufficient air is admitted to the boiler to complete
combustion but is minimized to reduce NOX formation.
Although NOX control could be obtained by these boiler modifications,
work was continued by boiler manufacturers and others to obtain lower levels
of NOX emission and to avoid operational problems such as corrosion,
erosion, combustion instability, and energy penalties from some of the
methods.
Babcock & Wilcox, Inc., (B&W) developed their dual register burner to
meet 1971 NSPS (0.7 Ib NC^/MBtu); however, test on pulverized coal units
demonstrated that the 0.6 Ib N02/MBtu level could also be obtained (3). The
dual register burner features limited turbulence, conical diffuser mixing, and
secondary air introduced around the primary air nozzle in two concentric air
zones that are independently controlled. Air flow to the burners is
controlled by use of a compartmented windbox.
A series of burners has been developed by the Foster Wheeler Energy
Corporation for NOX control. The latest is the controlled-flow/split-flame
burner (30, 31). This burner uses dual registers on the secondary air to
produce a substoichiometric air zone near the flame. A modified annular
primary air nozzle is also used to separate the coal-air mixture into four
concentrated streams that form four independent flames. Overfire air can be
used with the controlled-flow/split-flame burner for additional NOX control
but it is not recommended.
NOX control is being accomplished by the Riley Stoker Corporation using
a controlled diffusion flame and the Turbo® furnace (21) . The Turbo furnace
is designed with a venturi shaped bottom to utilize the lower furnace cooling
surface more effectively. Combined with the Turbo furnace, downward-tilted,
nonswirl burners are used to delay mixing of the secondary air with the fuel
and primary air. In this way the fuel and secondary air mix by diffusion,
which decreases the combustion rate and flame temperature.
Combustion Engineering, Inc., produces tangentially fired boilers, in
which the burners are directed tangent to an imaginary circle at the center of
the furnace. This provides a large amount of internal recirculation of the
combustion gases and slower mixing of fuel and air (16). These boilers are
inherently low in NOX formation. Additional reduction in emission levels
can be achieved using overfire air.
-------
These combustion modification techniques are described in greater detail
in other literature (14, 15, and 23).
Flue Gas Treatment
The FGT methods of NOX control applicable to coal-fired utility boilers
are well described in literature (1, 9, and 13). These postcombustion
processes can be divided into dry or wet types. The dry processes can be
further categorized into four subdivisions: catalytic reduction, noncatalytic
reduction, adsorption, and irradiation. The majority of the dry processes are
of the reduction type. These catalytic and noncatalytic reduction processes
can also be classified as selective or nonselective processes based on the
type of reducing agent used. The majority are selective and usually use NHj
as the reducing agent. If the NH3 is injected after the boiler economizer,
where temperature of the flue gas is about 700°F to 800°F, a catalyst is
necessary. These processes are described as SCR processes. If NH3 is
injected into the secondary superheater region of the boiler, where
temperature of the flue gas is 1,000°F to 1,800°F, a catalyst is not
necessary. These processes are described as SNR processes.
Selective Catalytic Reduction —
The SCR method is the most advanced FGT method, and the one on which the
overwhelming majority of NOX FGT processes are based. As with the majority
of all types of NOX FGT, most of the SCR processes were developed in
Japan. Since the presence of some oxygen improves the NOX reduction, the
reactions of NHj with NOX may best be expressed as follows:
4NH3 + 4NO + 02 + 4N2 + 6H20 (1)
4NH3 + 2N02 + 02 •*• 3N2 + 6H2<> (2)
In the presence of a catalyst and with the optimum reaction temperature,
usually 570°F to 840°F, an NH3:NO molar ratio of 1:1 typically reduces
NOX emissions by 90% with residual NH3 concentrations of 10 to 20 ppm or
higher. It should be noted that the Japanese seem to prefer 80% N0£ removal
in which N^NO molar ratios range from 0.81:1 to 0.9:1 with residual NH3
concentration usually less than 5 ppm. (This reduces capital and operating
costs as well as effects on downstream equipment from ammonium salt
deposition. )
The only equipment necessary is an NHg storage and injection system, a
reactor, and catalyst. Additional fan capacity is necessary because of the
pressure drop increase across the reactor of 2 to 5 in. 1^0 (4 to 9 mm Hg) .
The components and precise composition of most catalysts are proprietary.
However, catalysts composed mostly of titanium and vanadium oxides are
generally used, since these components are resistant to attack by S02 and
803. S(>2 oxidation to 863 can be a problem, especially with coal-fired
boilers where flue gas SC>2 concentrations are relatively high. Proprietary
additives to the catalyst can reduce the amount of S(>2 oxidation to less
than 0.5% to 1.5%.
-------
In addition to the different catalyst compositions, different reactor and
catalyst configurations have also been developed, primarily to handle .various
particulate loadings. With natural-gas-fired boilers, fixed packed-bed
reactors with spherical-, cylindrical-, or ring-shaped catalyst pellets are
used. Parallel flow reactors and catalysts are preferred for oil- and coal-
fired boilers to tolerate the particulate matter in the flue gas stream.
Parallel flow designs include tubular, honeycomb, corrugated, or parallel-
passage configurations. The catalyst may be an active material coated on a
metallic or ceramic carrier or may be a homogeneous material.
The major concerns of operating SCR processes include: plugging and
erosion of the catalyst by fly ash; emission of NH3 or ammonium salts;
increased 863 emissions from oxidation of S02; effects on operation of
downstream equipment such as the air heater, electrostatic precipitator (ESP),
flue gas desnlfurization (FGD) process, and baghouse; environmentally sound
disposal or reclamation of catalyst; lack of proven NH3 analytical control
systems; sensitivity of the process to temperature changes due to boiler load
variations; and reliability of the process and its effects on the boiler
system availability.
In spite of the potential problems, there are over 60 full-scale SCR
units successfully operating on gas- or oil-fired boilers in Japan. Over 10%
of these units are larger than 330 MW. Two commercial SCR units began
operating in 1980 on coal-fired boilers in Japan. The 175-MW retrofit unit
supplied by Mitsubishi Heavy Industries at the Shimonoseki Station of Chugoku
Electric Power Company was the world's first full-scale coal-fired SCR
system. It began operation in April 1980 and is operating at its designed 51%
NOX reduction efficiency with less than 1 ppm NH3 slip. The other SCR
unit is 90-MW capacity on a new 350-MW boiler at Tomakomai Station of Hakkaido
Electric Power Company. It was supplied by Babcock-Hitachi and began
operation in October 1980. Two other SCR units on coal-fired boilers are
under construction with planned startup by mid-1981.
In the United States, EPA and the Electric Power Research Institute
(EPRI) are evaluating SCR technology on coal-fired pilot-scale units. EPA
sponsored two 0.5-MW-size tests, each of which was recently concluded. The
UOP Shell Flue Gas Treatment process for simultaneous NOX and SOX control
was evaluated at Tampa Electric Company's Big Bend Station. The Hitachi Zosen
SCR process was tested at the Plant Mitchell Station of Georgia Power
Company. EPRI is currently operating a 2.5-MW pilot plant at the Arapahoe
Station of Public Service Company of Colorado using the Kawasaki Heavy
Industries, Ltd., process.
The first large-scale SCR demonstration unit in the United States is
being erected by Southern California Edision Company at the Huntington Beach
Station. It is a 107.5-MW capacity on a gas- and oil-fired unit.
Selective Noncatalytic Reduction—
Exxon Research and Engineering Corporation developed the SNR process in
which NH3 is injected into the boiler where proper flue gas temperatures
allow the reduction of NOX by reaction with NH3 to proceed without a
catalyst. Generally, 40% to 60% NOX reduction is achieved with
-------
molar ratios of 1 to 2:1. SNR may be more attractive than SCR in cases where
only 40% to 60% NOX control is needed since SNR is simple and does not
require expensive catalysts.
The general disadvantage of SNR is the limited NOX control achievable,
especially with large boilers, due to the difficulty of achieving rapid
uniform mixing and the variations of flue gas temperature and composition
usually present within the boiler region where the SNR is operated. NH3
consumption and unreacted NHg levels can also be high.
There are several large SNR units installed in Japan, between 30- and 100-
MW capacity, mostly supplied by Tonen Technology (a subsidiary of Toa Nenryo)
which has a license from Exxon. These are operated on gas- and oil-fired
boilers or furnaces. Practically all are only for emergency use during a
photochemical smog alert or when total plant emissions exceed the regulation.
There are presently two commercial SNR plants operating in the United
States. One is on a glass melting furnace and the other a petroleum refinery,
both located in California. The construction of five other industrial-scale
units is planned. The SNR process is also being installed by Exxon at the
No. 4 oil-fired unit of the Haynes Station of the Los Angeles Department of
Water and Power.
Other Flue Gas Treatment Techniques—
In addition to SCR and SNR, dry processes which are being developed for
simultaneous SOX and NOX control include:
1. Activated carbon processes where NH3 reduces NOX to N£.
2. Copper oxide processes where NH3 reduces NOX to %.
3. Electron beam irradiation processes in which NILj is added to produce
ammonium sulfate and nitrate.
The optimum temperature range for simultaneous SOX and NOX control
with activated carbon processes is 430°F to 445°F. Although NOX may be
adsorbed below 212°F, for treating large quantities of flue gas above
212°F the carbon is mainly useful as an NOX reduction catalyst.
Therefore, while NOX is converted to N2 by reaction with NH3 in the
presence of the activated carbon catalyst, S02 is simultaneously adsorbed by
the carVon to form I^SO^ The 1^804 may also compete for NH3 in
forming ammonium sulfate or bisulfate. The formation of these ammonium salts
increases NH3 consumption and also lowers catalyst activity. The carbon
must be regenerated, either by washing or thermal regeneration. Washing
produces a dilute solution, which requires much energy to concentrate for use
as a fertilizer. Thermal regeneration seems to be preferred. A concentrated
S02 gas is recovered, which can be used for sulfuric acid or elemental
sulfur production.
The major drawback of the activated carbon processes is the enormous
consumption of activated carbon, which is more expensive than ordinary carbon
10
-------
used only for SOX removal. Since carbon and ammonia consumption increases
with the S<>2 content of the flue gas, the process is best suited for flue
gases relatively low in S(>2. In Japan, Sumitomo Heavy Industries and
Unitika Company have operated activated carbon pilot plants of 0.6-MW and 1.5-
MW capacity respectively.
The UOP Shell Flue Gas Treatment process may simultaneously remove SOX
and NOX. SOX reacts with the copper oxide acceptor to form copper
sulfate. The copper sulfate and copper oxide are SCR catalysts for the NOX
reduction by NH3. Regeneration of the multiple catalyst beds by a reducing
gas, such as fL^' yields a S02-rich stream that can be used to produce
liquid S(>2» elemental sulfur, or sulfuric acid. By eliminating NHg
injection, the process is strictly an FGD process, whereas, eliminating
regeneration of the catalyst beds allows the process to be used for only NOX
control. The major disadvantages are the large consumption of fuel for making
hydrogen and the catalyst expense.
In addition to the EPA-sponsored pilot plant mentioned earlier, the
process has been installed in Japan on a 40-MW oil-fired boiler. The unit has
demonstrated 90% SOX removal and 70% NOX reduction.
Another process for simultaneous SOX and NOX control is the electron
beam process developed by Ebara Manufacturing Company in Japan. NH3 is
added to the flue gas, after which the gas stream is irradiated with an
electron beam in a reactor, promoting the conversion of SOX, NOX, and
N% to ammonium sulfate and ammonium nitrate. The ammonium sulfate and
ammonium nitrate may be collected downstream in an ESP or baghouse and
potentially sold as a fertilizer. The most economically practical removal
efficiency range appears to be 80% to 90% for each of NOX and SOX, though
higher removals can be achieved with much greater electron beam energy input.
The optimum temperature range is 160°F to 195°F.
Ebara has worked on the process since 1971. It has been tested at a
0.3-MW and 3-MW scale in Japan. Avco Corporation in the United States has
also examined this technique and has a cross-licensing agreement with Ebara in
sharing of technology and in marketing of the process. Although the process
appears attractive because of simplicity, simultaneous SOX and NOX
control, and byproduct formation, there are still many questions concerning
costs, byproduct quality, etc., which must be determined.
The wet processes normally involve simultaneous removal of SOX and
NOX. The major problem associated with wet NOX control processes is the
absorption of NOX by the scrubbing solution in which it can be concentrated
and converted into other forms. NOX in the flue gas is predominantly NO,
which is much less soluble than N02, whereas, N02 is even less soluble
than S02. The two common methods of removing the NOX in flue gas by wet
processes are: (1) direct absorption of the NOX in the absorbing solution
or (2) gas-phase oxidation to convert the relatively insoluble NO to NO^
followed by absorption of N02. Presently, development of the wet NOX FGT
processes has practically ceased because of the complexity and unfavorable
economics of these processes in comparison with the dry processes.
11
-------
STATUS OF PROCESSES EVALUATED
ADVANCED LOW-NOX BURNER
EPA instituted a program to develop a coal-firing burner capable of
reducing NOX formation to levels lower than achieved by present combustion
modifications. This program is directed toward developing burner prototypes
for industrial and utility appplication. The burner design at various stages
of development is shown in Figure 3. Only the utility ALNB prototype
development program is discussed in detail in this report.
Process Description
The ALNB is being designed to prevent the formation of both types of
NOX (thermal and fuel) while maintaining boiler efficiency and meeting
boiler demand. The primary objectives are: (1) to provide an initial fuel-
rich, i.e., oxygen deficient, zone which maximizes the conversion of organic
nitrogen compounds to N2> and (2) create an overall oxidizing atmosphere
around the fuel-rich zone to maximize burnout and to minimize the potential
for corrosion in the lower furnace section of the boiler. These objectives
are obtained by providing for the optimum interaction between the primary fuel
jet and the swirl-stabilized recirculation zone, together with delayed air
addition from the outboard staged air injectors (33).
For this study the ALNB concept was integrated into a B&W dual register
burner design with four tertiary air ports located a distance of about one
throat diameter around the central burner (3) . Air admitted through the
burner is made up of three air streams referred to as primary, inner, and
outer secondary air. Primary air is used as the medium to carry the entrained
pulverized coal. Secondary air is injected around the primary air through an
annnlus. Swirl vanes in the inner secondary air annulus impart swirl to
control mixing of the primary and inner secondary streams and to control flame
shape. The remaining secondary air is admitted through the outer secondary
air annulus, which is concentric to the inner secondary air annulus. Two
possible configurations for the tertiary air ports are shown in Figure 4. The
first has four ports per burner, the second shares ports between burners. The
ports are arranged to provide specific mixing rates with the other air streams
and avoid fuel-rich zones along the lower furnace wall. The central burner is
designed to operate at approximately 70% stoichiometry to minimize oxygen
availability for conversion of fuel nitrogen to NOX and the remaining 45% to
50% of the air is supplied through the tertiary ports to complete combustion.
12
-------
INITIAL BURNER DEVELOPMENT
FURNACE WALL
TERTIARY AIR -
- SWIRL BLOCKS
SECONDARY1. AIR
COAL & PRIMARY AIR
N,
COAL INJECTOR
IGNITER
INDUSTRIAL PROTOTYPE
(FOSTER WHEELER)
COAL INLET
DISTRIBUTED MIXING BURNER
SECONDARY AIR TERTIARY AIR
SWIRL VANES
COAL A
PRIMARY AIR
SECONDARY AIR TERTIARY AIR
FIELD DEMONSTRATION BURNERS
DISTRIBUTED MIXING CONCEPT
TERTIARY AIR
UTILITY PROTOTYPE
(BABCOCK & WILCOX)
-COAL
PRIMARY AIR
Figure 3. Major stages of the ALNB development (11)
-------
Tertiary Ports \^
^o O
o o o o
o O o
O O O O ^ Burners
o o o o o O
4 Tertiary Ports Per Burner Shared Tertiary Ports
Figure 4. Alternatives for injection of ALNB tertiary air.
At the time of the writing of this report, not all areas of the ALNB
design had been defined. However, the general description given above is
expected to be the basis for prototype construction. The research goal for
the NOX emission level from the ALNB is 0.2 Ib N02/MBtu.
Technical Considerations
In a study funded by EPA (18, 33) the effects of various burner and
operating variables on NOX emissions were explored by EERC. Both single and
multiple burner tests were performed and the results quantified to aid in
burner development.
Results of the test on one burner evaluated in the single burner test
program are shown in Figures 5, 6, and 7. The optimum operating conditions
for this burner are a burner zone stoichiometric ratio (SRg) of 0.5 to 0.7,
a primary swirl vane angle (Sp) of 45 degrees, and a secondary swirl vane
angle (Sg) of 60 degrees. These data were obtained at a primary
stoichiometric ratio (SRp) of 0.23 and 0.25 and a theoretical stoichiometric
ratio (SRj) of 1.25. These results are unique to a particular design
because optimum operating conditions will vary as burner design varies.
Mutiple burner test results summarized in Figure 8 show that lower NOZ
emissions are directly related to burner zone stoichiometry. The burner zone
stoichiometry used must be weighed against the CO level (combustion
14
-------
a
P.
ft
04
O
400
300
200
100
0
Utah coal
48 x 106 Btu/hr
SRp = 0.23
SS = 60
SRB =
SRB = 0.63
100 110 120 130 140
OVERALL STOICHIOMETRY (% Theoretical Air)
800
600
ft
ft
*v
s
n
400
CM
Q
O
-------
400
300
200
O
<2J
O
100
0
Utah coal
48 x 106 Btu/hr
SRP = 0.23
SRT =1.25
SS = 60
D Sp
o Sp
= 60°
= 45°
= 30°
40 50 60 70 80
BURNER ZONE STOICHIOMETRY (% Theoretical Air)
CN
O
O
<2J
O
800
600
400
200
0
40 50 60 70 80
BURNER ZONE STOICHIOMETRY (% Theoretical Air)
Figure 6. Effects of ALNB primary swirl, single burner (11)
16
-------
a
P.
P.
CN
O
P.
P.
erf
p
IN
O
o
CJ
400
300
200
100
Utah Coal
48 x 106 Btu/hr
SRp = 0.25
SRT =1.25
SP = 450
40
60 80 100 120
4 BURNER ZONE STOICHIOMETRY (% Theoretical Air)
800
600
400
200
40 60 80 100 120
BURNER ZONE STOICHIOMETRY (% Theoretical Air)
Figure 7. Effects of ALNB secondary swirl, single burner (11)
17
-------
I
ex
o
-------
efficiency). Also studied in the multiple burner tests was the effect of
removing rows or columns of tertiary air ports from service, as shown in
Figure 9. Results showed that NO emission decreased slightly but carbon
burnout (CO level) was more sensitive. Additional optimization is required.
A study by EERC on the effects of coal type^ was initiated to develop a
data bank on a full range of U.S. coals. Results from this study, of which
Figure 10 is an example, led to the conclusions that NOX emissions are
sensitive to fuel type, that the nitrogen content of the fuel does not
correlate with NOX emissions, and that burner efficiency is sensitive to the
type of coal.
Development Status
The first phase of low-NOx burner development by EPA was initiated in
1970 when EPA contracted with the International Flame Research Foundation
(IFRF) to study the influence of burner variables on NOX emissions using
pulverized coal. In the IFRF study the following burner parameters were
investigated:
• Method of fuel injection
• Throat velocity
• Geometry of the burner exit
• Position of the fuel injector
• Type of burner exit
• Proportion of primary air
• Swirl intensity of the combustion air
Results of the study show the two variables having the greatest effect on
NOX emissions to be the method of fuel injection and the degree of swirl.
Test work was performed on a 2 to 3 thermal MW (6-9 x 106 Btu/hr) refractory
tunnel.
To obtain burner data at a more realistic boiler size and firing rate,
scaleup was performed by EERC on a boiler and a large water tube simulator
capable of firing up to 40 thermal MW with single or multiple burners. A
smaller system simulating a D-type package water tube boiler was used to
evaluate various burner designs.
While the test program of IFRF was designed to determine the effect of
various burner parameters on NOZ emissions, the EERC program was designed to
quantify these effects for development of an experimental burner and boiler
system. The EERC program thus had a larger scope than the IFRF program,
covering not only single burner variables at greater firing rates, but also
19
-------
COLUMN
to
o
o
n
©
0
0
NORMAL LEVELS WITH ALL TERTIARY PORTS IN SERVICE
NO CO
o
O
H
13
O
1
PM
s
W
H
COLUMN 2
AND ROW 2
COLUMN 2
COLUMN 3
ROW 2
ROW 3
BURNER AND TERTIARY AIR PORT TEST ARRANGEMENT
200 400
NO, CO @ 0% 02, DRY, ppm
600
Figure 9. Effects of ALNB tertiary ports out of service, four burners fired at 12.5 x 10 Btu/hr each (11).
-------
400
300
200
CN
O
o
53
100
0
64 x 106 Btu/hr
SRp = 0.23
SPT = 1.25
W. Va
(N = 1.55% dry, ash free)
Utah (N = 1.71% dry, ash free)
Utah
(Baseline -
N = 1.65% dry, ash free)
I
I
40 60 80 100 120
BURNER ZONE STOICHIOMETRY (% Theoretical Air)
Figure 10. Effects of ALNB coal type, four burners fired at
12.5 x 106 Btu/hr each (11).
21
-------
the interactions of multiple burners and the effects of coal types. Figure 11
shows one of the burners used in the test program, called the simple double
concentric burner, and the variables evaluated. Tertiary air was used with
the simple double concentric burner and other burners used in the test
program.
While the EERC test program was being conducted, another possible use for
the low-NOx burner was being studied at a pilot-plant scale by EPA
(17) . This study was exploring sulfur capture using a dry sorbent. It is
hypothesized that after coal and sorbent are mixed in the pulverizer to
achieve intimate contact, the ALNB burner will create conditions (fuel-rich
burner zone and decreased peak flame temperature) that will be favorable for
SOX removal. Dry sorbents that have been tested are limestone, Na2CO£,
and NaHC(>3. Depending on pilot-plant results, funding will be sought for
field evaluation of dry sorbents. The stated goals are to obtain 50% SC>2
capture at a 2:1 sorbent:S02 stoichiometry and a research goal of 90% S(>2
removal at 3:1 sorbent:S(>2 stoichiometry.
Upon completion of the ALNB design, EPA contracted with B&W to perform a
field evaluation of the burner in a utility boiler. EERC has received the
subcontract. The industrial boiler contract which parallels the utility
program was awarded to EERC with the subcontract to Foster Wheeler. The nine
major task areas that make up the utility boiler field evaluation program are
(1) program definition, (2) prototype construction and performance evaluation,
(3) host-site boiler baseline evaluation, (4) burner installation, (5)
performance optimization, (6) industry coordination, (7) boiler restoration,
(8) data analysis, and (9) summary of program results (4, 18).
The field evaluation program is behind schedule because of delays in
obtaining an acceptable host site. Some boilers were omitted as possible host
sites because more spacing between burners was required for the addition of
tertiary ports than was available while others required major changes to the
windbox and boiler structural supports. Further delays resulted from an
unwillingness to participate by owners of possible host sites (4).
This program should provide information on areas of uncertainty such as
the effect of commercial application of the ALNB on boiler tube bending and
configuration, boiler wall structural requirements, and boiler efficiency.
EXXON PROCESS
Process Description
The Exxon process, developed by Exxon Research and Engineering
Company, controls NOX from flue gas by injection of NH3 and air through a
distribution grid, or grids, directly into the cavity of the secondary
superheater (29).
At high temperatures, NOX reacts with NH3 in the presence of oxygen
by the following overall reaction (28):
22
-------
to
u>
INPUT
• TEMPERATURE
• COMPOSITION
• VELOCITY
• LOAD
• COAL TYPE
FURNACE
• STAGED AIR
• SCALE
• BURNER ARRANGEMENT
SWIRL
• EXTENT
• GENERATION METHOD
THROAT
• STOICHIOMETRY
• MIXING
• VELOCITY
Figure 11. Variables evaluated in the single ALNB test program (11).
-------
NO + NH3 + 1/402 -> N2 + 3/2H20 (1)
R. K. Lyon of Exxon Research (28) has proposed the following mecha ism for
this reaction:
NH2 + NO -»• N2 + H + OH (2)
NH2 + NO -*• N2 + H20 (3)
H + 02 -> OH + 0 (4)
0 + NH3 -*• OH + NH2 (5)
OH + NH3 -> H20 + NH2 (6)
H + NH3 -> H2 + NH2 (7)
This reaction proceeds at a satisfactory rate in a narrow temperature range
around 1,740°F (950°C) as shown in Figures 12 and 13. Below 1,650°F
(900°C), the reaction rate is too slow for adequate NOX control, causing
NH3 and NO to flow through unreacted. Above 1,830°F (1,000°C), the
NH3 reacts with 02 to produce NOX, as illustrated by the reaction:
NH3 + 5/402 -> NO + 3/2H20 (8)
Because of this sensitivity, temperature gradients within the boiler reaction
zone caused by variable heat transfer and gas velocity or changes in the
boiler load have considerable effect on the process efficiency.
Residence time is as important as reaction temperature. The reactants
(NO, NH3, 02) must remain in the boiler injection zone for a sufficient
length of time for the reaction to go to the desired completion. A typical
residence time is 0.1 second.
Proper mixing of the reactants with the flue gas is crucial. Uneven
distribution can cause pockets of ammonia as well as NOZ to pass through the
boiler unreacted. The Exxon process uses proprietary Exxon gas-phase mixing
technology to disperse the small volume of reactants throughout the flue gas.
Ammonia addition is dependent on the NOX concentration in the flue
gas. Tests at optimum temperature conditions indicate that a rate of 0.6 to 2
moles of NH3 per mole of NOX will accomplish 50% NOX removal or above,
but the rate should be lowered to 1.5 or below when possible to minimize NH3
emissions (see Figures 14 and 15).
The NH3 injection grid is insulated and covers the entire cross-
sectional area of the flue gas flow path. Each grid is constructed of
separate injection zones with each zone having its own NH3 controls to deal
with temperature fluctuations in a plane normal to the flue gas flow. If
multiple grids are used, air is passed through idle grids to prevent plugging
24
-------
1.0
0.8
i
« 0.6
•H
4-1
•H
C
0.4
0.2
0
4% excess 02
300 ppm NO
1
I
I I
(NH3)/(NO), molar
0.3
0.5
1.6
I
I
1200 1400 1600 1800 2000 2200
TEMPERATURE, °F
Figure 12. Effect of temperature on NO reduction for various
levels of NH^ injection with the Exxon process (19)
25
-------
K3
S-i
td
^
i
•H
fi
n)
a
•H
1.0
0.8
0.6
0.4
0.2
0
Initial Initial
Natural gas
Utah coal
Navaho coal
Pittsburgh coal
Illinois coal
I
I
I
I
1500 1600 1700
1800
AVERAGE RADIAL TEMPERATURE, UF
1.0
o
0.6
•H
c
2 0.4
c
•H
O
IS
0.2
T . . , /NOT . . n
Initial Initial
5% Excess 02
I I
= 1.0
Natural gas
Utah coal
Navaho coal
Pittsburgh coal
Illinois coal
I
I
1900 2000
0
1500 1600 1700
I
I
1800 1900
AVERAGE RADIAL TEMPERATURE, °F
2000
Figure 13. Effect of temperature on NO reduction for various fuel sources with the Exxon process (19).
-------
1.0
0.8
J-l
n)
iH
O
e
0.6
n)
•H
0.4
0.2
O Natural gas
O Utah coal
D Navaho coal
& Illinois coal
O Pittsburgh coal
I
0.5
1.0
NH3
/NO
Initial Initial
1.5
, molar
2.0
Figure 14. Comparison of NO reductions at the optimum temperature condition with the Exxon process (19)
-------
0.4
i
n)
•H
j-j
•H
c
o
la
C
•H
on
0.3
0.2
0.1
O Utah
a Navaho
6 Illinois
O Pittsburgh
0 Natural gas
I
1.0
NH3
/NO
Initial Initial
2.0
, molar
3.0
Figure 15. Comparison of the NH3 emissions for all fuels tested at the
peak NO reduction temperature with the Exxon process (19).
28
-------
of grid holes (no nozzles are used) and to protect the grid from high flue gas
temperatures. To prevent loss of carrier air, multiple air compressors are
used; however, emergency steam can be connected for use if total air loss
occurs.
Since the main process control parameter for the Exxon process is boiler
load, initial NH3 injection rates are established based on calculated or
measured flue gas temperatures for a given boiler load or load range including
temperature variations across the plane of the injection grid. This is then
incorporated into the control system using a microprocessor for responding to
load changes. Temperature changes which are not a function of boiler load are
also incorporated into the control system allowing injection rates to be
optimized, NOX reduction maximized, and NHg breakthrough minimized.
Carrier air rate is held constant; therefore, it is not a control variable.
To check analyzer response to injection rate, optimization control response
logic is built into the microprocessor. Operator intervention is required if
large magnitude changes for the system are indicated without significant
change in boiler operating parameters.
Technical Considerations
Since flue gas temperature is the most important operating parameter of
the Exxon process, anything that causes the temperature to fluctuate can
affect the performance of the process. Slagging, changes in boiler load,
changes in excess Q^• and other operational variations can cause the
temperature profile to shift in relation to the firebox. Variations in flue
gas temperature are also present normal to the flue gas flow. Boiler tubes at
the walls absorb heat to produce steam; therefore, temperatures at the wall
are lower than at the center. In large utility boilers the temperature
variation can exceed 270°F.
Two techniques are used by Exxon to deal effectively with the temperature
variations. One method uses multiple NE^ injection grids with each grid
positioned in the boiler to correspond to a different load or combination of
loads. In this way, as the temperature profile of the boiler shifts due to a
load change, the required reaction temperature range, 1,650°F (900°C) to
1,830°F (1,000°C), is accessible by feeding the appropriate grid. The
second method uses one grid, instead of using multiple grids, and the reaction
temperature is manipulated. This is accomplished by injecting H2 along with
the NH3. Based on the H2:NH3 injection ratio selected, NOX reduction
will proceed at a satisfactory rate at a temperature range of 1,290°F
(700°C) to 1,830°F (1,000°C). Therefore, one grid can be used to
handle load changes because the reaction temperature can be adjusted to the
boiler conditions by controlling the H2 flow rate.
Exxon has developed a performance prediction procedure to optimize grid
location in the boiler and to estimate NOX removal. Variables used in the
calculations are: flue gas temperature and flow distribution, including
severity of cross-sectional variations, flow path geometry, available reaction
time, and suitability of the dimensions of the zone for grid placement, based
on manufacturer's data. The procedure will give an estimated NOZ removal
percentage at a particular location in the flue gas path.
29
-------
To optimize grid placement, the performance prediction procedure is used
at a number of locations within the boiler to generate a graph of location
versus NOX removal. This is done for 50%, 75%, and 100% boiler load
conditions. A combination-load grid, which will serve two of the boiler
loads, is situated at the maximum of the intersection points of the load
curves. The single-load grid is situated at the maximum of the remaining load
curve.
For an actual installation, measurements of the temperature and flow
patterns within the predicted zone would be used to confirm or adjust the
final grid location. The most frequently used load is emphasized in designing
grid placement, so that NOX emissions can be minimized at that load.
Potential problems with the Exxon process generally relate to NH3
emissions. Emission levels from a pilot-scale coal-burning test system
averaged 11 to 34 ppm NH3 at the boiler exit. The assessment of the Exxon
process using the performance prediction procedure for eight major utility
boiler types gave predicted NHg levels at the boiler outlet of 21 to 43 ppm
for an NE^NO mole injection rate of 1.0, and 64 to 129 ppm for an NI^iNO
mole injection rate of 1.5, although these might vary under actual operating
conditions. Operating experience on full-scale oil- and gas-fired units shows
NH3 levels of 10 to 40 ppm.
NHj emissions when combined with 863 can cause problems with
(NH4)2S04 and NH4HS04 formation. This is particularly true in coal-
fired applications, where sulfur content is relatively high. Studies indicate
substantial amounts of (NH4>2S04 and NH4HS04 are deposited in the
air heater. Although most of the deposits may be removed by soot blowing,
modifications in the air heater design may be required. Operating experience
in Japan with low-sulfur oil-fired boilers has shown that water washing of the
air heaters is necessary two or three times a year.
The presence of NH3 in the flue gas has mixed effects on ESP
performance. It has been used as a flue gas additive to neutralize condensed
H2S04 and reduce reentrainment losses by increasing fly ash cohesiveness
(5). However, NH3 can also cause excessive sparking between the electrodes,
especially with low-sulfur, high-resistivity, western coal fly ash. Reactions
of NH3 with S03 will deplete the 803 content of the flue gas which
lowers collection efficiency. The net effect of these factors on ESP
performance is unclear. It is also unclear at this time whether the NH3,
(NH4>2S04, or NH4HS04 will present additional operating requirements
for 802 scrubber systems, such as waste water treatment.
Development Status
The reaction mechanism of the Exxon process was developed by Exxon using
simulated flue gases in a bench-scale reactor. When the reaction mechanism
was established, evaluation of operating parameters such as reaction
temperature and NH3 injection rate was performed on a 0.3-MWt (10°
MBtu/hr) oil- and gas-fired boiler. Corrosion and fouling effects were tested
on a 9-Mfft (30 MBtu/hr) oil- and gas-fired boiler.
30
-------
Commercial application of the Exxon process is primarily limited to gas-
and oil-fired units in Japan, as shown in Table 2. The Los Angeles Department
of Water and Power's Haynes Electrical Generating Station is scheduled to be
the first U.S. electric utility to install the Exxon process (20) with startup
projected for May 1981. The Haynes Unit 4 is a 230-MW boiler burning either
No. 6 fuel oil containing 0.25% sulfur or gas. Predicted performance for the
Exxon process is 51% NOX reduction at full load using one NH3 injection
grid containing six zones. Exxon is guaranteeing the process for 90% of the
predicted performance which is 46% NOX reduction. At 50% boiler load, no
NOZ reduction is guaranteed.
Although no commercial application of the Exxon process on a coal-fired
boiler has yet been made, studies have been made to assess the possibility of
such an application. A coal-fired pilot-plant study, sponsored by Exxon
Research and Engineering and EPRI and run by KVB, Incorporated, was completed
in early 1978. The project utilized a fire tube boiler, 0.9 Mfft (3
MBtu/hr), modified to fire pulverized coal with preheated air. Three
bituminous and one subbiluminous coals were fired. A 65% NOX reduction
could be obtained for all coals with a 1:1 NH3:NOX molar ratio feed
rate. Optimum operation, that is, maximum NOX reduction and minimum NHj
emissions, was 55% NOX reduction and 10 to 35 ppm NH3 emitted. Hydrogen
injection was tested with one coal and found to increase NOX reduction and
decrease NH^ emissions.
Under contract with EPA, eight different pulverized-coal-fired boiler
designs were assessed by Exxon to predict the applicability of the Exxon
process. This was done using the Exxon-developed performance prediction
procedure and boiler design data supplied by the manufacturer. Each boiler
design was evaluated at 100%, 75%, and 50% load. As a basis two injection
grids were used and there was no % addition. The best predicted results of
any design, coal type, and load at an NH3:NOZ molar ratio of 1.5:1 was 63%
NOX reduction and the worst was 45% NOX reduction (28). Using updated
technology, Exxon now states that the predicted NOX reductions would be
10-20 percentage points higher.
HITACHI ZOSEN PROCESS
Process Description
Hitachi Zosen has developed a process for the dry SCR of NOX with NH3
(32). U.S. licensee for the Hitachi Zosen process is the Chemico Air
Pollution Control Corporation. The process is based on a catalyst and a
reactor design developed by Hitachi Zosen that permits treatment of the flue
gas with a high particulate loading. Therefore, flue gas from a coal-fired
boiler may be fed directly to the reactor, upstream of the air heater, without
previous particulate removal. The reactor pressure drop is 2 to 3 inches
H2<>. An NH3~air, N^-steam, or NH3 - flue-gas mixture is injected
into the flue gas upstream of the reactor at an NH3:NOX molar ratio of 1:1
to achieve 90% NOX removal. Automatic control of the NH3 flow rate is
31
-------
10
t-0
TABLE 2. SUMMARY OF COMMERCIAL APPLICATIONS OF
EXXON THERMAL DENOX PROCESS
Source
Industrial boiler
degeneration boiler
Cogeneration boiler
CO boiler
Petroleum heater
Petroleum heater
Industrial boiler
Petroleum heater
Petroleum heater
Oilfield steamer
Utility boiler
Utility boiler
Utility boiler
Refuse incinerator
Utility boiler
*Re finery heaters (14),
boiler (1)
**Refinery heaters (4),
boilers (2)
**Utility boiler
Glass melting furnace
**Petroleum heater
**Petroleum heaters (2)
**Refuse incinerator
Location
Japan
Japan
Japan
Japan
Japan
Japan
Japan
Japan
Japan
California
Japan
Japan
Japan
Japan
Japan
California
California
Cal if ornia
California
California
California
Cal if ornia
Fuel
Oil/gas
Oil/gas
Oil/gas
CO gas/gas
Gas
Gas
Oil
Oil/gas
Oil/gas
Crude oil
Oil
Oil
Oil
Refuse/gas
Oil
Oil/gas
Oil/gas
Oil/gas
Gas
Gas
Gas
Refuse/gas
Heat release,
HBtu/hr
215
1,135
1,135
400
515
190
340
250
250
50
1,210
3,000
1,500
-
2,900
647 (total)
349 (total)
2,100
150
150
47 (total)
160
Approximate
initial NOX
emissions, vtmm
185
140
140
160
130
130
135
79
85
270
160
150
100
100-180
110-140
100-125
100-150
200
1,500-2,000
75
82
216
Reduction
efficiency, %
55
60
60
50
63
63
53
51
53
65
45
33
35
20-70
40
50-60
50
51
>50
50
>60
>60
*Two heaters in operation, remainder not yet in operation.
**Not yet in operation.
-------
based on the flue gas flow rates and the inlet NOX concentration and it is
optimized using the outlet NOX and NH3 concentrations. NH3 is diluted
by air to 5% or by steam or flue gas to 5% to 20%. This enhances mixing and
places the NH3-air mixture outside of the flammability limits of 15.5% to
27.0%.
In the reactor, NOX is reduced to N2 by a reaction with NH3 in the
presence of a catalyst at an optimum temperature of 734°F. The reactions
are listed by Hitachi Zosen as follows:
4NO + 4NH3 + 02 -> 4N2 + 6H20 (1)
6NO + 4NH3 -> 5N2 + 6H20 (2)
2N02 + 4NH3 + 02 -»• 3N2 + 6H20 (3)
6N02 + 8NH3 •> 7N2 + 12H20 (4)
4NH3 + 302 •»• 2N2 + 6H20 (5)
Reactions (1) and (3) predominate when the molar ratio of NH3:NOZ is
approximately 1:1, which is required for 90% NOX reduction; however,
reactions (2) and (4) gain dominance as the NH3:NOX molar ratio becomes
less than equimolar. Reaction (5) represents the breakdown of NH3 by 02.
This undesirable reaction becomes a problem at higher than optimum
temperatures. At the optimum reactor temperature or below, it is
insignificant. Hitachi Zosen reports that the level of excess NH3 in the
flue gas leaving the system is low «10 ppm). After leaving the reactor, the
treated flue gas flows to the air heater.
Technical Considerations
The temperature range required to achieve 90% NOX reduction is 600°F
to 750°F. As can be seen in Figure 16, additional reduction can be obtained
at temperatures higher than the acceptable range but NH3 decomposition
becomes more prevalent. Below the acceptable temperature range reduction
efficiency becomes unsatisfactory. Since flue gas temperature will fluctuate
with the boiler load, some method must be employed at low load to raise the
flue gas temperature to acceptable reactor conditions. Four techniques that
can be used are to operate an auxiliary furnace, to bypass hot flue gas around
the economizer, to reduce water flow to the economizer, or use a split
economizer design. It might also be possible to design for low-load
conditions adequate to maintain the required NOX reduction efficiencies. It
may be possible that low loads will not have an adverse effect on NOX
reduction efficiency since reduced loads will decrease the amount of flue gas
to be treated, resulting in an increased residence time in the reactor.
Therefore, the decreased temperatures may be offset by the increased residence
time.
To reduce the effects of high fly ash loadings from coal-fired flue gas
on the catalyst, Hitachi Zosen uses a corrugated configuration (see Figure 17)
and the flue gas passes parallel to the catalyst surface. Because of this
33
-------
NOX REDUCTION, %
00
i-i
CD
PC 2!
H- O
rt X
(U
O ^
rr o>
H. Cb
C
tsi n
O rt
CO H-
ro o
3 3
i-i ro
O >-!
O CO
ro c
CO CO
en
i-i
to n
^ rt
• H-
O
3
0)
rt
e
o
i-!
ho
O
-p-
o
oo
o
o
o
NO
O
00
n -c-
H -P-
H
i
S 00
<-a o
Ul
to
00
00
n
H
I—I
O
H
W
tfl
o
o
OJ
o
o
u>
K)
O
OJ
-P-
O
00
o
-P-
o
o
-p-
to
o
O
C
i-i
o < ro
x ro ••
i-1
^-s O
goo
o H- o
I—' rt fu
ro ^ M
i-i n HI
fU )-i.
rt ho i-i
H-
O
fD
cr
o
i-i
I
HI
rt
K3
-------
geometric surface area) i
unique configuration, aroa velocity (volumetric flue gas flow rate/catalyst
ather than space velocity (volumetric flue gas flow
rate/reactor volume) is tsed to determine the quantity of catalyst required.
bhat the area velocity recommended for 90% NOX
8 and 33 ft^/hr-ft^, which corresponds to a space
Hitachi Zosen reports
reduction is between 22.
velocity of 5,000 to 10,000 ft3/hr-ft3. The stoichiometric requirement is
given primary consideration when determining the NE^tNOg molar ratio
required to achieve a given NOX reduction, but other factors must also be
considered. The stoichiometric requirement determines the amount of NH3
that must be injected if NH3 usage is 100% efficient. However,
inefficiencies result from incomplete mixing of NOX and NH3 and
consumption of NH3 by reactants other than NOX. Because of these
inefficiencies in NH3 usage, excess NH3 must be injected. Here, care
must be taken since too much additional NH3 will result in increased levels
of unreacted NH3 passing through the reactor. If this occurs, the
breakthrough NH3 becomes a pollutant. Figure 18 shows representative curves
of NOX removal efficiency and exhaust NH3 as a function of the
molar ratio.
Figure 17. Configuration of Hitachi Zosen NOXNON 500 and 600 series
catalyst (32).
The oxidation of S(>2 can be a problem in NOX catalytic
reduction. The catalyst composition may cause <1% of the S(>2 in the flue
gas to be oxidized to 803, which will combine with NHg to form
and NH4HS03. These deposit as sticky solids on equipment
35
-------
PERCENT REDUCTION OF NO,
o
a
01
0)
,c
4J
V-i
o
o
•H
4-1
Cfl
r-l
O
d
O
o
T3
0)
(-1
X
O
ro
33
CO
^-J
n)
0)
0)
en
c
o
•H
4-1
n)
CD
CO
0)
O
o
t-t
ft
a
eu
en
o
-H
a
oo
iH
(1)
isnvnxa
36
-------
at low temperatures. The sulfate and bisulfate formation temperatures are
dependent on the NHg and SOg concentrations as can be seen in
Figure 19. Air heater plugging, catalyst masking, and equipment corrosion are
the primary problems related to the deposits. However, Hitachi Zosen states
that the deposits can be removed by soot blowing or water washing.
Pressure drop for the reactor and catalyst is reported to be 2 to 3
inches I^O by Hitachi Zosen and should not increase significantly over the
catalyst life. Figure 20 shows a typical pressure drop versus operating time
profile for a Hitachi Zosen catalyst.
Catalyst life is dependent on a number of factors including the
resistance of the surface to abrasion by particulate matter, the masking of
the catalyst by fine particulates, and chemical attack by compounds in the
flue gas. Figure 21 is a plot of the percent NOX reduction versus catalyst
age. As can be seen NOZ reduction decreases with age. The loss in NOZ
reduction efficiency can be a result of abrasion gradually removing possible
reaction sites. Additional reaction sites can be lost as a result of fine fly
ash particles plugging pores in the catalyst surface. Fly ash plugging can be
reduced by regular soot blowing. S0$ and other trace elements in flue gas
can chemically attack the catalyst causing loss in reactivity. Hitachi Zosen
claims their catalyst is resistant to attack of this nature, however.
o
H
H
U
X
o
92 _
90 -
34567
CATALYST AGE, hours x 103
Figure 21. NOX reduction versus catalyst age for the Hitachi Zosen process (32)
37
-------
NH3 CONCENTRATION, ppm
00
H-
00
l-i
n>
CD n
e o
i-h P-
pj H-
rt rt
CD H-
--. O
cr1 3
H- CD
CD
C ^"h
h-' o
rt rt
(D &
rc
^^
U) Ml
Kl O
O
3
O
I-h
O
S
H-
-------
VO
0
CM
K
CO
QJ
fj
C
•H
ri
0
Pi
p
t=
C/l
H
PM
1.58
1.18
0 79
v/ • / j
0.39
0
Temperature: 734 F
Ash concentration: 8.28 gr/sft
• .
^
^-— ^-
•
' • • • I I 1
40
30
20
10
n
0 2000 4000 6000 8000
Tl
C/l
G
M
U
po
0
W
i
cc
K3
O
OPERATING TIME, hours
Figure 20. Pressure drop versus operating time for the Hitachi Zosen process (32).
-------
The reactor inlet NOX concentration has no effect on NOX redaction
efficiency in the range of 150 to 650 ppm of inlet NOX, as shown in
Figure 22, provided that the required NH3:NOX molar ratio is maintained.
As shown in Table 3 it can be seen that NOX reduction efficiency is
independent of inlet S(>2 and H^O concentrations. The presence of 1% to 2%
02 in the flue gas is necessary for NOX control, as shown in Figure 23.
TABLE 3. INFLUENCE OF S02 AND H20
CONCENTRATIONS
Inlet concentration Temperature/NOT reduction. %
S02. ppm H20. % 392QF 482QF 572OF 662OF
0
250
250
10
10
0
55
54
59
73
73
73
86
85
87
94
93
93
Basis: Temperature: 392OF to 662op
Area velocity: 34.7 sft3/hr-ft2
NH3:NOX (mole ratio): 1.0
Development Status
Hitachi Zosen has been developing an NOX control process and catalyst
since 1970. Six different catalyst series have been developed. They are
known as NOXNON 100, 200, 300, 400, 500, and 600. The 100 series is
nonselective, for use with CO, H2» and hydrocarbons as reducing
agents. NH3 is the reducing agent for the five remaining series. Series
200 is used for treating clean flue gas, that is, gas which does not contain a
significant amount of SOX or particulate matter. The 300 and 400 series are
resistant to SOX. The NOXNON 500 and 600 catalyst were developed to have
the following characteristics:
• High selectivity for adsorption and reaction activation of NH3 and
NOX
• Immunity to SOX attack, especially by 803
• A low pressure drop
• Tolerance to plugging from gases with high particulate loadings
A corrugated, honeycomb-type configuration was found to be the most
suitable for these conditions. The NOXNON 500 Type II catalyst was tested for
40
-------
o
M
H
X
o
100
80
60
40
20
Area velocity = 24.3
734°F
662°F
734°F
Area velocity = 82.85 sft3/hr-ft2
662UF
Temperature: 662°F, 734°F
NH3:NO (molar ratio): 1.0
I
200 400
INLET NOX CONCENTRATION, ppm
600
Figure 22. Influence of NOX concentration on NOX reduction with the
Hitachi Zosen process (32).
41
-------
NJ
52
O
M
H
O
&
100
80
60
40
20
I 1
O 9
Area velocity = 7.68 Nm /m »h
Temperature: 390°C
NH3:NOX (molar ratio): 1.0
I
5 10
02 CONCENTRATION, vol %
15
Figure 23. Influence of Q£ concentration on NOX reduction for the Hitachi
Zosen process (32).
-------
over 8,000 hours with a negligible increase in pressure drop, although
catalyst activity exhibited a sharp initial decrease due to abrasion of
protruding active sites on the surface. The NOXNON 500 Type III catalyst was
designed specifically for increased surface hardness, which was demonstrated
in over 3,500 hours of testing for abrasion resistance. The NOXNON 600
catalyst has the same composition as the 500 series catalyst but the solid
metal support used in the 500 series was replaced by a wire mesh support. The
NOXNON 600 catalyst was tested for over 14,000 hours with a negligible
increase in pressure drop and stable NOX removal efficiency.
Since the beginning of catalyst development, a total of 41 pilot plants
have been constructed by Hitachi Zosen (see Table 4). While a majority of
these pilot plants are at oil- and gas-fired installations, there are three
pilot plants at coal-fired installations. The most significant work in coal-
burning applications was performed at the Isogo Station of the Electric Power
Development Company of Japan. In 1978, EPA contracted with Hitachi Zosen to
build a 0.5-MW-size pilot plant at the coal-burning Mitchell Station of the
Georgia Power Company.
TABLE 4. HITACHI ZOSEN PILOT-PLANT EXPERIENCE
No. of plants
Heavy fuel-oil-fired boilers 21
LNG- and LPG-fired furnaces 6
Iron ore sintering 3
Heavy oil-fired cement kilns 2
Heavy oil-fired glass smelting furnaces 2
Coke ovens 3
Coal-fired power plants 3
LPG-fired simulation gas (for gas turbine) 1
Total 41
Total raw gas flow through test plants 35,130 Nm3/hr
There are no existing commercial applications of the Hitachi Zosen
process on coal-fired boilers. Table 5 lists nine commercial installations
using other fuels.
43
-------
TABLE 5. COMMERCIAL PLANTS USING THE HITACHI ZOSEN PROCESS
Customer
Treating
capacity,
Nm3/hr
Flue gas source
Process
Completion
1 Osaka Gas Company, 53,000
Sakai
2 Daiki Engineering, 5,000
Chiba
3 Idemitsu Eosan, 350,000
Chiba
4 Shin-Daikyowa Petro- 440,000
chemical, Tokkaichi
5 Hitachi Zosen, 6,000
Osaka
6 Toshin Steel Mill, 70,900
Himej i
7 Kawasaki Steel, 762,000
Chiba
8 Nippon Satetsu, 10,000
Himej i
9 Maruzen Petro- 150,000
chemical Company
(formerly Kansai
Oil Company), Sakai
LNG- or naphtha-fired
furnace
LPG-fired furnace
Co boiler
heater
Fuel-oil-
wet-type
Gas-fired
furnace
Kerosene-
heating
Iron ore
wet-type
Fuel-oil-
heating
Fuel-oil
and gas-fired
fired boiler with
desulfurization
annealing
fired steel
furnace
sintering plant with
desulfurization
fired steel
furnace
fired boiler
Ammonia reduction 1975
Ammonia reduction 1975
Ammonia reduction 1975
Ammonia reduction 1975
Ammonia reduction 1975
Ammonia reduction 1976
Ammonia reduction 1976
Ammonia reduction 1977
Ammonia reduction 1979
-------
PREMISES
The following premises were used for the comparative economic evaluations
of the processes in this study. The premises are for projects with a
construction schedule starting in 1981 and ending in 1983, with 1984 as the
first year of operation.
DESIGN PREMISES
Plant Size and Fuel
The power unit is a new, coal-fired, horizontally opposed, balance-draft
boiler burning pulverized coal. A 30-year life and a north-central location
(Illinois, Indiana, Ohio, Michigan, and Wisconsin) has been assumed. The unit
has a net output of 500 MW including all system energy requirements up to and
including the ESP's and the induced draft (ID) fans. This does not include
the energy usage of any flue gas treatment processes. The unit has a design
heat rate of 9,500 Btu/kWh. Utility basic mechanical equipment is shown in
Table 6.
TABLE 6. COAL PLANT BASE MECHANICAL EQUIPMENT
Total plant electrical capacity* 2,000 MW
Unit rated electrical capacity* 500 MW
Steam generator type Balanced draft
Horizontally fired, dry bottom
Pulverized coal
Design heat rate 9,500 Btu/kWh
Fly ash removal type Cold-side electrostatic
precipitator
Fly ash removal efficiency 99.8%
a. Net including system energy usage up to and including ESP's
and the ID fans (does not include F6D or F6T).
Fuel for the plant is a coal having a heating value of 11,700 Btu/lb as
fired and containing 3.5% sulfur and 15.1% ash. The coal composition and the
input coal requirements (based on a heat rate of 9,500 Btu/kWh) for the 500-MW
45
-------
boiler are listed in Table 7. The capacity factor is equivalent to 5,500
hours of operation at full load.
TABLE 7. BASE CASE COAL
COMPOSITION
AND INPUT FLOW RATE
(500-MW new unit,
9,500 Btu/kWh heat rate)
Component Wt %. as fired Lb/hr
c
H2
N2
02
S
Cl
Ash
H20
66.7
3.8
1.3
5.6
3.4
0.1
15.1
4.0
270,800
15,400
5,300
22,700
13,800
400
61,300
16,300
Flue Gas Composition
Flue gas compositions are based (2) on combustion of pulverized coal
assuming a total air rate after the air preheater equivalent to 139% of the
stoichiometric requirement. This includes 20% excess air to the
boiler and 19% additional air inleakage at the air preheater.
It is assumed that 80% of the ash present in the coal is emitted as fly
ash and that 92% of the sulfur is emitted as SOX. Three percent of the
sultur emitted as SOX is S03 and 100% of the chlorine is emitted as
HC1. NOX emissions are assumed to be 0.6 Ib equivalent N02/MBtu with 95%
being NO and 5% N02 . Table 8 gives the composition and flow rate at the
economizer outlet for untreated flue gas.
NOy Control System
Proper reheating credits or debits are assigned as applicable for NOZ
removal. For dry catalytic processes, catalyst replacement occurs during
boiler outages and does not affect boiler on-stream time. Spent catalyst
disposal costs are assumed to be zero, with the catalyst support salvage value
being equal to the catalyst coating removal and disposal costs. This
assumption may be unique to the Hitachi Zosen catalyst which employs a
46
-------
metallic support made of stainless steel. It may not apply to ceramic
supports or homogeneous catalyst. Even with salvage of the metallic support,
disposal costs may be greater than zero; however, Hitachi Zosen feels that
this is a reasonable assumption for their catalyst. A sufficient quantity of
catalyst to ensure the desired removal efficiency is maintained. Redundancy
and the number of modules for dry processes are based on the NOX removal
system module availability and the required NOX removal efficiency. For
this study, the NOX removal system availability is assumed to be 100% so
that no redundant trains are needed. Redundancy is achieved through sparing
vital equipment in the NHg vaporization and injection system.
TABLE 8. FLUE GAS COMPOSITION
AND FLOW RATE AT THE ECONOMIZER OUTLET
(500-MW new unit, coal fired, 9,500
Btu/kWh, 3.5% S, 11,700 Btu/lb HHV
as fired, 2,249,000 aft3/min at 705<>F)
Component Vol. %
Lb/hr
N2
02
C02
S02
S03
NO
N02
HC1
H20
Ash
74.86
3.27
14.21
0.24
0.008
0.037
0.002
0.007
7.36
-
3,326,900
165,800
992,300
24,600
953
1,765
143
416
210,200
49,000
100.00 4,772,100 (appro*.)
Since it is presently unclear whether the presence of nitrogen compounds
in waste water as a result of air heater washing to remove NHg compounds is
a significant problem for U.S. applications, no provisions are made for the
treatment of waste water.
Separate ID fans are not included for the systems. Rather, a larger
boiler ID fan is used to compensate for the higher pressure drop of the
combined boiler-FGT system and the increased costs are assigned to the FGT
process.
47
-------
Raw Materials
All raw materials used are assumed to be received by either rail or
truck. Thirty-day storage facilities at full load are provided.
ECONOMIC PREMISES
The capital structure is assumed to be 35% common stock, 15% preferred
stock, and 50% long-term debt. The cost of capital is 11.4% for common stock,
10.0% for preferred stock, and 9.0% for long-term debt. The weighted cost of
capital is 10.0% and the discount rate is 10% (8).
A 30-year economic life and a 30-year tax life are assumed for the
utility plant. Salvage value is less than 10% and is equal to removal costs.
The annual sinking fund factor for a 30-year economic life and 10.0% weighted
cost of capital is 0.61%. The use of the sinking fund factor does not
indicate that regulated utilities commonly use sinking fund depreciation. The
sinking fund factor is used because it is equivalent to straight-line
depreciation levelized for the economic life of the facility using the
weighted cost of capital. The depreciation schedule for other types of plants
or facilities is based on their expected useful life.
An annual interim replacement allowance of 0.56% is also included as an
adjustment to the depreciation account to ensure that the initial investment
will be recovered within the actual rather than the forecast life of the
facility (12). Since power plant retirements occur at different ages, an
average service life is estimated. The interim replacement allowance does not
cover replacement of individual items of equipment since these are covered by
the maintenance charge.
The sum of the years digits method of accelerated depreciation is used
for tax purposes. For a 50% tax rate, 30-year tax life, 30-year book life,
10.0% weighted cost of capital, and a 0.61% sinking fund factor, the annual
levelized accelerated depreciation credit is 1.36%. Using a 10% investment
tax credit rate the levelized investment tax credit is 1.92% annually. For a
10.61% capital recovery factor (weighted cost of capital plus sinking fund
factor), 0.56% allowance for interim replacements, 3.3% straight-line
depreciation, 50% debt ratio, 9.0% debt cost, and a 50% income tax rate, the
levelized income tax rate is 4.31%.
The levelized annual capital charges as shown in Table 9 are 14.7% of the
total investment. The annual capital charge includes charges for the capital
recovery factor, interim replacements, insurance and property taxes, State and
Federal income taxes, and credits for investment credit and accelerated
depreciation.
The annual capital charge is applied to the total capital investment. It
is recognized that land and working capital (except spare parts) are not
depreciable and that provisions must be made at the end of the economic life
of the facility to recover their capital value. In addition, investment
48
-------
credit and accelerated depreciation credit cannot be taken for land and
working capital (except spare parts). The cumulative effect of these factors
makes an insignificant change in the annual capital charge rate in most cases
and is therefore ignored.
TABLE 9. LEVELIZED ANNUAL CAPITAL CHARGES
FOR REGULATED UTILITY FINANCING
Capital charge. %
Capital recovery factor 10.61
Interim replacements 0.56
Insurance and property taxes 2.50
Levelized income tax 4.31
Investment credit (1.92)
Accelerated depreciation (1.36)
Total 14.70
Capital Investment Estimates
Capital investment estimates are based on a north-central location
(Illinois, Indiana, Ohio, Michigan, Wisconsin) and represent projects
beginning 1981 and ending 1983. Capital cash flows for a standard project are
assumed to be 25% the first year, 50% the second year, and 25% the third year
of the project life. Capital costs are projected to mid-1982, which
represents the midpoint of the construction expenditure schedule.
The preliminary capital cost estimates are considered to have a -20% to
+40% range of accuracy for the Hitachi Zosen and Exxon processes. They are
based on a process description, flowsheet, material balance, and equipment
list. Piping, ductwork, and instrumentation are factored. The range of
accuracy for the cost of the ALNB is -20% to +100%. It is based on the best
fixed capital investment estimate for incremental costs above present burner
costs now available from boiler manufacturers.
The total fixed capital investment consists of direct capital costs,
engineering design and supervision, construction expense, contractor fees, and
contingency. The total capital investment includes the total fixed capital
investment plus allowances for startup and modification, royalties, the cost
of funds during construction, plus the cost of land, working capital, and,
where applicable, cost of the initial catalyst charge.
49
-------
Direct Investment—
Direct capital costs cover process equipment, piping, insulation,
transport lines, foundations, structural, electrical, instrumentation, raw
material and byproduct storage, site preparation and excavation, buildings,
roads and railroads, trucks, and earthmoving equipment. Direct investments
are prepared using standard estimation techniques (22, 24) and the average
annual Chemical Engineering (6) cost indexes and projections as shown in
Table 10. A premium for 7% overtime is included in the construction labor.
Appropriate amounts for sales tax and for freight are included in the process
capital costs.
TABLE 10. COST INDEXES AND PROJECTIONS
Year:
1978
1979a 198Qa 198ia 1982a 1983a 1984a
Plant 218.8 240.2
Materialb 240.6 262.5
Laborc 185.9 209.7
259.4
286.1
226.5
278.9
309.0
244.6
299.8
333.7
264.2
322.3
360.4
285.3
344.9
385.6
305.3
a. TVA projections.
b. Same as index in Chemical Engineering
machinery, supports."
c. Same as index in Chemical Engineering
labor."
(24) for
(24) for
"equipment,
"construction
Necessary electrical substations, conduit, steam, process water, fire and
service water, instrument air, chilled water, inert gas, and compressed air
distribution facilities are included in the utilities investment. These
facilities are costed as increments to the facilities already required by the
power plant. Services, nonpower plant utilities, and miscellaneous are
estimated at 6% of the total process capital.
Indirect Investment—
Indirect capital investment consists of engineering design and
supervision, architect and engineering contractor costs, construction costs,
contractor fees, and contingency. Construction costs, which consist of costs
for mobile equipment, temporary lighting, construction roads, water supply,
construction safety and sanitary facilities, and other similar expenses
incurred during construction are considered as part of construction expenses
and are charged to indirect capital investment.
Listed below are the factors used to determine the indirect capital cost.
50
-------
% of direct investment
Engineering design and supervision 7
Architect and engineering contractor 2
Construction expense 16
Contractor fees 5
Total 30
A contingency of 20% has been included for unforeseen expenses. It is based
on the sum of the direct investment less waste disposal and the above indirect
investments.
Other Capital Investments—
Startup and modification allowances are 10% of the total fixed
investment. For proprietary processes, the actual royalty fees are charged.
Working capital is the total amount of money invested in raw materials,
supplies, accounts receivable, and monies on deposit for payment of operating
expenses. Working capital is the equivalent cost of 1 month's raw material
cost, 1.5 months' conversion cost, and 1.5 months' plant and administrative
overhead costs. In addition, it includes 3% of the total direct investment to
cover spare parts, accounts receivable, and monies on deposit to pay taxes and
accounts payable. Land cost is assumed to be $5,000 per acre. For the
Hitachi Zosen process, a 1982 initial catalyst charge of $600/ft3 is also
included.
Annual Revenue Requirements
Annual revenue requirements use 1984 costs and are based on 5,500 hours
of operation per year at full load.
Direct Operating and Maintenance Costs—
Direct costs include raw materials, labor, utilities, maintenance, and
analytical costs. Raw material, labor, and utility costs are listed in
Table 11.
Unit costs for steam and electricity are based on the assumption that the
required energy is purchased from another source. Unit costs ( $/kW,
mills/kWh) are calculated on the basis that the power unit size is the net
power output after the addition of the pollution control systems. Actually,
the electrical usage by the pollution control equipment after the ESP will
result in a derating of the utility plant. To minimize iterative
calculations, instead of derating the utility plant, the pollution control
equipment is charged with purchased electricity. Maintenance costs are
estimated to be 5% of the direct investment.
51
-------
TABLE 11. COST FACTORS
1984 Utility Costs
Electricity $0.037/kWh
Steam $2.70/MBtu
1984 Labor Costs
Operating labor
Analyses
1984 Raw Material Costs
Ammonia
Catalyst (Hitachi Zosen) $700.00/ft3
$15.00/man-hr
$21.00/man-hr
$155.007ton
Indirect Costs—
Indirect costs cover levelized annual capital charges and overheads. The
levelized annual capital charges consist of a sinking fund factor, allowance
for interim replacement, property taxes, insurance, weighted cost of capital,
income tax, credits for accelerated depreciation, and investment credit. the
levelized annual capital charge as shown in Table 9 is 14.7%.
Overheads consist of plant, administrative, and marketing expenses.
Plant and administrative overhead is 60% of conversion costs less utilities.
The plant and administrative overheads include plant services such as safety,
cafeteria, medical, plant protection, and general engineering (excluding
maintenance). Fringe benefits are included in the base wage rates.
First-year revenue requirements using the 14.7% levelized capital charges
are determined. In addition, levelized annual revenue requirements are
calculated using a 10%/year discount factor, a 6%/year inflation factor, and a
30-year economic life that gives a 1.886 levelizing factor (8).
52
-------
SYSTEMS ESTIMATED
Process descriptions, flowsheets, material balances, and major equipment
lists and descriptions were prepared for each of the NOZ control methods
evaluated in this study, with the exception of the ALNB. An equipment list is
not available for the ALNB because of its early stage of development.
Two levels of NOZ reduction are examined. With a baseline emission of
the 0.6 Ib N02/MBtu (450 ppm at 3% 02) NSPS, costs are determined to
achieve a 50% NOX reduction to 0.3 Ib N02/MBtu (225 ppm at 3% 02) for
each of the three processes although 50% NOZ reduction is not typical for
the Hitachi Zosen process. In addition, costs are determined for a 90%
reduction in NOZ to 0.06 Ib N02/MBtu (45 ppm at 3% 02). The 90% NOZ
reduction involves the following three process alternatives.
1. Moderate NOZ reduction achieved by the ALNB and the remaining NOZ
reduction achieved by the Hitachi Zosen process.
2. Moderate NOZ reduction achieved by the Exxon process and the
remaining NOZ reduction achieved by the Hitachi Zosen process.
3. 100% of the NOZ reduction achieved by the Hitachi Zosen process.
The levels of NOZ reduction assumed for each process and process
combination are graphically illustrated in Figure 24. The Hitachi Zosen and
Exxon processes are divided into major operational areas to facilitate cost
comparisons.
ADVANCED LOW-NOX BURNER
The flow diagram and material balance for the base case are shown in
Figure 25 and Table 12 respectively. For this study an NOZ reduction to
0.3 Ib N02/MBtu, from the baseline emission of 0.6 Ib N02/MBtu, is
believed to be a conservative objective compared with the research goal of
0.2 Ib N02/MBtu.
The 500-MW ALNB system has 40 burners, each with 4 available tertiary
ports. The central burner operates at 70% of the stoichiometric combustion
air with an additional 50% supplied at the tertiary ports. The central
burner, which is approximately 40 inches in diameter at the throat, contains
53
-------
u. / -
0.6 .
4J
CNl
o
& 0.4 .
rH
53
° n •}
H 0 . -J •
H
w 0.2.
o
S3
0.1.
0.06.
0
Single Processes Combination Processes
V* «M
•» ••
^
•••
,__
^
^
I
jjjjjg
ALNB Exxon Hitachi
Zosen
ALNB /Hit
•••M
SsS
1
•
•
m
•
^
:acl
.
li EXJ
^|
M
(
•
1
1
m
i
ui
con/
•:*;•!•§
•
•X'X*
:^
I
«
W
^
H:
tachi Zosen
Zosen Hitachi Zosen
CASES STUDIED
Figure 24. Assumed NOX reduction for the six cases studied.
54
-------
Ui
Ui
ELECTROSTATIC
PRECIPITATOR
10
-g-
COAL
Figure 25. Advanced low-NOx burner flow diagram.
-------
TABLE 12. ADVANCED LOW-NOX BURNER
MATERIAL BALANCE
Stream
J
2
j
4
5
ft
7
8
9
10
Description
sft3/min (600E)
1
406.000
2
Coal and air
1.318 400
3
Air feed to
air heater
5 072 900
1.121.300
80
4
Combustion air
to boiler
4 37Q *nn
968.100
Anm
Description
Total stream. Ib/hr
sft3/min (600F)
Temperature, op
6
Air to burners
912.400
201,700
535
7
Flue gas to
economizer
4.772.100
1.003.700
^
Flue gas to
air heater
4.772.100
1.003.700
705
q
Flue gas to
ESP
5.465.500
1.157.000
a oo
in
Flue gas to
F6D unit
5.41«,fion
1.157.000
5 on
1
i
j
4
5
ft
7
8
9
10
1
i
\
4
5
ft
7
8
9
10
Stream
Description
Total stream. Ib/hr
sft3/min (60OF)
Temperature. OF
-
11
Fly ash
from ESP
48.900
56
-------
the primary annulus, the inner secondary annulus, and the outer secondary
annulus. The primary annulus is approximately 20 inches in diameter and
contains a conical diffuser fabricated from an abrasion-resistant metal to mix
the primary air and entrained coal particles. The inner secondary annulus,
which is constructed of stainless steel, contains stainless steel swirl
vanes. These swirl vanes are used to control mixing between primary and
secondary air. Additional combustion air is admitted through the outer
secondary annulus, which is also fabricated from stainless steel. The
tertiary ports, made of stainless steel, are located approximately one burner
throat diameter from the central burner. A compartmented windbox constructed
of carbon steel is used to aid in control of the combustion air flow.
The ALNB design using tertiary ports may create installation and
structural support problems when retrofitted on existing boilers. However, on
a new boiler, as is the case evaluated in this study, these problems may be
avoided by design of a unit compatible with the ALNB.
EXXON PROCESS
In this study air is used as the carrier for the NH3 (2%
NH3~in-air). Proprietary Exxon gas-phase mixing technology is used to
disperse the NH3 throughout the flue gas. An NH3:NOX molar ratio of
1.5:1 is used to obtain a 50% NOX reduction efficiency. N^ breakthrough
(loss in the flue gas) is approximately 50 ppm. The Exxon process is not
expected to increase the pressure drop of the boiler system, therefore no
additional flue gas fan capacity is included. The main control criteria are
based on boiler load.
The flow diagram and material balance for the base case are shown in
Figure 26 and Table 13 respectively. Hydrogen addition is not used. The
Exxon process is divided into two processing sections and the equipment
assigned to the appropriate section. The equipment list and descriptions by
area are presented in Table 14. The total land requirement is one acre.
NH3 Storage and Injection
In the NH3 storage and injection section a compressor (and spare) for
unloading liquid NH3 from truck or rail transport and a sufficient quantity
of 250 psig storage tanks for a 30-day supply are included. Before NHg is
injected into the flue gas, it is vaporized in a shell-and-tube steam-heated
vaporizer and mixed with 25 psig air supplied by a compressor. There is one
vaporizer and there are three air compressors, two operating and one spare.
Each compressor handles 50% of the required capacity. Two zoned injection
grids are included to accommodate flue gas temperature changes resulting from
load changes as well as flue gas temperature variations across the injection
plane.
Air Preheater Section
In the air preheater section two modified air heaters are provided.
These incorporate possible changes in air heater design required to prevent
adverse operational effects from (^4)2804 and NH4HS04 deposits
57
-------
RAIL
OR
TRUCK
HOOK-UP
00
AMMONIA
STORAGE TANK
UNLOADING
COMPRESSOR
COAL
i
STEAM
II
NH3
VAPORIZER
8 *_
Figure 26. Exxon process flow diagram.
AIR
I.
VvV
-------
TABLE 13. EXXON
MATERIAL BALANCE
•>.
i
t,
5
h
7
K
q
in
Stream
Description
sft3/min (60OF)
Temoeratnre. op
Pressure, psig
1 .
Coal to boiler
2
Combustion air
to air heater
1,121,300
80
3
Combustion air
to boiler
968,100
S3S
4
Gas to
economizer
5
Gas to
air heater
1.048,400
705
Stream
Description
1
2
')
4
5
6
7
H
9
If)
Total stream, Ib/hr
sft3/min (60OF)
Temperature, op
Pressure, psig
«
Gas to ESP
5,666,700
1,201,700
300
7
Gas to
FGD unit
5,«17,800
1,201,700
300
8
Air to idle
injection grid
66,500
9
NH3 and
air to active
injection grid
134,700
10
NH3 from
storage
1.582
110
225
Stream
Description
1
2
j
4
b
b
7
8
9
10
Total stream, Ib/hr
sft3/min (600F)
Temperature, op
Pressure, psig
11
Steam to
vaporizer
1,100
298
50
12
Fly ash
from ESP
48,900
4
5
6
7
8
9
10
59
-------
TABLE 14. EXXON
EQUIPMENT LIST
Item (number); description
Total equipment
cost. 1982 $
Area 1—NH3 Storage and Injection
1. Compressor. NH3 unloading (2): Single cylinder, double 61,500
acting, 300 sft3/min at 250 psig, 30 psig suction, 125 hp,
cast iron
2. Tank. NH3 storage (9): Horizontal, 9 ft dia x 66 ft long, 311,900
30,000 gal, 250 psig, carbon steel
3. Vaporizer. NHq (1): Steam at 298°F, tube type, 29 ft2, 6,900
0.50 HBtu/hr, carbon steel
4. Compressor, air (3): 22,900 aft3/min at 25 psig, 14.7 1,132,800
psia suction, single stage, 2,250 hp
5. Injection grid. NH3 and air (2): 320,400
6. Pump. NH3 (2): 6 gpm, 0.5 hp, 28 ft head, carbon steel 3.000
Subtotal 1.836.500
Area 2—Air Preheater Section
1. Air preheater (2): Modified, size 29.5 Ljungstrom air
heater
2. Soot blower, steam (2): 20 ft, retractable, hot side
of air heater, 120 Ib/min steam
Subtotal
509,000
26.400
535.400*
Total, Areas 1-2
2,371,900
a. Incremental cost resulting from modification of the air preheater system.
60
-------
resulting from NH3 breakthrough. Two soot blowers, one per air heater, are
included for hot-side air heater cleaning. The modified design also allows
for the combination of the intermediate and cold temperature sections into one
continuous element as well as a potentially different element design.
However, only the incremental cost of the modified air heater, above costs of
the standard air heater, is included in the cost estimate. These costs may
change as further testing and evaluations are completed with coal-fired flue
gas.
HITACHI ZOSEN PROCESS
This is a dry NOX FGT process for the SCR of NOX with
NH3. Catalyst for the Hitachi Zosen process is designed to handle high
particulate loading with a pressure drop of 2 to 3 inches of %() across the
reactor. Therefore, flue gas from a coal-fired boiler may be fed directly to
the reactor upstream of the air heater before particulate removal.
Two base cases for the Hitachi Zosen process are examined for FGT on a
500-MW coal-fired boiler with a 0.6 Ib N02/MBtu baseline emission. One is
90% NOX reduction from 0.6 Ib/MBtu to 0.06 Ib/MBtu by treatment of the
entire flue gas stream. The other case is a 50% overall NOX reduction from
0.6 Ib/MBtu to 0.3 Ib/MBtu by treatment of a portion of the flue gas at a 90%
NOX reduction level. (Economics for an 80% NOX reduction case, 0.3
Ib/MBtu to 0.06 Ib/MBtu, are scaled from the 90% NOX reduction case. The
80% reduction case is used in combination with the ALNB and the Exxon process
to achieve an overall NOX reduction of 90% from the 0.6 Ib N02/MBtu
baseline emission.)
For 90% NOX reduction NH3 is injected into the flue gas upstream of
the reactor at an NH3:NOX molar ratio of 1:1. Air is mixed with the N^
(5% NH3 in air) to obtain improved mixing with the flue gas. In the reactor
NOX is reduced by NH3 to N2 in the presence of the catalyst at a
temperature between 600°F and 750°F. The area velocity (flow rate of
gas/surface area of catalyst) is 24.3 ft3/hr-ft2 for 90% NOX removal
requiring 10,734 ft3 of catalyst.
To obtain a net 50% NOX reduction, 56% of the flue gas leaving the
economizer is treated for 90% NOX reduction; the remaining 44% of the flue
gas is bypassed around the reactor and recombined with the treated flue gas.
Operating conditions are the same as the 90% reduction case stated above
except catalyst requirements are reduced to 6,105 ft3.
The NH3 flow rate is automatically controlled based on the flue gas
flow rate to the reactor, reactor inlet and outlet NOX concentration, and
NH3 outlet concentration. NH3 level leaving the reactor is assumed to be
less than 10 ppm. To prevent formation of (^4)804 and NH4HS03 at low
boiler loads the catalyst bed temperature is controlled by bypassing a part of
the high-temperature flue gas flow around the economizer to the reactor.
61
-------
The catalyst is manufactured as individual units which are joined to
form the required catalyst bed. Flue gas passes parallel to the catalyst
surface. The catalyst composition has not been revealed for proprietary
reasons; however, Hitachi Zosen does state that it is constructed of common
material. The catalyst life is assumed to be one year because the guarantee
by Hitachi Zosen is only one year (the actual life may be longer).
Process Description (90% NOx reduction)
The flow diagram and material balance for the 90% NOZ reduction case
are shown in Figure 27 and Table 15 respectively. The Hitachi Zosen 90% NOX
reduction process is divided into four processing sections and the equipment
assigned to the appropriate section. The equipment list and descriptions by
area are presented in Table 16. The total land requirement is 1.5 acres.
NH3 Storage and Injection—
A compressor (and spare) for unloading liquid NH3 from truck or rail
transport and a sufficient quantity of 250 psig storage tanks to maintain a
30-day NH3 supply are included. Before NH3 is injected into the flue gas,
it is vaporized in a she11-and-tube steam-heated vaporizer and mixed with air
supplied by a small blower to form a 5% N^-in-air mixture. This is below
the flammability limits for NH3 in air (15.5% to 27.0%) and improves flow
control and mixing. There is one vaporizer and one air-NH3 blower for each
of the two reactor trains with one additional air-NH3 blower which serves as
a spare. A pump is placed between the NH3 storage tanks and the vaporizer
of each train to aid in the control of NH3 flow. A third pump serves as a
spare. Two NH3 injection grids, one per train, provide even distribution of
the NH3 in the flue gas before it enters the reactor.
Reactor Section—
Two reactors are provided, each handling 50% of the total flow. The
reactors are fixed-bed type and constructed of carbon steel. Each reactor is
provided with fly ash hoppers for collection of deposited fly ash and with two
soot blowers for periodic cleaning. The catalyst consists of corrugated
plates contained in units of 1 x 1 z 1 meter and 0.5 x 1 x 1 meter. These
units are joined to form the appropriately sized catalyst bed. Two monorail
cranes, one mounted on each reactor, are supplied to load and unload catalyst
units.
Flue Gas Handling—
A larger ID fan is provided downstream of each ESP to compensate for the
increased pressure drop created by the FGT system. However, only the
incremental cost attributed to the NOX removal system is included in the
cost estimates. Because of the larger pressure drop, additional costs for
bypass ducting around the ID fan to prevent boiler implosion are included with
gas handling.
Air Preheater Section—
This area description is the same as previously given for the same area
with the Exxon process.
62
-------
U)
Figure 27. Hitachi Zosen process (90% NOX reduction) flow diagram.
-------
TABLE 15. HITACHI ZOSEN (90% NOX REDUCTION)
MATERIAL BALANCE
1
2
j
4
b
6
7
«
9
10
Stream
Total stream. Ib/hr
sft3/min (600F)
Temperature, op
Pressure. psi«
1
406.000
2
Combustion air
5,072,900
1,121,300
80
3
Combustion air
4,379,500
968,100
535
4
Gas to
4,772,100
1,003,700
890
5
Flue gas
4,772,100
1,003,700
705
Stream
2
i
4
5
6
/
«
9
111
Description
Total stream. Ib/hr
sft3/min (6flop)
Temperature, °F
Pressure, psl«
6
Flue gas - NH3
mixture to
reactor
4,807,600
1,011,700
697
7
Gas to
air heater
4,807,600
1,011.800
705
8
Gas to ESP
5,501,000
i, ids, 100
300
9
Gas to
FGD unit
5,452,100
1, 1
6
7
8
9
10
Description
Total stream, Ib/hr
*ft3/min (60<>F)
Temperature, op
Pressure, psig
11
NH3 from
storage
1,055
110
225
12
Steam to
vaporiz-er
730
298
50
13
Fly ash
from ESP
48,900
A
5
6
T
8
9
10
64
-------
TABLE 16. HITACHI ZOSEN (90% NOX REDUCTION)
EQUIPMENT LIST
Total equipment
Item (number); description cost. 1982 $
Area 1—NH3 Storage and Injection
1. Compressor. NH3 unloading (2): Single cylinder, double 61,500
acting, 300 sft3/min at 250 psig, 30 psig suction, 125 hp,
cast iron
2. Tank , NH3 storage (6): Horizontal, 9 ft dia x 66 ft long, 207,900
30,000 gal, 250 psig, carbon steel
3. Vaporizer. NH3 (2): Steam at 298°F, tube type, 10 ft2, 7,400
0.33 MBtu/hr, carbon steel
4. Blower. NH3 and air (3): 3,950 aft3/min, AP 15 in. H20, 17,900
15 hp, carbon steel
5. Injection grid. NH3 and air (2): 74,600
6. Pump. NH3 (3): 2 gpm, 0.5 hp, 28 ft head, carbon steel 4.600
Subtotal 373.900
Area 2—Reactor Section
1. Reactor (2): 55 ft x 33 ft x 41 ft high, 700°F 2,054,200
operating temperature, carbon steel, insulated, with
fly ash hoppers
2. Soot blower, steam (4): 33 ft, retractable, 870 Ib/min 58,000
steam
3. Reactor crane and hoist (2): 33 ft monorail, 2,500 Ib 56.200
capacity, 40 ft lift
Subtotal 2.168.400
(continued)
65
-------
TABLE 16 (continued)
Total equipment
Item (number); description cost. 1982$
Area 3—Flue Gas Handling
1. Blower, flue gas (4): 468,261 aft3/min, AP 22 in. HfcO, 467.700
2,500 hp, 316 stainless steel
Subtotal 467.70Qa
Area 4—Air Preheater Section
1. Air preheater (2): Modified, size 29.5 Ljungstrom air 509,000
heater
2. Soot blower, steam (2): 20 ft, retractable, hot side of 26.400
air heater, 120 Ib/min steam
Subtotal 535.40Qt>
Total, Areas 1-4 3,545,400
a. Incremental cost resulting from increased flue gas pressure drop.
b. Incremental cost resulting from modification of the air preheater system.
66
-------
Process Description (50% NOx reduction)
The flow diagram and material balance for the 50% NOX reduction case
are shown in Figure 28 and Table 17 respectively. The Hitachi Zosen 50% NOX
reduction process is divided into four processing sections and the equipment
assigned to the appropriate section. The equipment list and descriptions by
area are presented in Table 18. The total land requirement is 1.5 acres.
NH3 Storage and Injection—
The NH3 storage and injection system for this case is similar to that
previously described for the same area with the Hitachi Zosen 90% NOX
reduction case.
Reactor Section—
One reactor is provided handling 56% of the flue gas leaving the boiler.
This reactor is the same as previously described for the same area in the
Hitachi Zosen 90% NOX reduction case.
Flue Gas Handling—
This area is the same as previously described for the Hitachi Zosen 90%
NOX reduction case.
Air Preheater Section—
This area is the same as that described for the Exxon process.
67
-------
oo
I~~7 AMMONIA
I STORAGE TANK
RAIL
TRUCK
HOOK-UP UNLOADING
COMPRESSOR
NH3
VAPORIZER
Figure 28. Hitachi Zosen process (50% NO reduction) flow diagram.
-------
TABLE 17. HITACHI ZOSEN (50% NOX REDUCTION)
MATERIAL BALANCE
Stream
J-
2
)
t,
5
6
7
8
-------
TABLE 18. HITACHI ZOSEN (50% NOX REDUCTION)
EQUIPMENT LIST
Total equipment
Item (number); description cost. 1982 $
Area 1—NH3 Storage and In.iection
1. Compressor. NH3 unloading (2): Single cylinder, double 61,500
acting, 300 sft3/min at 250 psig, 30 psig suction, 125 hp,
cast iron
2. Tank, NH3 storage (3): Horizontal, 9 ft dia x 66 ft long, 104,000
30,000 gal, 250 psig, carbon steel
3. Vaporizer. NH3 (1): Steam at 298°F, tube type, 11 ft2, 3,900
0.371 MBtu/hr, carbon steel
4. Blower. NH3 and air (2): 4,385 aft3/min, AP 15 in. H20, 12,400
15 hp, carbon steel
5. Injection grid. MBh and air (1): 37,300
6. Pump. NH3 (2): 2 gpm, 0.5 hp, 28 ft head, carbon steel 3.100
Subtotal 222.200
Area 2—Reactor Section
1. Reactor (1): 55 ft x 37 ft x 36 ft high, 700°F 1,027,100
operating temperature, carbon steel, insulated, with
fly ash hoppers
2. Soot blower, steam (2): 37 ft, retractable, 870 Ib/min 31,200
steam
3. Reactor crane and hoist (1): 37 ft monorail, 2,500 Ib 29.100
capacity, 35 ft lift
Subtotal 1.087.400
(continued)
70
-------
TABLE 18 (continued)
Item (number); description
Total equipment
cost, 1982$
Area 3—Flue Gas Handling
1. Blower, flue gas (4): 466,814 aft3/min, Ap 21 in. H20, 222.700
2,250 hp, 316 stainless steel
Subtotal 222.70Qa
Area 4—Air Preheater Section
1. Air preheater (2): Modified, size 29.5 Ljungstrom air 509,000
heater
2. Soot blower, steam (2): 20 ft, retractable, hot side of 26.400
air heater
Subtotal 53S.400b
Total, Areas 1-4
2,067,700
a. Incremental cost resulting from increased flue gas pressure drop.
b. Incremental cost resulting from modification of the air preheater system.
71
-------
RESULTS AND COMPARISON
Based on the design and economic premises and on the process equipment
for each process, the capital investment, in 1982 dollars, and the annual
revenue requirements, in 1984 dollars, were determined for the three 50% and
three 90% NCX reduction processes. Annual revenue requirements were
determined foi the first full year of operation. Levelized annual revenue
requirements were determined using a 10%/year discount factor, a 6%/year
inflation factor, and a 30-year economic life. The individual capital
investment and annual revenue requirement tables for each process are shown in
Appendix A. Costs for the ALNB are differential costs representing the
additional capital investment and annual revenue requirements as compared with
a boiler design using standard burners. Because of the different sources of
data, simplifying assumptions made, and the necessity of projecting costs into
the future, these estimates are considered to be accurate to an overall
variation of -20% to +40% for the Exxon and Hitachi Zosen processes. For the
ALNB the estimates are considered to be accurate to an overall variation of
-20% to +100%.
The ALNB, Exxon, and Hitachi Zosen processes were evaluated at 50% NOX
reduction. This degree of NOX reduction corresponds to the reduction of
NOX emissions *!rom 0.6 to 0.3 Ib N02/MBtu.
The ALNB/Hitachi Zosen, Exxon/Hitachi Zosen, and Hitachi Zosen processes
were evaluated at 90% NOX reduction. This degree of reduction lowers the
NOX emissions from 0.6 to 0.06 Ib NOx/MBtu.
CAPITAL INVESTMENT
The capital investment results for the 50% and 90% NOX reduction cases
are shown in Table 19.
Fifty Percent NOY Reduction
Advanced Low-N0x Burner—
The capital investment for the ALNB is $2.4M ($4.8/kW), as shown in
Table A-l. Because of the burner's early stage of development, a breakdown of
the direct and indirect investment is not available. This value is based on
the assumption that the ALNB incremental costs are similar to the incremental
costs incurred for design and application of the dual register
burner/compartmented windbox system instead of a high-turbulence circular
burner/single windbox system. Since development of the ALNB is being
sponsored by EPA, a royalty fee is not charged for the technology.
72
-------
TABLE 19. SUMMARY OF CAPITAL INVESTMENTS
Capital
investment,
mid-19 82$
Process M$ $/kW
50% NOX reduction
ALNB 2.4 4.8
Exxon 9.9 19.7
Hitachi Zosen 15.7 31.4
90% NOX reduction
ALNB/Hitachi Zosen 25.9 51.8
Exxon/Hitachi Zosen 32.1 64.2
Hitachi Zosen 25.5 50.9
Exxon—
The capital investment for the Exxon process is $9.9M ($19.7/kW), as
shown in Table A-3. Approximately one-third of the total capital investment
is for the NH3 storage and injection equipment ($3.3M). Royalties, at
$1.5M, rank second to the NI^ storage and injection system in contribution
to the total capital investment. The remaining $5.1M of the capital
investment is made up of various smaller charges, such as air preheater
section ($0.6M), construction expenses ($0.6M), allowance for startup and
modifications ($0.6M), and interest during construction ($1.0M).
Hitachi Zosen—
The capital investment for the Hitachi Zosen process is $15.7M
($31.4/kW), as shown in Table A-5. Equipment costs for the reactor and flue
gas handling sections and the initial catalyst charge cost are the major
contributors. The reactor section requires $2.2M for a reactor, soot blowers,
and a monorail crane and hoist with the related accessories. The flue gas
handling section contributes $2.1M to the total capital investment to cover
the incremental cost for larger fans resulting from the increased flue gas
pressure drop caused by the reactor system and for additional ductwork
required to route the gas to and from the reactor.
Ninety Percent NOY Reduction
ALNB/Hitachi Zosen—
To obtain the capital investment for the process combination, the capital
investment for the ALNB (Table A-l) is added to the capital investment for the
Hitachi Zosen 80% NOX reduction case (Table A-7) to obtain the sum for the
combined processes (Table A-9). The capital investment estimated for the
combined ALNB and Hitachi Zosen processes is $25.9M ($51.8/kW). Breakdown of
73
-------
the direct and indirect investments is not shown in Table A-9 since these
items are not available for the ALNB (see Table A-l) . The initial catalyst
charge cost at $5.0M is the largest cost item in the capital investment.
Exxon/Hitachi Zosen—
To obtain the capital investment for the combined processes, the capital
investment for the Exxon process (Table A-3) is added to the 80% reduction
Hitachi Zosen process capital investment (Table A-ll) to obtain the sum for
the two processes (Table A-13).
The capital investment for the combined Exxon and Hitachi Zosen
processes is $32.1M ($64.2/kW). The four primary cost areas of the
Exxon/Hitachi Zosen process are NH3 storage and injection ($3.9H), the
reactor section ($4.2H), flue gas handling ($3.1M), and the initial catalyst
charge ($5.0M).
Hitachi Zosen—
The capital investment for the Hitachi Zosen process is $25.5M
($50.9/kW), as shown in Table A-15. Three areas contribute the majority of
the capital investment. The reactor section is $4.2M, the flue gas handling
section is $3.1M, and the initial catalyst charge cost is the largest of the
three at $6.5M.
Overall Capital Investment Comparison
A comparison of the capital investments for each of the 50% and 90% NOX
reduction alternatives is shown in Table 20, along with identification of the
major cost components.
For 50% NOX reduction, the wide variation in capital investment is the
result of the varied approaches to NOX reduction and the equipment
employed. The ALNB requires different burners and different windbox and
boiler wall construction. The incremental cost differences for these changes
are thought to be small. The Exxon process requires NH3 storage,
vaporization, and carrier air supply equipment and installation of two NH3
injection grids in the boiler. The Hitachi Zosen process requires similar
NHj storage, vaporization, and carrier air supply equipment and an injection
grid. In addition it requires a reactor, catalyst, and additional fan
capacity and ductwork for routing flue gas to and from the reactor.
While Hitachi Zosen has the highest capital investment for 50% reduction,
it has the lowest capital investment for 90% reduction. The ALNB/Hitachi
Zosen process is similar in cost for 90% reduction because the Hitachi Zosen
capital investment for 80% reduction (0.3 reduced to 0.06 Ib N(>2/MBtu) is
not significantly less than for 90% reduction. As in all Hitachi Zosen
capital investments, the initial catalyst charge cost is a major contributor.
In the 80% and 90% reduction cases the difference in capital investment is
primarily a result of the difference in catalyst requirements.
As shown in Table 20, the Exxon process requires more capital investment
for the NH3 storage and injection system than does the Hitachi Zosen
74
-------
TABLE 20. CONTRIBUTION OF DIRECT INVESTMENT, ROYALTIES, AND CATALYST TO CAPITAL INVESTMENT
Ln
50% reduction
Direct investment
NH3 storage and injection
Reactor section
Flue gas fans
Air preheater section
Total process capital
Other capital charges
Royalties
Catalyst
Subtotal of process capital,
royalties, and catalyst
Other investments
Total capital investment (TCI)
ALNB a Exxon
% of % of
MS TCI MS TCI
3.3 34
-
-
0.6 6
3.9 40
1.5 15
_ _ _ _
5.4 55
- 4.5 45
2.4 9.9
Hitachi Zosen
% of
MS TCI
0.5
2.2
2.1
0.6
5.4
0.5
3.7
9.8
5.9
15.7
3
14
13
4
34
3
25
62
38
90% reduction
ALNB/ Exxon/
Hitachi Zosena Hitachi Zosen
% of % of
M$ TCI M$ TCI
3.9
4.2
3.1
0.6
11.8
0.5 2 2.0
5.0 19 5.0
5.5 21 18.8
20.4 79 13.3
25.9 32.1
12
13
10
2
37
6
16
59
41
Hitachi Zosen
% of
MS TCI
0.8
4.2
3.1
0.6
8.7
0.5
6.5
15.7
9.8
25.5
3
17
12
2
34
2
26
62
38
a. Breakdown of the direct investment for processes containing the ALNB is not available.
-------
process. This is because of the larger NHj storage requirements, larger
carrier air supply equipment, and the more complex injection grid system with
the Exxon process.
The costs of air heater modifications, which are not an inherent part of
the processes but may be required for process application to coal-fired
boilers when NIL} injection is used in NOX control, are also included with
the Exxon and Hitachi Zosen processes. These modifications may be required
because of ammonium salt deposition in the air heater. The cost for air
heater modifications used in this study are the same for both the Exxon and
Hitachi Zosen processes, although the Exxon process has a much higher level of
breakthrough NH^ and there is more potential for problems from the ammonium
salts deposition.
It should be noted that the royalties are a significant portion of the
capital investment for the Exxon process. As shown in Table 20, the royalties
are three times higher than those for the Hitachi Zosen process.
The capital investment for 50% NOZ reduction is considerably less than
that for 90% reduction. The ratio of the lowest 90% reduction capital
investment to the lowest 50% reduction capital investment is approximately
eleven to one.
ANNUAL REVENUE REQUIREMENTS
The annual revenue requirements for each of the NOX reduction processes
are summarized in Table 21.
TABLE 21. SUMMARY OF ANNUAL REVENUE REQUIREMENTS
Annual revenue requirements,
1984$
First year
Process
50% NOX reduction
ALNB
Exxon
Hitachi Zosen
90% NOZ reduction
ALNB/Hitachi Zosen
Exxon/Hitachi Zosen
Hitachi Zosen
MS
0.45
3.4
8.0
11.5
14.2
13.3
Mills/kWh
0.17
1.2
2.9
4.2
5.2
4.9
Levelized
M$
0.54
5.1
13.0
18.4
22.6
21.9
Mills/kWh
0.20
1.9
4.7
6.7
8.2
7.9
76
-------
Fifty Percent NQy Reduction
Advanced Low-N0x Burner—
The first-year annual revenue requirements for the ALNB are $0.45M (0.17
mills/kWh), as shown in Table A-2. Levelized annual revenue requirements are
$0.54M (0.20 mills/kWh). These are incremental costs above that required for
presently available dual register burner systems capable of meeting the 1979
NSPS. Direct costs are low for the ALNB since utilities and raw materials,
which normally contribute most of the direct cost, are not required. The only
direct cost is $0.06M for maintenance, labor, and materials. Capital charges
are the largest cost item of the annual revenue requirements.
Exxon—
The first-year annual revenue requirements for the Exxon process are
$3.4M (1.2 mills/kWh), as shown in Table A-4. Levelized annual revenue
requirements are $5.1M (1.9 mills/kWh). Raw materials and utilities are
significant portions of the annual revenue requirements. A total of 4,351
tons of NH3 is used each year at a yearly cost of $0.67M. Utilities at
$0.76M are also a major factor. However, capital charges of $1.5M constitute
the largest portion of the annual revenue requirements.
Hitachi Zosen—
The first-year annual revenue requirements for the Hitachi Zosen process
are $8.0M (2.9 mills/kWh), as shown in Table A-6. Levelized annual revenue
requirements are $13.OM (4.7 mills/kWh). Raw materials cost ($4.5M),
particularly annual catalyst replacement cost ($4.3M), is the major cost item
for the Hitachi Zosen process. Capital charges are also high at $2.3M, but
this is significantly lower than the annual cost for replacement catalyst.
Ninety Percent NO* Reduction
ALNB/Hitachi Zosen—
The annual revenue requirements for the combined ALNB and Hitachi Zosen
processes are determined by the same method used to determine the capital
investment. The ALNB annual revenue requirements (Table A-2) are added to the
annual revenue requirements for Hitachi Zosen 80% NOX reduction (Table A-8)
to obtain the total annual revenue requirements (Table A-10) for a 90% overall
NOX reduction.
The first-year annual revenue requirements for the ALNB/Hitachi Zosen
process are $11.5M (4.2 mills/kWh). The levelized annual revenue
requirements are $18.4 (6.7 mills/kWh). Annual catalyst replacement at$5.8M
is the largest contributor to the annual revenue requirements and the
levelized capital charges are the next highest at$3.8M.
Exxon/Hitachi Zosen—
The annual revenue requirements for the combined Exxon and Hitachi Zosen
processes are determined by adding the Exxon annual revenue requirements
(Table A-4) to the annual revenue requirements of the Hitachi Zosen 80% NOX
reduction case (Table A-12) to obtain the total annual revenue requirements
(Table A-14) for a 90% overall NOX reduction.
77
-------
The first-year annual revenue requirements for the Exxon/Hitachi Zosen
process are $14.2M (5.2 mills/kWh). The levelized annual revenue requirements
are $22.6M (8.2 mills/kWh). The annual catalyst replacement cost and the
levelized capital charges are the two largest costs associated with the annual
revenue requirements, $ 5.8M and $4.7M, respectively, but NH3 at $0.9M and
utilities at $1.4H are also significant.
Hitachi Zosen—
The first-year annual revenue requirements for the Hitachi Zosen process
are $13.3M (4.9 mills/kWh), as shown in Table A-16. The levelized annual
revenue requirements are $21.9M (7.9 mills/kWh). Raw material consumption is
the major contributor to the total. Of the $7.9M for raw materials, annual
catalyst replacement is $7.5M and is the single largest cost. Other
significant cost items are the levelized capital charges ($3.7M) and utilities
($0.8M).
Overall Annual Revenue Requirements Comparison
A comparison of the annual revenue requirements for each of the 50% and
90% NOX reduction alternatives is shown in Table 22, along with
identification of the major cost components.
Annual revenue requirements for 50% NOX reduction processes follow the
same trend as capital investment, i.e., the ALNB has the lowest, the Exxon
process is intermediate, and the Hitachi Zosen process has the highest. The
ALNB has the lowest annual revenue requirements because no raw materials or
utilities are required. Capital charges constitute most of the annual revenue
requirements for the ALNB. This is in contrast to the Hitachi Zosen process
in which raw materials, primarily replacement catalyst, are $4.5M and
utilities are $0.5M.
As in the 50% reduction case, the annual revenue requirements are lowest
for the 90% reduction process using the ALNB/Hitachi Zosen process. The
Hitachi Zosen process by itself follows and the Exxon/Hitachi Zosen process
has the highest annual revenue requirements. By using the ALNB in combination
with the Hitachi Zosen process, the annual revenue requirements can be reduced
below those of the Hitachi Zosen process alone. The decrease is a result of
reduced NHg consumption, reduced replacement catalyst requirements, and
reduced utility requirements.
Catalyst replacement requirements have a large effect on revenue
requirements as can be seen when comparing the three 50% NOX reduction
cases. The cost is high and replacement must be made annually based on the
Hitachi Zosen guarantee of a one-year catalyst life for applications to coal-
fired flue gas. Since a one-year catalyst life is guaranteed it was used as a
basis for the cost estimate; however, actual catalyst life could be longer.
Should a two-year life be obtainable for the Hitachi Zosen process a
significant savings can be realized in the annual revenue requirements.
Table 23 shows that a two-year catalyst life will reduce the levelized annual
revenue requirements of the Hitachi Zosen (50% NOX reduction), ALNB/Hitachi
78
-------
TABLE 22. CONTRIBUTION OF RAW MATERIALS AND UTILITIES TO ANNUAL REVENUE REQUIREMENTS
50%
reduction
ALNB Exxon
Hitachi Zosen
% of % of
Raw materials
NH3
Catalyst
Total raw material cost
Utilities
Steam
Electricity
Subtotal of raw materials
and utilities
Other costs
MS RR MS
0.7
-^—
0.7
0.1
0^1
1.5
0.5 100 1^9
RR
21
—
21
3
21
44
56
M$
0.2
4_,3
4.5
0.1
0^
5.1
-L.9
% of
RR
3
54
57
1
_6
64
36
ALNB/
Hitachi Zosen
% of
MS
0.2
5^8
6.0
0.1
0^
6.7
JL*
RR
2
50
52
1
_5
58
42
90% reduction
Exxon/
Hitachi Zosen
M$
0.9
6.7
0.2
8.2
6.0
% of
RR
6
42
48
1
58
42
Hitachi
M$
0.4
7.9
0.1
0^6
8.6
4,1
Zosen
% of
RR
3
56
59
1
_5
65
35
First-year annual revenue
requirements (RR)
Levelized annual revenue
requirements
0.5
0.5
3.4
5.1
8.0
13.0
11.5
18.4
14.2
22.6
13.3
21.9
-------
Zosen, and Hitachi Zosen (90% NOX reduction) processes by 30% and the
Exxon/Hitachi Zosen process by 24%. Even though this is a significant
reduction in levelized annual revenue requirements for the processes
containing catalyst, it is not sufficient to change the cost relationship of
the processes. For 50% NOX reduction the ALNB is still lowest followed by
Exxon and Hitachi Zosen and for 90% NOX reduction the ALNB/Hitachi Zosen
remains the lowest followed by Hitachi Zosen and Exxon/Hitachi Zosen.
TABLE 23. THE EFFECT OF CATALYST LIFE ON
ANNUAL REVENUE REQUIREMENTS
Process
Annual
Hitachi Zosen catalyst
replacement cost. M$
1-year 2-year
catalyst catalyst
life life
Levelized
annual revenue
requirements. M$
1-year 2-year
catalyst catalyst
life life
50% reduction
Hitachi Zosen
90% reduction
ALNB/Hitachi Zosen
Exxon/Hitachi Zosen
Hitachi Zosen
4.3
5.8
5.8
7.5
2.1
2.9
2.9
3.7
13.0
18.4
22.6
21.9
9.0
12.9
17.1
14.8
Since the Exxon process is a large consumer of NHg and electricity
relative to the ALNB, the Exxon/Hitachi Zosen process has annual revenue
requirements that are higher than the ALNB/Hitachi Zosen process.
The NH3 consumption for the Exxon process is one and one-half times
that of the Hitachi Zosen 90% NOX reduction process and almost three times
that of the Hitachi Zosen 50% NOX reduction process. Therefore, as shown in
Table 22, the annual NH3 costs are greater for the Exxon process than for
the Hitachi Zosen process in direct proportion to the consumption of NH3.
As shown in Table 22, there are no significant differences in steam and
electricity costs for the Exxon and Hitachi Zosen processes. The ALNB has no
charges for utilities.
Annual revenue requirements for 50% NOX reduction are considerably less
than that for 90% reduction. The ratio of the lowest 90% reduction levelized
annual revenue requirements to the lowest 50% reduction levelized annual
revenue requirement is approximately 34 to 1.
80
-------
As can be seen in Figure 29, 50% NOX reduction cost is also lower in
dollars per pound of N(>2 removed, with the exception of the Hitachi Zosen
process. The main reason for the lower cost with 50% NOZ reduction, as
explained earlier, is that the ALNB and Exxon processes do not require the
expensive annual catalyst replacement needed for the Hitachi Zosen
process. The Hitachi Zosen 50% NOX reduction case has a slightly higher
dollar per pound of N(>2 removed cost than the 90% reduction case for two
reasons. First, there is some economy of scale in the capital investment for
the 90% reduction case compared with the 50% reduction case. Therefore, the
capital charges and maintenance, which are factored from the capital
investment, are a smaller portion of the annual revenue requirements for the
90% reduction case. Also, certain cost items, such as labor, are the same for
both 50% and 90% reduction cases making them a smaller cost (per pound of
NOX basis) for the 90% reduction case.
OVERALL CAPITAL INVESTMENT AND ANNUAL REVENUE REQUIREMENTS COMPARISON
Comparisons of the capital investment and levelized annual revenue
requirements for each of the six NOX control processes are shown in
Figures 30 and 31. Also included in the figures are the effect of the
accuracy range on the capital investment and levelized annual revenue
requirements.
For 50% NOX reduction the ALNB has the lowest capital investment and
levelized annual revenue requirements. The Exxon process has the second
lowest and Hitachi Zosen has the highest capital investment and levelized
annual revenue requirements.
The capital investment of the ALNB/Hitachi Zosen and the Hitachi Zosen
processes is almost equal for 90% NOX reduction, but the levelized annual
revenue requirements are lower for the ALNB/Hitachi Zosen process. In
comparison with the above two processes, the Exxon/Hitachi Zosen process
capital investment is substantially higher; however, the Exxon/Hitachi Zosen
levelized revenue requirement is comparable.
Capital investment and levelized annual revenue requirements are
significantly higher for 90% NOX reduction than for 50% NOX reduction.
ENERGY CONSUMPTION
Energy consumption for all of the NOX reduction cases studied is less
than 1% of the boiler capacity, as shown in Table 24. Energy requirements for
the three 50% reduction cases range from none for the ALNB to 0.4% of the
boiler capacity for the Exxon process. The range for 90% reduction is from
0.4% of the boiler capacity for the ALNB/Hitachi Zosen and Hitachi Zosen
processes to 0.7% of the boiler capacity for the Exxon/Hitachi Zosen process.
81
-------
H-
•8
i-i
(D
PROCESS
hO
l*O
,
O
0
en
rt
t-h
O
hi
hj
CD
C
n
ft ^-.a
H. °
0
s, ?
pi M
1 ?
3 r^
&•
O
Ml O
a ON
X -o>
cf CH o
en oo
0) o
o a H-
3 0
K3 O
h-»
ft a H"
i i _| B
H- O N5
N H
fD O
i — »
01
3 -P-
3
C
Oi
h-1 H-.
t
i™i QV,
ft)
0)
g
f 00
'""i
fD
p
H- 0
h{
fD
3
ro
3
•
"
•
"
m
.
PC
H-
rt
01
O
H-
N
O
cn
ro
3
0
s^s
••••
1
X
X
0
3
PS
H-
rt
n
H-
IN
0
en
(D
3
•••
1
*£>
h-<
a
w
•^^
H-
rt
O
3"
H-
N
0
en
3
^•H
i
|
1
o
0
a
o
*
"73
fD
C
O
rt
H-
0
3
1
[T]
H-
rt
01
O
H-
N
O
cn
fD
3
On
0
5-9
1
M
o a
3 td
LJ
•
.
o
0
a
o m
w
fD
O "
rt
H-
O
1 1
-------
CO
00
OJ
ALNB
Exxon
Hitachi Zosen 50%
ALNB/Hitachi Zosen
Exxon/Hitachi Zosen
Hitachi Zosen 90%
I I
I I
\ I I I I I
50% NOY Reduction
A.
I I
90% NOX Reduction
I I I
10 20 30 40 50 60 70 80 90 100 110 120
CAPITAL INVESTMENT, S/kW
Figure 30. Capital investment comparison and accuracy range (based on a -20% to +40% range for
Exxon an.c Hitachi Zosen processes and a -20% to +100% range for the ALNB).
-------
00
w
o
P-I
ALNB
Exxon
Hitachi Zosen 50%
ALNB/Hitachi Zosen
Exxon/Hitachi Zosen
Hitachi Zosen 90%
ID
50% NOX Reduction
90% NOY Reduction
A.
123456789 10 11
LEVELIZED ANNUAL REVENUE REQUIREMENTS, mills/kWh
12
Figure 31. Levelized annual revenue requirements and accuracy range (based on a -20% to +40%
range for Exxon and Hitachi Zosen processes and a -20% to +100% range for the ALNB)
-------
TABLE 24. COMPARISON OF ENERGY REQUIREMENTSa
Process
MBtu/hr
Electricity,
MBtu/hr
Total equivalent
energy consumption,b
percent of
boiler capacity
50% Reduction
Advanced Low-N0x Burner
Exxon
Hitachi Zosen
90% Reduction
0.0
5.7
6.0
0.0
11.5
7.7
0.0
0.4
0.3
ALNB/Hitachi Zosen
Exxon/Hitachi Zosen
Hitachi Zosen
8.0
11.7
10.9
10
21
10
.2
.7
.3
0.4
0.7
0.4
a
b
Does not include energy
. Based on a 500-MW boiler
tion of electricity, and
steam.
requirement represented
, a gross heat rate of 9
a boiler efficiency of
by raw materials.
,500 Btu/kWh for genera-
90% for generation of
Two of the three 90% NOX reduction processes, the ALNB/Hitachi Zosen
and the Hitachi Zosen, have estimated energy consumptions equivalent to the
Exxon process (50% NOX reduction) but the Exxon/Hitachi Zosen process is
higher at 0.7% of the boiler capacity.
The NOX control alternatives containing the Exxon process are the
highest energy consumers at both the 50% and 90% NOX reduction levels
because of energy consumption of the large air compressors in the
storage and injection section.
85
-------
CONCLUSIONS
The economic conclusions of this study are based on NOZ control
technology at various early stages of development applied to a new power
plant. Further development and retrofit applications could greatly alter hoth
the absolute and relative costs of the processes. To develop accurate and
timely economics in this rapidly evolving field, continued monitoring of
developments in NOX control technology is necessary.
For moderate NOX reduction of 50%, the ALNB is by far the most
economical alternative, even if its costs are to increase several times
relative to the other processes.
The Hitachi Zosen process has a higher capital investment than the
Exxon process at the 50% reduction level because of the initial catalyst
charge, reactor, additional ductwork, and additional fan capacity required.
It also has higher revenue requirements, primarily because of annual catalyst
replacement costs, although its NH3 requirements are much less than those of
the Exxon process. Changes in catalyst cost or NH3 consumption or cost
would appreciably affect the cost relationship of these processes.
The royalties for the Exxon process are a significant portion of the
capital investment.
For high NOX reductions of 90% the combination of the ALNB/Hitachi
Zosen process is the most cost effective alternative. Although the capital
investment for the ALNB/Hitachi Zosen process is slightly higher than the
capital investment for the Hitachi Zosen process, the annual revenue
requirements are substantially lower. The magnitude of the difference in
annual revenue requirements is large enough to overcome the slightly higher
capital investment and make the ALNB/Hitachi Zosen process the most
economically attractive.
The energy requirements for achieving 50% and 90% NOX reduction are
greater than that required for particulate removal except for the ALNB.
However, energy requirements for NOX reduction are still modest, much less
than 1% of the boiler output, in comparison with the energy needed for removal
of S02 from flue gas.
Catalyst cost is a very important economic factor with an SCR-type
process. With the Hitachi Zosen process, the catalyst cost may represent as
much as 25% of the capital investment and 35% of the levelized annual revenue
requirement.
86
-------
Catalyst life is also a very important economic factor. A two-year
catalyst life will reduce the levelized annual revenue requirements of the
Hitachi Zosen process by about 30%.
Since current technology requires SCR-type systems to achieve low
emission levels (0.06 Ib N02/MBtu), the cost for obtaining these low levels
versus more moderate emission levels (0.3 Ib N02/MBtu) are substantially
greater. To achieve low emission levels as compared with moderate levels,
based on the most economical alternatives, could require about a tenfold
increase in capital investment and about a thirtyfold increase in levelized
annual revenue requirements.
87
-------
REFERENCES
1. Jumpei Ando. NOZ Abatement for Stationary Sources in Japan.
EPA-600/7-79-205, U.S. Environmental Protection Agency, Washington, D.C.,
1979.
2. B&W. Steam/Its Generation and Use. B&W, New York, 1975.
3. E. J. Campobenedetto. B&W, Barberton, Ohio, Private Communication, April
1980.
4. E. J. Campobenedetto. Field Evaluation of Low Emission Coal Burner
Technology on a Utility Boiler, in: Proceedings of the Joint Symposium
on Stationary Combustion NOX Control, Vol. I, IERL RTP-1083, U.S.
Environmental Protection Agency, Washington, D.C., 1980, pp. 209-250.
5. C. Castaldini, K. G. Salvesen, and H. B. Mason. Technical Assesment of
Thermal DeNOx Process. EPA-600/7-79-117, U.S. Environmental Protection
Agency, Washington D.C., 1979.
6. Chemical Engineering. Economic Indicators. Vol. 83 to 86, McGraw-Hill,
Inc., New York, 1976-1979.
7. J. N. Driscoll. Flue Gas Monitoring Techniques and Manual Determination
of Gaseous Pollutants. Ann Arbor Science Publishers, Ann Arbor,
Michigan, 1974, pp. 219-263.
8. EPRI. Technical Assessment Guide. EPRI PS-866-SR, Special Report,
Electric Power Research Institute, Palo Alto, California, 1978.
9. H. L. Faucett, J. D. Maxwell, and T. A. Burnett. Technical Assessment
of NOX Removal Processes for Utility Application. TVA Bulletin Y-120,
Tennessee Valley Authority, Muscle Shoals, Alabama; EPA-600/7-77-127
(NTIS PB 276 637/6WP), U.S. Environmental Protection Agency, Washington,
D.C. ; EPRI FP-1253, Electric Power Research Institute, Palo Alto,
California, 1977.
10. M. H. Heap, T. M. Lowes, R. Walmsley, H. Bartelds, and P. LeVaguerese.
Burner Criteria for NOX Control, Vol. 1, EPA-600/2-76-061a, U.S.
Environmental Protection Agency, Washington, D.C., 1976.
11. M. H. Heap. Unpublished data presented at the Second Technology Transfer
Panel Meeting, Energy and Environmental Research Corporation, Newport
Beach, California, November 7, 1979.
88
-------
12. P. -H. Jeynes. Profitability and Economic Choice, First Edition, The
Iowa State University Press, Ames, Iowa, 1968.
13. G. D. Jones. Selective Catalytic Reduction and NO* Control in Japan,
A Status Report, DCN 81-203-001-27-07, Radian Corporation, Austin, Texas;
EPA-600/7-81-030, U.S. Environmental Protection Agency, Washington, D.C.,
1981.
14. KVB, Incorporated. Assessment of NOX Control Technology for Coal-Fired
Utility Boilers, Appendix D. 1977 Technology Status Report, ANL/ECT-3,
Argonne National Laboratory, Argonne, Illinois, 1977.
15. K. J. Lim, L. R. Waterland, C. Castaldini, Z. Chiba, E. B. Higginbotham.
Environmental Assessment of Utility Boiler Combustion Modification NOx
Controls, Vol. 1, Technical Results/Vol. 2, Appendices, EPA-600/7-80-
075a&b, U.S. Environmental Protection Agency, Washington,
D.C., 1980.
16. J. J. Marshall and A. P. Selker. The Role of Tangential Firing and
Fuel Properties in Attaining Low NOx Operation for Coal-Fired Steam
Generation, in: Proceedings of the Second NOX Control Technology
Seminar, EPRI FP-1109-SR, Electric Power Research Institute, Palo Alto,
California, 1979, pp. 5-21 to 5-28.
17. G. B. Martin. U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina, Private Communication, April 1980.
18. G. B. Martin. Field Evaluation of Low NOx Coal Burners on Industrial
and Utility Boilers, in: Proceedings of the Third Stationary Source
Combustion Symposium, Vol. 1, EPA-600/7-79-050a, U.S. Environmental
Protection Agency, Washington, B.C., 1979, pp. 213-231.
19. L. J. Muzio, J. K. Arand, and K. L. Maloney. Noncatalytic NOx Removal
with Ammonia, EPRI FP-735, Electric Power Research Institute, Palo Alto,
California, 1978.
20. Pollution Control Industry News. Air/Water Pollution Report,
December 24, 1979, p. 518.
21. A. H. Rawdon, R. A. Lisauskas, and F. J. Zone. Design and Operation
of Coal-Fired Turbo Furnaces for NOx Control, in: Proceedings of the
Second NOX Control Technology Seminar, EPRI FP-1109-SR, Electric Power
Research Institute, Palo Alto, California, 1979, pp. 6-1 to 6-9.
22. The Richardson Rapid System. Process Plant Estimation Standards,
Vol. I, III, & IV, Richardson Engineering Services, Inc., Solano Beach,
California, 1978-1979 Edition.
23. R. E. Thompson. Nitric Oxide Controls for Coal-Fired Utility Boilers
from an Application Viewpoint, in: Proceedings of the Second NOX
Control Technology Seminar, EPRI FP-1109-SR, Electric Power Research
Institute, Palo Alto, California, 1979, pp. 9-1 to 9-27.
89
-------
24. V. W. Uhl. A Standard Procedure for Cost Analysis of Pollution Control
Operations, Vol. I & II, EPA-600/8-79-018a&b, U.S. Environmental
Agency, Washington, D.C., 1979.
25. U.S. Environmental Protection Agency. Standards of Performance for New
Stationary Sources, Federal Register, Vol. 36, No. 247, December 23,
1971, pp. 24876-24895.
26. U.S. Environmental Protection Agency. New Stationary Source
Performance Standards; Electric Utility Steam Generation Units,
Federal Register, Vol. 44, No. 113, June 11, 1979, pp. 33580-33624.
27. U.S. Environmental Protection Agency. National Air Pollutant Emission
Estimates, 1970-1978, EPA-450/4-80-002, 1980.
28. G. M. Varga, Jr., M. E. Tomsho, B. H. Ruterbories, 6. J. Smith, and
W. Bartok. Applicability of the Thermal DeNOx Process to Coal-Fired
Utility Boilers, EPA-600/7-79-079, U.S. Environmental Protection Agency,
Washington, D.C., 1979.
29. G. M. Varga, Jr., Exxon Research and Engineering Company, Linden, New
Jersey, Private Communication, June 1979.
30. J. Vatsky. Experience in Reducing NOx Emissions on Operating Steam
Generators, in: Proceedings of the Second NOZ Control Technology
Seminar, EPRI FP-1109-SR, Electric Power Research Institute, Palo Alto,
California, 1979, pp. 7-1 to 7-17.
31. J. Vatsky. Foster Wheeler Energy Corporation, Livingston, New Jersey,
Private Communication, March 1980.
32. P. W. Winkler. Chemico Air Pollution Control, New York, Private
Communication, October 1979.
33. D. M. Zallen, R. Gersham, M. P. Heap, and W. H. Nurick. The
Generalization of Low Emission Coal Burner Technology, in: Proceedings
of the Third Stationary Source Combustion Symposium, Vol. II,
EPA-600/7-79-050b, U.S. Environmental Protection Agency, Washington,
D.C., 1979, pp. 73-109.
90
-------
APPENDIX A
CAPITAL INVESTMENT AND ANNUAL REVENUE REQUIREMENT TABLES
Appendix A contains the capital investment and annual revenue requirement
tables for each of the processes evaluated in this study.
Process Page
Advanced Low-N0x Burner 92
Exxon Thermal DeNOx 94
Hitachi Zosen (50% NOX reduction) 96
Hitachi Zosen (80% NOX reduction) to be combined with the ALNB 98
Advanced Low-N0x Burner/Hitachi Zosen 100
Hitachi Zosen (80% NOX reduction) to be combined with Exxon 102
Exxon Thermal DeNOx/Hitachi Zosen 104
Hitachi Zosen (90% NOX reduction) 106
91
-------
TABLE A-l. CAPITAL INVESTMENT SHEET
ADVANCED LOW-NOX BURNER
Investment. $
Total fixed investment 1,888,000
Other Capital Investments
Allowance for startup and modifications 189,000
Interest during construction 295,000
Royalties 0
Land 0
Working capital 49.000
Total Capital investment 2,421,000
Dollars of total capital per kW of generating capacity 4.8
Basis: New 500-MY? north-central pulverized-coal-fired power unit;
9,500 Btu/kWh heat rate; 3.5% sulfur, 15.1% ash coal; 0.6 Ib/MBtu N02
equivalent uncontrolled emission; 20% excess air to furnace, 39%
total excess air; 1982 cost basis. Costs are the difference between
those of the ALNB design and those of a boiler with standard burners.
92
-------
TABLE A-2. ANNUAL REVENUE REQUIREMENTS
ADVANCED LOW-NOj BURNER
Total
annual
Direct Costs - First Year
Conversion costs
Operating labor and supervision
Utilities
Steam
Electricity
Maintenance
Labor and material
Analyses
Total conversion costs
Total direct costs
Indirect Costs - First Year
Overheads
Plant and administrative (60% of
conversion costs less utilities)
Marketing
Byproduct credit
Total first-year operating and
maintenance costs
Levelized capital charges (14.7% of total
capital investment)
Total first-year annual revenue require-
ments
Levelized capital charges (14.7% of total
capital investment)
Levelized first-year operating and mainte-
nance costs (1.886 first-year 0 and M)
Levelized annual revenue requirements
0
0
61,000
0
61,000
61,000
98,000
356.000
454,000
First-year annual revenue requirements
Levelized annual revenue requirements
0.45
0.54
Mills/tWh
0.17
0.20
Basis: Power unit as described in capital investment table, operating 5,500 hr/yr at full
load; 1984 cost basis. Costs are the difference between those of the ALMS design and those
of a boiler with standard burners.
93
-------
TABLE A-3. CAPITAL INVESTMENT SHEET
EXXON THERMAL DENOZ
Investment. $
Direct Investment
NH3 storage and injection 3,268,000
Air preheater section 585.000
Total process capital 3,853,000
Services, utilities, and miscellaneous 231.000
Total direct investment 4,084,000
Indirect Investment
Engineering design and supervision 286,000
Architect and engineering contractor 82,000
Construction expense 653,000
Contractor fees 204,000
Contingency 1.062.000
Total fixed investment 6,371,000
Other Capital Investments
Allowance for startup and modifications 637,000
Interest during construction 994,000
Royalties 1,526,000
Land 5,000
Working capital 337.000
Total capital investment 9,870,000
Dollars of total capital per VH of generating
capacity 19.7
Basis: New 500-MW north-central pulverized-coal-fired
power unit; 9,500 Btu/kWh heat rate; 3.5% sulfur, 15.1%
ash coal; 50% NOZ reduction from 0.6 Ib/MBtu N02 equiva-
lent uncontrolled emission; and 20% excess air to furnace,
39% total excess air; 1982 cost basis.
94
-------
TABLE A-4. ANNUAL REVENUE REQUIREMENTS
EXXON THERMAL DENOX
Direct Costs - First Year
Raw materials
NH3
Total raw materials cost
Conversion costs
Operating labor and supervision
Utilities
Steam
Electricity
Maintenance
Labor and material
Analyses
Total conversion costs
Total direct costs
Annual quantity
Unit
cost. $
4,351 tons
4,380 man-hr
27888.7 MBtu
18.460 z 106 kVh
2,190 man-hr
155/ton
15/man-hr
2.70/MBtu
0.037/kWh
21/man-hr
Total
annual
cost. S
674.400
674,400
65,700
75,300
683,000
204,200
46.000
1,074,200
1,748,600
Indirect Costs - First Year
Overheads
Plant and administrative (60% of
conversion costs less utilities)
Marketing
Byproduct credit
Total first-year operating and
maintenance costs
Levelized capital charges (14.7% of total
capital investment)
Total first-year annual revenue require-
ments
Levelized capital charges (14.7% of total
capital investment)
Levelized first-year operating and mainte-
nance costs (1.886 first-year 0 and M)
Levelized annual revenue requirements
189,500
0
0
1,938,100
1.450.900
3,389,000
1,450.900
3.655.300
5,106,200
First-year annual revenue requirements
Levelized annual revenue requirements
3.4
5.1
Mills/kWh
1.2
1.9
Basis: Power unit as described in capital investment table, operating 5,500 hr/yr at full
load; 1984 cost basis.
95
-------
TABLE A-5. CAPITAL INVESTMENT SHEET
HITACHI ZOSEN (50% NOX REDUCTION)
Investment. $
Direct Investment
NH3 storage and injection 494,000
Reactor section 2,179,000
Flue gas fans 2,131,000
Air preheater section 585.000
Total process capital 5,389,000
Services, utilities, and miscellaneous 323.000
Total direct investment 5,712,000
Indirect Investment
Engineering design and supervision 400,000
Architect and engineering contractor 114,000
Construction expense 914,000
Contractor fees 286,000
Contingency 1.485.000
Total fixed investment 8,911,000
Other Capital Investments
Allowance for startup and modifications 891,000
Interest during construction 1,390,000
Royalties 458,000
Land 8,000
Working capital 339,000
Catalyst 3.681.000
Total capital investment 15,678,000
Dollars of total capital per kW of generating
capacity 31.4
Basis: New 500-MW north-central pulverized-coal-fired
power unit; 9,500 Btu/kWh heat rate; 3.5% sulfur, 15.1%
ash coal; 50% NOZ reduction from 0.6 Ib/MBtu N02 equiva-
lent uncontrolled emission; and 20% excess air to furnace,
39% total excess air; 1982 cost basis.
96
-------
TABLE A-6. ANNUAL REVENUE REQUIREMENTS
HITACHI ZOSEN (50% NOZ Reduction)
Direct Costs - First Year
Raw materials
NH3
Catalyst
Total
Unit annual
Annual quantity cost, $ cost, S
1,612 tons 155/ton 249,900
4.254.100
Total raw materials cost
Conversion costs
Operating labor and supervision
Utilities
Steam
Electricity
Maintenance
Labor and material
Analyses
Total conversion costs
Total direct costs
4,380 man-hr
29922.2 MBtu
12.368 x 106 kWh
2,190 man-hr
15/man-hr
2.70/MBtn
0.037/kWh
21/man-hr
4,504,000
65,700
80,800
457,600
285,600
46.000
935,700
5,439,700
Indirect Costs - First Year
Overheads
Plant and administrative (60% of
conversion costs less utilities)
Marketing
Byproduct credit
Total first-year operating and
maintenance costs
Levelized capital charges (14.7% of total
capital investment)
Total first-year annual revenue require-
ments
Levelized capital charges (14.7% of total
capital investment)
Levelized first—year operating and mainte-
nance costs (1.886 first-year 0 and M)
Levelized annual revenue requirements
238,400
0
0
5,678,100
2.304.700
7.982,800
2.304,700
10.708.900
13,013,600
First-year annual revenue requirements
Levelized annual revenue requirements
_MJ_
8.0
13.0
Mills/kWh
2.9
4.7
Basis: Power unit as described in capital investment table, operating 5,500 hr/yr at full
load; 1984 cost basis.
97
-------
TABLE A-7. CAPITAL INVESTMENT SHEET
HITACHI ZOSEN (80% NOX REDUCTION)
(To be combined with the ALNB)
Investment. S
Direct Investment
NH3 storage and injection 582,000
Reactor section 4,195,000
Flue gas fans 3,051,000
Air preheater section 585.000
Total process capital 8,413,000
Services, utilities, and miscellaneous 505.000
Total direct investment 8,918,000
Indirect Investment
Engineering design and supervision 624,000
Architect and engineering contractor 178,000
Construction expense 1,427,000
Contractor fees 446,000
Contingency 2.319.000
Total fixed investment 13,912,000
Other Capital Investments
Allowance for startup and modifications 1,391,000
Interest during construction 2,170,000
Royalties 458,000
Land 8,000
Working capital 485,000
Catalyst 5.034.000
Total capital investment 23,458,000
Dollars of total capital per kW of generating
capacity 46.9
Basis: New 500-MW north-central pulverized-coal-fired
power unit; 9,500 Btu/kWh heat rate; 3.5% sulfur, 15.1%
ash coal; 80% NOX reduction from 0.3 Ib/MBtu N02 equiva-
lent emission after 50%' reduction from 0.6 Ib/MBtu N02
equivalent with the ALNB; 20% excess air to furnace, 39%
total excess air; 1982 cost basis.
98
-------
TABLE A-8. ANNUAL REVENUE REQUIREMENTS
HITACHI ZOSEN (80% NOZ Reduction)
(To be combined with the ALNB)
Direct Costs - First Year
Raw materials
NH3
Catalyst
Total raw materials cost
Conversion costs
Operating labor and supervision
Utilities
Steam
Electricity
Maintenance
Labor and material
Analyses
Total conversion costs
Total direct costs
Indirect Costs - First Year
Overheads
Plant and administrative (60% of
conversion costs less utilities)
Marketing
Byproduct credit
Total first-year operating and
maintenance costs
Levelized capital charges (14.7% of total
capital investment)
Total first-year annual revenue require-
ments
Levelized capital charges (14.7% of total
capital investment)
Levelized first-year operating and mainte-
nance costs (1.886 first-year 0 and M)
Levelized annual revenue requirements
Total
Unit annual
Annual quantity cost. $ cost. $
1,306 tons 155/ton 202,400
5,817,700
6,020,100
4,380 man-hr 15/man-hr 65,700
39,532.8 MBtn 2.70/MBtn 106,700
16.471 z 106 kWh 0.037/kWh 609,400
445,900
2,190 man-hr 21 /man-hr 46.000
1,273,700
7,293,800
334,600
0
0
7,628.400
3.448.300
11,076,700
3,448,300
14,387,200
17,835,500
First-year annual revenue requirements
Levelized annual revenue requirements
JLL
11.1
17.8
Mills/kWh
4.0
6.5
Basis: Power unit as described in capital investment table, operating 5,500 hr/yr at full
load; 1984 cost basis.
99
-------
TABLE A-9. CAPITAL INVESTMENT SHEET
ADVANCED LOW-NOX BURNER/HITACHI ZOSEN
Investment. $
ALNB fixed investment 1,888,000
Hitachi Zosen fixed investment 13.912.000
Total fixed investment 15,800,000
Other Capital Investments
Allowance for startup and modifications 1,580,000
Interest during construction 2,465,000
Royalties 458,000
Land 8,000
Working capital 534,000
Catalyst 5.034.000
Total Capital investment 25,879,000
Dollars of total capital per kW of generating capacity 51.8
Basis: New 500-MW north-central pulverized-coal-fired power unit;
9,500 Btu/kWh heat rate; 3.5% sulfur, 15.1% ash coal; 90% NOX
reduction from 0.6 Ib/MBtu uncontrolled emission; 20% excess air to
furnace, 39% total excess air; 1982 cost basis.
100
-------
TABLE A-10. ANNUAL REVENUE REQUIREMENTS
ADVANCED LOW-NOX BURNER/HITACHI ZOSEN
Direct Costs - First Year
Raw materials
NH3
Catalyst
Total raw materials cost
Conversion costs
Operating labor and supervision
Utilities
Steam
Electricity
Maintenance
Labor and material
Analyses
Total conversion costs
Total direct costs
Indirect Costs - First Year
Overheads
Plant and administrative (60% of
conversion costs less utilities)
Marketing
Byproduct credit
Total first-year operating and
maintenance costs
Levelized capital charges (14.7% of total
capital investment)
Total first-year annual revenue require-
ments
Levelized capital charges (14.7% of total
capital investment)
Levelized first-year operating and mainte-
nance costs (1.886 first-year 0 and M)
Levelized annual revenue requirements
Total
Unit annual
Annual quantity cost. $ cost. $
1,306 tons 155/ton 202,400
5,817,700
6.020,100
4,380 man-hr 15/man-hr 65,700
39,532.8 MBtn 2.70/MBtn 106,700
16.471 x 106 kWh 0.037/kWh 609,400
506,900
2,190 man-hr 21 /man-hr 46.000
1,334,700
7,354,800
371,600
0
0
7,726,400
3.804,300
11,530,700
3,804,300
14.572.200
18,376,500
First-year annual revenue requirements
Levelized annual revenue requirements
M$
11.5
18.4
Mills/tWh
4.2
6.7
Basis: Power unit as described in capital investment table, operating 5,500 hr/yr at full
load; 1984 cost basis.
101
-------
TABLE A-ll. CAPITAL INVESTMENT SHEET
HITACHI ZOSEN (80% NOX REDUCTION)
(To be combined with Exxon)
Investment. $
Direct Investment
NH3 storage and injection 582,000
Reactor section 4,195,000
Flue gas fans 3.051.000
Total process capital 7,828,000
Services, utilities, and miscellaneous 470.000
Total direct investment 8,298,000
Indirect Investment
Engineering design and supervision 581,000
Architect and engineering contractor 166,000
Construction expense 1,328,000
Contractor fees 415,000
Contingency 2.158.000
Total fixed investment 12,946,000
Other Capital Investments
Allowance for startup and modifications 1,295,000
Interest during construction 2,020,000
Royalties 458,000
Land 8,000
Working capital 457,000
Catalyst 5.034.000
Total capital investment 22,218,000
Dollars of total capital per kW of generating
capacity 44.4
Basis: New 500-MW north-central pulverized-coal-fired
power unit; 9,500 Btu/kWh heat rate; 3.5% sulfur, 15.1%
ash coal; 80% NOX reduction from 0.3 Ib/MBtu N02 equiva-
lent emission after 50% reduction from 0.6 Ib/MBtu N02
equivalent with Thermal DeNOz; 20% excess air to furnace,
39% total excess air; 1982 cost basis.
102
-------
TABLE A-12. ANNUAL REVENUE REQUIREMENTS
HITACHI ZOSEN (80% NOZ Reduction)
(To be combined with Exxon)
Direct Costs - First Year
Raw materials
NH3
Catalyst
Total
Unit annual
Annual Quantity cost. $ cost. $
1,306 tons 155/ton 202.400
5.817.700
Total raw materials cost
Conversion costs
Operating labor and supervision
Utilities
Steam
Electricity
Maintenance
Labor and material
Analyses
Total conversion costs
Total direct costs
4,380 man-hr
29,464.2 HBtu
16.471 x 106 kWh
2,190 man-hr
15/man-hr
2.70/MBtu
0.037/kwh
21/man-hr
6,020,100
65,700
79,600
609,400
414.900
46.000
1,215,600
7,235,700
Indirect Costs - First Year
Overheads
Plant and administrative (60% of
conversion costs less utilities)
Marketing
Byproduct credit
Total first-year operating and
maintenance costs
Levelized capital charges (14.7% of total
capital investment)
Total first-year annual revenue require-
ments
Levelized capital charges (14.7% of total
capital investment)
Levelized first-year operating and mainte-
nance costs (1.886 first-year 0 and M)
Levelized annual revenue requirements
316,000
0
0
7,551.700
3.266.000
10,817,700
3,266,000
14.242.500
17,508,500
First-year annual revenue requirements
Levelized annual revenue requirements
10.8
17.5
Mills/kWh
3.9
6.4
Basis: Power unit as described in capital investment table, operating 5,500 hr/yr at full
load; 1984 cost basis.
103
-------
TABLE A-13. CAPITAL INVESTMENT SHEET
EXXON THERMAL DENOX/HITACHI ZOSEN
Investment. $
Direct Investment
NH3 storage and injection 3,850,000
Reactor section 4,195,000
Flue gas fans 3,051,000
Air preheater section 585.000
Total process capital 11,681,000
Services, utilities, and miscellaneous 701.OOP
Total direct investment 12,382,000
Indirect Investment
Engineering design and supervision 867,000
Architect and engineering contractor 248,000
Construction expense 1,981,000
Contractor fees 619,000
Contingency 3.220.000
Total fixed investment 19,317.000
Other Capital Investments
Allowance for startup and modifications 1,932,000
Interest during construction 3,014,000
Royalties 1,984,000
Land 13.000
Working capital 794,000
Catalyst 5.034.000
Total capital investment 32,088,000
Dollars of total capital per kW of generating
capacity 64.2
Basis: New 500-HW north-central pulverized-coal-fired
power unit; 9,500 Btu/kWh heat rate; 3.5% sulfur, 15.1%
ash coal; 90% NOZ reduction from 0.6 Ib/MBtu N02 equiva-
lent uncontrolled emission; 20% excess air to furnace,
39% total excess air; 1982 cost basis.
104
-------
TABLE A-14. ANNUAL REVENUE REQUIREMENTS
EXXON THERMAL DENOZ/HITACHI ZOSEN
Annual quantity
Unit
cost, $
Total
annual
cost, $
Direct Costs - First Year
Raw materials
NH3
Catalyst
Total raw materials cost
Conversion costs
Operating labor and supervision
Utilities
Steam
Electricity
Maintenance
Labor and material
Analyses
Total conversion costs
Total direct costs
5,657 tons
8,760 man-hr
57,352.9 MBtn
34.931 i 106 kWh
4,380 man-hr
155/ton
15/man-hr
2.70/MBtu
0.037/kWh
21/man-hr
876,800
5.817.700
6,694,500
131,400
154,900
1,292,400
619,100
92.000
2,289,800
8,984,300
Indirect Costs - First Year
Overheads
Plant and administrative (60% of
conversion costs less utilities)
Marketing
Byproduct credit
Total first-year operating and
maintenance costs
Levelized capital charges (14.7% of total
capital investment)
Total first-year annual revenue require-
ments
Levelized capital charges (14.7% of total
capital investment)
Levelized first-year operating and mainte-
nance costs (1.886 first-year 0 and M)
Levelized annual revenue requirements
505,500
0
0
9,489,800
4.716.900
14,206,700
4,716,900
17.897.800
22,614,700
First-year annual revenue requirements
Levelized annual revenue requirements
14.2
22.6
Mills/tWh
5.2
8.2
Basis: Power unit as described in capital investment table, operating 5,500 hr/yr at full
load; 1984 cost basis.
105
-------
TABLE A-15. CAPITAL INVESTMENT SHEET
HITACHI ZOSEN (90* NOX REDUCTION)
Investment. $
Direct Investment
NH3 storage and injection 840,000
Reactor section 4,195,000
Flue gas fans 3,051,000
Air preheater section 585.000
Total process capital 8,671,000
Services, utilities,, and miscellaneous 520.000
Total direct investment 9,191,000
Indirect Investment
Engineering design and supervision 643,000
Architect and engineering contractor 184,000
Construction expense 1,471,000
Contractor fees 460,000
Contingency 2.390.000
Total fixed investment 14,339,000
Other Capital Investments
Allowance for startup and modifications 1,434,000
Interest during construction 2,237,000
Royalties 458,000
Land 8,000
Working capital 522,000
Catalyst 6.473.000
Total capital investment 25,471,000
Dollars of total capital per kW of generating
capacity 50.9
Basis: New 500-MW north-central pulverized-coal-fired
power unit; 9,500 Btu/kVh heat rate; 3.5% sulfur, 15.1%
ash coal; 90% NOZ reduction from 0.6 Ib/MBtn N02 equiva-
lent uncontrolled emission; 20% excess air to furnace,
39% total excess air; 1982 cost basis.
106
-------
TABLE A-16. ANNUAL REVENUE REQUIREMENTS
HITACHI ZOSEN (90% NOZ Redaction)
Annual Quantity
Unit
cost, $
Total
annual
cost, $
Direct Costs - First Year
Raw materials
NH3
Catalyst
Total raw materials cost
Conversion costs
Operating labor and supervision
Utilities
Steam
Electricity
Maintenance
Labor and material
Analyses
Total conversion costs
Total direct costs
2,901 tons
4,380 man-hr
53860.6 MBtn
16.533 z 106 kWh
2,190 man-hr
155/ton
15/man-hr
2.70/MBtu
0.037/kWh
21/man-hr
449,700
7.479.600
7.929,300
65,700
145,400
611,700
459,600
46.000
1,328,400
9,257,700
Indirect Costs - First Year
Overheads
Plant and administrative (60% of
conversion costs less utilities)
Marketing
Byproduct credit
Total first-year operating and
maintenance costs
Levelized capital charges (14.7% of total
capital investment)
Total first-year annual revenue require-
ments
Levelized capital charges (14.7% of total
capital investment)
Levelized first-year operating and mainte-
nance costs (1.886 first-year 0 and M)
Levelized annual revenue requirements
342,800
0
0
9,600,500
3.744.200
13,344,700
3,744,200
18.106.500
21,850,700
First-year annual revenue requirements
Levelized annual revenue requirements
13.3
21.9
Mills/kWh
4.9
7.9
Basis: Power unit as described in capital investment table, operating 5,500 hr/yr at full
load; 1984 cost basis.
107
-------
APPENDIX B
CALCULATION OF PROCESS CAPITAL
Below is an illustration of how the process capital is obtained for the
capital investment sheet from the equipment list. The NB^ storage and
injection area of the Hitachi Zosen 90% NOX reduction process is used as an
example.
The equipment list shown in Table B-l represents the major equipment
items in the NH3 storage and injection section. These equipment items are
costed on an erected basis, that is, the cost includes the labor required to
place the equipment in position ready for operation. The total ($373,900)
represents the total untaxed cost of the process equipment and is used as the
basis for estimating the field equipment cost shown in Table B-2.
As can be seen in Table B-2, the untaxed field equipment cost (column B)
is estimated as a percentage (column A) of the NH3 storage and injection
process equipment subtotal ($373,900). Untaxed field equipment cost is then
broken down into material and labor by using the labor to material ratio
(column C) . Materials (column D) are then taxed at 4% to obtain sales tax
(column E) which is added to the untaxed field equipment cost (column B)
giving the field equipment cost (column F).
No taxes are charged to paint since material cost is insignificant
compared with labor cost. Concrete foundations and excavation are not
factored. Concrete is estimated for each equipment item and then
totaled. Excavation is based on the quantity of land required.
Freight and sales tax on process equipment are also added to obtain the
area direct investment. The weight of each piece of equipment is approximated
and the total freight cost is calculated on a cost per weight basis. A sales
tax of 4% is applied to the materials portion of process equipment.
The costs for process equipment, field equipment, freight, and process
equipment sales tax are summed to give the area investment for NHg storage
and injection.
Process equipment cost $373,900
Field equipment cost $395,500
Freight (based on estimated equipment weight) $ 57,300
Process equipment sales tax $ 12.900
Area investment $839,600
The area investment is then rounded to $840,000 which is the NH3
storage and injection investment listed in Table A-15.
109
-------
TABLE B-l. HITACHI ZOSEN (90% NOX REDUCTION)
EQUIPMENT LIST
Total equipment
Item (number); description cost. 1982 $
Area 1—NH3 Storage and Injection
1. Compressor. NH3 unloading (2): Single cylinder, double 61,500
acting, 300 sft3/min at 250 psig, 30 psig suction, 125 hp,
cast iron
2. Tank . NH3 storage (9): Horizontal, 9 ft dia x 66 ft long, 207,900
30,000 gal, 250 psig, carbon steel
3. Vaporizer. NH3 (1): Steam at 298°F, tube type, 10 ft2, 7,400
0.33 MBtu/hr, carbon steel
4. Blower. NH3 and air (3): 3,950 aft3/min, AP 15 in. H20, 17,900
15 hp, carbon steel
5. Injection grid. NH3 and air (2): 74,600
6. Pump. NH3 (2): 2 gpm, 0.5 hp, 28 ft head, carbon steel 4.600
Subtotal 373,900
110
-------
TABLE B-2. BREAKDOWN OF FIELD EQUIPMENT COST
Field equipment
Piping and insulation
Concrete foundations
Excavation, site
preparation roads
Structural
Electrical
Instrumentation
Duct, chutes, expansion
joints
Paint and miscellaneous
Total
A
% of
area 1
subtotal
20
-
-
5
20
15
10
4
B
Untaxed field
equipment cost
(labor + material)
74,780
103,300
-
18,695
74,780
56,085
37,390
14,956
C
Labor /material
1.2
2.756
-
1.7
2.0
0.484
7.0
-
D
Material
33,99ia
27,503
-
6,924
24,927
37,793
4,674
-
E
Sales tax on
material (4%)
1,360
1,100
-
277
997
1,512
187
-
F
Field
equipment
cost
76,100
104,400
10,000
19,000
75,800
57,600
37,600
15.000
395,500
a. Example calculation of material:
material + labor = untaxed field equipment cost
material + 1.2 material = $74,780
material = $74,780/2.2 = $33,991
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/7-81-120
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Evaluation of the Advanced Low-NOx Burner, Exxon,
and Hitachi Zosen DeNOx Processes
5. REPORT DATE
July 1981
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
J.D. Maxwell and L.R. Humphries
8. PERFORMING ORGANIZATION REPORT NO.
TVA/OP/EDT-81/28
9. PERFORMING ORGANIZATION NAME AND ADDRESS
TVA, Office of Power
Division of Energy Demonstration and Technology
Muscle Shoals, Alabama 35660
10. PROGRAM ELEMENT NO.
INE829
11. CONTRACT/GRANT NO.
EPA-IAG-79-D-X0511
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final; 11/79-6/81
14. SPONSORING AGENCY CODE
EPA/600/13
15. SUPPLEMENTARY NOTES
541-2578.
IERL-RTP project officer is J. David Mobley, Mail Drop 61, 919/
16. ABSTRACT
The report is a technical discussion and preliminary economic evaluation
of six NOx control methods: three at 50% NOx reduction, and three at 90%. The base-
case power plant is a new 500-MW coal-fired unit emitting 0. 6 Ib NO2/million Btu in
the flue gas. The three 50% NOx reduction processes are the EPA-sponsored advan-
ced low-NOx burner (ALNB), the Exxon Thermal DeNOx process, and the Hitachi
Zosen process, which have capital investments of $4.8, $19. 7, and $31.4/kW, re-
spectively, and levelized annual revenue requirements of 0. 20, 1. 9, and 4. 7 mills/
kWh, respectively. For 90% NOx reduction, the ALNB process is combined with the
Hitachi Zosen process, the Exxon process is combined with the Hitachi Zosen pro-
cess, and the Hitachi Zosen process is used alone. Capital investment and levelized
annual revenue requirements for these three processes are $51. 8/kW and 6.7 mills/
kWh for the ALNB/Hitachi Zosen process $64. 2/kW and 8. 2 mills AWh for the
Exxon/Hitachi Zosen process, and $50. 9/kW and 7.. 9 mills/kWh for the Hitachi
Zosen process alone. The ALNB, a combustion modification, is the least expensive
NOx control method. As expected, the costs for obtaining high levels of NOx reduc-
tion (90%) are significantly greater than for more moderate levels (50%).
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Pollution
Nitrogen Oxides
Coal
Combustion
Pollution Control
Stationary Sources
Advanced Low-NOx Bur-
ner (ALNB)
Exxon Thermal DeNOx
Hitachi Zosen
Combustion Modification
13 B
07B
08G,21D
21B
13. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
140
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
112
-------
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Environmental Research Information Center
Cincinnati. Ohio 45268
OFFICIAL BUSINESS
PENALTY FOR PRIVATE USE S3OO
AN EQUAL OPPORTUNITY EMPLOYER
POSTAGE AND FEES PAID
US ENVIRONMENTAL PROTECTION AGENCY
EPA-335
// your address is incorrect, please change on the above label
tear off- and return to the above address.
If you do not desire to continue receiving these technical
reports, CHECK HERE O; tear off label, and return it to the
above address.
------- |