United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park NC 2771 1
EPA-600/7-78-163
August 1978
Effect of SO2
Emission
Requirements on
Fluidized-bed
Combustion
Systems:
Preliminary
Technical/Economic
Assessment
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.
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-600/7-78-163
August 1978
Effect of SO2 Emission
Requirements on Fiuidized-bed
Combustion Systems:
Preliminary Technical/Economic
Assessment
by
R.A. Newby, N.H. Ulerich, E.P. O'Neill,
D.F. Ciliberti, and D.L. Keairns
Westinghouse Research and Development Center
1310 Beulah Road
Pittsburgh, Pennsylvania 15235
Contract No. 68-02-2132
Program Element No. EHE623A
EPA Project Officer; D. Bruce Henschel
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Ofice of Research and Development
Washington, DC 20460
-------
PREFACE
The Westinghouse R&D Center is carrying out a program to provide
experimental and engineering support for the development of fluidized-
bed combustion systems under contract to the Industrial Environmental
Research Laboratory, U. S. Environmental Protection Agency (EPA), at
Research Triangle Park, NC. The contract scope includes atmospheric and
pressurized fluidized-bed combustion processes as they may be applied
for steam generation, electric power generation, or process heat. Specific
tasks include work on calcium-based sulfur removal systems (e.g., sorp-
tion kinetics, regeneration, attrition, modeling), alternative sulfur
sorbents, nitrogen oxide emissions, particulate emissions and control,
trace element emissions and control, spent sorbent and ash disposal, and
systems evaluation (e.g., impact of new source performance standards on
fluidized-bed combustion system design and cost).
This document contains the results of a brief technical effort,
defined and completed under a technical directive, from August 1977 to
January 1978. The work reported represents a preliminary application of
prior work completed by Westinghouse under contract to EPA. This prior
work on fluidized-bed combustion is found in references 1 through 7.
This study.was conducted in close coordination with EPA's Office of
Energy, Minerals and Industry headquarters office; EPA's Office of Plan-
ning and Evaluation; and a related study for OPE by Energy Resources Co.
of the appropriateness and differential effects of alternative regulatory
approaches to the standards-setting process. Westinghouse provided them
with cost and performance data and a methodology for performing FBC
parametric studies.
iii
-------
ABSTRACT
An evaluation has been performed to project the impact of up to
90 percent S0~ control on the capital and energy costs oi7 atmospheric-
pressure fluidized-bed combustion (AFBC) and pressurized fluidized-bed
combustion (PFBC) power plants. The ability of AFBC and PFBC to achieve
reduced emissions of particulates and nitrogen oxides are also considered.
Performance and economic projections are presented for current emission
standards and for a set of hypothetical emission requirements represent-
ing a more stringent degree of control, and are compared with equivalent
projections for conventional boilers with flue gas desulfurization. The
projections of fluidized-bed combustion performance and system cost show
that AFBC and PFBC plants can achieve up to 90 percent S02 control and
still remain economically competitive with conventional plants with flue
gas scrubbers. The selection of fluidized-bed plant design and operating
parameters, however, is critical to the achievement of high levels of
control at competitive energy costs. The projections of FBC performance
at up to 90 percent SO- control must be confirmed in the future on FBC
units sufficiently large to be representative of commercial-scale
combustors.
-------
TABLE OF CONTENTS
Page
PREFACE iii
ABSTRACT v
TABLES ix
FIGURES xi
ABBREVIATIONS & SYMBOLS xv
I. INTRODUCTION 1
II. SUMMARY 3
III. METHODOLOGY OF EVALUATION 6
IV. BASIS OF EVALUATION 8
Study Parameters 8
Costing Assumptions 9
Power Plant Base Designs 9
Performance Projections Basis 12
Cost Projection Basis 14
Cost Sensitivity and Uncertainties 15
V. SULFUR OXIDE CONTROL . 17
Power Plant Overall Energy Conversion Efficiency 17
Power Plant Residue Generation 20
Power Plant Investment and Energy Cost 20
Impact of Combustor Operating Conditions 22
Advanced Sulfur Removal Systems 31
VI. PARTICULATE CONTROL 32
VII. CONTROL OF OXIDES OF NITROGEN 34
VIII. CONCLUSIONS 35
IX. REFERENCES 37
APPENDICES - A. Sulfur Oxide Removal Data Base and Model 39
B. FBC Power Plant Design Bases .81
VII
-------
APPENDICES - C. Relationships for FBC Cost Evaluations 113
D. Particulate Control Projections 141
E. Conventional Power Plant Design and Cost Basis 171
Vlll
-------
LIST OF TABLES
Table Title Page
1 Costing Assumptions 11
2 Performance Projections 13
3 FBC Plant Operating Conditions 26
4 Preliminary Comparison of the Environmental Impact
from the Leaching of FBC and FGD Residues 29
Al Range of Atmospheric TG Sulfations 49
A2 Range of Pressurized TG Sulfations - 10 atm
(1013 kPa) 50
A3 FBC Plant Operating Conditions 51
A4 Fluid-Bed Combustor Data at Conditions near the
1000 pm AFBC Base Case Projections 62
A5 .Fluid-Bed Combustor Data at Conditions near the
500 ym AFBC Base Case Projections 63
Bl Summary of Characteristics of AFBC Conceptual
Designs 82
B2 Summary of Characteristics of PFBC Conceptual
Designs 90
B3 Characteristics of Selected Base FBC Designs 97
B4 Solids Handling Major Equipment for AFBC 101
B5 Hot Gas and Air Major Equipment for AFBC 100
B6 Tower Components of AFBC 102
B7 AFBC Base Plant Capital Cost Breakdown 103
B8 Major Component Costs Comparisons in PFBC 104
IX
-------
LIST OF TABLES (Cont'd)
Table Title Page
B9 Solids Handling Major Equipment for PFBC 105
BIO Factors that Affect Comparisons of Solids
Injection with Petrocarb Systems 106
Bll Differences in Hot-Gas Cleanup with Cyclone
Separators and Ducon Filters 108
B12 Hot Gas Cleanup and Air Major Equipment for PFBC 109
Bl3 Heat Exchange Major Equipment for PFBC 110
B14 PFBC Base Plant Capital Cost Breakdown 111
Cl Auxiliary Loss Breakdown for AFBC 116
C2 Scaling of Solids Handling Major Equipment for AFBC 120
C3 Scaling of Hot Gas and Air Major Equipment for AFBC 122
C4 tower Components for AFBC 122
C5 Scaling of Balance-bf-Plant Costs for AFBC 124
C6 Auxiliary Loss Breakdown for PFBC 129
C7 Scaling of Solids Handling Major Equipment for PFBC 13l
C8 Scaling of Hot Gas Cleanup and Air Major Equipment
for PFBC 132
C9 Scaling of Heat Exchange Major Equipment for PFBC 133
CIO Scaling of Balance-of-Plant Costs for PFBC 134
Dl Effluent Loadings from Bed 153
D2 Primary Cyclone Efficiencies 155
D3 Overall Efficiencies of Primary Cyclone 155
D4 Projected Efficiencies of Tan Jet for Each Species 162
D5 Projected Overall Efficiencies for the Tan Jet 162
x
-------
LIST OF TABLES (Cont'd)
Table Title Page
D6 Overall Fractional Penetration for Each Species 165
D7 Overall System Particulate Fractional Penetration 167
El Wet Lime Absorber System Parameters, Conventional
Furnace-Steam Cycle 173
E2 Wet Lime Absorber Parameters, Conventional 173
Furnace Steam Cycle
XI
-------
LIST OF FIGURES
Figure Title Page
1 FBC Cost Evaluation Logic Diagram 7
2 Required Sulfur Removal Efficiency 18
3 FBC Power Plant Energy Conversion Efficiency 19
4 Capital Investment for FBC with Eastern Coals 23
5 Cost of Electricity for FBC with Eastern Coals 23
6 Cost of Electricity for FBC with Western Coals
and Lignite 23
7 Sulfur Removal Performance for Typical Sorbents 24
8 Comparison between AFBC, PFBC, and FGD for S02
Emission Standard of 1.2 Ib/MBtu (516 ng/J) 28
9 Comparison between AFBC, PFBC, and FGD for SO™
Emission Requirement of 90% Sulfur Removal 28
Al Required Reaction Rate as a Function of Sulfur
Removal Efficiency and Gas Residence Time 43
A2 The Impact of Sulfur Generation Pattern and
Desulfurization Requirement on Required Reaction
Rate 44
A3 Polynomial Fit of TG Rate Data from Run 231 52
A4 Polynomial Fit of TG Rate Data from Run 241 53
A5 Polynomial Fit of TG Rate Data from Run P17 54
A6 Polynomial Fit of TG Rate Data from Run 75-281 55
A7 Polynomial Fit of TG Rate Data from Run P96 57
A8 Polynomial Fit of TG Rate Data from Run P104 58
XII
-------
LIST OF FIGURES (Cont'd)
Figure Title Page
A9 Predictions of the Ca/S Feed Ratios Required
for Desulfurization Using Westinghouse TG Data
for Three Sorbents 60
A10 Sulfur Removal Projected from TG Data and Measured
on the Exxon Miniplant 60
All Fluid-Bed Combustor Data at Conditions near the
1000 ym AFBC Base-Case Projection Conditions 64
A12 Fluid-Bed Combustor Data at Conditions near the
500 pm AFBC Base-Case Projection Conditions 65
A13 Polynomial Fit of the TG Rate Data from Run 224 67
A14 Polynomial Fit of the TG Rate Data from Run 213 68
A15 Polynomial Fit of the TG Rate Data from Run 296 69
A16 Polynomial Fit of the TG Rate Data from Run 381 70
A17 Polynomial Fit of the TG Rate Data from Run 85 71
A18 Polynomial Fit of the TG Rate Data from Run 298 72
A19 Ca/S Molar Feed Required to Maintain 90% Sulfur
Removal in AFBC with Carbon Limestone 73
A20 The Sulfation Rate of Carbon Limestone 74
A21 Ca/S Molar Feed Required to Maintain 90% Sulfur
Removal in AFBC with Limestone 1359 76
A22 Comparison of Sorbent Capacity Obtained from
Westinghouse TG Data with B&W Fluid Bed Results 77
Dl Coal and Char Size Distributions (PFBC) 143
D2 Coal and Char Size Distributions (AFBC) 143
D3 Sorbent Size Distribution (PFBC) 144
D4 Sorbent Size Distribution (AFBC) 144
Xlll
-------
LIST OF FIGURES (Cont'd)
Figure Title Page
D5 Pressure Drop vs. Flow for Different Size
Fractions of Limestone 1^5
D6 Size of Ash Elutriated from Bed for both PFBC
and AFBC 147
D7 Total Solids from Combustor (PFBC) 152
D8 Total Solids from Combustor (AFBC) 152
D9 Schematic of Particle Control System 154
D10 Grade Efficiency for Primary Cyclone 156
Dll Combined Solids Leaving Primary Cyclone (PFBC) 157
D12 Combined Solids Leaving Primary Cyclone (AFBC) 157
D13 Combined Solids Leaving CBC (PFBC) 159
D14 Combined Solids Leaving CBC (AFBC) 159
D15 Grade Efficiency for Tan Jet 160
D16 Combined Solids Leaving CBC Cyclone 160
D17 Combined Solids to Final Control Device (PFBC) 161
D18 Combined Solids to Final Control Device (AFBC) 161
D19 Grade Efficiency for ESP 164
D20 Baghouse Performance at Sunbury Steam Electric
Station 164
D21 Grade Efficiency for Granular Bed Filter 166
D22 Projected Particulate Emitted vs. Calcium to
Sulfur Ratio 166
xiv
-------
ABBREVIATIONS AND SYMBOLS
EPA Environmental Protection Agency
OPE Office of Planning and Evaluation
FBC Fluid-bed combustion
AFBC Atmospheric fluid-bed combustion
PFBC Pressurized fluid-bed combustion
NSPS New source performance standards
TGA Thermogravimetric analysis
FGD Flue gas desulfurization
TVA Tennessee Valley Authority
ECAS Energy Conservation Alternatives Study
TG Thermogravimetric
B&W Babcock and Wilcox
ANL Argonne National Laboratories
NCB National Coal Board
FB Fluid bed
PER Pope, Evans and Robbins
NASA National Aeronautics and Space Administration
NSF National Science Foundation
DOE Department of Energy
GE General Electric
EST Electrostatic Precipitator
BOP Balance of Plant
A/E Architect/Engineer
xv
-------
APPENDIX A
h bed height, expanded
R Sulfur removal, fractional
Z gas residence time, expanded bed height/interstitial gas velocity
K average rate constant for SO- sorption in the bed
K first order rate constant for the reaction
CaO + S02 + 1/2 02 —* CaS04
y volume fraction of bed bubbles
e ' bed voidage in emulsion phase
F fraction of emulsion volume occupied by inerts
U sorbent utilization, fractional
C mole S0_/cc in TG reaction gas
P solids density, mole Ca/cc
eTG TG voidage
C Ca/S molar feed ratio
d particle diameter
xvi
-------
APPENDIX C
E coal energy input rate
F coal rate
c
H coal heating value, as received
F fresh sorbent feed rate
s
X coal sulfur content
s
M sorbent weight per mole of calcium
F, rate of spent solids
X weight fraction of sorbent released as CO- in calcination
L
-------
I. INTRODUCTION
Current fluidized-bed combustion (FBC) power plant development pro-
grams, and FBC power plant engineering studies, are focusing attention
on the achievement of current new source performance standards (NSPS)
for sulfur oxides, particulates, and oxides of nitrogen. In addition,
the FBC design philosophy expressed in these programs is generally directed
toward minimizing the fluid-bed combustor cost by applying compact designs
resulting from the use of high fluidizing velocities and correspondingly
large sorbent particle diameters.
Under the 1977 Amendments to the Clean Air Act, periodic revisions
to the emission standards will be evaluated by EPA. It is appropriate,
therefore, that the impact of potentially more stringent standards on
FBC power plant design and economics be addressed in development programs.
The relative merits of minimizing the fluid-bed combustor cost versus
maximizing its performance (e.g., maximum combustion efficiency, minimum
calcium-to-sulfur ratio, minimum particle attrition) must be assessed.
Although compact design could reduce combustor cost to a minimum, the
resulting plant energy costs could be higher than optimal and the potential
for economically meeting more stringent environmental emission goals
could be lower.
The objective of this evaluation is to project the impact of more
stringent emission requirements on FBC power plant economics, with emphasis
on achieving up to 90 percent S0_ control. The impact of fluid-bed com-
bustor design and operating conditions on FBC environmental control capa-
bilities and on plant economics is assessed. The projected cost of
achieving both current standards and more stringent degrees of control is
compared with the estimated cost of achieving equivalent degrees of con-
trol in a conventional coal boiler equipped with a flue gas scrubber.
-------
The current EPA new source performance standards for large coal-fired
boilers are: sulfur oxides, 1.2 Ib S0'2 MBtu (516 rig/J); particuiates,
Oil Ib/MBtu (43;0 rig/J); and oxides of nitrogen, 0.7 Ib N0'2 MBtii (301 fig/j) ;
The more stringent degrees of control considered in this study are:
o Sulfur oxides - 90 percent removal of coal sulfur content
9 Particuiates - 0.03 Ib/MBtu (12.9 ng/J)
• Oxides of nitrogen - 0.6 Ib/MBtu (258 ng/J).
The selection of these control/emission levels as the "more stringent
degree of control" should riot be construed as suggesting that these are
the most stringent levels that FBC can achieve or that FBC would ever be
required to achieve. These levels have been selected for the current study
drily because they are one set of values that were considered during the
planned revision of the NSPS for utility boilers.
The evaluation incorporates and utilizes the results of previous
FBC engineering studies arid modeling studies in developing preliminary
performance and cost projections for atmospheric-pressure fluidized-bed
combustion (AFBC) arid pressurized fluidized-bed combiistibri (PEBC); this
work has been performed ori a technical directive from the industrial
Environmental Research Laboratory, U; S; Environmental Protection Agency
(EPA)i during the period of August 1977 to Jariuary 1978.
-------
II. SUMMARY
Projections of AFBC and PFBC power plant performance and economics
have been developed through the assimilation of previous FBC power plant
design studies, FBC performance models, and data assessments. The key
parameters in the evaluation are the sorbent calcium-to-sulfur ratio,
the coal sulfur content and the fluid-bed combustor design and operating
conditions.
The projections of FBC power plant energy costs indicate that—for
both the existing S02 emission standard and for 90 percent sulfur
removal—FBC has considerable potential for being cost competitive with
conventional coal-fired power plants using lime slurry scrubbing. The
competitiveness of FBC depends upon proper selection of fluid-bed com-
bustor operating conditions—i.e., sufficiently long gas residence time
in the bed (sufficiently low gas velocity, and sufficiently deep beds)
and sufficiently small sorbent particle size. This selection of variables
will result in a less compact combustor, but the cost savings resulting
from decreased sorbent requirements would more than compensate for
increased combustor costs.
In the design of FBC power plants emphasis should be placed on
maximizing the fluid-bed combustor performance rather than on minimizing
the combustor cost through compact design. The combustor cost represents
a small portion of the FBC power plant investment and is also relatively
insensitive to changes in design and operating conditions. On the other
hand, the overall FBC power plant cost of electricity is strongly
dependent on the combustor performance.
The calcium-to-sulfur molar ratio—that is, the moles of sorbent
calcium fed to the fluid-bed combustor divided by the moles of sulfur
-------
fed in the coal—is the single, most important performance factor rela-
tive to FBC power plant cost and performance for high-sulfur Eastern coals
(2 to 5 wt% sulfur). The calcium-to-sulfur ratio has a dramatic impact
on the FBC power plant thermal efficiency, capital investment, and cost
of electricity. Increased calcium-to-sulfur ratio, if required for lower
sulfur oxide emissions, results in increased auxiliary power consumption
for solids handling and significant sorbent calcination energy losses.
The resulting reduced net plant efficiency and slightly increased equipment
costs for solids handling, crushing, drying, feeding, and spent solids
disposal lead to increased capital investment and energy costs. In
addition, the increased cost of raw sorbent at increased feed rates
significantly increases the energy cost.
The all-important projection of sorbent feed requirements was
accomplished in this study using a kinetic model for SC>2 capture developed
3
by Westinghouse. This model—using rate constants measured in laboratory
thermogravimetric analysis (TGA) equipment, and confirmed where possible
using available data from experimental fluidized-bed combustors—is
capable of projecting sorbent requirements, where TGA data have been
generated, as a function of key combustor operating/design conditions.
Further confirmation of the accuracy of model projections is necessary,
especially at high levels of S0« removal, where very little combustor
data are currently available; data from larger combustors, more repre-
sentative of commercial-scale units, are also necessary to confirm model
projections.
While cost and performance uncertainties exist for several sub-
systems in the FBC power plants (for example, solids feeding and particu-
late control), these are expected to be resolved through proper design
and specification of materials and operating conditions and maintenance
and operating procedures. The overall financial impact of these cost/
performance uncertainties.will probably be small relative to the uncer-
tainties .in such site factors as sorbent availability, sorbent cost,
coal cost, solid waste disposal feasibility or utilization markets,
local emission standards, and so on.
-------
For low-sulfur Western coals and lignites the impact of increased
calcium-to-sulfur ratio is greatly reduced, because of the relatively
small quantities of sorbent involved. Uncertainties associated with
sorbent selection and cost are also less significant.
Projections of particulate control and emissions of oxides of
nitrogen for FBC power plants indicate that the more stringent emission
requirements considered here of 0.03 Ib/MBtu (12.9 ng/J) and 0.6 Ib/MBtu
(285 ng/J), respectively, are economically feasible and of lower cost
impact than the more stringent sulfur oxide requirement. Conventional
fabric filter (baghouse) techniques should permit achievement of this
requirement, depending on particle size and future environmental standards,
PFBC plants are projected to require two stages of particulate control
equipment operating at the combustor temperature and pressure: e.g.,
conventional cyclones followed by a filter system. Nitrogen oxide
levels from the assessment of FBC experimental results have been shown
to be generally less than 0.6 Ib/MBtu (258 ng/J) without special control
efforts.
On the basis of available information, the projections developed
indicate that both AFBC and PFBC should be able to achieve the higher
levels of control considered in this evaluation economically if proper
selection of combustor design and operating conditions is made. Develop-
ment programs should focus attention on developing large-scale informa-
tion on the relationship between combustor operating conditions and FBC
plant emissions, while engineering evaluation of this information should
be performed to provide assessment of FBC pollution control capabilities.
-------
III. METHODOLOGY OF EVALUATION
The evaluation has utilized the results of several previous in-depth
engineering studies to develop quantitative cost and performance projec-
tions. An evaluation logic-diagram (Figure 1) summarizes the sequence
of steps. Previous FBC engineering studies were reviewed, and base
designs (for AFBC and PFBC) and costing ground rules were selected.
Relationships for scaling plant performance and plant cost were developed
and applied to the base designs in order to generate plant performance
and cost results with respect to the study parameters (calcium-to-sulfur
ratio, coal properties, sorbent properties, coal cost, sorbent cost).
Projections of sorbent performance and sulfur removal efficiency versus
calcium-to-sulfur ratio, as a function of the combustor design and oper-
ating conditions, were applied to generate specific cost estimates for
FBC power plants.
These costs were then compared with costs for conventional boilers
8 9
with scrubbers, employing cost estimates generated in previous studies. '
Results from the evaluation are reported in three sections covering
each of the potential revised emission standards categories: sulfur
oxide control, particulate control, and control of oxides of nitrogen.
The report body is followed by appendices that detail the sulfur oxide
removal data base and model, the FBC power plant design bases and cost
projections, particulate control projections, and the conventional power
plant design basis and cost projections.
-------
16S2B6C
FBC Design Review
Ground Rules
Operating Conditions
Configuration
Component Design
Performance
Cost
Cost Sensitivity
Ground Rules
Operating Conditions
Technical Design Uncertainties
Performance Uncertainties
Raw Material Costs
Performance
Scaling
Relationships
Cost Scaling
Relationships
Sorbent
Performance
Projections
("Calcium- to-Sulfur Ratio
I Coal Sulfur-Content, Ash-
\ Content. Heating Value
Sorbent Properties
Base Design Selection
Ground Rules
Operating Conditions
Configuration
Component Design
Performance
Breakdown
Cost
Breakdown
Parametric
Performance
Plant Efficiency
Component
Capacities
Particulate
Emissions
fCalcium-to-Sulfur Ratio
Coal Sulfur-Content. Ash-Content, Heating Value
< Sorbent Properties
Sorbent Cost
I Coal Cost
Sulfur Removal Efficiency
Coal Sulfur-Content. Ash-Content. Heating Value
Sorbent Properties
Sorbent and Coal Cost
Figure 1 - FBC cost-evaluation logic diagram
-------
IV. BASIS OF EVALUATION
The basis applied in this evaluation is briefly described. The
study parameters, the costing assumptions, the FBC power plant base-
designs, the FBC plant performance-pro jee'tioh basis, arid the FBC cost-
projectibh basis are summarized.
F.BC .S.tudy:..Parameters.
The following parameters arid conditions were considered in the
evaluation:
• Concepts - once-through sorbent operation of AFBC and PFBC
• Coals
- Eastern (2 to 5 wt % sulfur, id wt % ash, 13 wt % moisture)
— Western subbitiimihdus coalj capable of complying with
1:2 ib/MBtu (516 fig/J) without further SO^'control
(6:6% suif'ur; 8% ash; 23,260 kJ/kg heating value)
- Lignite (0;4% sulfur} 6% ashj 18,600 kJ/kg heating value)
• Soirbents
Representative limestone for AFBC (average sorbent reactivity)
Representative dolomite for PFBG (average sorbent reactivity)
o Emission Standards
- Sulfur oxide standard - 1.2 Ib/MBtu (516 ng/J) (current
standard) and 90% sulfur removal
Particulate standard - 0.1 (43.0) (current standard) and
0.03 Ib/MBtu (12:9 ng/J)
Oxides of nitrogen - 0.7 (301) (current standard) and
0.6 Ib/MBtu (258 ng/J)
-------
• Operating Conditions
- Calcium-to-sulfur ratio - 1 to 10
- Fluid-bed combustor - 2 sets of conditions (temperature,
pressure, velocity, bed depth, excess air, sorbent particle
diameter) illustrated for each of the AFBC and PFBC power
plants
• Sorbent Cost - $5-60/Mg ($5-66/short ton)
• Coal cost - $l/MBtu ($0.95/GJ).
The effect of coal price, coal ash-content, and heating value were also
examined and found to be of secondary importance with respect to the
relative economics of AFBC, PFBC, and conventional power plants.
Costing Assumptions^
Table 1 summarizes the costing assumptions applied. These costing
assumptions were adapted from the recent Energy Conversion Alternative
Study (EGAS) ground rules, and were applied identically for both AFBC
and PFBC.8'10'11
Power Plant Base Designs
Previous FBC power plant designs for AFBC and PFBC were
, 1-3,8,9,11,12 _,...,. . . . .
reviewed. Plant design comparisons are summarized in
Appendix B. Base power plant designs were adapted from these studies,
drawing heavily from the Westinghouse and General Electric plant designs
8 11
resulting from the EGAS programs. ' The General Electric designs for
AFBC and PFBC power plants were applied with several modifications
because they represent a consistent basis on which a conventional power
o
plant design was also developed. Details of the selected design bases
are discussed in Appendix B.
The major FBC power plant design characteristics selected for the
base plant were:
• Combustor configuration - modular design with vertical stacking
of beds and separate steam generation functions for each bed;
horizontal steam tubing and waterwalled vessels.
-------
© Control of carbon utilization - carbon burnup cell (1100°C,
30% excess air)
e Coal and sorbent feeding - in-bed feeding with coal and sorbent
premixed
• Particulate control equipment - final stage cleaning by bag-
house (AFBC) and granular bed filter (PFBC)
• Spent solids handling - air cooling of spent solids with
recovered energy used for coal and sorbent drying
• Spent solids storage - 15 day on-site storage with final,
long-term, off-site storage in lined containment
• Steam cycle - 24,000 kPa, 538°C superheat temperature, and
538toC reheat temperature
• Gas turbine compression ratio (PFBC) - 10:1
• Cooling tower - wet mechanical draft
• Base plant capacity - the base plants had electrical capacities
of about 800 MWe for AFBC and 900 MWe for PFBC, depending mainly
upon.the sorbent calcium-to-sulfur ratio and coal properties.
Coal energy input was fixed at 2.262 GJ/s for the AFBC power
plant and 2.308 GJ/s for the PFBC power plant to comply with
Q
the EGAS study basis. These energy inputs remained fixed
as coal characteristics and operating conditions were varied
in the evaluation.
A multitude of potentially acceptable plant design options exist,
as is apparent from Appendix B. Various degrees of modularity can be
applied in an attempt to maximize shop fabrication of the combustors.
Numerous combustor configurations are possible, ranging from horizontal
arrangements of beds placed side-by-side to the vertical stacking con-
cepts. Alternative steam circuitry and tubing layouts have been
conceived.
The utilization of carbon can be maintained at high levels by
several schemes: separate bed for combustion of elutriated char fines,
operated at higher excess air and temperature (carbon burnup cell);
high excess air operation of the combustor beds (up to 100% excess air
10
-------
Table I
COSTING ASSUMPTIONS
Site - midwest
Cost base year - mid-1975
Construction start - mid-1975
Construction time, yr - 5.5
Escalation, %/yr - 6.5
Interest during construction, %/yr - 10.0
A&E services, % of indirect cost - 15.0
Labor rate, $/hr - 11.75
Indirect field labor cost, % of direct field labor - 90
Contingency, % of direct, indirect, and A&E services - 20
Plant availability target, % - 90
Plant capacity factor, % - 65
Capitalization rate, %/yr - 18
Operating and maintenance cost, %/yr - 1.95
Coal price, $/GJ - 0.95
11
-------
for a pressurized fluid-bed boiler); recycle of elutriated char fines
back to the combustor beds; and minimization of char elutriation by low-
velocity operationj higher bed temperatures, control of coal particle
size, etc.
The feeding of coal and sorbent to the combustor may be performed
either separately or in combination, with feeding being in bed or above
bed. Several techniques for distributing the solids to a multitude of
feed points and injecting the solids are available.
Final-stage particulate control can be achieved by several methods.
For AFBCj high-efficiency cyclones; electrostatic precipitatdrs (high or
low temperature), or fabric filters may be applied. For PFBC, high-
efficiency cyclones, a hot electrostatic precipitator, granular bed fil-
ters i ceramic filters, or low-temperature cleaning schemes combined
with a regenerative heat exchanger may be applicable.
Spent solids from FBC power plants may find specific forms of
utilization or alternative disposal application's. Lorig-terin off-site
storage-may be required in some cases* depending oh plant location.
Th'usj numerous FBC power plant design and operating options have
been cdhsidered for both AFBG and PFBC plants utilizing the basic design
concept selected.-* The cost data for these options indicate that the
FBC power plant equipment investment will not be greatly sensitive to
the choice of these options^ but the performance and operability of the
plant coiild be significantly influenced by their choice (see Appendix B) ;
The options selected for this evaluation provide reasonable plant cost
projections and assume that any significant performance and operability
problems have been resolved.
Performance Projection Basis
the FBC power plant performance factors (equipment capacities, plant
efficiency, sulfur removal, etc.) that were judged to be the most signifi-
*Alternative FBC concepts are being considered that may have an appreci-
able impact on system cost (e.g., recirculating bed concepts^); This
report does not include an analysis of the impact of emission standards
on these systems.
12
-------
Table 2 ^a 1698P76
APPROACH FOR CONSIDERATION OF FBC PERFORMANCE VARIABLES
Scaling Parameters - calcium-to-sulfur ratio, coal properties (sulfur
content, ash content, heating value), sorbent
properties, and combustion conditions
Equipment Capacities Scaled (Appendix C) -
• Sorbent handling, crushing, drying and feeding
• Fines handling
• Spent sorbent cooling, handling, and on-site storage
• Spent solids off-site disposal
Power Plant Thermal Efficiency -contributing factors scaled (Appendix Cl
• Auxiliary power consumption
• Combustion efficiency
• Energy losses-sorbent calcination, sensible heat losses, etc.
Sulfur Removal Performance (Appendix A)
e Extensive kinetic data collected on TG equipment, and comprehensive
experimental FBC data bank utilized
• Sulfur removal model utilized to project FBC sulfur removal efficiency
as a function of: 3
- Calcium-to-sulfur ratio
- Bed depth
- Velocity
- Sorbent particle size
- Bed voidage and density
- Sorbent calcination conditions
• Results of model compare well with experimental FBC data from data bank
for broad range of operating conditions and large number of data sources,
but data from large combustors and data for high levels of S02 removal are
limited.
Particulate Control Projections (Appendix D)
• Experimental fresh feed size distribution selected
• Coal-ash elutriation rate projected as 100% of coal ash content
• Sorbent attrition rate and elutriation rate projected from basic
attrition model 13
• Experimental size distribution of elutriated coal-ash and sorbent
fines applied, as available
• Experimental grade efficiencies for cyclones, electrostatic
precipitators, baghouses, and granular bed filters applied
• Computer model projects particulate emission from FBC par-
ticulate control system
NOX Control Projections
• Statistical evaluation of FBC NOX emission data performed
» Projections based on representative NOX performance without
specific control efforts
13
-------
cant in influencing the power plant cost were scaled from the base plant
o
performance values (from General Electric base plants ) with respect to
the study parameters. The nature of the performance projections are
summarized in Table 2. The relation developed and applied in this
evaluation to estimate the effect of these variables on plant performance
are simplified to provide practical insight and sufficient accuracy for
plant cost projections, these simplified relations are detailed in
Appendix C.
CtiST PROJECTION BASIS
Investment and eriefgy costs were estimated by scaling the base plant
Q
equipment costs (from the modified General Electric base plant ) and plant
performance (electric generating efficiency), derived in Appendix B, with
respect to the calcium-to-sulfur ratio and the coal properties, this
procedure for scaling is detailed in Appendix C.
For the FBC cost projections, seven plant categories were used to
Q jl
describe the FBC capital investment breakdown; ' Five of these cate-
gories were assumed to be fixed iri cost over the ranges of parameters
considered (turbine generators,' electrical; civil and structural; process
piping and instrumentatiori, and yardwdfk and miscellaneous), and two
categories were scaled according to changes iri capacity (steam generators
and process mechanical equipment).
Component costs, direct labor costs, indirect field costs', arid
materials costs were scaled according to capacity ratios with a 0;85
power factor generally appreciable for solids handling items, a 0.-6
power factor for cyclones and particulate control auxiliary components,
and a 0.68 power factor for fans and motors.
The process mechanical equipment category includes the scaled items
coal drying and crushing, limestone drying and crushing, coal and lime-
stone blendirig arid feeding, spent bed material cooling, arid spent solids
material on-site and off-site storage. Solids handling controls are
assumed to be constant in cost. Complete listings are presented in
Appendix C.
14
-------
The steam generator category includes hot gas cleanup items, air
supply, and the ccmbustor and steam generator items. The hot gas cleanup
items, which are assumed to be unchanged in cost with respect to the
study parameters, are the bed cyclone units, cyclone air lock valves,
fines injection system, carbon burnup cyclone air lock valves, and
cooler air lock valves, these items being generally more strongly
influenced by gas-handling capacity than by the variables considered in
this study. A baghouse (AFBC) has been substituted for the base design
ESP. A granular bed filter (PFBC) based on previous Westinghouse designs
has been used in place of the General Electric design. The fluid-bed
combustor/steam generator is assumed to be constant in cost for the
parameters of interest in this study, based on previous sensitivity
analyses which are reviewed in Appendix B. The maximum increase in com-
bustor cost resulting from reduced gas velocities would be about 10 per-
cent expressed as $/ton of coal but may be reduced when expressed as
$/kw if the plant performance improves with reduced velocity.
The cost of electricity is obtained from the total plant capital
investment and operating cost, and the plant generating capacity. The
categories considered are capitalization, operating labor, indirect
operating costs and maintenance costs, the cost of coal and the cost of
sorbent.
COST SENSITIVITY AND UNCERTAINTIES
The sensitivity of the FBC power plant investment to the design and
operating conditions of the fluidized-bed combustion system (not including
solids feeding and particulate control) has been shown to be rather small
2
(no more than +10%). This is partly because the FBC power plant is
largely made up of conventional power plant components (see Appendix B).
The solids feeding system, the' fluid-bed combustor, and the hot gas
cleaning system represent the developmental components which account for
20 to 30 percent of the total plant cost. The fluid-bed combustor
itself accounts for 10 to 15 percent of. the plant investment and is very
15
-------
2
insensitive to changes in design and operating conditions. The combustor
should be designed to maximize performance (e.g., carbon utilization,
sulfur removal efficiency, particle carry-over, etc.) rather than to
minimize the vessel cost. For example, low combustor fluidization
velocities will probably result in a power plant producing cheaper
electricity (due to improved process performance) than does a FBC power
plant designed with a compact fluid-bed combustor (high fluidization
velocities and large sorbent particles).
The greatest technical design uncertainties appear to be associated
with the solids feeding and hot gas cleaning systems, especially for
PFBC. These systems could have a large impact on plant reliability.
The greatest performance uncertainty is undoubtedly associated with
the rate of sorbent feeding required to achieve the specified sulfur
removal efficiency. Significant variability in sorbent properties and
performance exists. Sorbent performance data from large FBC units,
and for high degrees of SCL removal, are lacking.
The costs of coal and sorbent .are subject' to great uncertainty and
variability, depending on power plant location. These uncertainties
could have a greater impact on the plant energy cost than do technical
design uncertainties or performance uncertainties.
Finally, the cost assumptions, or ground rules, applied (Table 1)
may provide substantial cost uncertainties. Plant site is very influen-
cial in this respect. Power plant construction time could vary signifi-
cantly from one plant to another. Even though plant costs (AFBC, PFBC,
and conventional) are compared, with the same ground rules applied to all
of them, impacts of differing site conditions could be significantly dif-
ferent for each of them. The concept of "economy of scale" is also very
uncertain because of the impact of plant construction time, interest
during construction and escalation, and regulatory restrictions, and
different plant capacities may modify the comparison of costs.
Within the framework of the set of costing assumptions stated, the
FBC power plant costs projected are expected to be broadly representative
of FBC power plant cost potential.
16
-------
V. SULFUR-OXIDE CONTROL
The control of sulfur oxide emissions from the FBC power plant is
addressed independently of particulate control and oxides of nitrogen
in this section. Plant performance and cost is expressed parametrically
as a function of calcium-to-sulfur ratio and coal properties. The impact
of combustor operating conditions on the required calcium-to-sulfur
ratio is discussed and related to specific projections of performance
and plant cost. Costs are compared with a conventional coal-fired power
plant using lime slurry scrubbing for sulfur oxide control, as a function
of sorbent cost.
Two sulfur oxide emission requirements are considered, the current
standard of 1.2 lb SO^/MBtu (516 ng/J) and a more stringent requirement
of 90 percent sulfur removal. The relationship between the sulfur oxide
standard expressed in lb SO«/MBtu and expressed in percent removal is
shown in Figure 2. For a 4.0 wt % sulfur Eastern bituminous coal
(10 wt % ash, 13% moisture, heating value of 25,300 J/g) the required
sulfur removal efficiency to satisfy the current standard is about
83 percent.
POWER PLANT OVERALL ENERGY CONVERSION EFFICIENCY
The rate of sorbent feeding to the FBC power plant can have a sig-
nificant impact on the power plant thermal efficiency, due mainly to the
energy requirement for calcination of the sorbent. Increased auxiliary
power consumption for solids handling is of secondary importance. Plant
efficiency is expressed as a function of calcium-to-sulfur ratio and coal
sulfur content in Figure 3, considering efficiency losses due to calcina-
tion and auxiliary power consumption. Both PFBC and AFBC are considered.
The figure assumes a constant coal combustion efficiency and a constant
rate of heat release due to the sulfation reaction for all calcium-to-
sulfur ratios equivalent to meeting the current sulfur oxide standard
for each coal sulfur content. The PFBC plant efficiency drops more
17
-------
Curve 693150-A
100
90
2?
'o
80
E
o>
^
13
70
Eastern Bituminous Coals
10 Wt % Ash
13 Wt % Moisture
Parameter- Wt % Sulfur in Coal
60
I
I
0.2 0.4 0.6 0.8 1.0 1.2
SCL Emission Standard (Ib/M Btu)
Figure 2 - Required sulfur removal efficiency
18
-------
Curve 693149-A
o>
c
o
CD
>
c
o
o
40
39
38
37
36
35
= 34
= 33
fO
0>
O
32
31
30
I i i i i
Parameter: Wt % Sulfur in
Coal \
\
\
1 2 3 4 5 6 7 8 9 10
Calcium-to-Sulfur Ratio (Molar)
Figure 3 - FBC power plant energy conversion efficiency
19
-------
swiftly than does the AFBC plant efficiency because the dolomite sorbent
for PFBC has a greater calcination energy loss than does the limestone
sorbent used for AFBC.
The energy efficiency of a conventional coal-fired power plant with
lime slurry scrubbing is about 34 percent for a 4 wt % sulfur coal, with
0
a stack-gas reheat temperature of 79°C. Then, from Figure 3, PFBC
power plants should always be more energy efficient than conventional
power plants over the range of calcium-to-sulfur ratio considered, and
AFBC power plants should be more efficient for calcium-to-sulfur ratios
less than 6.
POWER PLANT RESIDUE GENERATION
The rate of spent solids generation for disposal is an important
environmental factor for which specific standards have not yet been
developed. The ratio of spent solids to coal input (kg/kg) for both
AFBC and PFBC will range between 0.3 and 0.7 for a 4 wt % sulfur Eastern
coal (not including the ash content of the coal) and depending on the
combustor operating conditions. The ratio of (spent sorbertt) to the coal
ash in the solid residue will range from 2 to 10 for typical Eastern
coals.
For a conventional coal-fired power plant using lime slurry scrubbing
the ratio of scrubber sludge to coal input (kg/kg) is about 0.4 for a
4 wt % sulfur Eastern coal and assuming a water content of 50 wt %. This
is a value intermediate to the range for AFBC and PFBC. A comparison
between the environmental impacts of the two solid waste sources is
being developed.
POWER PLANT INVESTMENT AND ENERGY COST
Total capital investment for PFBC and AFBC power plants are shown
in Figure 4 as a function of the calcium-to-sulfur ratio and the coal
sulfur content for Eastern coals. The derivation of the base plant
capital costs is shown in Appendix B; the method by which the base plant
costs were varied to account for variations in calcium-to-sulfur ratio
is detailed in Appendix C. The base particulate control equipment is
assumed constant for all calcium-to-sulfur ratios (baghouse for AFBC and
granular bed filter for PFBC). The increasing investment with increasing
20
-------
Curve 693^5-A
900
•w-
"c
I/I
o>
CO
800
700
600
I I I
I I
PFBC
AFBC
Basis
"•Mid-1975Cost Base
• 5.5 Year Construction Time
• Escalation to End of 1981
at 6.5% per Year
• Interest during Con-
struction at 10%
per Year
-•20% Contingency
Wt% Sulfur
in Coal
i I I
1 i
I I
123 4 56789 10
Calcium- to-Sulfur Ratio
Figure A - Capital investment for FBC with Eastern coals
21
-------
calcium-to-sulfur ratio is due to the combined effects of increased equip-
ment cost for solids handling, crushing, drying, feeding and disposal,
and reduced power plant efficiency (Figure 3).
Figure 5 shows the resulting parametric FBC power plant energy cost
for Eastern coals. A sorbent delivered cost of $5/Mg is assumed in the
figure. PFBC and AFBC energy costs are very similar at identical values
of the calcium-to-sulfur ratio. Higher coal costs will favor PFBC
because of higher plant efficiencies. Again, the calcium-to-sulfur
ratio and coal sulfur content have very significant impacts on the energy
cost with high-sulfur Eastern coals.
FBC with Western coals and lignite (Figure 6), on the other hand,
are not significantly influenced by the calcium-to-sulfur ratio, because
of their low sulfur contents, and, therefore, are not discussed further
in this evaluation.
IMPACT OF COMBUSTOR OPERATING CONDITIONS
The sulfur removal efficiency of a fluid-bed combustor depends on a
multitude of design and operating conditions that influence the distri-
bution and release of sulfur in the bed, the contacting of sulfur oxides
and sorbent particles, the calcination of the sorbent particles, the
temperature and concentration distributions in the bed, the reaction
kinetics, the interparticle diffusion resistance, the attrition and
elutriation of sorbent particles, etc. The major operating conditions
have been identified and correlated (for dense-phase fluidization sys-
tems) with sulfur removal efficiency (see the development of the Westing-
hosue kinetic model in Appendix E of Reference 4, and in Appendix A of
this report). Figure 7 shows performance curves for representative
sorbents and specific operating conditions projected using the Westing-
house kinetic model. The curves are representative of typical limestones
and dolomites, though significant variability in sorbent performance
exists (see Appendix A). The specific sorbent types that served as the
basis for Figure 7 (Carbon limestone and 1337 dolomite) are among the
more reactive of the sorbents that have been tested on the Westinghouse
22
-------
Curve 693148-A
60
_ 50
I I I I I I I I '
PFBC
AFBC
Parameter: Wt % Sulfur in Coal
Basis
• Capitalization Rate 18% Per Year
• 65% Plant Capacity Factor
-•Coal Price $0.95/GJ
• Sorbent Price $5/Mg
40
o
CJ
30
I I
1 2 3 4 5 6 7 8 9 10
Calcium-to-Sulfur Ratio
50
40
o
CJ
30
Western Coal - 0.6 Wt* Sulfur, 8Wt%Ash,
23,260 kJ/kg Heating Value
Lignite-0.4 Wt% Sulfur, 6Wt%Ash,
18,600 kJ/kg Heating Value
Cost of Coal $0.95/GJ
Cost of Sortent $5/Mg
Westejrn __——.-
456 7 8 9 10 11 12
Calcium-to-Sulfur Ratio
Figure 5 - Cost of electricity for FBC
with Eastern coals
Figure 6 - Cost of electricity for FBC with
Western coal and lignite
-------
Curve 693151-B
90
,o
c
—
'o
- .80
1
cc
L_
^
l/l
70 -
I
Operati
Sorbent Type
ing Conditions
A.FBC P.FBC
Limestone Dolomite
a b c d
Average Diameter. |jm 500 1000 500 2000
Pressure, kPa 101 101 1013 1013
Bed Temperature, °C 840 840 950 95.0"
Excess Air, % 20 20 20 20
Velocity, m/s 1.83 3.05 1.52 1.52
Bed Depth, m 1.22 1.22 3.05 3.05
I
I
_L
2 3 4 5 6 7 8 9 10
Calcium-to-Sulfur Ratio (Molar)
Figure 7 - Sulfur removal performance for typical sorbents
(projected using Westinghouse kinetics model)
24
-------
thermogravimetric apparatus and are expected to be representative of
sorbents than will be available to FBC power plants. The performance
projections in Figure 7 require confirmation on operating fluidized-bed
boilers sufficiently large to provide data representative of commercial-
scale units.
Table 3 lists two sets of combustor operating conditions and pro-
jected calcium-to-sulfur ratios required to meet the two levels of sulfur
oxide emissions (1.2 Ib/MBtu [516 ng/J] and 90% sulfur removal efficiency)
for a 4 wt % sulfur Eastern coal. The calcium-to-sulfur ratios were taken
from Figure 7. The sorbent average particle diameter and fluidization
velocity (gas residence time in the bed) are the key conditions varied
between the two sets, and they clearly have a dramatic impact on the
required calcium-to-sulfur ratio. The high velocity-large particle
sorbent conditions are typical of many designs that have been proposed.
With these sets of operating conditions to illustrate the impact of
the operating conditions selection and the potential performance of FBC
power plants, a comparison may be made between the energy cost of AFBC,
PFBC, and conventional coal-fired power plants. None of the sets of con-
ditions listed in Table 3 are meant to represent optimum operating
conditions.
The energy cost projections for the conventional coal-fired power
plant with lime slurry scrubbing have been taken from two sources. The
cost for the current sulfur oxide .emission of 1.2 Ib/MBtu (516 ng/J) was
extracted from a recent report by TVA which applied the same cost ground
9
rules as were applied in this report. The cost agrees closely with those
from previous TVA cost studies for FGD systems when placed on the same
14
basis. The cost for 90 percent sulfur removal was taken from the
General Electric EGAS cost by applying cost modifications suggested by
TVA. The details of these projections are presented in Appendix E.
Energy costs for PFBC, AFBC, and a conventional plant with lime
slurry scrubbing are plotted together in Figures 8 and 9 as a function
25
-------
Table 3
FBC PLANT OPERATING CONDITIONS
Pressure, kPa
Bed temperature, °C
Excess air, %
Sorbent type
Sorbent diameter, ym
Bed velocity, m/s
.Bed depth, m
Gas residence time in bed, s
Ca/S for 1.2 Ib/MBtu S02
(516 ng/J) emission3*^
Ca/S for 90% sulfur removal
efficiencyb
AFBC
7.0
2.9
PFBC
101
840
20
Limestone
1000
3.05
1.22
0.4
5.5
500
1.83
1.22
0.67
2.4
2000
1.52
3.05
2.0
2.0
1013
950
20
Dolomite
500
1.52
3.05
2.0
1.2
4.5
1.7
a - 83% sulfur removal for 4 wt % sulfur coal.
b - From Figure 7, projections based upon Westinghouse kinetic model.
For sorbents of representative reactivity, based upon about 25 sources
of sorbents that have been tested by Westinghouse on their laboratory
thermogravimetric analysis apparatus in several hundred characteriza-
tion tests (see Appendix A).
26
-------
of the sorbent cost for the two sulfur oxide emission standards:
1.2 Ib/MBtu (516 ng/J), the current standard, and 90 percent sulfur
removal efficiency (equivalent to 0.74 Ib/MBtu [318 ng/J]), both for a
4 wt % sulfur Eastern coal. Calcium-to-sulfur ratios in FBC are taken
from the projections in Table 3.
For the current sulfur oxide standard, the break-even cost of
sorbent ranges between $7 and $65/Mg for AFBC and is $57 to >$100/Mg
for PFBC, depending upon the selected operating conditions. For 90 per-
cent sulfur removal, the break-even cost of sorbent ranges between $3 and
$55/Mg for AFBC and $11 to $70/Mg for PFBC, depending on the selected
operating conditions. Typical sorbent costs in this country today range
from $5 to $10 per Mg (at quarry), suggesting that FBC should be competi-
tive with FGD.
The equivalent cost of spent solids disposal (for off-site disposal)
that is included in Figures 8 and 9 is $3/Mg of fresh sorbent for AFBC
and $4/Mg of fresh sorbent for PFBC. Potential markets for the spent
solids would have some impact on the comparison in Figures 8 and 9 but,
more importantly, could significantly reduce the environmental impact of
the FBC spent solids.
A definitive analysis of the environmental impact of the disposal
of spent sorbent from FBC processes cannot be carried out since environ-
mental impact criteria are not established for land disposal. Data from
leaching tests, however, provide a preliminary comparison of the environ-
mental impact of fluidized-bed combustion and limestone scrubbing residues.
Results are presented in Table 4 which indicate the FBC spent sorbent may
have a reduced impact from leaching. Codisposal of the FBC spent sorbent
and ash is recommended since it may further reduce the environmental
impact. The FBC spent sorbent and ash will form solid compacts with the
addition of water, which results in an effective low-cost method to reduce
the permeability and thus the environmental impact. '
The cost for disposal will depend on the alternative sorbent selected
Q
and the specific location. TVA projected costs for AFBC, PFBC, and lime-
stone scrubber residues, providing perspective on the disposal costs.
27
-------
Curve 696927-A'
70
60
KJ
00
O
CJ
40
30 L
Eastern Coal (4'wt% Sulfur, 10 wt% Ash):
PFBC
AFBC
Conventional Plant with Lime
Slurry Scrubbing (Ca/S = 1.05)
Stack Reheat of 79°C
Ca/S = 1.2
10 20 30 40 50
Sorbent Cost ($/Mg)
60
70
Figure 8 - Comparison between AFBC, PFBC and
FGD for S02. emission standard of
1.2 Ib/MBtu (516 ng/J) (current
standard)
Curve 696928-A'
60
t 50
o
O
40
30
i i \ \ I i
Eastern>Coal.(4wt% Sulfur, 10wt% Ash)-
PFBC
AFBC
Conventional Plant with Lime .
Slurry Scrubbing (Ca/S =1.1) /
Stack Reheat of 79°C /
Ca/S =7
= 2.9
= 1.7
10 20 30 40 50
Sorbent Cost ($/Mg)
60
Figure 9 - Comparison between AFBC, PFBC
and FGD for SC>2 emission
requirement of 90% sulfur
removal
-------
Dwg. 169868*+
TABLE 4
PRELIMINARY COMPARISON OF THE ENVIRONMENTAL IMPACT FROM THE LEACHING OF FBC* AND FGD RESIDUES
Chemical
Property
NJ
FBC
Solubility of Major Compounds: Ca, SO.andTDS
Contributing to Potential Environmental Concern
Major Component CaSO, Relatively Inert
High Alkalinity in Leachate: pH <0 to 13
Trace Elements: Not Expected to Cause
Environmental Problem. Most Leachates Meet
Drinking Water Standards
TOC in Leachate: Low
FGD
High Concentrations of Mg. Cl in Addition
to Ca. S04. and TDS. (F
of Double-Alkali System)
to Ca. S04. and TDS. ( Plus Na in the Case
Major Component CaSO.
1/2 H20: Potential
Environmental Hazard for Untreated Sludge
pH= 5 to 10 for Lime or Limestone
Scrubbing Systems
pH =12 to 13 for Double-Alkali System
)• Several Elements in Liquor and Leachate, e.g.
As. Se, Cd. Mn NO^. and F. Exceeding the
Drinking Water Standards Including the Ponded
and Oxidized Sludges
•Improvement with Stabilization
TOC in Leachate: Low
TOC in Liquors: Low
* AFBC and PFBC
-------
The comparative cost for on-site disposal based on fixation of the
scrubber sludge with a fixing agent and formation of solid compacts with
the FBC spent sorbent using water has been calculated on the basis of
TVA report costs. The capital and operating cost result in a cost of
about $7/ton ($8/Mg) dry residue for AFBC and PFBC plants and $12/ton
($13/Mg) sludge for limestone wet scrubbing plants. The quantity of
spent sorbent from FBC processes will generally range from 0.3 to 0.7
kg/kg coal for a once-through sorbent utilization and will depend on
the sorbent, coal, FBC design, and emission requirement. The comparative
quantity of solids from the limestone wet bcrubber will be approximately
0.4 kg sludge/kg coal or 0.2 kg dry solids/kg coal.
The impact of averaging time criteria related to the sulfur oxide
emission standard has not been factored into this evaluation. .Varia-
tions in coal sulfur content, sorbent properties, or combustor operability
(coal and sorbent feeding, bed maldistribution, etc.) with time could
reduce FBC sulfur removal reliability.
The effect of coal ash-content and coal heating value on the-
economics of FBC with high-sulfur Eastern coals is of secondary importance
compared to the effect of coal sulfur-content and calcium-to-sulfur ratio.
These secondary variables are not discussed in this report.
ADVANCED SULFUR REMOVAL SYSTEMS
More stringent sulfur oxide emission standards may lead to increased
economic incentives to develop advanced sulfur removal systems for FBC.
The small-scale investigation of several advanced concepts has been or
is being performed. For example:
« Sorbent regeneration
- Reductive decomposition
- Two-step sulfide generation, H2S generation
- Sulfide-sulfates solid-solid reaction
a Precalcination of once-through sorbents for improved reactivity
and sorbent utilization
30
-------
e Additives for improved sorbent utilization
« Reconstitution of sorbent fines by agglomeration
• Alternative metal oxide sorbents.
All of these advanced concepts could potentially reduce the rate of
sorbent consumption and improve FBC economics and environmental impact.
31
-------
VI. PARTICULATE CONTROL
Particulate control in AFBC and in PFBC are quite different in
nature. For AFBC conventional technology exists - hot electrostatic pre-
cipitators (°-400°C) or fabric filter techniques (baghouses) - which
should be capable of controlling particulate emissions to the 0.03
Ib/MBtu (12.9 ng/J) level. Testing of these control options on operating
fluidized-bed combustors, however, is necessary before actual control
performance and any operating problems can be firmly defined. No such
testing has yet been conducted.
For PFBC the protection of the gas turbine from erosion and deposi-
tion damage may require that particulate emissions be controlled to
levels probably less than 0.03 Ib/MBtu, depending on size distribution
and.composition. Granular:bed filters, ceramic filters, and high- •
temperature precipitatprs1 have been proposed for hot gas cleaning for
PFBC but have not yet been demonstrated.
The total loading and particle size distribution of particulates
attrited and elutriated from a fluidized-bed combustor have been pro-
jected, using a Westinghouse model; these projections have been com-
pared with the expected performance of particle control equipment that
might reasonably be employed (see Appendix D). The projections suggest
that, over the range of calcium-to-sulfur ratios considered' (up to. 10),
within the variability of sorbent attrition performance expected, and
based on the current data available, particulate emission levels of 0.03
Ib/MBtu (12.9 ng/J) shou-ld be economically achievable for AFBC and PFBC,
based upon the particulate control equipment costs included by Westing-
house in ±his study (see Append-ix B).
For AFBC the major uncertainties involve the details of the equip-
ment specifications (filter materials, face velocities, etc.). and oper-
ating and maintenance procedures. The uncertainties involved in the
32
-------
PFBC particulate control system are greater, but the level of the
emission standard may have little impact upon the development of these
systems because of the turbine protection requirements, which are expected
to be controlling unless the particulate size distribution is relatively
fine. If future emission requirements are established that require con-
trol of fine particulates , (<10 pm), then environmental concerns may
become controlling.
As with sulfur removal control, the combustor design and operating
conditions have a great impact on the control of particulate emissions.
With little cost penalty the combustor can be designed and operated
(e.g., at low velocities) to limit particle attrition and elutriation,
though very efficient final-stage particulate control devices will still
be required. The coal ash-content and coal heating value have little
effect on the particulate control requirements for FBC relative to the
coal sulfur-content and the calcium-to-sulfur ratio.
33
-------
VII. CONTROL OF OXIDES OF NITROGEN
A statistical evaluation of FBC NO emissions data has been per-
X
formed under contract to EPA (EPA contract number 68^02-2132). Modeling
of the formation of oxides of nitrogen in fluid-bed combustors is also
being attempted •. The data from PFBC units show a high probability of
emissions significantly lower than (h6 lb N62/MBtu (258 ng/J). AFBC
emissions should satisfy a 0.6 lb NCL/MBtu (258 ng/J) requirement but
are hot generally as low as in PFBC-.
For the control of sulfur oxides and particulates, the control
techniques and relation to FBC operating and design parameters are
relatively clear. For nitrogen oxides, however, the relation between
emissions and operating and design parameters in FBG are not clearly
understood; Efforts are under way to improve understanding of these
relationships and to develop NO, control strategies that can reduce FBC
X
NO. emissions below their inherently low levels.
34
-------
VIII. CONCLUSIONS
On the basis of available information, the more stringent emission
requirements considered in this study (sulfur oxides: 90%
sulfur removal; particulates: 0.03 Ib/MBtu (12.9 ng/J); oxides
of nitrogen: 0.6 Ib N02/MBtu (258 ng/J) should be economically
achievable for both AFBC and PFBC power plants.
The economical realization of these environmental goals depends
critically on the proper selection of fluid-bed combustor design
and operating conditions. In particular, the gas residence time
in the bed (as determined by gas velocity and bed height) should
be sufficiently long, and sorbent particle size should be suf-
ficiently small. In this assessment residence times of 0.67 to
2.0 s (gas velocities of 1.5 to 1.8 m/s) and partial size averag-
ing 500 pm appeared to offer effective SO- removal performance,
although these conditions are not necessarily optimal.
The high level of sulfur oxides emission control considered
has a greater impact on the FBC power plant energy cost than
do the revised particulate and oxides of nitrogen standards
considered. The most critical process parameter, with respect
to FBC power plant cost and performance, is the calcium-to-
sulfur ratio.
The fluid-bed combustor cost is not strongly dependent on changes
in design and operating conditions. The fluid-bed combustor
should be designed to minimize plant energy cost rather than to
minimize the combustor cost. For example, low - rather than
high - fluidization velocities will probably result in lower
FBC power plant energy cost.
35
-------
e Particulate control to levels as low as 0.03 Ib/MBtu (12.9 ng/J)
should be economically achievable for AFBC using commercially
available techniques. Baghouses seem most suitable for this
duty. However, no testing of any type of final-stage particle
control device on an AFBC unit has yet been conducted.
• Particulate control to levels below 0.03 Ib/MBtu (12.9 ng/J) may
be dictated for PFBC by turbine protection requirements, depending
on particle size. Projections indicate that 0.03 Ib/MBtu
should be achievable, but the technology to meet this control
at high temperature and pressure has not yet been demonstrated.
• .Oxides of nitrogen will generally be emitted by FBC at levels
below the 0.6 Ib NO^/MBtu (258 ng/J) requirement considered in
this evaluation. No direct control techniques for oxides of
nitrogen have been clearly demonstrated on fluidized-bed com-
bustors to date, although several options are under study.
• The greatest FBC power plant uncertainties presently involve
reliability questions - e.g., solids feeding, particulate
control (especially for PFBC),' material erosion/corrosion/
deposition, and process control. The impact of emission
standard averaging time basis and system reliability have
not been evaluated.
a AFBC and PFBC development programs should focus on more
stringent emission standards and their relation to combustor
design and operating conditions.
® Advanced FBC sulfur removal concepts, for example, sorbent
precalcination, sorbent regeneration, sorbent fines recon-
stitution, additives for improved sorbent utilization,
alternative metal oxide sorbents should be evaluated with
respect to more stringent emission standards.
36
-------
IX. REFERENCES
1. Archer, D. H., et al., "Evaluation of the Fluidized-Bed Combustion
Process," Vols. I-III, report to EPA, Westinghouse Research and
Development Center, Pittsburgh, Pa., November 1971, Contract 70-9,
NTIS PB 211-444, 212-916, and 213-152.
2. Keairns, D. L., et al., "Evaluation of the Fluidized-Bed Combustion
Process," Vol. I-III, report to EPA, Westinghcuse Research
Laboratories, Pittsburgh, Pa., 15235, December 1973, EPA-650/2-73-048a,
b, and c, NTIS PB 231-162, PB 231-163, and 232-433.
3. Keairns, D. L., et al., "Fluidized-Bed Combustion Process Evaluation -
Phase II - Pressurized Fluidized-Bed Coal Combustion Development,"
report to EPA, Westinghouse Research Laboratories, EPA-650/2-75-027c,
September 1975, NTIS PB 246-116.
4. Newby, R. A., and D. L. Keairns, "Alternatives to Calcium-Based SO
Sorbents for Fluidized-Bed Combustion: Conceptual Evaluation,"
report to EPA, January 1978, EPA-600/7-78-005.
5. Newby, R. A., S. Katta, and D. L. Keairns, "Calcium-Based Sorbent
Regeneration for Fluidized-Bed Combustion: Engineering Evaluation,"
report to EPA, March 1978, EPA-600/7-78-039.
6. Alvin, M. A., E. P. O'Neill, L. N. Yannopoulos, and D. L. Keairns,
"Evaluation of Trace Element Release from Fluidized-Bed Combustion
Systems," report to EPA, March 1978, EPA-600/7-78-050.
7. Sun, C. C., C. H. Peterson, R. A. Newby, W. G. Vaux, and D. L. Keairns,
"Disposal of Solid Residue from Fluidized-Bed Combustion: Engineering
and Laboratory Studies," report to EPA, EPA-600/7-78-049, March 1978.
8. Energy Conversion Alternative Study (ECAS), General Electric Phase II
Final Report, Vol. II, NASA CR-134949, 19.77, NTIS .PB 269-379.
37
-------
9. Utility Boiler Design/Cost Comparison: Fluidized-Bed Combustion
versus Flue Gas Desulfurization, report by TVA to EPA, EPA-600/
7-77-126, November 1977.
10. Evaluation of Phase 2 Conceptual Designs and'; Implementation Assess-
ment Resulting from the Energy Conversion Alternatives Study (EGAS),
submitted by NASA, Lewis Research Center, to ERDA and NSP, April
1977, NTIS PB 270 017.
11. Beecher, D. T., et al., Energy Conversion Alternatives Study (EGAS),
Westinghouse Phase II, Final Report, "Summary and Advanced Steam Plant
with Pressurized Fluidized Bed Boilers," Vol. Ill, prepared for NASA
October 1976, NASA CR-134942, NTIS PB 268-558.
12. Garner, D. N., et al., "A Comparison of Selected Design Aspects of
Three Atmospheric Fluidized Bed Combustion Conceptual Power Plant
Designs," Radeon Corp., presented at the 5th International Fluidized
Bed Combustion Conference, Washington, DC, December 1977.
13. Vaux, W. G. , and D. L. Keairns, "Particle Attrition in.Fluidized-
Bed Combustion Systems," 27th Canadian Chemical Engineering Con-
ference, Calgary, Alberta, October 23-27, 1977.
14. McGlamery, G. G. and R. L. Torstrick, "Cost Comparison of Flue Gas
Desulfurization Systems," Proceeding of Symposium on Flue Gas
Desulfurization, Atlanta, GA, November 1974, Vol. I, p. 149, NTIS
PB 242-572.
15. Sun, C. C., C. H. Peterson, and D. L. Keairns, "Environmental Impact
from the Disposal of Unprocessed and Processed FBC Spent Sorbent and
Ash." Proceedings of the 5th International Conference on Fluidized
Bed Combustion, Washington, DC, December 12-14, 1977.
38
-------
APPENDIX A
SULFUR OXIDE REMOVAL DATA BASE AND MODEL
SUMMARY
The projections of calcium-to-sulfur molar feed ratio required to
achieve any selected degree of desulfurization in a fluidized-bed com-
bustor are derived using a simplified model for fluidized-bed desulfuri-
zation, with kinetics rate constants developed using laboratory thermo-
gravimetric (TG) data. For confirmation, where possible, the TG-supported
model has been compared with available data from fluidized-bed combustors.
The TG data base, the model, and its agreement with fluidized-bed data
are discussed. Operating and design conditions under which a typical
limestone can be used to maintain 90 percent sulfur removal in AFBC are
outlined. Areas for improvement of the projection technique are
identified .
INTRODUCTION
The feed ratio of calcium to sulfur required for desulfurization in
fluidized-bed combustion depends on the system design, the operating
conditions, the particular calcium-based sorbent, and the sorbent particle
size used. In assessments of the overall FBC system it has been con-
venient in the past to select calcium/sulfur feed ratios of about 3/1
for AFBC and about 2/1 for PFBC as base cases, and, over the years,
these estimates have acquired a more permanent status, without reference
to any particular sorbent, system design, or operating conditions. Desul-
furization in the bed, however, is essentially a competition between two
processes: once sulfur dioxide is released in the bed it may either
remain in the gas and escape or react with calcium oxide (CaO) or
39
-------
carbonate (CaCO~) in the bed to form calcium sulfate (CaSO,). The
balance between these processes can be drastically altered by enhancing
the probability of escape (shallow bed, high gas velocity) or by
changing the reaction rate of the sorbent (e.g., decreasing the sorbent
particle size, raising the bed temperature at atmospheric pressure).
In order to predict the calcium-to-sulfur molar rate required for
a given level of desulfurization, a model of the desulfurization system
that encompasses the effects of the relevant variables is required.
Such a model must succeed in correctly projecting the effects of key
variables on the system. In particular, since the following effects
have been observed in experimental combustors, it should demonstrate
• That an optimum temperature for desulfurization exists at
atmospheric pressure using sorbent particle of size 500 ym or
larger
• That sorbent utilization improves at higher pressures
('VLOOO kPa)
• . That there is.no marked, temperature .effect at higher pressures
(0,1000 kPa).
Attempts to model fluidized-bed desulfurization to the extent of
predicting calcium-to-sulfur molar feed ratios may be: a) statistical
correlations based on analysis of fluidized-bed combustion data (Babcock
i r,
A3
A2
& Wilcox [B&W], Battelle ); b) models which attempt to use sulfation rate
data obtained in the laboratory (Argonne National Laboratories [ANL],
A4 A5
Westinghouse, National Coal Board [NCB] ; or c) fundamental models
in which the gas/solid reaction model is matched with the fluid dynamics
of the bed (no successful model known).
The correlation models are generally untrustworthy outside the
range of data correlated, and since the objective is to project desulfur-
ization under novel constraints, they are not useful here. Of the other
models, the simplest and most direct is that developed at Westinghouse,
and it was used here because it has given satisfactory results in the
past.
40
-------
It may be noted that recent fundamental gas-solid modeling by
A6
Hartman and Coughlin has yielded a sufficiently close prediction of
effects observed in practice to be considered as a starting point for
the development of a fundamental model: development and analysis of a
reasonable body of self-consistent rate data has not yet been reported.
THE WESTINGHOUSE MODEL
The model used here has been described previously (Appendix E of
Reference A4). The following assumptions are made in deriving the model:
1. Release of sulfur from coal as sulfur dioxide (SO ) occurs with
equal probability at any place in the bed.
2. Sulfur dioxide released at a distance, h, below the upper bed
surface passes in plug-flow through a homogeneous bed of sorbent
of height h. If a mean reaction rate, k, can be assigned to the
absorption of gas in the bed, the fraction of the sulfur
-kt
retained is 1 - e , where t is the gas residence time in the
bed of height h. Integration for all points,h along the bed
height yields an expression for the fractional sulfur removal,
1 —kz
R, R = 1 - — (1-e ), where z is the residence time of gas
entering the bed from below.
3. The mean rate constant for gas absorption in the bed, k, is
obtained at any fixed mean utilization of the sorbent from
thermogravimetric data reduced to unit gas concentration. The
rate constant for any given sorbent type and chemical form will
depend on the mean sorbent utilization, sorbent particle size,
bed temperature, bed pressure, and gas composition in the bed.
The model of sulfur generation in the bed assumes that sulfur is
released from coal with equal probability at all places in the bed.
While this is a reasonable general assumption, it implies that, for
example, .10 percent of the sulfur is released within the top 10 percent
of bed height. Assume that all the sulfur generated in the bed below the
top 10 percent is captured with 100 percent efficiency, and the overall
41
-------
desulfurization target is 97 percent. Then the top 10 percent of the
bed must capture the sulfur liberated within it with at least 70 percent
efficiency, despite the small remaining bed height and effective gas
residence time.
As the emission standard becomes more stringent, the reaction
:ant term, kz, ;
lowing table shows:
constant term, kz, required between SO- and CaO increases as the fol-
Rate Constant/Rate
Constant at 80%
Effective Sulfur Dioxide (at constant gas
Removal Standard, % residence time)
80 1
90 2
95 4
97.5 8
98.75 16
The rate constant for limestone sulfation required for varied sulfur
removal efficiencies is projected in Figure Al as a function of gas
residence time. In practice, the rate constant could be varied by chang-
ing, e.g., the sorbent particle size or bed temperature. Alternatively,
bed residence time could be adjusted by varying gas velocity or bed
depth.
If all of the sulfur is actually liberated in the lower two-thirds
of the bed, then a different picture emerges. The top third of the
bed behaves only as an absorber into which SO is fed from the bottom.
The reaction rates required for desulfurization with this kind of sulfur
generation pattern are illustrated in Figure A2.
A3
A generalized sulfur generation model has been published by ANL,
and this could be incorporated into the model discussed here, if desired.
The choice of a particular sulfur generation pattern, however, in any
case is speculative.
42
-------
Curve 69M87-A
100
90
C
o
°
-------
Curve 694388-A
C
.2
03
GO
i_
£
"c
C
o
o
o>
•+-*
ro
45
40
35
30
^ 25
20
15
10
75
Residence time t = l s
Uniform
Generation
Model
Lower Bed
Generation
Model
80
85
90
95
100
SCL Removal Required
Figure A2 - The impact of sulfur generation pattern and
desulfurization requirement on required
reaction rate
-------
Operating conditions that would disturb the uniform sulfur genera-
tion pattern are: the location of coal feed points and their number, the
coal particle size, and the extent of vertical mixing in the operating
fluidizing mode. Thus, fine particles of coal fed near the base of a
tall bed at low fluidization would probably cause disproportionate
release of SCL near the base of the bed and improve sulfur removal, rela-
tive to the prediction of a uniform sulfur generation model.
TG DATA
The use of TG data to model the fluidized-bed sulfation depends on
several assumptions, which are listed below:
1. The rate-limiting process is governed by diffusion within the
sorbent itself.
2. The residence time of the sorbent in the bed does not affect
its reactivity (although the degree of sulfation does affect
reactivity).
3. The fluidized--bed atmosphere surrounding the limestone particles
is oxidizing with respect to the CaS/CaSO transition.
The implication of the first assumption is that while mass transfer
(e.g., of SO- from the gas phase to the sorbent particle) in the fluidized
bed may be much greater than in the TG apparatus, mass transfer dominates
the actual reaction rate over a small extent of reaction. Once the cal-
cium utilization has reached about 10 percent, the rate is usually dom-
inated instead by inter- and intragranular diffusion of the solid. (In
same pressurized cases comparatively low mass transfer depresses the
observed TG rate over a significant portion of the rate curve). The same
internal pore structure must be provided in the sorbent during the TG
test as would prevail during a fluid-bed combustor operation. Thus, the
calcination conditions under which the sorbent pore structure is generated
must be carefully controlled before the TG sulfation experiment if the TG
data are to provide a valid simulation of the effect of sorbent utiliza-
tion on reaction rate.
45
-------
Concerning the second assumption above, the mean residence time of
the sorbent may be much longer in a fluidized bed (up to 24 hours) than
in a TG experiment (^2 hours), unless the latter is carried out under
conditions of low-SO~ concentration (0.05%). The sorbent pore structure
may be modified by this long exposure to temperature, and, unfortunately,
this may either enhance or retard the sulfation rate, depending on the
particular sorbent. This effect is not accounted for in the TG-based
kinetic model.
It is often postulated that the existence of local reducing areas
in the bed may cause sulfur capture as CaS, which is subsequently oxidized
to sulfate. The assumption that this mechanism is unimportant in deter-
mining the overall rate of reaction is justified by the fact that oxida-r
tion of CaS in limestone is limited in the same way that utilization of
CaO is limited in sulfation. Oxidized sulfided limestone contains an
inner layer of sulfided stone which has not been oxidized. Since CaS is
not found in the product stone from FBC, it is clear that there is no
significant additional reaction rate component from sulfidation of the
sorbent.
The TG data, which yield rate as a function of sulfur loading, or
sorbent utilization, vary greatly according to the sorbent properties
and combustor operating conditions. The variables that affect the reac-
tion rate and can be studied on the TGA are
1. Type of sorbent
2. Form of sorbent (e.g., limestone, lime, hydrated lime)
3. Particle size of sorbent
4. Calcination conditions (in particular the temperature and
partial pressure of C0« during calcination)
5. Residence time (time at temperature)
6. System pressure
7. System temperature
8. Percent excess air in the system
9. SCL partial pressure and other gas composition variables.
46
-------
In making a projection of the effect of operating conditions on the
calcium-to-sulfur molar feed ratio, it is advisable to use TG data
obtained under experimental conditions that correspond closely to those
that will be encountered in the fluidized bed.
For any set of design input:
Bed height
Gas velocity
Bed voidage
Bed emulsion and bubble phase volume
Coal:
Ash content (heating value)
Sulfur content
the rate data can be used in the model to give calcium-to-sulfur molar
feed ratio projections. A data base of over 300 atmospheric-pressure and
70 pressurized TG runs exists at Westinghouse over the range of condi-
tions listed in Tables Al and A2.
BASE OPERATING CONDITION PROJECTIONS
Sorbent feed projections were obtained for the base operating condi-
tions (Table A3) using TG runs 231, 241,, P17 and 75-281. The rate data
were fitted with polynomial equations for use in the computer generated
projections. Figures A3-A6 illustrate the polynomial fits used for TG
rate data.
The average rate constant for S0« sorption in the bed, K, that is
needed to maintain a given level of sulfur retention, R, was calculated
, . . . . ,„ expanded bed height N . ,
at a defined gas residence time (Z = - - *——. — :— : - ~ - : - ) using the
6 interstitial gas velocity b
equation
Note that K is related to the first-order rate constant for the reaction
CaO + S02 + 1/2 02 •> CaS04, K1^ by
- 1 (1 - Y)e(l ~ F)
-
47
-------
where
K1 = f(U)
Y = volume fraction of bed bubbles
e = bed voidage in emulsion phase
F = fraction of emulsion volume occupied by inerts.
For these projections, the following were assumed:
e = e TG = 0.5
Y = 0.5
F = 0 .
In other words, it was assumed that the volume fraction of active emulsion
phase in the bed is 0.5 (one-half of the bed volume is tubes, bubbles, and
inerts). Thus, the first-order reaction rate constant can be calculated
for any sulfur retention level.
The sorbent utilization, U, at which the reaction rate constant,
K , applies was determined from TG data. A mass balance on the TG
system gives the molar rate of reaction, -j— , in terms of the reaction
note constant as
dU K C eTG
dt
where
p = solids density, mole Ca/cc
C = mole SO /cc in TG reaction gas.
The polynomial equations representing TG rate data U = f(—) were then
dt
used to calculate the sorbent utilization. The required calcium-to-
sulfur molar feed ratio, C, is then defined by
C -^
L u '
The dolomite calcium-to-sulfur ratio projections in Table A3 for the
PFBC case were derived from Figure A5 and A6, using available TG data
that were not at the exact conditions of the base PFBC case. As indicated,
48
-------
Table Al
RANGE OF ATMOSPHERIC TG SULFATIONS
Stone(s)
Ames
1359
Size,
U.S. Mesh
% Excess
Air
Temp . ,
°C
Comment
, Brownwood, 16/18 20 815 Varied calcination
, Greer, Carbon 5—100/200
1337, Western,
Mississippi
Bellefonte
Greer
Ames
Ames, Brownwood
Greer, Carbon, 1359
Ames, 1359,
Greer, Carbon
1359 (2263)*
Carbon
Tymochtee
Canaan, Kaiser
16/18'
35/40
35/40
5/6-35/40
20
20'
20
20
780-950 Varied calcination
750-940
900
850
Calcined at 900°C in
60% C02
Scattered particle
sizes
16/18
-325
16/18
100/200
20
20
2-16% 0,
^
20
=815
800-950
, 815
815
Calcined at 900°C in
60% CO.
*Residence time varied (0.1-0.5% SO. in sulfating atmosphere)
49
-------
Table A2
RANGE OF PRESSURIZED TG SULFATIONS, 10 atm
Stone
Size j
U.S. Mesh
% Excess
Air
Temp . ,
°C
Comment
Canaan
1337
1337
Greer
Greer
Greer
Greet
Greer
Greer*
fyihochtee
1359 (2263)*
35/40
35/40
8/10
35/40
6/8
40/100
100/200
16/18
14/18
16/18
16/18
300
300
20
300
300
300
300
15-100
20
Calc. 815
(0.15 atm C0_)
0.75-16% 02
20
843
843-1010
900
843-1010
900-1010
90'6'-1010
950-1000
815
815
815
815
Half-calcined
Carbonated
FB Calcine 815,
15% CO,,
*Residence time varied (0^05-0.5% SCL in sulfating atm).
50
-------
Pressure, kPa
Bed Temperature, °C
Excess Air, %
Sorbent Type
Sorbent Diameter, pm
Bed Velocity, m/s
Bed Depth, m
Ca/S for 1.2 Ib/MBtu SC>2
Emission3
Ca/S for 90% Sulfur Removal
Efficiency
Table A3
FBC PLANT OPERATING.CONDITIONS
AFBC
101
840
20
Limestone
1000
3.05
1.22
5.5
500
1.83
1.22
2.4
7.0
2.9
PFBC
1013
950
20
Dolomite
2000 500
1.52 1.52
3.05 3.05
2.0 1.2
4.5
1.7
83% sulfur removal for 4 wt % sulfur coal.
51
-------
4-1
c
•H
e
•g 3
o
o
n)
BJ
Carbon Limestone, 1000-1190 ym particles
815°C, 0.5% S02> 4% 02
Calcined nonisothermally up .to 815°C in 15% CO,
* .Polynomial Fit (fraction =0.2063 - 10.87 • rate +
275.9
? 3
rate - 2807 • rate )
Run 231 data points
0 0
*
.*
..*
o*
*
*
•••*
A
0 of 00
**
*..
J 1 *
0.04 0.08 0...12 0.16 0.20
Fraction 'sulfated
Figure A3 - Polynomial fit of TG rate data from run 231
-------
Ul
Carbon Limestone, 420-500 pm particles
815°C, 0.5% SO , 4% 02
Calcined nonisothermally up to 815°C in 15% CO
Polynomial Fit (fraction = 0.3794 - 6.351 •
^
CO
-------
3
C
•H
e
c
•1-1
o
<=>
Dolomite 1337, 420-500 ym particles
10 atm, 927°C, 4% C02, 16% 02, 0.5% S02
Calcined nonisotherraally up to 927°C in 16% 02, 4% C02
* Polynomial Fit (fraction = .5485 - 4.632-rate)
• Run P17 - data points
oo
o
-1
-2
4- 1-
.1 .2 .3 .4 .5 .6 .7 .8
Fraction Sulfated
Figure A5 - Polynomial fit of TG rate data from Run P17
-------
Dolomite 1337, 1680-2000 urn particles
10 atm, 800°C, 0.4% S00, 10% 0_, 12% CO,
2. 2. i
Half-calcined
Ul
Ln
/— s
.H
1
CO
i-l
3
C
*rH
a
c
•H
O
O
i— 1
X
01
n)
f*5
4
3.5 •
3
2.5 •
2
1.5 •
1
.5 •
0
0
* Polynomial Fit (fraction = 0.
* 624.4 rate2)
• Run 75-281 data points
*
. *
*
*
•
*
*
\
*
.*
*
' *-
. •
1 I • I t t * 1
0.1 0.2 0.3 0.4 0.:5 0.6 Or..7 0.8
Fraction sulfated
Figure A6 - Polynomial fit of TG rate data from run 75-281
-------
the run P17 TG data in Figure A5 are for 300 percent excess air (16 per-
cent 0~ in the flue gas), whereas the base case assumes 20 percent excess
air; the run 75-281 data are for an 800°C half-calcined dolomite of
2000 ym. Since the technical/economic analysis presented in the
body of this report was completed in the basis of Table A3, subsequent
TG runs have been made with dolomite 1337 at the exact conditions assumed
for the PFBC base case (runs:P96 and P104, see Figure A7 and A8). It was
.found that using runs P17 and 75-281 to approximate the base case condi-
tions was not completely accurate; both the 500 ym and, in particular, the
2000 ym particles were more reactive than had been assumed. The effect
is especially apparent at 90 percent S0_ removal. The revised calcium-
to-sulfur projections (for comparison with Table A3) for PFBC are:
Sorbent type Dolomite
Bed velocity, m/s 1.52
Bed depth, m 3.05
Sorbent diameter, ym 2000 500
Ca/S for 1.2 Ib/MBtu SO 1.9 1.1
emission3
Ca/S for 90% sulfur removal 2.8 1.3
efficiency
o
83% sulfur removal with a 4 wt % sulfur coal
If the technical/economic assessment discussed in the body of this
report were repeated using these new calcium-to-sulfur projections, the
PFBC cases would appear more economically attractive than they did in the
original assessment.
CONFIRMATION OF THE MODEL
The model used to project the effects of more stringent S0? standards
has been used previously to show that TG data accurately demonstrate
important features of desulfurization phenomena in fluidized beds. Desul-
furization phenomena that have been observed in fluidized beds and
A4
demonstrated on the TG include:
56
-------
CO
a)
4-1
•H
e
0)
4-1
cti
C
O
4J
n)
3
CO
Dolomite 1337, 420-500 ym
950°C, 3.5% 02, 0.5% S02
Calcined nonisothermally up to 950°C in
14% C02, 3.5% 02
* Polynomial Fit (fraction = 0.8657 - 4.730
rate + 1.451 • rate )
• Run P96 data points
.1
.2 .3 .4
Fraction sulfated
.5
.6
.7
.8
.9
Figure A7 - Polynomial Fit of TG Rate Data from Run P96
-------
00
.1 •
X
H
•/
CO
0)
4-J
3
c
•H
^
Ol
CO
M
C
0
•H
CO
U-l
3
C/3
Dolomite 1337, 2000-2380 yin
*° :950°C, 3.5% 00, 0.5% S0~
*• z z
\ Calcined nonisbthefmally up to 950°C in
** 14% CO., 3.5% 0.
.* 2 2
\
• * * Polynomial -fit
^ * (fraction^ .898 -30.6 rate +415. 7 rate
\ -• Run P104
\
*^
*$?
*•.
* •
**.
* •
*••
*••
^v
-^k
^
****,
*"*»•>
"*"*-M.^
-• • | | ........ | i -. | . . - 1 *"lft
!) !i '2 !s !A Is !e !? Is ,9
- 2019 rate )
Fraction sulfated
Figure A8 - Polynomial Fit of TG Rate Data from Run P104
-------
o The occurrence of an optimum temperature for desulfuriza-
tion in AFBC with sorbent particle sizes of greater than
500 ym
o No marked temperature effect at higher pressures (^1000 kPa)
• Improved sorbent utilization at higher pressures (^1000 kPa)
• Improved sorbent utilization with precalcination.
The specific data that the TG has been used to model include data from
the NCB, ANL, B&W, Pope, Evans and Robbins (PER), and Westinghouse.
Model projections of the calcium-to-sulfur molar feed ratios required
for various levels of desulfurization in AFBC, as a function of limestone
type, are compared to the data collected from the ANL and British Coal
(A2)
Research fluid-bed units for limestone 1359 in Figure A9. Conditions
for the fluid-bed runs were:
1 atm, limestone 1359
490-630 ym particles in the.feed
2.6-2.8 ft/s
788-798°C
2 ft bed height
3% 02, 15% C02 in the flue gas.
To obtain the projections, TG rate data from sulfation at 815°C in 0.5%
SO,,, 4% 0?, and N_ were utilized. The sulfations were carried out with
420 to 500 ym particles of.limestone, calcined at 815°C in 15% C02 and
nitrogen. The gas residence time (as determined by input bed height and
velocity) was 0.66 s. ANL operated with a gas residence time of 0.74 s.
This longer residence time may account for the slightly lower calcium-to-
sulfur molar ratio requirements in the ANL 1359 data.
Projections of the calcium-to-sulfur molar feed ratios required
for various levels of desulfurization in PFBC are compared with data
collected from the Exxon miniplant in Figure A10. It should be noted
that the scatter of miniplant data is partially due to the range of opera-
ting conditions chosen for the plant.
59
-------
Curve 691792-0
c
o
§
"8
ce
"c
o
Fluid-Bed
Operating Conditions:
1 atm, 101. 3 kPa
420- 500 urn limestone particles
Bed Height 4 ft
Velocity 6 ft/ s
815°C
-20% excess air
Carbon Limestone
Greer Limestone
Limestone 1359
ANL best fit of data collected
for Limestone 1359(1971 )
Curve 693259-A
-100
-90
-70
I -60
a:
a1* -50
-40
-30
-20
-10
o Exxon Miniplant Data
for Pfizer Dolomite
(Jan177) [830-950°C1
TG Data for Run 73 -150
Using:
Voidage
BedHt
Sup. Vel
Particle Size
Temp
Pressure
0.67
4m
1.6m/s
420-500 urn
871°C. (1144K) '
lOAtm, 1013kPa
Sorbent
i
Pfizer
Dolomite
3 4 5
Ca/ S Molar Ratio
1 2
Ca/S Molar Feed Ratio
Figure A9 - Predictions of the Ca/S feed
ratios required for desul-
furization using Westinghouse
TG data for three sorbents
Figure A10 - Sulfur removal projected from TG
data and measured on the Exxon
miniplant
-------
VALIDITY OF THE ASSESSMENT PROJECTIONS
A computerized file of fluid-bed combustor data from PER, the NCB,
ANL, Exxon, and B&W was searched for combustor. runs carried out at operat-
ing conditions near the base case conditions for the technical/
economic assessment. The limited bench-scale and pilot plant data avail-
able at the base case conditions indicate that the TG projections are
representative of the fluid-bed data.
Atmospheric combustor data at operating conditions near the base
case are given in Tables A4 and A5. Note that, under base case conditions,
80 percent sulfur removal has not been obtained at a calcium-to-sulfur
ratio of less than four with 1000 ym particles. Figure All compares the
calcium-to-sulfur feed requirements for desulfurization from Table A4
with the TG projections for 1000 ym particles. The 500 ym data combustor
runs are plotted in Figure A12 with TG projections. The scatter of com-
bustor data is partially due to the range of operating conditions plotted,
i.e., 799 to 899°C temperatures, and varied sorbent types.
Pressurized combustor data at operating conditions near the base case
with a defined sorbent particle size were not in the data file. Recent
A9
results from the Exxon miniplant, however, indicate that greater
than 97 percent sulfur removal can be obtained with calcium-to-sulfur
molar feed ratios of less than 2. This is in line with PFBC projections
for the base case.
DESIGN CONDITIONS FOR 90 PERCENT SULFUR REMOVAL IN AFBC
The Westinghouse model was used to project the design and operating
conditions required to maintain 90 percent sulfur removal in AFBC. TG
rate data from finely divided particle size fractions of limestone, with
diameter, d, were used to generate calcium-to-sulfur molar ratio require-
ments for 90 percent sulfur removal as a function of gas residence time
61
-------
Table A4
FLUID-BED COMBUSTOR DATA AT CONDITIONS NEAR THE 1000 urn
AFBC BASE CASE PROJECTIONS
Study
Sorbent
Particle
Size,
Mm
Bed
Temperature,
°C
Approximate
Gas
Residence
Time , sa
Ca/S
Molar
Ratio
% S02
Removal
Continuous-Feed Unit Tests:
ANLV '
ANL(A10)
ANL(A10)
NCB(AU)
B&W(A12)
1360
1360
1359
18
Carbon
1000-1400
1000-1400
1400
<3175
<6350
871
871
871
777
819
0.42
0.42
0.67
0.38
0.19
2.5
4.2
1.6
2.9
2.8
72
86
18
73
58
Batch Unit Tests:
PER(A13)
,Exxon(A14)
.(A14)
Exxon
(A14)'
Exxon v '
(A14)
Exxon
Westinghouse
p\
Ha Q P£* c -f r( on
1359
1359
1359, .
1359
1359
840-2381
930
930.
930
930
860
871
871
982
927
0.14
0.25
0.25
0.25
0.28
16.0
6.8
5.4
5..1
5.9
93
90
90
90
90
Projection:
Carbon
Carbon
r p T -i m o
1000-1190
1000-1190
Bed height
840
840
expanded
0.4
0.4
5.5
7.0
83
90
Superficial Velocity
62
-------
Table A5
FLUID-BED COMBUSTOR DATA AT CONDITIONS NEAR THE 500 ym
AFBC BASE CASE PROJECTIONS
(Continuous-Feed Unit Tests)
Study
Sorbent
Particle
Size,
pm
Bed
Temperature,
°C
Approximate
Gas
Residence
Time, sa
Ca/S
Molar
Ratio
% S02
Removal
ANLV
ANL(A3)
ANL(A3)
ANL(A3)
ANL(A3)
ANL(A3)
ANL(A3)
ANL(A15)
ANL(A15)
NCB(A11)
NCB(AU)
NCB(AU)
NCB(A11)
NCB(AU)
NCB(AU)
NCB
-------
Curve 696679-A
cu
o:
100
90
80
70
60
50
40
30
20
10
•o* _
o Flu id-Bed Date of AN. L, NCB,_
B&W, PER,and Exxon ( See
Table A4>
* Westinghouse Projeetions -
(Run 231)
840 °C
0.4 s Gas Residence Time-
1
2345
Ca/S Molar Feed Ratio
Figure All - Fluid-bed combustor data at conditions near the
1000 pm AFBC base-case projection conditions
64
-------
100
_ 90
o
o>
Ul
a so
LO
70
Curve 696680-A
o
00
o Fluid-Bed Data of ANL and NCB
(See Table A5)
* Westinghouse Projections! Run 241)
500 Mm Carbon Limestone
840 °C
0.4 s Residence Time
1
3 4
Ca/S Molar Feed Ratio
Figure A12 - Fluid-bed combustor data at conditions near the
500 ym AFBC base-case projection conditions
-------
and sorbent particle size.* The projections are valid for a once-through
AFBC system using no sorbent enhancement options (such as precalcination),
operating at 815°C with 20 percent excess air. For these calculations,
it was assumed that no sulfur is released from the carbon burnup cell
arid that fines' residence time does not limit their utilization. The
technique by which the model was utilized to make these projections was
described in detail in a previous section. The TG data used, and the
polynomials derived from the data, are shown in Figures A13 through A18
(TG data from Figures A3 and AA were also used).
Results for Carbon limestone are shown in Figure A19. The figure
shows that for a 0.5 s gas residence time, 90 percent sulfur removal can
be maintained with a calcium-to-sulfur molar feed ratio of 5/1 when the
limestone particle size is less than 1000 vim. The calcium-to-sulfur
molar ratio can be reduced to 3/1 if the particle size is less than
600 urn.
Using 500 ym particles of Carbon limestone, a 5/1 calcium-to-sulfur
molar feed will1 be sufficient., to remove 90 percent of. the sulfur generated.
with a gas residence time of <0.1 s. Increasing the gas residence time
to 0.18 s will reduce the required calcium-to-sulfur ratio to 3/1. Further
increasing the gas residence time to 1 s will only reduce the calcium-to-
sulfur molar feed ratio to 2.5/1.
It should be noted that the proper selection of design and operating
conditions is an important, but limited, means of reducing calcium con-
sumption. For example, 500 jam Carbon limestone particles will not meet
90 percent sulfur removal requirements with a calcium-to-sulfur molar
feed ratio of 2/1 at any operating or design condition. The TG rate
curve in Figure A20 shows that at 38 percent sulfation of 500 ym Carbon
limestone, the rate drops rapidly as diffusion of S0~ through product
sulfate limits the stone's utilization. If greater than 40 percent
*It was assumed that projections developed from a particle size, d, are
representative of a feed particle size distribution with a mean diam-
eter, d. Past projections based on this assumption have been accurate.
However, distributions in which d does not represent d may exist. No
systematic study on representing feed particle size distributions has
yet been done.
66
-------
Carbon Limestone, 74-149 ym particles
815°C, 0.5% S02, 4% 02
Calcined nonisothermally up to 815°C in 15% CO
* Polynomial Fit (fraction = 0.4968 - 5.25 • rate
7
6
5
^
— i
i
»x
en
OJ
u 4 •
3
C
'e
•5 3
o
o
, — I
* 2
0)
u
ni
Prf
1
0
- 47.21 • rate2)
* .. • Run 224 data points
••*"
• • A
• • •« £
• •&••
*
• ••«*• ••
' .*
. ^* *
• ^k •• •
* ...
• ***...
*•• '
* ••••
A •
*' .
* .
*
..*
• *
...*
" *
*. *
•. *
*.. *
'•. *
"**•
! 1 1 ! ! 1 1 1 1 1 ••••—,
0 0.10 0.20 0,30 0.40 0.50
Fraction sulfated
Figure AJ.3 - Polynomial fit of the TG rate data from run 224
-------
oo
en
0)
4-1
3
a
•H
e
c
•H
o
o
X
HI
4-1
n)
1.6
1.4
1.2 •
.8 ..
.6 ••
0
Carbon limestone, > 4000 ym particles
815°C, 0.5% S02, 4% 02
Calcined nonisothermally up to 815°C in 15%
* Polynomial Fit (fraction = 0.137 - 25.4 • rate +
2053
rate2 - 62858 • rate3)
» Run 213 data points
0.02 0.04 0.06, 0..08 0-.10, 0.12 0;.14 0.16
Fract ion- sulfated
Figure: A14 - Polynomial fit of. the TG rate, data from: run 213
-------
ON
to
4% 02
Calcined nonisothermally up to 815°C
in 15% C02
* Polynomial Fit (fraction = 0.3755-2.714
2
rate r- 104.7 • rate )
e Run 296 data points
* •
4-
??*•
isp-
-»•
0 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45
Fraction sulfated
Figure A15 - Polynomial fit of the TG rate data from run 296
-------
o
en
0)
H
•H
n)
Pi
Limestone 1359, 420-500 ym .particles
., '0.:5% S0, '4% 0
Calcined nonisothermally up to 815°C in 15% C02
* Polynomial Fit (fraction = 0.2776-7.59 ' rate
•'*
-Run 381 data points
0
*»
. * ..
*„
V.
•*-
-•*..
*•*
•".*
-*
X
•••*.
•-*
4-
4-
4-
4-
^"v
0.25
0.05 0.10 =0,.1'5 0.20
Fraction sulfated
Figure A16 - Polynomial fit of the TG rate .data from run 381
0.30
-------
Limestone 1359, 1000-1190
815°C, 0,5% S0, 4% 0
particles
— >
H
1
^
w
J-J
3
•H
B
c
•H
O
O
i — 1
X
HI
4J
cfl
Pi
4.5
4
3.5
T
2.5 •
2
1.5 •
1
.5 •
0
Calcined nonisothermally up to 815°C
, * * Polynomial Fit (fraction = 0.1282-9
.* 23
... * 409 • rate - 5898 • rate )
*
...•a a Run 85 data points
•*«
*
. •*
*
A ...
•ft
* . .
*
*._...
*
4
• * *
. . a. — 9
1 1 1 1 1 "*"*"•••—•»•«.... .
0 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16
Fraction sulfated
Figure A17 - Polynomial fit of the TG rate data from run 85
-------
Limestone 1359, 4000 ym particles
815°C, 0.5% S0_, 4% 0_
s
1
s
to
4-1
C
•rH
e
c
•H
0
O
r— f
X
-------
Dwg. 6424A03
Solids Density =2.70 * 10 mole Ca/cc
Bed Voidage = 0.5
Volume Fraction of Active Emulsion
Phase in Bed-0.5
Bed Temperature-815°C
% Excess Air-20%
T Expanded Bed Height
Superficial Gas Velocity
Figure A19 - Ca/S molar feed required to maintain 90% sulfur
removal in AFBC with Carbon limestone
73
-------
Cfl
w
I-"
0
O
O
w
H
< -1
OE!
-2
RUN 0241
CARBON LIMESTONE : 420/500 um
815 C ; 200 ML/MIN.
4% 02 ; 0.5% S02 ; N2 BAL.
CALCINED NONISOTHERMALLY UP TO
815 C IN 15% CO2
o
.8
.9
FRACTION SULFATED
Figure A20 - The sulfation rate of Carbon limestone
-------
utilization of a 500 ym sorbent is required, a calcine with a more open
pore structure must be used. The pore structure of Carbon limestone can
be modified by a sorbent activation technique, or another sorbent with
a greater utilization potential can be chosen.
The variation in calcium feed requirements with sorbent type is
illustrated by comparing Figures A19 with the results for Grove limestone
(limestone 1359) in Figure A21. Grove limestone may require twice the
calcium molar feed as Carbon limestone to meet 90 percent sulfur removal
requirements.
AMENDMENTS TO THE MODEL
Future amendments to be made on the model include a) integration
of sorbent particle size distribution, b) including the impact of attrition
and elutriation on particle size distribution, and c) discerning the effect
of different mass transfer rates on the TG and in fluid beds.
Instead of using a TG curve based on one particle size representing
the estimated average particle size in the bed, it would be preferable
to use a composite TG curve representing the fuel particle size distri-
bution within the bed. Westinghouse has already calculated such a
composite curve for one particular case - the atmospheric-pressure
' A12
experiments of :B&W, with excellent results..
The Westinghouse TG data for different sizes of Carbon limestone
were combined to derive a projected composite utilization for the size
consistancy in the bed of the Babcock and Wilcox 3 ft by 3 ft (0.9 m by
0.9 m) AFBC unit. The average operating parameters of 20 B&W runs (8 ft
[2.4 m] bed, 1.5 ft [0.5 m]/s, 50% sulfur removal) using Carbon limestone
were used to estimate a molar reaction rate on the TGA (0.1% Ca/min)
which corresponds to the average bed rate constant. A weighted average,
based on the B&W sorbent feed size distributions, of the sorbent
utilization obtained on the TGA at 0.1 percent calcium reacting per
minute was calculated for TG runs on finely divided size fractions of
Carbon limestone at temperatures and gas compositions similar to those
in the B&W combustor. The results of this comparison are shown in
75
-------
Dwo. 6424A02
CD
U_
1—
ro
O
00
r25
20
_
Solids Density 2.70 * 10 mole Ca/cc
Bed Voidage 0.5
Volume Fraction of Active Emulsion
Phase in Bed-0.5
BedTemperature=815°C
% Excess Air = 20%
T Expanded Bed Height
Superficial Velocity
Figure A21 - Ca/S molar feed required to maintain 90% sulfur
removal in AFBC with limestone 1359
76
-------
Curve 690412-A
60
50
CO
§» 40
13
Q_
CO
•E 30
"c
OJ
° 20
o
ro
M
10
i i 1 i I
Carbon Limestone
TG Data
CalcinedNonisothermally upto815°C in 15% CO
Sulfated at 815°C in 4% 02§ 0.5% S02
B&W Data
Bed Temperature 774-858°C
3% SCoal
3-5.3% Excess 02
12-14%CCL in Flue Gas
10 20 30 40 50
Utilization Calculated from TG Data (0.1% Ca/min Rate Criterion)
Using B&W Feed Particle Size Distribution
Figure A22 - Comparison of sorbent capacity obtained from
Westinghouse TG data with B&W fluid-bed
results(A12)
77
-------
Figure A22. Westinghouse feels that the agreement is excellent. This
technique to account for particle size distributions is time consuming,
however, for many TG experiments are required to cover the full range of
particle sizes. A fundamental model that would permit the construction
of the TG curves for different particle size ranges from a base curve
would be of inestimable value.
As another amendment, a more fundamental correction to the model
would result from generation of a mean rate value for the bed given the
residence time distribution of the sorbentin the bed. This mean rate
would consider particle attrition within the bed and elutriation from
the bed.
The third required amendment would adjust the initial rates of
reaction to account for mass transfer effects in the fluidized bed. To
some extent this effect can be calculated from batch fluidized-bed
results. ' Mass transfer should only be significant in cases for sorbent
utilization of 10 percent or less, corresponding to calcium-to-sulfur
molar feed' ratios-:of 10/T or higher. Such high--sorbent feed.-ratesj,lie
outside the range of practical interest, except for generative systems.
REFERENCES
Al Attig, R. C., et al., Additive Injection for Sulfur Dioixide Control,
A Pilot Plant Study, APTP-1176, The Babcock and Wilcox Company,
Alliance, OH, 1970, NTIS PB 226 761.
A2 Liu, C. Y., Personal Communication.
A3 Jonke, A. A., et al., Reduction of Atmospheric Pollution by the
Application of Fluidized-Bed Combustion, ANL ES-CEN-1002, Argonne
National Laboratory, Argonne, IL, 1970.
A4 Keairns, D. L., et al., Fluidized-Bed Combustion Process Evaluation -
Phase II - Pressurized Fluidized-Bed Coal Combustion Development,
report to EPA, EPA-650/2-75-!027c, Westinghouse Research Laboratories,
Pittsburgh, PA, 1975, NTIS PB 246 116.
78
-------
A5 Bethell, F. U., D. W. Gill, and B. B. Morgan, Mathematical Modeling
of the Limestone-Sulfur Dioxide Reaction is a Fluidized Bed
Combustor, Fuel 52; 1973.
A6 Hartman, M. and R. W..Coughlin, Reaction of Sulfur Dioxide with
Limestone and the Grain Model. AIChE J., 22(5); 1976.
A7 Jonke, A. A., et al., Recuction of Atmospheric Pollution by the
Application of Fluidized-Bed Combustion, ANL ES-CEN-J1004, Argonne
National Laboratory, Argonne, IL, 1971.
A8 Gregory, M. W., et al., A Regenerative Limestone Process for
Fluidized-Bed Coal Combustion and Desulfurization, Monthly Report
83 to EPA, Contract 68-02-1312, Exxon Research and Engineering
Company, Linden, NJ, 1977.
A9 Bertrand, R. R., et al., A Regenerative Limestone Process for
Fluidized Bed Coal Combustion and Desulfurization. Monthly Report
98 to EPA, Contract 68-02-1312, Exxon Research and Engineering
Company, Linden, NJ, 1978.
A10 Carls, E. L. , et al., Reduction of Atmospheric Pollution by the
Application of Fluidized Bed Combustion, ANL ES-CEN-1001, Argonne
National Laboratory, Argonne, IL, 1969.
All Reduction of Atmospheric Pollution. DHB 060971,. National Coal
Board, 1971.
A12 Lange, H. B., and C. L. Chen., S09 Absorption in Fluidized Bed
Combustion of Coal-Effect of Limestone Particle Size. EPRI
FP-667, The Babcock & Wilcox Company, Alliance, OH, 1978.
A13 Gordon, J. S., et al., Study of the Characterization and Control of
Air Pollutants from a Fluidized-Bed Boiler - The S02 Acceptor
Process. EPA R2-72-021, Pope, Evans and Robbins, Inc., Alexandria
VA, 1972, NTIS PB 229 242.
A14 Hammons, G. H., and A. Skopp, A Regenerative Limestone Process for
Fluidized Bed Coal Combustion and Resulfurization, Final Report for
Contract CPA 70-19, Esso Research and Engineering Company, Linden,
NJ, 1971.
79
-------
A15 A Development Program on Pressurized Fluidized-Bed Combustion,
ANL ES-CEN-1007, Argonne National Laboratory, Argbnne, IL, 1974,
80
-------
APPENDIX B
FBC POWER PLANT DESIGN BASES
In this appendix the various comprehensive conceptual plant designs
that have been developed for AFBC and PFBC electric utility power plants
are reviewed, and base designs are selected and described. Additional
design options are discussed. All of these plants were designed to
satisfy current EPA new source performance standards for sulfur oxides
(516 ng/J), particulates (43 ng/J), and nitrous oxides (301 ng/J),
although, based on current data and Westinghouse projections for sulfur
oxide removal, some of the designs will not indeed meet these standards
for the specific operating conditions applied (calcium-to-sulfur ratio,
bed depth, velocity, sorbent particle size, etc.).
Available cost projections for these designs are compared. The
sensitivity of the costs and uncertainties involved in them are dis-
cussed, and the base FBC power plant costs are estimated.
CONCEPTUAL DESIGNS FOR AFBC
Comprehensive conceptual designs of AFBC power plants have been
•Q "I
reported: by Westinghouse/Foster Wheeler in 1971 to EPA; by General
Electric/Bechtel in 1977 to NASA, ERDA, and NSF (under the Energy
o o
Conversion Alternatives study, or "EGAS"); and most recently by
Burns and Roe/Combustion Engineering, by Stone & Webster/Foster Wheeler,
B3
and by Stone & Webster/Babcock & Wilcox to DOE. As will be discussed
later, the GE/Bechtel designs/costs have been utilized and expanded
R7
upon somewhat by TVA. Table Bl presents a comparison of the key
characteristics of these designs. Complete information on all of the
designs is not available, but the diversity of the design bases is
evident. Combustion Engineering, Babcock & Wilcox, and Foster Wheeler
81
-------
Table Bl
SUMMARY OF CHARACTERISTICS OF AFBC CONCEPTUAL
ENGINEERING STUDIES
POWER PLANT CONFIGURATION
Type Plant construction and
philosophy
Capacity (MWel
Number of boiler modules
Coal handling
Sorbent handling
Spent solids handling
Spent solids storage and disposal
Bed-overflow Arrangement
Cooling tower type
Ffl STEAM GENERATOR DESIGN .
Number of main beds per module
Arrangement of beds
Steam circulation/arrangement
Steam cycle
Pressure, kPa
Superheat Temperature. °C
Reheat Temperature. °C
Vessel dimensions
Height, m
Cross-section, m
Bed area, per bed. m
Structural support
Heat transfer surface
Configuration
Diameters (cm)
Pitch (cm) (horizontal)
CONTROL OF CARBON UTILIZATION
technique
Carbon burnup cell ICBCI design
Bed area, m'
Location
Heat transfer surface
Fly-ash i njection. m per point
Westinghouse/Foster
Wheeler (EPA) - 1971
Grassroots; maximum
shop fabrication ;
modular boilers
300 land 6351
4
Crushing; drying in
coal-fueled fluid-bed
dryer with primary al'r
tor feeder-distributor
Handling combined with
coal handling system
Crushed and double1
screened sOrbent
.purchased
Dry conveying ; no cap-
ture of solids heat con-
tent
Short-term on-slte silo.
rail disposal with no
oil-site disposal defined
FBC bed-and CBC bed-
overflow to spent solids
handling
None ~ once-through
cooling water
5 ' "- '
Vertically stacked;
separate functions for
each bed
Once1* rough :
waterwalls and horizon-
tal tubes
Subcritical
16,548
538
538
33.2
6. 55 (rectangular)
14.5
Top supported
Horizontal
5.08, 3.8
10.2
Carbon burnup cell
-15
Integral bed in main
combustor: lowest bed
None in bed: in free-
hoard
Petrocarb;0.9
General Electric/
Bechtel (NASA. ERDA.
NSF) -1977
Grassroots; modular
boilers
814
4
Crushing: drying with
waste h'bt air from spent
solids
Handling 'per formed
separately from coai
handling system; sbr-
bent crushed oh site
Dry conveying; solids
air-cooled to I50°C, with
air for coal andsorbent
drying .
15 day on-site storage.
rail disposal with no
off-site disposal defined
FBCberi-dverfldw'toCBC;
CBC bed-overflow to
spent solids handling
Wet mechanical
draft-24 calls
6
Vertically/stacked:
separate functions !6'r
each bed
Once-through ;
waterwalls and horizon-
tal tubes
Supercritical
24. 133
538
538
58.6
11 x 4 (rectangular)
37.9
Horizontal
5 08. 3. 18
15.24. 7.6
Carbon burnup cell
-38
Integral bed in main
combustor, lowest bed
Water tubes in free-
hoard
Burns and Roe/ Combustion
Engineering IDOEI
Maximum shop fabrica-
tion of bed cells; grass-
fboti
553
1
• •
Spetit solids energy
recovery
^_
^—
20 cells
Beds on single level
Drum-type, assisted cir-
culation; waterwalls and
horizontal tubes
Supercritical
17. 930 (at superheater)
541
541
.42.4 .
1784m (rectangular)
30.1 (per cell)
Bottom supported ; top
supported convection
pass
Horizontal '
Carbon burnup cell
100.3
Detached from main beds
- —
1.67
Stone & Webster/Foster
Wheeler IDOEI
Mini muni land area;
Conventional cohvectlve
heat recovery zone ;
expansion
538
1
_^
Spent solids energy
recovery
i
.
-^-^-
5
Vertically, stacked:
separate functions for
each bed
Once-through :
waterwalls and horizon-
tal tubes
Supercritical
18.070lat superheater!
541
541
55.5
368m Irectangularl
18
Complete top support
Horizontal
Carbon burnup cell
111.5
Integral bed in main
combustor
— •—
1.0
Stone & Webster/Babcodi
& Wllcox (DOE!
Maximum similarity
with conventional
boilers : expansion
552
1
j
No spent solids energy
recovery
^—L.
__^^
-=^-
5
2 parallel vertical pairs
5th bed below and be-
tween the pairs
Drum-type assisted .cir-
culation; waterwalls
and horizontal tubes
Supercritical
18. 070 (at superheater)
541
541
49.7
685m (rectangular)
158.3
Complete top support
Horizontal
Carbon burnup cell
200.7
Detached and to the
side of the main
combustor
0.84
82
-------
Table Bl (Cont'd)
COAL FEEDING
Technique
Bed area per injector point, m2
Feed systems per module
SORBENT FEEDING
Technique
Bed area per injector point, nr
Feed systems per module
BED OVERFLOW
Technique
Bed area per overflow point m2
PART ICULATE CONTROL
Equipment
Arrangement
COMBUSTOR OPERATING
CONDTIONS
Temperature. °C
Pressure. kPa
Excess air. *
Velocity, m/s
Bed depth, m
Static
Expanded
Calcium-to-sullur ratio
Sorbent particle diameter, pm
Coal particle diameter, urn
SORBENT
Type
Source
Size received, ym
Moisture. %
COAL
Type
Source
Composition
Wt.* sulfur
Wt.tash
Wt tmoislure
Heating value. Lower, J/g
PERFORMANCE PROJECTIONS AND
ASSUMPTIONS
SO. emission target. ng/J
Sullur removal efficiency. %
Particulate target. ng/J
Availability Target, %
Boiler efficiency. %
Combustion efficiency, »
Plant efficiency. »
Auxiliary power, % of gross
Slack temperature. °C
CBC hrat duty. % of total
Westinghouse/Foster
Wheeler IEPAI - 1971
Combined dilute-phase
pneumatic feeding of
coal and sorben! in bed
0.93
2
Combined with coal
feeding
Same as coal
Same as coal
Gravity drain
14.5
High efficiency multi-
clone cyclones : ESP
Cold ESP final cleaning.
primary cyclone drains
CBC. secondary cyclone
drains to disposal
871 (1038 CBC)
102
10I70.5CBCI
2.1-3. 35 13.3 CBCI
0.3
6.0
<6350
<6350
Limestone
1359
7.3
12!
IC.f)
Burns and Roe/ Combustion
Engineering (DOE)
Fuller-Kinyon pneumatic
in bed feeding: combined
1.67
36 stream splits per
coal hopper
Combined with coal
feeding
Same as coal
Same as coal
Gravity drain
10.2
Mullitube cyclones;
baghouse
Bughouse for linal
cleanup
843 11093 CBC)
20I25CBCI
3.66 12. 74 CBCI
0.61
1.22
2.3
Limestone
Western
0.89
7.8
8.0
18.731
516
43.0 required
43
84.2
34.2
Stone & Webster/Foster
Wheeler (DOE)
Stoker feeder above bed:
10.2
15 splits per coal hopper
Gravity feeder above
bed
60
1
Gravity drain
15.3
Multitude cyclones ;
haghouse
Bughouse for final
cleanup
843 11038 CBC)
20I30CBCI
3.05 (2.44 CBCI
0.61 10. 70 CBC)
1.22
2.1
Limestone
Eastern
2.55
10.0
13.6
25.774
516
73.9 required
43
85.93
14.1
Stone & Webster/Babcock
& Wllcox IDOE)
Pneumatic in bed
0.84
40 splits per coal
hopper
Combi ned with coal
feeding
Same as coal
Same as coal
Gravity drain
35.7
No mechanical collectors:
hot ESPs
ESPs for primary and
final cleaning
843 (1093 CBCI
15 (50 CBCI
2.44 12.44 CBCI
0.46 10.61 CBCI
1.22
2.5
Limestone
Eastern
2.55
10.0
13.6
25.774
516
73. 9 required
43
85.83
34.6
83
-------
Table Bl (Cont'd)
Weslinghouse/Foster General Electric/ Burns and Roe/Combustion Stone & Webster/Foster S<.vtvA Weubter/Batco*
Wheeler (EPA) - 1971 BechteMERDA) - 1977 Engineering (DOE) Wheeler IDOE) & Wilcox (DOEI
Air Supply
FD fan pressure. kPa 113 (preheat to 390°CI 118 (no preheat)
ID (an pressure. kPa none 2.1
System pressure drop, kPa 11.8 18.3
Heat Transfer coefficients. J/sm2 °C
In bed 283 226 256 227 256-284
Above bed 226 (in splash reqioni 74
Paniculate loading basis
Ash, % of ash input 100 100 100 75
Sorbent, * of fresh feed 3
84
-------
each are performing 200 MWe fluidized-bed boiler demonstration unit
designs for TVA, from which commercial plant designs are being scaled,
but these designs were not available at the time this report was prepared.
Major differences between the conceptual designs exist in: the
type of plant construction (grass roots or expansion); the design philos-
ophy (boiler modularity, maximum shop fabrication, minimum land area,
maximum similarity with conventional boilers - especially in the con-
vective pass); type of cooling tower (once-through cooling water or
mechanical draft tower, type of steam circulation, coal and sorbent
handling [combined or separate handling]); solids feeding technique
(in bed or above bed, bed area per feed point, combined or separate
coal and sorbent feeding); particulate control equipment (cyclones and
electrostatic precipitator [ESP], cyclones and hot ESP, cyclones and
baghouse, hot ESP with no mechanical collectors). Major differences in
the fluid-bed combustor operating conditions are found in the superficial
velocity (2.4 to 3.66 m/s) and the excess air levels (10 to 20% for the
main beds and 25 to 70.5% for the carbon burnup cell). Significant
differences in the coal properties for each design also exist. The
most significant differences in plant performance are in the projected
boiler efficiency (84.2 to 87.9%) and the projected plant efficiency
(34.1 to 35.8%). These would be influenced by the auxiliary power con-
sumption, the assumed carbon combustion efficiency and the recovery of
energy from the hot spent solids.
For every design category in Table Bl there are a multitude of
options in addition to those selected in the five conceptual design
studies listed. For the dense-phase fluidization concepts currently
being developed for AFBC with once-through sorbent operation, the most
important additional options not considered in these conceptual designs
are related to the combustion efficiency of carbon (carbon burnup cell,
fines recycle to main beds, low carbon elutriation by low gas velocity,
high excess air, high bed temperature, coal particle size), and the
85
-------
disposal of spent sorbent (landfill, ocean dumping, long-term storage,
processing for environmental acceptability or for utilization as building
material, agricultural uses, etc.).
Several alternative concepts for the FBC sulfur removal system are
also of potential interest but are not considered here. For example,
• Sorbent regeneration
- reductive decomposition
- two-step, sulfide generation, H~ generation
- sulfide-sulfate solid-solid reaction
• Freealeination of once-through sprbents for improved reactivity
and utilization
• Additives for improved limestone/dolomite utilization
• Reconstitution of sorbent fines by agglomeration
• Alternative metal oxide sorbents.
These and other sulfur removal system concepts are under study but have
not been considered in the first-generation AFBC design studies. Their
impact on FBC cost and performance could ultimately be significant.
Bl B2
Cost projections for the Westinghpuse ' and General Electric con-
ceptual designs are the major AFBC designs available at the time of this
writing. To enable comparison the Westinghouse cost estimates were con-
verted, as required, to the costing assumptions used in the body of this
report (see Table 1, page 11); the GE estimates, having been prepared
under EGAS, were already based upon the assumptions in Table 1. The
1971 Westinghouse costs have been escalated to mid-1975 dollars using
the Chemical Engineering Cost Index, which is very approximate in nature.
The costs are $571/kW for the Westinghouse design (converted to the
Table 1 assumptions) and $632/kW for the General Electric design. This
is only a 10% difference in capital investment, even with the differ-
ences in plant design basis (crushed sorbent purchased, no spent solids
heat recovery, no cooling tower in Westinghouse AFBC power plant design)
and the approximate nature of the above escalation of the Westinghouse
cost. A probable reason for the fairly close agreement is that the AFBC
86
-------
power plant is at least 75 percent conventional equipment with respect
to capital investment. For example, the General Electric AFBC plant
costs break down to 16.1 percent for the fluid-bed combustor, 1.6 per-
cent for solids feeding, 3.0 percent for spent solids cooling, and
4.1 percent for particulate control. These four systems may be con-
sidered to be nonconventional, although they all represent systems com-
prised of largely conventional components. Because of this, AFBC power
plant capital investment projections by different plant designers
applying different AFBC design concepts are expected to be reasonably
close together (±15%) if the costs are placed on the same basis and the
power plant scopes are identical.
The sensitivity of the capital cost of the fluid-bed boiler and
the AFBC power plant to changes in design and operating conditions is
expected to be small. The fluid-bed boiler can be minimized in cross-
sectional area by operating with high superficial velocities, using large
sorbent particle diameters and low excess air rates. This reduction in
cross-section will require a simultaneous increase in bed depth in order
to provide sufficient volume for the immersed heat transfer surface and
sufficient gas-solid contacting time to meet combustion efficiency and
sulfur removal targets. The increase in bed depth, and the resulting
increase in freeboard height, will counteract, to a large extent, any
cost reduction due to decreased vessel cross-section. Ultimately,
changes in design and operating conditions that increase the AFBC power
plant efficiency, such as increased carbon combustion or reduced sorbent
consumption by increasing bed depth or reducing superficial velocity,
will have a greater impact on the cost of electricity than will the com-
bustor cross-sectional area.
Cost and design uncertainties for AFBC exist mainly in the areas of
performance and reliability of the solids feeding system, the fluid-bed
boiler, and the particulate control system. Long-term erosion, corrosion,
plugging and deposition phenomena (occurring to tubes, distributor plates,
solids feed lines, cyclones, etc.) could affect the plant reliability but
may not ultimately have a great cost impact if these phenomena can be
87
-------
controlled by proper selection of operating conditions and hardware
design. The number of solids feed points required to promote uniform
bed conditions, high carbon combustion efficiencies, reliable boiler
operation, and the like is an example of an uncertainty in AFBC that
should be resolved through early large-scale operation and will, it is
hoped, have little cost impact on AFBC. Innovations such as above-bed
feeding conceivably could simplify the solids feeding system and
improve reliability, though performance problems in sulfur removal and
carbon utilization may result.
B7
TVA has applied cost uncertainty factors (an add-on multiplier)
of 20 percent to the General Electric costs for spent-bed coolers and
cyclones, coal and limestone blending and feeding, hot gas cleanup, air
supply and fines injection, and the AFB module (including tower com-
ponents, controls and ducts). TVA claims that, on the basis of its
experience, such factors account for potential increases in the cost of
these undemonstrated components. The TVA uncertainty factors increase
the. General Electric base .investment by only- 2.7 percent, ,an amount that:.
is probably not significant, based on the expected accuracy of conceptual
design cost estimates. TVA also suggested other modifications to the
AFBC power plant design, such as in the area of solid residue disposal
(see Appendix C).
CONCEPTUAL DESIGNS FOR PFBC
Westinghouse/Foster Wheeler, Westinghouse (update of Refer-
•n c
ence Bl), Westinghouse/C. R. Main (under EGAS), and General Electric/
B2
Bechtel (under EGAS) have reported on comprehensive conceptual designs
for PFBC. These designs are characterized and compared in Table B2.
The GE/Bechtel design has been utilized and somewhat expanded upon by
B7
TVA. The concept dealt with in these studies is the fluidized-bed
boiler (with steam tubes in the bed) with an open-cycle gas turbine
expanding the exhaust gas. Many other cycles are possible, ranging from
88
-------
the adiabatic combustor to air-cooled FBC cycles to various closed gas-
turbine cycles, along with several other fluid-bed combustor concepts,
such as recirculating FBC concepts. Other conceptual designs for PFBC
power plants are being developed by Burns & Roe/United Technologies/
Babcock & Wilcox, Curtiss-Wright/Dorr-Oliver/Stone & Webster, General
Electric/Foster Wheeler under DOE sponsorship and privately by American
Electric Power.
Many similarities exist in the four PFBC designs summarized in
Table B2. The major differences are found in the solids feeding system,
the particulate control equipment design, and the combustor operating
conditions. Even though Westinghouse/C. T. Main and General Electric/
Bechtel use widely different gas turbine inlet temperatures, 960 and
871°C, respectively, their projected power plant efficiencies are nearly
identical at 39.0 and 39.2%, respectively. Westinghouse and General
Electric also scaled their designs to a 904°C turbine temperature. The
General Electric/Bechtel design used coarse sorbent particle sizes and
high fluidization velocities relative to the Westinghouse/C. T. Main
design. It is not surprising that the systems that differ most between
the Westinghouse ECAS and General Electric ECAS designs, the solids
feeding system and the particulate control system, represent the areas
of greatest design uncertainty. As with the AFBC design, many options
exist with the PFBC design. Several methods for controlling carbon utili-
zation are possible, such as carbon burnup cells, fines recycle to the
main combustor beds, operation with high excess air and/or low fluidiza-
tion velocities, operation with high combustor temperatures, and control
of coal feed-size.
Solids feeding could be performed by either in-bed or above-bed
techniques. As in AFBC, the number of feed points per unit of bed area,
the impact of above-bed feeding, and the mechanical reliability of avail-
able feeding systems are not understood. Current conceptual designs use
a number of feed points which, the designers feel, have been sufficiently
demonstrated in development units to provide a conservative basis.
89
-------
Table B2
SUMMARY OF CHARACTERISTICS OF PFBC
CONCEPTUAL ENGINEERING STUDIES
POWER PLANT CONFIGURATION
Type of lant onslruction
and hilosophy
Capacity IMWel
Number of boiler modules
Number of gas turbines
Gas turbine compression ratio
Coal handling
Sorbent handling
Spent solids handling
Spent solids storage and disposal
Bed-overflow arrangement
Coaling tower type
FB COMBUSTOR DESIGN
Number of main beds per module
Arrangement of beds'
Steam circulation/arrangement
Steam cycle
Pressure, kPa
Superheat temperature. °C
Reheat temperature. °C
Vessel dimensions
Height, m
Cross-section, m
Bed area, per bed, m
Structural support
Heat transfer surface
Configuration
Diameler. cm
Horizontal pitch, cm
CONTROL OF CARBON UTILIZATION
Technique
Carbon burnup cell ICBCI design
Bed area, m?
Location
Heat transfer surface
FK ash injection, m' per point
COAL FEEDING
Technique
Bed area per injector point, m?
Feed systems per module
Westinghouse/Foster
Wheeler IEPAI - 1971
Grassroots; modulai with max-
imum shop fabrication
318 land 6351
4
2
10:1
Crushing; drying in coal-
fueled fluid-bed dryer
No crushing (purchased sized);
drying combined with coal
handling
Dry conveying with no
energy recovery
Short-term on-slte silo; rail
disposal ; no off-site storage
FBC bed and CBC bed overflow
to spent solids handling
None --once-through
cooling water
4
Vertically stacked; separate
functions (or.each'bed
Once-through
Waterwalls and horizontal
tubes
Subcrilical
16.548
538
538
37
3.7(cylindrical)
3.25
Top supported
Horizontal
5.1. 3.9
10.2
Carbon burnup cell
1.3
Integral bed in combuslor
Waterwalls
0.33
Petrocarb : combined coal
and sorbent feeding
0.8
i
Westinghouse Update
(EPA) - 1975
Grassroots; modular
construction
635
4
2
10:1
Crushing; drying in coal-
fueled fluid-bed dryer
No crushing (purchased
sized) : drying combined
with coal handling
Dry conveying with no
energy recovery
Short-term on-site silo; rail
disposal ; no off-site storage
FBC bed and CBC bed overflow
to spent solids handling
None -- once-through
cooling water
4
Vertically stacked ; separate
fun£tions for each bed :
Once-through
Waterwalls and horizontal
lubes
Subcritical
16.5*
538
538
37
5.2(cylindricall
6.5
Top supported
Horizontal
5.1. 3.9
10.2
Carbon burnup cell
2.6
Integral bed in combustor
Waterwalls
0.66
Petrocarb ; combined coal
and sorbent feeding
0.8
4
Weslinghouse/C.T. Main
(NASA.ERDA.NSF) -1977
Grassroots; modular
construction
619
4
2
10:1
Combined coal and sorbent
crushing and drying to 3%
moisture in hot-air dryer
Combined with coal
Cooing with energy recovery
15 day on-site storage; no
off-site storage
FBC bed and CBC bed overflow
to spent solids handling
Mechanical draft evaporator
4
Vertically stacked ; separate - '
functions lor each bed
Once- through
Waterwalls and horizontal
tubes
Supercritical
24. 133
538
538
28.3
6.1 (cylindrical)
8.5
Horizontal
5.1. 3.8
8.9
Carbon burnup cell
3.0
Integral bed in conbustor
Waterwalls and in-bed surface
-1.0
Petrocarb : combined coal
and sorbent feeding
1.07-2.14
t
General Electric/ Bechtel
INASA.ERDA.NSFI -1977
Grassroots: modular
construction
904
4
4
10:1
Separate crushing and drying
to 3» moisture in hot-air
dryer
Separate crushing and drying
to 3% moisture in hot air
dryer
Cooling with energy reto;ery
15 day on-site storage; no
off-site storage
FBC bed overflow to CBC; CBC
bed overflow to spent solids
handling
Wet mechanical draft
evaporator
6
• Vertically stacked'- separate .
functions' for each bed
Once- through;
Waterwalls and horizontal
luoes
Supercritical :
24. 133
538
538
36.6
4.0 (cylindrical I
4.0
Horizontal
4.4. 3.2
11.4. 7.6
Carbon burnup cell
2.8
Integral bed in combustor
No surface
Petrocarb
0.55
6
90
-------
Table B2 (Cont'd)
SOrtBENT FEEDING
Technique
Bed area per injector point, m^
Feed systems per module
BED OVERFLOW
Technique
Bed area per overflow point, m
PARTICIPATE CONTROL
Equipment
Arrangement
COMBUSTOR OPERATING CONDITIONS
Temperature, °C
Pressure, kPa
Excess air, *
Velocity, m/s
Bed depth (expanded) ; m
Calcium-to-sulfur ratio
Sorbent particle diameter, urn
Coal particle diameter, urn
SORBENT
Type
Source
Size received, urn
Moisture. *
COAL
Type
Source
Composition
Wt% sulfur
W1% ash
WtH moisture
Heating value, lower, J/g
PERFORMANCE PROJECTIONS AND
ASSUMPTIONS
SO emission target, ng/J
Sulfur removal efficiency, *
Paniculate target. ng/J
Availability target, %
Boiler efficiency. %
Combustion efficiency. %
Plant efficiency. *
Power distribution, *gas turbine
Auxiliary power, % of gross
Combuslor exit temperature, °C
Gas turbine inlet temperature. °C
Slack temperature, °C
CBC heat duty, % of total
Air Supply
Pressure, kPa
Pressure drop, kPa
Heat Transfer Coefficients. Jlsm?°cr
In bed ,
Above bed
Westinghouse/Foster
Wheeler (EPA) - 1971
Combined with coal
Same as coal
Same as coal
Gravity flow
3.5
Two stages of cyclones
Primary 11 Ducon per module)
drains to CBC ; secondary
12 Aerodynes) drains to
disposal
954 (1093 CBC 1
1013
10I79CBO
1.7-2.7(1. 74 CBCI
3.7
6.0
2300 (average)
<6350
Dolomite
1337
-------
Table B2 (Cont'd)
Paniculate Loading Basis
Ash, * of ash input
Sorbent, * of tresh feed
Granular Bed Rlter Basis
Face velocity, m/min.
Number of niters per boiler
module
Westinghouse/ Foster
Wheeler (EPA) - 1971
100
5
no granular bed
filter system
Westinghouse Update
IEPAI - 1975
100
5
Westinghouse/C.T. Main
IERDAI - 1977
37, 460 ppm atcombustor
exit; 750 ppm to GBF
75
4
General Electric/Bechlel
IERDAI - 1977
43, 870 ppm at combustor
exit : 1790 ppm to GBF
39
16
92
-------
The conceptual designs reviewed used granular bed filters for final-
stage particulate control. The critical specification, such as face-
velocity, number of granular bed modules, and the granular bed filter
reliability, is currently not known, though test programs are proceeding
that deal with these questions. In addition to granular bed filters,
other high-temperature filters, such as ceramic filters and even high-
temperature electrostatic precipitators, are being proposed for final-
stage particulate control. Low-temperature devices such as electrostatic
precipitators or baghouses with recuperative heat exchange systems are
probably too complex and thermally inefficient to be considered for PFBC.
In contrast with AFBC, a much greater range of operating conditions
may be applicable to PFBC power generation systems than are included in
the conceptual designs reviewed. Higher gas-turbine pressures, such as
in the Americal Electric Power PFBC concept, alternative combustor tem-
perature ranges, and higher excess air levels, such as in the adiabatic
combustor concepts, may lead to quite different PFBC power plant per-
formance and economics.
Alternative sorbent system concepts are under development for PFBC,
in small-scale efforts that may improve PFBC economics and environmental
impact by reducing sorbent consumption. These include sorbent regenera-
tion schemes, precalcination of sorbents, additives that improve sorbent
utilization, and alternative sorbents.
The four available cost projections for the PFBC conceptual designs
have been converted to an identical set of cost ground rules (Table 1)
for comparison. The Westinghouse and GE estimates under EGAS were
already prepared according to the assumptions in Table 1 and needed no
conversion. Significant equipment and performance assumption differences
account for the differences in the power plant investments:
,. General
, Westinghouse Electric^2
Westinghouse Westinghouse 1977 NASA/ 1977 NASA/
Investment 1971 (EPA) 1975 (EPA) ERDA/NSF ERDA/NSF
$/kW 527 580 610 723
93
-------
For example, in the Westinghouse EGAS and General Electric EGAS studies
the 18.5 percent difference in investment can be attributed mainly to
differences in cost estimates for solids feeding and particulate control.
The change in energy cost due to this investment difference would be
about 10 percent, and the significance of a 10 percent difference between
conceptual designs would generally be considered to be low.
Differences in the 1971 and 1975 Westinghouse cost estimates and
the Westinghouse EGAS cost are attributed mainly to plant basis and
equipment selection. For example, Table B2 indicates that the Westinghouse
1971 and 1975 conceptual PFBC design purchased double-screened sorbent
rather than crushing and screening on site and used once-through cooling
water rather than a cooling tower as in the 1977 Westinghouse conceptual
design. The 1971 Westinghouse PFBC design used multiple stages of high-
efficiency cyclones for particulate control, and the 1975 Westinghouse
update added a final-stage granular bed filter. In general, then, these
costs are consistant and represent the only four comprehensive PFBC con-
ceptual designs currently available.
The breakdown of the capital investment of PFBC indicates, as with
AFBC, the power plant is largely conventional equipment (70-80% with
respect to investment). The pressurized fluidized-bed boiler represents
only about 6 to 8 percent of the total plant investment, so efforts to
reduce its cost further will probably not be effective in reducing the
cost of PFBC power generation.
TtS
In addition, in 1973 Westinghouse carried out a detailed cost
sensitivity analysis of PFBC. This study showed that the plant capital
investment would vary only over a range of about 8 percent of the base
investment because of conceivable changes in the pressurized boiler
operating conditions (other than the calcium-to-sulfur ratio), such as
bed temperature, velocity, bed depth, excess air, and pressure; changes
in particulate carry-over from the boilers; changes in the heat transfer
surface configuration, tube materials, and heat transfer coefficient; and
changes in the gas-turbine inlet temperature and steam superheat
94
-------
temperature. The cost of the fluid-bed boiler itself was especially
insensitive to design and operating changes. This is a very important
conclusion with respect to efforts to reduce the fluid-bed combustor
cost by reducing its diameter with higher operating velocities. Since
design and operating changes produce little impact on cost, the performance
of the combustor (e.g., combustion efficiency, particle elutriation,
calcium utilization) should be maximized in order to move in the direc-
tion of optimum PFBC economics.
Uncertainties for PFBC rest mainly with the solids feeding and
particulate control systems and their impact on PFBC power plant relia-
bility. Additional uncertainties, as with AFBC, are based on general
erosion, corrosion, deposition, and agglomeration phenomena that could
occur and could reduce PFBC plant performance and reliability. This
latter class of uncertainties may be resolved through proper materials
selection, operating conditions selection, and minor engineering modifica-
B7
tions. TVA reviewed the General Electric EGAS PFBC conceptual design
and additional costs to components that they considered unconventional.
These cost factors, TVA claims, based on their experience with develop-
mental processes, reflect the uncertainties related to these components.
They applied factors of 20 percent to increase the investments associated
with coal injection, sorbent injection and particulate control, and factors
of 40 percent to the investment for spent bed material handling components
and the pressurized fluid-bed boiler modules. These cost increases
resulted in an increased power plant investment of about 12 percent over
the General Electric base cost. As with the TVA review of PFBC, the cost
allowances applied by TVA for PFBC do not account for the possibility
of cost reductions with the components they considered to be uncertain.
The TVA assessment of the GE/Betchel design includes considerations of
spent bed material disposal; this aspect of the TVA assessment is more
meaningful than are the cost allowances for residue disposal included in
other conceptual designs.
95
-------
SELECTED FBC POWER PLANT BASES
FBC power plant bases to be used in this study for AFBC and PFBC
have been selected from the conceptual engineering studies characterized
in Tables Bl and B2. The characteristics of the selected base FBC power
plants are summarized in Table B3. The AFBC and PFBC power plants were
T) O
taken mainly from the General Electric EGAS conceptual designs.
General Electric designs provided the basic framework because of the
subsequent conventional power plant design that was performed by GE
B7
under the same framework and because of the review by TVA. As dis-
cussed below, some major components of the GE design have been adjusted
.e West:
B1.B4.
B5
on the basis of the Westinghouse PFBC design under ECAS and previous
Westinghbuse work.
the AFBC base plant design applied is essentially taken directly
from the General Electric ECAS study, but TVA has added off-site storage
in a lined pond to the plant, and Westinghouse has substituted a bag-
house system for final-stage particulate contrbl for the original
electrostatic precipitator.
The AFBC power plant cost is categorized into seven groups: steam
generators; turbine generators; process mechanical equipment; electrical;
civil, and structural; process piping and instrumentation; yardwork and
miscellaneous; and costs are reported as major component costs and
balance-of-plant costs. From these seven groups, changes were made only
to the steam generator and the process mechanical equipment categories
for the AFBC base plant. In the process mechanical equipment category,
the cost of off-site spent solids storage was added to the solids
R7
handling equipment, shown in Table B4 and based on TVA estimates for
off-site storage in a clay-lined, diked impoundment. The steam
generator category includes hot gas cleanup equipment, air supply
equipment, and the fluid-bed boiler tower components. The hot gas
cleanup category is modified by substitution of a baghouse filter
system for the original electrostatic precipitators. These systems
are assumed to be identical in investment cost for the purposes of this
96
-------
study, as is shown in Table B5. No changes are made in the original
tower component costs, Table B6. Also, no changes are made in the
balance-of-plant costs (direct labor, indirect field labor, and mate-
rial(s). Table B7 summarizes the AFBC base plant cost, comparing it with
the original GE plant estimate. It must be stressed that the base design
conditions (Table B3) and costs (Table B7) have no particular signifi-
cance — they do not represent optimum conditions or a representative
cost, but are developed only as a means for scaling the plant costs to
parametric operating conditions (Appendix C).
The PFBC base plant design is the GE EGAS estimate modified to sub-
stitute the Westinghouse EGAS designs for coal and sorbent handling,
solids feeding and final-stage granular bed filter particulate control
designs/cost estimates in place of the GE estimates for those items.
The PFBC combustor operating temperature is also increased from the
General Electric base-value of 899°C to 954°C, without changing any of
the General Electric base costs, and off-site storage in a lined pond
has been added by TVA.
As with AFBC, seven cost categories were used to describe the PFBC
power plant cost and only two of these were modified from the GE EGAS
costs — the process mechanical equipment category and the PFB steam
generator category. The major cost differences between the GE and West-
inghouse PFBC designs are summarized in Table B8. The major items, which
can be explained by differences in basic design philosophy, are coal and
dolomite injection and the granular bed filter.
Table B9 lists the major equipment costs in the solids handling
equipment segment of the process mechanical equipment category, with
both the GE original cost estimates and the base PFBC plant cost esti-
mates being listed. Tables B2 and BlO summarize the differences in
design bases for coal and sorbent injection equipment adopted in the two
designs, the Westinghouse basis being applied for the base design. In
general, the GE basis is much more conservative than the Westinghouse.
97
-------
Table B3
Own. 1694B61
CHARACTERISTICS OF SELECTED BASE FBC DESIGNS
AFBC PFBC
Base Design Base Design
POWER PLANT CONFIGURATION
Philosophy
Type of construction
Capacity
Number of boiler modules
Number of gas turbines
Gas turbine compression
Coal handling
Sorbent handling
Spent solids handling
Spent solids storage
Bed-overflow arrangement
Cooling tower type
FB COMBUSTOR DESIGN
Number of main beds per
module
Arrangement of beds
Steam circulation/
arrangement
Modular construction
Grassroots
814a
4
N.A.
N.A.
Crushing; drying with waste
hot air from spent solids
energy recovery
Handling separate from coal;
sorbent crushing and drying
on site
Dry conveying ; solids air
cooled to I50°C, with air
for coal and sorbent drying
15 day on-site storage,
off-site hauling to lined pond
.FBC main bed overflow to CBC;
CBC overflow to spent solids
handling
Wet mechanical draft
Modular construction
Grassroots
904a
4
4
10:1 c
Combined coal and sorbent
crushing and drying to 3%
moisture in hot-air dryer
Combi ned wi th coal
Cooling with energy recovery
for coal and sorbent drying
15 day on-site storage,
off-site hauling to lined pond
FBC main bed overflow to CBC;
CBC overflow to spend solids
handling
Wet mechanical draft
Vertically stacked; separate
functions for each bed
Once-thro ugh; waterwalls
and horizontal tubes
Vertically stacked; separate
functions for each bed
Once-through; waterwalls
and horizontal tubes
Steam cycle
Pressure, kPa
Superheat temperature, °C
Reheat temperature, °C
Vessel dimensions
Height, m
Cross-section, m
2
Bed area, per bed, m
Heat transfer surface
configuration
diameter, cm
horizontal pitch, cm
Supercritical
24,133
538
538
58.6"
37. 9
(rectangular)
Horizontal
5.08, 3.18
15.24, 7.6
Supercritical
24,133
538
538,
36.6U
4.0 (cylindrical)
4.0b
Horizontal
4.4, 3.2
11.4, 7.6
98
-------
Table B3 (Cont'd)
Own. 1694B62
CONTROL OF CARBON
UTILIZATION
Technique
Carbon burnup cell design
Bed area, m2
Location
Heat transfer surface
COAL FEEDING
Technique
Bed area per injector,
m2
Feed systems per module
SORBENT FEEDING
Technique
Bed area per injector, m2
Feed systems per module
BED OVERFLOW
Technique
Bed area per overflow
point, m2
PARTICULATE CONTROL
Equipment
Arrangement
COMBUSTOR OPERATING
CONDITIONS
Temperature, °C
Pressure, kPa
Excess air, %
Velocity, m/s
Bed depth (expanded!, m
Calcium-to-sulfur ratio
Sorbent particle diameter.um
Coal particle diameter, Mm
SORBENT
Type
Source
Size received, urn
Moisture, %
AFB
Base Design
Carbon burnup cell
~38b
Integral bed in combustor
Tubes in freeboard
Combined vibrational, in-
bed, feeding of coal and
sorbent
0.8
7
Combined wi th coal feeding
Same as coal
Same as coal
Gravity flow
2 cyclone stages and
baghouse system
Primary cyclone drains to
CBC; secondary cyclone
and baghouse drain to dis-
posal
PFBC
Base Design
Carbon burnup cell
2.8b
Integral bed in combustor
No surf ace
Petrocarb; combined coal
and sorbent feedingc
1.07
4
Combi ned with coal
Same as coal
Same as coal
Gravity flow
4.0*
Two stages of cyclone pi us
Ducon Granular Bed Filters
(GBF) c
Primary cyclone drains to
CBC; secondary and GBF drain
to disposal
843 (1093 CBC)
102
20 (30 CBC) .
3.05 (2. 74 CBC) °
1.22
2.0d'6H
H
<-3180Q
< 12, 700
954(1093 CBC)
1013
20 (30 CBC)
2. 1(1. 7 CBC) °
2.4 (1.8 CBC) °
2.0°,
n
<6350°
<6350
Limestone
e
No preclassification
10
Dolomite
e
No preclassification
10
99
-------
Table B3 (Cont'd)
Dug. 1694B63
AFBC
Base Design
PFBC
Base Design
COAL
Type
Source
Composition
Wt% sulfur
Wt%ash
Wt% moisture
Heating value, lower, J/g
PERFORMANCE PROJECTIONS
AND ASSUMPTIONS
SO. emission target, ng/J
Sulfur removal efficiency, %
Particulate target, ng/J
Availability target, %
Boiler efficiency, %
Combustion efficiency, %
Plant efficiency, %
Auxiliary power, % of gross
Combustor exit temperature,
°C
Gas turbine inlet tempera-
ture, °C
Stack temperature, °C
CBC heat duty,' % of total
Power distribution, %gas
turbine
Air supply
Supply pressure, kPa
ID fan pressure, kPa
Pressure drop, kPa
Heat transfer coefficient
Bituminous
Illinois No. 6d
9.6
13.0
23,793
516U
84 required
43"
90
87.9
98.4*
N.A.
N.A.
121
10. Ob
N.A.
118
2.1
18.3
Bituminous
Illinois No. 6d
3-'!
9.6d
13.0
23,793
516"
84 required6
43d
90
98.51
4.2*
926
904
149
4"
22
1003
N.A.
51.7b
J
°C)
-1
In bed
Above bed
Particulate loading Basis
Ash, %of ash input
Sorbent, % of fresh feed
Granular bed filter basis
Face velocity, m/min
Number per boiler
module
226
74
100
(See Appendix D)
N.A.
283
96
100
(See Appendix Dl
75
a. This represents the reference value for this quantity. Parametric values for this quantity
have also been determined as a function of the study parameters.
b. This represents the reference value. The value is modified in the parametric studies, but
is not specifically reported.
c. For further design details see Reference B5.
d. This is a major parameter in this study and the value shown is the reference value.
e. Specific projections for this quantity are reported. See Appendix A.
N. A .= Not applicable
100
-------
Table B4
SOLIDS HANDLING MAJOR EQUIPMENT FOR AFBC
Subsystems
CE Cost
Estimate,
1975 M$
Number
Original GE
Estimate,
M$
Estimate Used
in Base
Plant Cost,
M$
COAL DRYING & CRUSHING
1 - Dryer System-189 TPH
1 - Coal Crusher & 2 Screens
1 - Distribution Box
2 - Vibrating Feeders @ 94-1/2 TPH
2 - Surge Bins @ 7600 ft3
2 - Bin Activators
2 - Weigh Belt Feeders @ 94-1/2 TPH
0.910
0.159
0.031
0.012
0.060
0.021
0.053
Section Subtotal = 1.246
3.74
3.74
LIMESTONE DRYING & CRUSHING
1 - Dryer System @ 48.6 TPH
1 - Crusher & 2 Screens
1 - Distribution Box
2 - Vibrating Feeders @ 24.3 TPH
2 - Surge Bins
4 - Bin Activators
2 - Air Lock Valves
2 - Vibrating Feeders
2 - Weighbelt Feeders @ 24.3 TPH
0.359
0.033
0.011
0.008
0.022
0.026
0.006
0.008
0.018
Section Subtotal = 0.541
1.62
1.62
COAL & LIMESTONE BLENDING & FEEDING
2 - Blenders @ 115 TPH
14 - Air Lock Valves
14 - Vibrating Feeders
56 - Vibrating Tables
0.130
0.037
0.091
0.657
Section Subtotal = 0.915
2.74
2.74
SPENT BED MATERIAL COOLING
2 - Air Lock Valves @ 15 TPH
2 - Surge Bins @ Cooler
4 - Air Lock Valves @ 30 TPH
2 - Solids Cooler & Cyclones
0.027
0.042
0.068
0.312
Section Subtotal = 0.449
0.90
0.90
SOLIDS HANDLING CONTROLS
1 - Control System
Section Subtotal = 0.149
•0.45
0.45
OFF-SITE SPENT SOLIDS STORAGE
Total for Solids Handling
0.0
9.45
3.79
13.24
101
-------
Table B5
HOT GAS AND AIR MAJOR EQUIPMENT FOR AFBC
HOT GAS CLEANUP AND AIR SUPPLY
GE Cost
Est imate,
1975 M$
Number
Original GE
Estimate,
MS
Estimate Used
in Base
Plant Design,
MS
4 - Bed Cyclone Units
12 - Cyclone Air Lock Valves
4 - Fines Injection Systems
2 - CBC Cyclone Units
2 - CBC Cyclone Air Lock Valves
2 - Surge Bins @ Dust Cooler
2 - Coolers for CBC Dust
4 - Cooler Air Lock Valves
4 - Electrostatic Precipitators*
1 - Baghouse Filter System*
2 - Air Preheaters
2 - ID Fans and Motors
2 - FD Fans and Motors
2 - PA Fans and Motors
Section Subtotal
0.392
0.078
0.260
0.073
0.026
0.040
0.260
0.051
2.406
(2.406)
2.393
0.881
0.965
0.091
= 7.916
15.83
15.83
*Base design substitutes a baghouse filter system for the original electrostatic
precipitators and assumes identical costs for them.
Table B6
TOWER COMPONENTS FOR AFBC
Subsystems
Items
GE Cost
Estimate ,
1975 M$
Number
Original GE
Estimate ,
M$
Estimate UspH
in Base
Plant Design,*
M$
Heat Exchange and Pressure Parts
Injector and Air Parts of AFB
Control System
Flues, Ducts, Insulation, etc.
5.063
0.781
0.724
0.893
7.461
29.84
29.84
*No changes made in the base tower components.
102
-------
o
u>
Table B7
AFBC BASE PLANT CAPITAL COST BREAKDOWN*
Costs (Millions of Dollars)
Major Direct
Categories Components Labor (l)
Steam Generators 45.68 10.16
Turbine Generator 27.00 1.53
Process Mechanical Equipment 13.24 (9.45) 6.17
Electrical 12.34
Civil and Structural 10.40
Process Piping and
Instrumentation 8.28
Yardwork and Miscellaneous 1.59
85.92 (82.13) 50.47
Indirect Balance-of-Plant
Field(2) . Materials^) Total
9.15 6.30 71.29
1.37 0.10 30.00
5.55 29.40 54.36 (50.57)
11.10 12.30 35.74
9.36 13.70 33.46
7.46 10.10 25.84
1.43 . 1.70 4.72
45.42 73.60 255.41 (251.62)
BOP Labor, Materials & Indirects (Sum of 1 + 2 + 3) 169.49
A/E Home Office & Fee @ 15% 25.42
Contingency @ 20% 56.17 (55.41)
Escalation & Interest during Construction 184.68 (182.18)
Total, M $
Total, $/kWe
521.68 (514.63)
640.7 (632.0)
*GE original cost estimate in parentheses where they differ from the base costs.
-------
Table B8
MAJOR COMPONENT COSTS COMPARISONS IN PFBC (From Ref. B8)
Subsystem Gener
Goal and Dolomite Handling:
Injection
Preparation arid other
Subtotal
Hot-Gas Cleanup :
Granular bed filters
Cyclones and other
Subtotal
PFB Furnace Modules0
Gas. Turbine-Generators'
Stack Gas/Feedwater Heat Exchangers
Miscellaneous
Total
Cost, $/kWta
•al Electric Westinghouse
9;58 1.70
,.2,46 1;47
12.04 3.17
18.11 2.81
9.85 3.31
27.96 6,12
6 . 36 A i 34
iiiii 9*71
1.09 3.74
...1>.93 .2.27
60.50 29.36
KWt is the kilowatts of thermal power associated with the HHV of the
input coal. The thermal power inputs are 2307 MWt for the GE case
and 1742 MWt for the Westinghouse case;
Excluding hot-gas piping from major-component cost.
cExcluding auxiliary air compressors.
^Including auxiliary air compressors, solids waste handling, etc.
104
-------
Table B9
SOLIDS HANDLING MAJOR EQUIPMENT FOR PFBC
Subsystem
Estimate,
1975 M$
Number
GE Original
Cost Estimate,
M$
Estimate Used
In Base Plant
Cost, M$
COAL PROCESSING & FEEDING
1 - Dryer System @ 182-1/2 TPH
1 - Crusher 4 2 Screens
1 - Distribution Bin @ Crusher
2 - Vibrating Feeders @ 92 TPH
2 - Surge Bins @ 7600 ft3
2 - Bin Activators
2 - Feeders (vibrating) @ 365 TPH
2 - Coal Distribution Boxes
2 - Petrocarb Coal Injector System
Section Subtotal =
DOLOMITE PROCESSING i> FEEDING
1 - Dryer System @ 84 TPH
1 - Crusher & 2 Screens
1 - Distribution Bin @ Crusher
2 - Vibrating Feeders @ 42 TPH
2 - Surge Bins @ 1867 ft2
2 - Bin Activators
2 - Feeders (vibrating)
2 - Dolomite Distribution Boxes
2 - Petrocarb Injection Systems
Section Subtotal =
SPENT BED MATERIAL
2 - Throttle Valves for Bed Drains
2 - 190 ft3 Surge Hoppers
4 - Lockhopper Valves
2 - 580 ft3 lockhoppers
2 - High Temp. Feeders (vibrating) 34 TPH
2 - Surge Bins @ Solids Coolers
2 - Solids Coolers with Fans and Cyclones
for 24 MBtu/hr
4 - Air Lock Valves for Coolers @ 34 TPH
Section Subtotal =
CONTROL SYSTEMS
Solids Handling' Equipment
Section Subtotal =
OFF-SITE SPENT SOLIDS STORAGE
Total for I'FBC Solids Handling
0.884
0.159
0.031
0.012
0.060
0.046
0.014
0.010
7.198
8.414
0.468
0.093
0.015
0.010
0.028
0.042
0.013
0.008
3.851
4.528
0.024
0.087
0.069
0.170
0.029
0.040
0.299
0.073
0.791
0.149
0.149
0.000
18.04
2.555
6.20
9.73
J.367
3.40
1.58
0.30
0.00
29.65
1.58
0.30
4.845
16.3T
105
-------
Table BIO
FACTORS THAT AFFECT COMPARISON OF SOLIDS INJECTION
WITH PETROCARB SYSTEMS (From Ref. B8)
General Electric
Westinghousec
Mode of Injecting Coal and
Dolomite Separate
Bed Area per Injeetor(s)
Mixed
2 2
6 ft /injector pair 23 ft /injector
Number of Injectors per
Petrocarb Unit 22
Number of Petrocarb Units 24
4
16
Component Cost Per Injec-
tion Rate, $/(lb/hr)
22
a .
Westihghouse has increased the number of injectors from 4 to 8^ With
8 irijectors/Petrdcarb unit^ the bed area to be served by ah injector
becomes 11.5 ft^i Westinghouse estimated that the cost associated
with this change is negligiblei .
106
-------
Off-site spent solids storage in a diked, clay-lined impoundment is
also included in the base design (Table B9) as specified and estimated by
TVA.
Design factors for the granular bed filter system are compared in
Table Bll, the major cost factors being the higher face velocity selected
in the Westinghouse design and the fact that GE used a multitude of small
Ducon filter units (16 per boiler module) relative to the number used in
the Westinghouse design (4 per boiler module)—see Table B2.
Table B12 shows the cost breakdown in the hot gas cleanup section
(part of the PFB steam generator category). Both the original GE cost
estimate and the base plant estimates are listed. Also shown are the
high-pressure air equipment cost breakdown and the heat exchange equip-
ment (Table B13), where no cost modifications have been performed. Also
no changes in balance-of-plant costs (direct labor, indirect field costs,
and materials) have been made.
The total PFBC power plant cost is broken down by categories in
Table B14, with both the original GE costs and the base plant costs being
listed. The TVA cost adders for cost uncertainties in developmental
components were not included either for AFBC or PFBC.
In the parametric studies performed in this study, the performance
values and operating conditions are modified from the base values in
Table B3, as discussed in the body of this report. Equipment design
changes that occur as a result of these modifications are accounted for
in the cost model (see Appendix C) to the degree of accuracy required here.
The assumptions applied in the equipment design changes are based largely
on the finding of previous Westinghouse conceptual design studies. The
base designs selected in Table B3 and the resulting cost projections are
believed to be meaningful and of general significance because of the
general agreement found between the existing comprehensive conceptual
designs when they are placed on the same cost basis and when the technical
differences are evaluated.
107
-------
Table Bll
DIFFERENCES IN HOT-GAS CLEANUP WITH CYCLONE
SEPARATORS AND DUCON FILTERS (From Ref. 8)
General Electric
Westinghouse
Combustion Gas from Main Bed Cell and
Carbon Burnup Cell:
Particulate loading, ppm
Particle size distribution
Ducbn Filters:
Particulate loading, ppm
Face velocity, ft/min
2
Bed area per volume flow, ft per
actual ft^/min
Component cost* dollars per actual
ft3/min
43,900
More small
particles
1,790
39
0.026
60;5
37,500
More large
particles
750
75
0.013
8.9
108
-------
Table B12
HOT GAS CLEANUP AND AIR MAJOR EQUIPMENT FOR PFBC
HOT
16 -
4 -
32 -
8 -
4 -
4 -
4 -
2 -
2 -
4 -
4 -
2 -
2 -
2 -
4 -
2 -
32 -
4 -
HIGH
2 -
4 -
4 -
4 -
4 -
2 -
1 -
1 -
GAS CLEANUP
Two-in-One Cyclones for
Beds
145 ft3 Collecting Hoppers
Trickle Valves & Dip Legs
Lock Hopper Seal Valves
290 ft3 Lock Hoppers
Fines Injection Systems
Injection System Valves
Two-in-One Cyclones for CBC
290 ft3 Collecting Hoppers
Trickle Valves & Dip Legs
Lock Hopper Seal Valves (^
580 ft3 Lock Hoppers
High Temp. Feeders
(Vibrating)
Surge Bins for Dust Coolers
Airlock Valves for Dust
Coolers
CBC Dust Coolers
Granular Bed Filters
Fines Letdown & Removal
Systems for Granular Beds
Section Subtotal =
PRESSURE AIR
Air Dryer Precoolers
Air Dryers
Booster Compressors
400 ft3 Receivers 800//
Air Compressors for Granular
Beds
400 ft3 Receivers 400#
Booster Compressor Spare
Air Compressor Spare
Section Subtotal --
GE Cost
Estimate,
1975 M$ Number
$8.128
0.107
0.229
0.137
0.173
0.520
0.069
0.532
0.087
0.029
0.069
0.170
0.031
0.040
.0.057
0.260
20.894
0.721
32.253 2
0.174 2
0.350 2
0.416 2
0.051 2
0.193 2
0.029 2
0.104 .1
0.043 1
1.365
GE Estimate
Original Used in
Estimate, Base Plant
M$ Cost, M$
7.339
64.51 30.06
2.58 2.58
109
-------
Table B13
HEAT EXCHANGE MAJOR EQUIPMENT FOR PFBC
GE Cost n
Estimate,
M$ per
Module
GE
Original
Estimate,
M$
Estimate
u sed in
Base Plant
Cost, M$
PFB Heat Exchange and Pressure Parts
PFB Containment Shell and Nozzles
PFB Fuel Injection and Air Parts
PFB Controls
PFB Petrocarb Cooler Economizer E
x
PFB Module
Gas Turbine Economizer E_
2.213
0.566
0.237
0.567
0.089
3.67
0.627
14.68
2.51
14.68
2.51
110
-------
Table BlA
PFBC BASE PLANT CAPITAL COST BREAKDOWNS*
Categories
PFB Steam Generators
Turbine Generators
Process Mechanical
Equipment
Electrical
Civil and Structural
Process Piping and
Instrumentation
Yardwork and
Miscellaneous
Major
Components
44.74
(79.19)
50.62
21.42
(34.74)
116.78
(164.55)
Site Labor
Direct Indire
Labor (1) Field
4.35 3.91
1.70 1.53
Costs, M$
ct Ba lane e-of -Plant
(2) Materials (3)
3.10
0.20
6.29 5.66 26.20
9.52 8.57
9.99 8.99
13.51 12.16
1.59 1.43
46.95 42.25
11.00
11.20
20.10
1.70
73.50
B.O.P. Labor, Materials & Indirects
Total
56.10
(90.55)
54.05
59.57
(72.89)
29.09
30.18
45.77
4.72
279.48
(327.25)
162.70
*General Electric cost estimates
in parentheses where they dif-
fer from the base plant costs.
(sum of 1 + 2 + 3)
A/E Home Office & Fee fl 15%
Contingency ® 20%
Escalation & Interest during Construction
Total (M$)
Total ($/kWe)
24.41
60.78 (70.33)
199.84 (231.25)
564.5 (653.2)
624.6 (722.7)
-------
REFERENCES
B'l. Archer,. D. H. et al., "Evaluation of the Fluidized-Bed Combustion
Process", Vol. II. Report to Office of Air Programs, EPA, b.y
Westinghouse Research and Development Center, Pittsburgh, PA..,
November 1971, Contract 70-9, NTIS PB 212-916.
B2.. Energy Conversion Alternatives Study (ECAS) General Electric Phase II
Final Report, Volume. II,. NASA CR-134949,, 1977, NTIS PB 269-379.
B3. Garner, D. N. , et al. , "A Comparison, of Selected Design Aspects of
Three Atmospheric Fluidiz.ed; Bed? Combustion- Conceptual Power Plant
Designs," Radeon Cb.rp...,. presented1 at the- 5th International
Fluidized Bed Combustion Conference, Washington, DC, December 1977.
B4. Keairns, D. L., et al., "Fluidized-Bed^ Combustion Process Evaluation-
Phase II- Pressurized Fluidized-Bed Coal Combustion Development."
Report to. EPA,. Westinghou-se Research Laboratories,, Pittsburgh,. PA,
S.ep.tember 197/5:, EPA.-650/-2-75-Q27c, NTIS' PB: 2.4,6-116..
B'5.. Energy Conversion!. Al!te.rna£iv/e;S; Stud'y ('EGAS); Westinghou'se Phase, II
Final Report, Vbliume III,, NASA CR-1314.942',, 19.7-7, NTIS' P-B: 268-558..
B6i.. Keairns;, D.. L., et al., "Evaluation, of the Fluidized-Bed Combustion
Process," Vol. I and; II.. Report to Office of Research, and Develop-
ment, EPA,. Westingho.use Research Laboratories, Pittsburgh, PA.,
December 1973, EPA-650/2-73-048a and b, NTIS PB 231-162 and:
PB 231-163.
B:7. Utility Boiler Design/Cost Comparison: Fluidized-Bed Combustion
versus Flue Gas Desulfurization. Report to EPA, Tennessee- Valley
Authority, November 1977, EPA-600/7-77-126..
B8., Evaluation of Phase 2 Conceptual Designs and Implementation
Assessment Resulting from the Energy Conversion Alternatives
Study (EGAS), submitted by NASA,. Lewis Research Center, to
ERDA.and' NSF, April 1977, NTIS: PB. 270 017.
112
-------
APPENDIX C
RELATIONSHIPS FOR FBC COST EVALUATIONS
First-order approximations have been used to scale the costs and
performance of AFBC and PFBC power plants. The base plant costs (Appendix B)
have been broken down to permit scaling with respect to several parameters:
sorbent calcium-to-sulfur ratio, coal properties (sulfur content, ash
content, heating value), sorbent properties (molecular weight, mass
fraction weight reduction on calcination, heat of calcination), cost of
sorbent, and cost of coal. Both equipment size/capital costs and operat-
ing costs are considered.
ATMOSPHERIC-PRESSURE FLUIDIZED-BED COMBUSTION
In order to scale the cost of an AFBC power plant with respect to
the parameters listed, the rate of coal consumption, the rate of fresh
sorbent feeding, the rate of spent solids, and the net plant power are
estimated with respect to the base values of these quantities. For the
AFBC power plant the base value of the coal-energy input rate, E, is
fixed for all the cases considered at 2.26 GJ/s (7713 x 10 Btu/hr).
Thus, the coal rate, F , will be
c
Fc = I • . (1)
where H is the coal heating value (as received, wet coal). The fresh
sorbent rate, F , is
s
x
Fs = (Ca/s) at Fc M '
113
-------
where X is the coal sulfur-content (weight fraction), M is the sorbent
s
molecular weight (sorbent weight per mole of calcium) and Ca/S is the
molar calcium-to-sulfur ratio. The rate of spent solids, F,, disposal
is given approximately by
o sn
(3)
i Xs 30
F— TT /fTa/CM — m C\ Y ^ 4- Y 4- V
j r 1 V. v>ci / ^ / _ _ 111 V.-J- A-r, r\/A OO~-^A
d c I 32 C0_ s 32 A
where X^QO is tne weight fraction of the sorbent released as C0£ on
calcination, and X. is the coal ash-content (weight fraction). For
A
simplicity, in this expression for F, all of the coal sulfur is assumed
to be found in the spent sorbent or coal ash as calcium sulfate.
The net plant power output, P, is given approximately by
n(f - f) n H (F - F )
P = PK - Q/?/ / F H * *' - AP. , (4)
b 3414.4 c m 3414.4 A
where P is the base net-power, n is the gross generation efficiency
for the plant, and f and f, are the combustion efficiencies for the
b
parametric plant' and the"base plant, respectively; YL is the sorbent
heat of calcination, F , is the fresh sorbent feed rate for the base
s,b
plant, and AP is the change in auxiliary power between the base plant
A
and the parametric plant. In developing equation 4, several simplifying
assumptions were applied. For example, the combustor exit temperature,
the steam superheat and reheat temperatures and the stack gas temperature
are assumed identical with the base case values for all parametric cases.
Other exothermic or endothermic reactions occurring in the combustor are
assumed to generate energy at the base rates for all parametric cases.
For example, the exothermic sorbent sulfation reaction that occurs in
the combustor is assumed to generate energy at the base rate for 83 percent
sulfur removal from a 3.9 wt % sulfur coal. Higher sulfur-content coals
and/or higher sulfur removal efficiencies would actually increase the
plant power slightly, but this represents a small contribution. Energy
losses, above the base value, from the disposal of hot spent-solids
114
-------
are neglected, with a hot-air energy recovery system for coal and sor-
bent drying assumed to function with fixed thermal efficiency under all
parametric conditions. System pressure-drops, such as the bed pressure
drop, are assumed not to change significantly for the set of parameters
considered.
The specifications applied in this study are as follows:
a The coal higher heating value, for high-sulfur Eastern coal of
rank similar to the base Illinois No. 6 coal and containing about
13 wt % moisture, is expressed as
H = 13,922 (1 - X.) + 4,100 X - 1,810 (Btu/lb) (5)
A S
• For a typical limestone that might be applied for AFBC,
XC09 = 0.41, HR = 78,700 Btu/mole, and m = 111.0 Ib/mole
calcium.
• Base values for AFBC are (from the General Electric EGAS
C2
study and Appendix B)
F = 7.0578 x 105 Ib/hr
c,b
F = 1.9096 x 105 Ib/hr
s, b
F, , = 2.4923 x 105 Ib/hr
d, b
P, = 814,000 KWe
b
f. = 0.984
b
n = 0.439 .
• Auxiliary power consumption terms, as listed by General Electric,
are presented in Table Cl. The primary air fans (PA fans) and
the two solids handling terms are sealed according to the total
coal and sorbent rate to yield an auxiliary power term of
CF + F \
^ S— - l]
F K + F u /
c,b s,b /
APA = 3,800 I-— + F - ll + AP (kWe) , (6)
\ /i K *-> V> / *
115
-------
Table Cl
AUXILIARY LOSS BREAKDOWN FOR AFBC (from Ref. C2)
Item
Furnace
SA Fans
FD Fans (4)
PA Fans (4)*
ID Fans (4)
ESP (8)
Solids handling*
Turbine
Auxiliaries
24" A P,
65" A P,
16" A P,
16" A P,
0.33% of
No.
Assumptions Uni
0.82 eff 4
0.82 eff 4
0.82 eff 4
0.78 eff 4
8
4
Gross kW 1
of Total,
ts MWe
0.8
21.1
0.5
5.3
3.3
1.4
32.5
2.9
Major Pumps
Service
Booster
Cohdensate
Circ. water
Water Intake
Solids Handling*
"Hotel" Loads
A/E estimate ,
600 psi, 5.44 x 10 Ib, 75% x 90%
185 psi, 4.86 x 106 Ib, 70% x 90%
Proportion to cooling .heat duty
A/E estimate
Based on rates and lifts
A/E estimate 1% of generation
Cooling Tower Fans Proportional to heat duty
Transformers
0.5% of gross generation
2
2
2
3
2
1
1
24
4
0.9
3.4
1.1
4.7
10.1
1.0
1.9
8.8
2.8
4.4
TOTAL AUXILIARY POWER =64.4
^Auxiliary power scaled for these items.
116
-------
where AP represent additional auxiliary power for particulate
control above the base value.
© The net plant power becomes, from equation 4,
/ F + F \
P = 831,408 - AP -3,800 I _ . ,, II- 0.3162(Ca/S) X F . (7)
p \ J1 u + *" u / sc
\ c,b s,b /
In equation 7 it is assumed that the combustion efficiency, f in
equation 4, remains fixed at the base value, f, , for all values of the
b
parameters considered. Various models of the combustion losses were
applied to express f as a function of the design and operating parameters.
For example, if the combustion rate is first order with respect to the
carbon content of the bed, if the CBC is increased in size in direct pro-
portion to the carbon overflow rate, and if carbon elutriation can be
neglected, then it can be shown that
and as the sorbent feed rate increases the combustion efficiency drops.
Since carbon elutriation is believed to be the main mechanism of carbon
losses, the above model is probably a poor approximation. It is also
found when applying this model that the carbon losses are negligible in
impact on the plant power when compared to the impact of the calcination
energy loss occurring with increased calcium-to-sulfur ratio. Because
the calcination energy loss is so dominant, the assumption of constant
combustion efficiency is justified in these parameteric studies.
• If the sorbent particle diameter is reduced from the base value
of about 1000 urn to an average of 500 pm, an additional 1320 kWe
is subtracted from the base power to account for power require-
ments for additional grinding; the auxiliary power than becomes
/ F + F \ /F \
APA - 3278 F-TTF-- - : + 132° FT - x + APP ' (9)
\ c,b s,b / V s,b /
resulting in
117
-------
F + F
P = 830,088 - AP - 3278 I C S - 1|-1,320 [-^- - 1
P V c,b s,b
- 0.3162(Ca/S) X F . (10)
s c
These simple and approximate relationships express the required
major scaling quantities as a function of the parameters of interest.
The General Electric costs are described in a report prepared by
C4
TVA, and these have been applied with several modifications in this
evaluation (see Appendix B). Seven plant categories from the EGAS cost
format are used to summarize the FBC capital investment breakdown, and
five of these categories are assumed to be fixed in cost over the range
of parameters considered (turbine generators, electrical, civil and
structural, process piping and instrumentation, and yardwork and miscel-
laneous) , while two categories are scaled according to changes in
capacity (steam generators and process mechanical equipment).
Component costs, direct labor costs, indirect field costs and mate-
rials costs, are scaled .according .to the applicable capacity ratios F /F ,
*— C j D
(F + F )/(F , + F , ) F , and F./F, , , with a 0.85 power factor for
c s c,b s,b s,b d d,b
solids handling items, with a 0.6 power factor for cyclones and auxiliary
components, and a 0.68 power factor for fans and motors. These factors
were selected from previous plant cost studies performed by Westinghouse.
The process mechanical equipment category includes the scaled items
coal drying of crushing, limestone drying and crushing, coal and lime-
stone blending and feeding, spent bed material cooling, and spent bed
material off-site storage. Solids handling controls are assumed to be
constant in cost. The investment to haul spent solids off site and store
them in a clay-lined, diked impoundment is taken from TVA projections.
The steam generator category includes hot gas cleanup items, air
supply, and the combustor and steam generator items. The hot gas cleanup
items that are assumed to be unchanged in cost by the study parameters
are the cyclone units for the primary combustion beds, cyclone air lock
valves, fines injection system, carbon burnup cyclone air lock valves,
118
-------
cooler air lock valves, and the baghouse filter system. The carbon burnup
cell cyclone units, the surge bins and dust cooler, and the cooler for
the carbon burnup cell dust are scaled according to capacity. The addi-
tional cost of increased particulate control capability above the base
design is considered separately (Appendix D).
The PA (primary air) fans and motors, the air preheaters, the ID
fans and motors, and the FD fans and motors are all scaled according to
increased capacity.
The combustor steam generator is assumed to be constant in cost for
the parameters of interest in this study, based on previous comments on
cost sensitivity.
Tables C2 through C5 list the major equipment and balance-of-plant
items, indicating which items are scaled and with what power factor they
are scaled.
This method results in a parametric capital investment, I, given by
V0.85
I($ x 10 ) = 453.4 + 25.0
(11)
The energy cost, E , in mills/kWh is related to the investment, I,
the net plant power, P, the coal cost, p($/MBtu) and the sorbent cost,
S($/ton), by
3.5037 x 104 I + p F H x 10~3 + 0.5 S F
<
(12)
The above equations for I and E were used in calculating the AFBC costs
in Figure 4 through 6, 8, and 9 in the main text.
119
-------
Table C2
SCALING OF SOLIDS HANDLING MAJOR EQUIPMENT FOR AFBC
Subsystems
Cost
Estimate,
1975 M$
Number
Cost Scaling
Factor Used
Estimate
Used in
Base Plant
Cost, M$
COAL DRYING & CRUSHING
1 - Dryer System-189
TPH 0.916'
1 - Coal Crusher &
2 Screens 0:159'
1 - Distribution Box 0.031
2 - Vibrating Feeders
@ 94-1/2 TPH 0.012
2 - SUrge Bins @
7600 ft3 0.060
2 - Bin Activators 0.021
2 - Weigh Belt
Feeders @ 94-1/2
TPH . 6..-.053.
Section Subtotal = 1;246
LIMESTONE DRYING & CRUSHING
1 - Dryer System @
43:6 TPH 0.359
1 - Crusher & 2 Screens 0.083
1 - Distribution Box 0.011
2 - Vibrating Feeders
@ 24.3 TPH 0.008
2 - Surge Bins 0.022
4 - Bin Activators 0.026
2 - Air Lock Valves 0.006
2 - Vibrating Feeders 0.008
2 - Weigh Belt Feeders
@ 24.3 TPH 0.018
Section Subtotal = 0.541
COAL & LIMESTONE BLENDING & FEEDING
2 - Blenders @ 115 TPH 0.130
14 - Air Lock Valves 0.037
14 - Vibrating Feeders 0.091
56 - Vibrating Tables 0.657
Section Subtotal = 0.915
-------
Table C2 (Cont'd)
Subsystems
Cost
Estimate,
1975 M$
Number
Cost Scaling
Factor Used
Estimate
Used in
Base Plant
Cost, M$
SPENT BED MATERIAL COOLING
2 - Air Lock Valves
@ 15 TPH 0.027
2 - Surge Bins @ Cooler 0.042
4 - Air Lock Valves
@ 30 TPH 0.068
2 - Solids Cooler &
Cyclones 0.312
Section Subtotal = 0.449
SOLIDS HANDLING CONTROLS
1 - Control System
Section Subtotal = 0.149
OFF-SITE SPENT SOLIDS STORAGE
Total for Solids
Handling
0.0
2
0.85
None
(F./F )
d d ,b
0.85
0.90
0.45
3.79
13.24
121
-------
Table C3
SCALING OF HOT GAS AND AIR MAJOR EQUIPMENT FOR AFBC
Cost
Estimate,
Subsystems 1975 M$ Number
HOT
4 -
12 -
4 -
2 -
2 -
2 -
2 -
4 -
1 -
2 -
2 -
2 -
GAS CLEANUP
Bed Cyclone
Cyclone Air
Valves
AND AIR SUPPLY
Units 0.392
Lock
0.078
Estimate
Used in
Cost Scaling Base Plant
Factor Used Design, M$
Fines Injection
Systems
CBC Cyclone
CBC Cyclone
Valves
Surge Bins
Cooler
Coolers for
Cooler Air
Valves
0.260
Units 0.073
Air Lock
0.026
@ Dust
0.040
CBC Dust 0.260
Lock
0.051
0 (\
(F /F )
^ d d,b'
/i-t /T-I \ U • 0
( F / F )
v V d,b;
Baghouse Filter
System
2.406
Air Preheaters 2.393)
ID Fans and
FD Fans and
Motors 0.881)
Motors 0.965)
[0.9 + 0.1
, ,,0.68
d d,b
2 - PA Fans and Motors
0.091
(Fc,b+Fs,b)]
0.68
Section Subtotal = 7.916
15.83
Table C4
TOWER COMPONENTS FOR AFBC
Subsystems Items
Heat Exchange and Pressure Parts
Injector and Air Parts of AFB
Control System
Flues, Ducts, Insulation, etc.
M$
5
0
0
0
7
.063
.781
.724
.893
.461
Number
Cost Scaling
Factor Used
Estimate
Used in
Base Plant
Design, M$
4 None 29.84
122
-------
PRESSURIZED FLUIDIZED-BED COMBUSTION
The development for PFBC is similar to that for AFBC. For PFBC
the coal energy input rate is fixed as
E = F H = 7874 x 1Q6 (Btu/hr) (13)
from the General Electric EGAS conceptual design. Relationships 2, 3,
and 4 are applied for PFBC with a dolomite sorbent having characteristics
given by Xcc,2 = 0.474, HR = 100,800 Btu/mole and m = 188.0 Ib/mole cal-
cium. The base values for PFBC are assumed to be
F , = 7.2005 x 105 Ib/hr
c,b
F = 3.2996 x 105 Ib/hr
s,b
FJ . = 3.1289 x 105 Ib/hr
d , b
Pb = 904,000
n = 0.465
f. = 0.985
b
In these relations the combustor temperature has been raised from the
General Electric base value of 899°C to 954°C, while all other factors
are assumed to remain constant. Previous design studies have shown this
to have little impact on the power plant thermal efficiency, but the
higher temperature yields conditions under which the dolomite sorbent
will be fully calcined in the combustor. At 899°C and the base excess
air level calcination would not be expected, based on thermodynamics.
The auxiliary power term is given (based on Table C6) by
,b
123
-------
Table C5
SCALING OF BALANCE-OF-PLANT COSTS FOR AFBC
Direct Manual
Field Labor
IIH 1000
Balance of
Plant Material,
$ 1000
Cost Scaling
Factor Used
1.0 AFB STEAM GENERATORS (3)
1.1 Steam Generator Erection
Erect only (supply by others): 343
includes heat transfer surface and pressure
parts; buckstays, braces and hangers;
attached flue and duct; fluid-bed components;
control equipment and valves; miscellaneous
piping and small valves; insulation for above
Supply and erect: 158
Includes support steel for above; access
steel; miscellaneous materials and labor
operations
1.2 Steam Generator Auxiliaries
Erect only (supply by others): 70
includes PA fans; FD fans; ID fans; air pre-
heaters; coal and limestone feed tables;
pipes and air lock valves
Supply and erect:- 198
includes external flue and duct; support
steel for ductwork; insulation for external
ductwork
1.3' Stack Gas Cleanup:
Erect only (supply by others): 82
includes cyclones and electrostatic
precipitators
Supply and erect: • 14
includes support steel for cyclones and
precipitators
None
3,020
Hone
100
2,570
None
None
610
None
None
865
130
2.0 TURBINE GENERATOR (3)
Install only (supply 4y others)•
includes 883 MWe steam turbine; generator;
exciter; auxiliary equipment; integral steam
and auxiliary piping; insulation; miscel-
laneous labor operations
3.0 PROCESS MECHAiUCAL EQUIPKEIIT (3)
3.1 Boiler Feedwater Pumos
Supply and install: 10
Includes turbine-driven main feedwater
pumps and drivers (3 @ $940,000); feedwater
booster pumps and motors (2 @ $125,000)
3.2 . .'lain Circ. Water Pumps (3)
Supply and install: 3
includes main circ. v;ater pumps and motors
(3 @ $250,000)
6,300
100
None
3,200
Hone
790
None
124
-------
Table C5 (Cont'd)
Direct Manual
Field Labor,
!:H' 1000
Balance of
Plant Material,
3 1000
Cost Scaling
Factor Used
3.3 Other Pumps (3)
- Supply and install:
includes condensate pumps and motors (2 @
$100,000); and other pumps and drivers not
listed elsewhere
3.4 Main Condenser (3)
Supply and install:
includes shells; tubes; air ejectors
3.5 Heaters, Exchangers, Tanks and Vessels (3)
Supply and install:
includes l.p. feedwater heaters (4): i.p.
feedwater heater; h.p: feeduater heater;
deaerating heater and storage tank; miscel-
laneous heaters and exchangers; tanks and
vessels
3.6 Stacks and Accessories (3)
Supply and erect:
Includes concrete stacks and liners; lights
and marker painting; hoists and platforms;
stack foundations
3.7 Turbine Hall Crane (1)
Supply and erecc:
includes crane and accessories
3.8 Coal Handling (2)
Erect only (supply by others):
includes coal dryers (3); support and access
steel for dryers; coal grinders (3); screens
at grinders
Supply and erect:
includes railcar dumping equipment; dust col-
lectors; primary crushing equipment; belt
scale; sampling station; magnetic cleaners,
mobile equipment; conveyors to pile; reclaim-
ing feeders; conveyors to cascade; coal cas-
cade; conveyors and bucket elevators to dryer
and grinders; recirculating conveyors at
grinders; conveyors to blenders
3.9 Limestone Handling (2)
Erect only (supply by others):
includes limestone dryers (3); support and
access steel for dryers; limestone grinders
(3); screens at grinders; limestone surge
bins at AFS modules
670
lione
27
2,520
2,580
'lone
110
1,700
•lone
73
420
10
5,700
None
,0.85
.0,85
11
10
0.85
125
-------
Table C5 (Cont'd)
Direct "anual
Field Labor ,
!M 1000
Balance of
Plant Material,
$ 1000
Cost Scaling
Factor Used
Supply and erect: 45 1,960 (F /F b>°'85
limestone pile; ieclairning feeders; con-
. veyors to cascade; limestone cascade; con-
veyors and bucket elevators to dryers and
grinders; recirculating conveyors at
grinders; conveyors to blenders; pneumatic
transport feeders, hoppers, blowers; arid
piping to AFB modules
3.-10 Coal arid Limestone Slehd Handling (2)
Erect only (supply By others):
includes blenders (3); surge bins; bin
unloaders and feeders
- Supply and erect:
includes conveyors and bucket elevators to
AFB modules
3:11 Spent Solids Handling (2)
Erect only (support by others):
includes high-temperature vibrating feeders
and conveyor to solids cooler; solids cooler
- Supply and erect:
includes fly-ash handling system for pirecip-
itators and'air preheater; solids cooler
accessories arid bucket' elevator; ash con-
veyors; ash storage silos (6) with feeders,
unloaders and foundations; railcar loading
equipment
3.12 Cooling Tov;ers (3)
Supply and erect:
includes mechanical draft towers with fans
and motors
3.13 Other Mechanical Equipment (3)
Supply and erect:
includes water treatment and chemical injec-
tion; air compressors and auxiliaries; fuel
oil ignition and warm-up; screenwell; mis-
cellaneous plant equipment; equipment
insulation
4.0 ELECTRICAL (5)
4.1 'lain Transformers
4.2 Other Transformers and Main Bus
includes start-up transformer; station ser-
vice transformers; generator main bus
4.3 Switchgear and Control Centers
includes switchgear and load centers; motor
control centers; local control stations;
distribution panels, relay and meter boards
13
93
10
920
10
4,480
0.85
0.85
(F./F, ,)
d a ,b
0.85'
,0.85
63
30
525
4
15
37
2,680
1,720
29,400
2,030
1,160
3,020
ilorie
None
126
-------
Table C5 (Cont'd)
Direct Manual.
Field Labor,
MH 1000
' Balance of
Plant Material,
$ 1000
Cost Scaling
Factor Used
lt.lt Other Electrical Equipment
includes communications, grounding; cathodic
and freeze protection; lighting; preopera-
tional testing
4.5 Auxiliary Diesel Generator
Includes diesel generator, batteries and
associated dc equipment
4.6 Conduit, Cable Trays, Wire and Cable
5.0 CIVIL AND STRUCTURAL
5.1 Concrete Substructures and Toundations (1)
includes turbine building substructure; AFB
base mats; coal, limestone and ash handling
foundations, pits and tunnels; miscellaneous
equipment foundations; auxiliary buildings
substructures; miscellaneous concrete
5.2 Superstructures (1)
includes turbine building; auxiliary yard
buildings
5.3 Earthwork (1)
Includes building excavations; coal, line-
stone and ash handling excavations; circ.
water system excavations; AFB foundation
excavations; miscellaneous foundation excava-
tions; deuatering and piling
5.4 Cooling Tower Basin and Circ. Water System (3)
includes circ. water pumps pads, riser and
concrete envelope for pipe; cooling tower
basin; circ. water pipe; cooling tower mis-
cellaneous steel and fire protection
5.5 AFB Boiler Enclosures (1)
includes structural steel; noninsulated walls
and roofing; building services; elevators
428
564
1,050
350
230
135
115
55
885
6.0 PROCESS PIPING AND INSTRUMENTATION
6.1 Steam and Feedwater Piping (3) 55
includes main steam; extraction steam; hot
reheat; cold reheat; feedwater and condensate
large piping, valves and fittings
6.2 Other Large Piping (3) 165
includes auxiliary steam; process water;
auxiliary systems
2,400
110
2,890
6,540
300
1,800
2,170
13,700
2,680
2,670
None
Ncme
127
-------
Table C5 (Cont'd)
Direct Manual
Field Labor,
:m 1000
Balance of
Plane Material,
$ 1000
Cost Scaling
Factor Used
6.3 Small Piping (3)
includes all piping, valves and fittings of
2-in. diameter and less
6.4 Hangers and iiisc. Labor Operations (3)
- includes all hangers and supports; material
handling; scaffolding; misc. labor operations
6.5 Pipe Insulation (3)
6.6 Instrumentation and Controls (5)
7.0 YARDWORK AND MISCELLANEOUS (1)
7.1 Site Preparation and Improvements
- includes soil testing; clearing and grubbing;
rough grading; finish grading; landscaping
7.2 Site Utilities
includes storm and sanitary seuers; nonprocess
service water
7.3 Roads ar.d Railroads
includes railroad spur; roads, walks, and
parking areas
7.4 Yard Fire Protection, Fences, and Gates
7.5 Water Treatment Ponds
includes earthwork; corapacted-clay lining;
off-site pipeline
7.6 Lab, Machine Shop, and Office Equipment
85
245
40
115
705
38
27
52
12
1
135
760
920
430
None
10
50
750
600
10
280
1,700
128
-------
Item
Table C6
AUXILIARY LOSS BREAKDOWN FOR PFBC
Assumptions
No. of
Units
Total
MWe
Furnace
Fans for dryers
(coal dolomite, solids)
Coal crusher
Dolomite crusher
Injector compressor
Filter
Turbines
Pumps
Condensate
Circ. water
Service water
Intake water
Solids Handling
"Hotel" Load
Cooling Tower Fans
Transformer Loss
0.33% of steam turbine, kW
1% of gas turbine, kW
A P = 185 psia
Proportion to cooling duty
A/E estimate
A/E estimate
Based on rates and lifts
0.88% of generated kW
Vendors value
0.5% of generated kW
2
2
2
8
8
1
4
2
3
2
2
1
1
22
2
4.18
0.74
0.33
4.03
1.11
2.37
2.16
9.95
4.40
0.89
0.94
2.17
8.34
2.53
4.72
TOTAL AUXILIARY POWER = 39.86
*Auxiliary power scaled for these items.
129
-------
and the plant power is given by
F + F
P = 928,094 - AP - 11,450 I-—C . -S— -
p VFc,b + Fs,b
- 0.429(Ca/S) X F (15)
s c
for the base sorbent particle size of about 2000 ym. In relation 15 it
has been assumed that the combustion efficiency remains fixed at its
base value for all parametric cases, while significant increases in the
combustion efficiency above the base value of 98.5% are expected.
For the case of finer sorbent particles of 500 ym average diameter,
modifications to the power requirements for grinding (1320 kWe) result
in
P = 926,774 - AP - 10,130
- 1 - 1320 ~- - 1
- 0.429(Ca/S) X F . (16)
S C
The plant costs have been scaled, as for the AFBC case, using the
General Electric cost breakdown with several modifications as detailed
in Appendix B.
The scaling of cost items are summarized in Tables C7 through C9,
major equipment, and CIO, balance-of-plant equipment. The factors are
analogous to the AFBC case.
If the modified General Electric EGAS costs are combined from these
tables the plant investment, I ($ x 10 ) , is given by
(17)
130
-------
Table C7
SCALING OF SOLIDS HANDLING MAJOR EQUIPMENT FOR PFBC
Subsystems
COAL PROCESSING S, FEEDING
1 - Dryer System @ 182-1/2 TPH
1 - Crusher & 2 Screens
1 - Distribution Bin !? Crusher
2 - Vibrating Feeders @ 92 TPH
2 - Surg* Bins @ 7600 fr.3
2 - Bin Activators
2 - Feeders (vibrating) @ 365 TPH
2 - Coal Distribution Boxes
2 - Petrocarb Coal Injector System
Cost
Estimate
1975 M$
0.884
0.159
0.031
0.012
0.060
0.046
0.014
0.010
2.555
Section Subtotal = 3.771
DOLOMITE PROCESSING & FEEDING
1 - Dryer System @ 84 TPK
1 - Crusher & 2 Screens
1 - Distribution Bin @ Crusher
2 - Vibrating Feeders 0 42 TPH
2 - Surge Bins ? 1867 ft2
2 - Bin Activators
2 - Feeders (vibrating)
2 - Dolomite Distribution Boxes
2 - Petrocarb Injection Systems
0.468
0.093
0.015
0.010
0.028
0.042
0.013
0.008
1.3.67
Section Subtotal = 2.044
SPENT BED MATERIAL
2 - Throttle Valves for Bed Drains
2 - 190 £t3 Surge Hoppers
4 - Lock Hopper Valves
2 - 580 ft3 Lockhoppers
2 - High-Temp. Feeders (vibrating)
34 TPH
2 -'Surge Bins @ Solids Coolers
2 - Solids Coolers with Fans &
Cyclones for 24 MBtu/hr
4 - Air Lock Valves for Coolers 3
34 TPH
Section Subtotal
0.791
CONTROL SYSTF.HS
Solids Handling Equipment
OFF-SITE SPENT SOLIDS STORAGE
Total for PFBC Solids Handling
0.149
Section Subtotal = 0.149
4.845
Number
0.024
0.087
0.069
0.170
0.029
0.040
0.299
0.073
2
2
2
2
2
'2
2
2
Cost Scaling
Factor Used
bstimate Used
in Base
Plant Cost
MS
,0.85
6.20
(VFs,b>
0.85
3.40
(Fd/Fd,b>
None
0.85
,0.85
1.58
0.30
4.845
16.33
131
-------
Table C8
SCALING OF HOT GAS CLEANUP AND AIR MAJOR EQUIPMENT FOR PFBC
Cost
Estinate
1975 MS
Number
Cost Scaling
Factor Used
Estimate Used
in Base
Plant Cost
M$
HOT GAS CLF^NUP
16 - Two-in-One Cyclones for Beds
4 - 145 ft3 Collecting Hoppers
32 - Trickle Valves 4 Dip Legs
8 - Lockhopper Seal Valves
4 - 290 ft3 Lockhoppers
4 - Fines Injection Systems
4 - Injection System Valves
2 - Two-in-One Cyclones for CBC
2 - 290 ft3 Collecting Hoppers
4 - Trickle Valves & .Dip Legs
4 - Lockhopper Seal Valves
2 - 580 ft3 Lockhoppers
2 - High Terno. Feeders (vibrating)
2 - Surge Bins for Dust Coolers
4 - Airlock Valves for Dust Coolers
2 - CBC Dust Coolers
32 - Granular Bed Filters
4 - Fines Letdown & Renoval Systems
for Granular Beds 0.721
Section Subtotal = 15.03
HIGH PRF.SSURE .AIR
2 - Air Dryer Precoolers 0.174
4 - Air Dryers 0.350
4 - Booster Compressors 0.416
4 - 400 ft3 Receivers 800C 0.051
• 4 - Air Compressors for Granular Beds 0.193
2 - 400 ft3-Receivers 400// 0.029
1 - Booster Compressor Spare 0.104
1 - Air Compressor Spare _ 0.048
Section Subtotal = 1.365
8.128
0.107
0.229
0.137
0.173
0.520
0.069
0.532
0.087
0.029
0.069
0.170
0.031
0.040
0.057
0.260
7.339
(FH''FH h>
'
None
30.06
2.58
132
-------
Table C9
SCALING OF HEAT EXCHANGE MAJOR EQUIPMENT FOR PFBC
PFB Heat Exchange and
Pressure Parts
PFB Containment Shell and
Nozzles
PFB Fuel Injection and Air
Parts
PFB Controls
PFB Petrocarb Cooler
Economizer
PFB Module
Gas Turbine Economizer
Cost
Estimate
M$ per
Module
2.213
0.566
0.237
0.567
0.089
3.67
0.627
Estimate Used
in Base
Cost Scaling Plant Cost,
M$
Number Factor Used
4
4
None
14.68
2.51
133
-------
Table CIO
SCALING OF BALANCE-OF-PLANT COSTS FOR PFBC
Direct Manual
Field Labor,
M:-: 1000
Balance of
Plant Material,
$ 1000
Cost Seal ing
Factor Used
1.0 PFB STEAM GENERATORS (3)
1.1 Steam Generator Erection
- Erect only (supply by others): includes
PF3 tower skirt; PFB towers; piping
connections at tower; insulation
Supply and erect:
includes tower access steel;
miscellaneous materials and labor
operations
1.2 Steam Generator Auxiliaries
Erect only (supply by others):
includes coal and dolomite Petrocarb
injection systems with injection air
compressors and auxiliaries
Supply and erect:
includes support uteel for Petrocarb
systems.; coal and dolomite piping
from Petrocarb systems to PFB towers
1.3 Hot Gas Cleanup
Erect only (supply by others):
includes cyclones; hoppers and surge
bins; valves; feeders and injection
systems; dust coolers; granular bed
filters; granular bed blowback air
compressors and auxiliaries
Supply and erect:
includes support steel for hot gas
cleanup equipment; access steel
2.0 TURBINE CL', '"IRATORS (3)
2.1 Stcai'. Turbine Generator
Erect only (supply by others):
includes 732 MWe steam turbine:
generator; exciter; auxiliary equip-
ment; integral steam and auxiliary
piping; insulation; miscellaneous
labor ooerations
22
None
None
22
221
100
890
Mono
Hone
33
100
None
61
370
105
1940
3100
100
Non
134
-------
Table C-10 (Cont'd)
Direct Manual
Field Labor,
MH 1000
Balance of
Plant Material,
$ 1000
Cost Scaling
Factor Used
2.2 Gas Turbine/Compressor/Generators
Erect only (supply by others):
includes gas turbine compressors
with 50 MWe generators (4)
145
100
200
None
3.0 PROCESS MECHANICAL EQUIPMENT
3.1 Boiler Feedwater Pumps (3)
Supply and install:
includes turbine-driven main feedwater
pumps and drivers (3 @ $770,000)
3.2 Main Circ. Water Pumps (3)
Supply and install:
includes main circ. water pumps and
motors (3 @ $265,000)
3.3 Other Pumps (3)
Supply and install:
includes condensate pumps and motors -
(2 @ $100,000); and other pumps and
drivers not listed elsewhere
3.4 Main Condenser (3)
Supply and install:
includes shells; tubes; air ejectors
3.5 Heaters, Exchangers, Tanks, and
Vessels (3)
Erect only (supply by others);
includes heat recovery economizers;
insulation; support steel (supply and
erect)
Supply and erect:
includes low-pressure feedwater heaturs
(2); deaerating heater and storage tank;
miscellaneous heater and exchangers;
tanks and vessels
3.6 Stacks and Accessories (3)
Supply and erect:
includes concrete stacks and liners;
lights and marker painting; hoists and
platforms; foundations
27
2430
830
590
2370
None
None
None
None
20
23
180
1450
270
135
-------
Table C-10 (Cont'd)
Direct Manual
Field Labor,
MH 1000
Balance of
Plant Material,
$ 1000
Cost Scaling
Factor Used
3.7 Draft Ducts (3) 23
Supply and erect:
includes ducts from heat recovery
economizers to stacks
3.8 Turbine-Hall Crane (1) 3
Supply and erect:
includes crane and accessories
3.9 Coal Handling (2)
Erect only (supply by others): 22
includes coal dryers (3); support and
access steel for dryers; coal grinders
(3); screens at grinders
Supply and erect: 80
includes railcar dumping equipment; dust
collectors; primary crushing equipment;
belt scale; sampling station; magnetic
cleaners; mobile equipment; conveyors to
pile; reclaiming feeders; conveyors to
cascade; coal cascade; conveyors and
bucket elevators to dryers and grinders;
recirculating conveyors at grinders;
conveyors to Petrocarb; bucket elevators
at Petrocarb
3.10 Dolomite Handling (2)
Erect only (supply by others): .13
includes dolomite dryers (3); support and
access steel for dryers; dolomite
grinders (3); screens at grinders
Supply and erect: 43
includes magnetic cleaners; conveyor to
dolomite pile; reclaiming feeders; con-
veyors to cascade; dolomite cascade;
conveyors and bucket elecators to
dryers; conveyors and bucket elevators
to grinders; conveyors and bucket
elevators to Petrocarb
3.11 Spent Solids Handling (2)
Erect only (supply by others); 11
includes spent solids valves; hoppers;
feeders; fans; cyclones; conveyor to
coolers; ash coolers (2)
270
380
10
6150
(Fc/Fc,b)
0.85
(Fc/Fc,b)
0.85
10
2240
(Fs/Fs,b)
0.85
(Fs/Fs,b)
0.85
30
(Fd/Fd.b)
0.85
136
-------
Table C-10 (Cont'd)
Direct Manual
Field Labor,
MH 1000
Balance of
Plant Material,
$ 1000
Cost Sealing
Factor Used
Supply and erect:
includes ash cooler accessories and
bucket elevators; ash conveyors; ash
storage silos (6) with feeders,
unloaders and foundations; railcar
loading equipment
3.12 Cooling Towers (3)
Supply and erect:
includes mechanical draft towers with
fans and motors
3.13 Other Mechanical Equipment (3)
Supply and install:
includes water treatment and chemical
injection; air compressors and
auxiliaries; fuel oil ignition and
warm-up; screenwell; miscellaneous
plant equipment; equipment insulation
4.0 ELECTRICAL (5)
4.1 Main Transformers
includes main power transformers and
transformers at gas turbine generators
4.2 Other Transformers and Main Bus
includes start-up transformer; station
service transformers; generator main
bus
4.3 Switchgear and Control Centers
includes switchgear and load centers;
motor control centers; local control
stations; distribution panels, relay
and meter boards
4.4 Other Electrical Equipment
- includes communications; grounding;
cathodic and freeze protection;
lighting; preoperational testing
4.5 Auxiliary Diesel Generator
includes diosel generator, batteries
and associated dc equipment
85
3540
(Fd/Fd,b)
0.85
57
34
2450
1820
None
None
535
11
18
25
26,200
2960
1240
2050
None
373
2140
110
137
-------
Table C-10 (Cont'd)
4.6 Conduit, Cable Trays, Wire and Cable
Direct Manual
Field Labor,
MH 1000
Balance of
Plant Material,
$ 1000
381 2500
810 11,000
Cost Sealing
Factor Used
5.0 CIVIL AND STRUCTURAL
5.1 Concrete Substructures and Foundations
(1)
includes turbine building substructure;
PFB base mats; coal, dolomite, and ash
handling foundations, pits and tunnels;
miscellaneous equipment foundations;
auxiliary buildings substructures;
miscellaneous concrete
5.2 Superstructures (1)
includes turbine building; auxiliary
yard buildings
5.3 Earthwork (1)
includes building excavations; coal,
dolomite,and ash handling excavations;
circ. water system excavations; PFB
foundation excavations; miscellaneous
foundation excavations; dewatering and
piling
5.4 Cooling Tower Basin and Circ. Water
System (3)
includes circ. water pump pads, riser
and concrete envelope for pipe; cooling
tower basin; circ. water pipe; cooling
tower miscellaneous steel and fire
protection
6.0 PROCESS PIPING AND INSTRUMENTATION
6.1 Steam and Feedwater Piping (3)
includes main steam; extraction steam;
hot reheat; cold reheat; feedwater and
condensate large piping, valves and
fittings
None
395
3260
220
135
6130
300
100
1510
850
50
11,200
2400
None
138
-------
Table C-10 .(Cont'd)
Direct Manual
Field Labor,
MH 1000
Balance of
Plant Material,
$ 1000
Cost Sealing
Factor Used
6.2 Hot Gas Large Pipe (3)
includes PFB compressed air feed, PFB
hot gas discharge, cyclone and granular
bed filter piping; gas turbine inlet
piping
6.3 Other Large Piping (3)
includes auxiliary steam; process water;
auxiliary systems; spent solids piping
6.4 Small Piping (3)
includes all piping, valves and fittings
of 2-in diameter and less
6.5 Hangers and Misc. Labor Operations (3)
includes all hangers and supports;
material handling; scaffolding; misc.
labor operations
6.6 Pipe Insulation (3)
6.7 Instrumentation and Controls (5)
7.0 YARDWORK AND MISCELLANEOUS (1)
7.1 Site Preparation and Improvements
includes soil testing; clearing and
grubbing; rough grading; finish grading;
landscaping
7.2 Site Utilities
includes storm and sanitary sewers;
non-process service w;itcr
7.3 Roads and Rail roads
includes railroad spurs; roads, walks and
parking areas
7.4 Yard Firt: Protection, Fences ;ind Gates
7.5 Water Treatment Ponds
includes earthwork; compacted-clay
lining off-site pipeline
7.6 Lab, Machine Shop and Office Equipment
120
5700
195
125
390
40
230
1,150
38
27
52
12
1
135
3350
1100
1900
450
5200
20,100
10
50
750
600
10
280
1,700
None
139
-------
The cost of electricity for PFBC is represented by the same general
relationship used for AFBC, equation 12.
While these relationships for AFBC and PFBC are based on oversimpli-
fying assumptions, they provide a means of understanding the major impact
of the critical design and operating parameters.
REFERENCES
Cl. Energy Conversion Alternatives Study (EGAS) Westinghouse Phase II
Final Report, Volume III, NASA CR-134942, 1977, NTIS PB 268-558.
C2. Energy Conversion Alternatives Study (EGAS) General Electric Phase II
Final Report, Volume II, NASA CR-134949, 1977, NTIS PB 269-379.
C3. Evaluation of Phase 2 Conceptual Designs and Implementation Assess-
ment Resulting from the Energy Conversion Alternatives Study (EGAS).
Report to ERDA and NRF, NASA, Lewis Research Center, April 1977,
NTIS PB 270-017.
C4. Utility Boiler Design/Cost Comparison: Fluidized-Bed Combustion
versus Flue Gas Desulfurization. Report to EPA, TVA, November 1977,
EPA-600/7-77-126.
140
-------
APPENDIX D
PARTICULATE CONTROL PROJECTIONS
In this work we have studies the effect of lower SO standards on
particulate effluent. By increasing the calcium-to-sulfur ratio used in
fluidized-bed combustion processes, the S0_ emissions can be reduced. A
larger amount of particulate matter is carried over from the bed, however,
since there is a larger amount of sorbent to be lost by attrition and
elutriation per unit mass of coal burned.
A consistent methodology has been formulated and carried out for
analyzing the effect of the increased carry-over on FBC flue gas particu-
late emissions, for both the pressurized and atmospheric cases. A dis-
cussion of the method and its assumptions follows. The ability to control
these emissions to the requirement of 12.9 ng/J (0.03 Ib/MBtu) is assessed.
The validity of any particulate removal study depends on the accuracy
with which particle loadings and size distributions at the source can be
estimated. In most previous work it has been assumed that an "average"
particle represents the dust being elutriated from the -bed.- The properties
of the three main sources of flue gas particulates in FBC (char, ash, and
sorbent), however, are quite different, with the result that calculations
based on such "average" particles can only be rough estimates. This work
estimates loadings and size distributions for each of the three components
and accounts for the differences in physical properties (density and
shape) in the particulate removal calculations.
PROJECTION OF PARTICULATE EMISSIONS
Solids Feed to Fluidized Bed
The solids fed to the process consist of coal and dolomite. The
coal size distribution for both the pressurized and atmospheric cases
141
-------
are shown as the bottom curve in Figures Dl and D2. These size distri-
butions are typical 6350 um and 12,700 pm distributions, respectively,
from a hammer mill. The s.prbent size distributions were obtained by
screen analysis of an available sample of single-screened limestone from
Greer. The sorbent for the pressurized case is a 6350 ym stone and for
the atmospheric case it is a 3180 ym material. The very fine material
from the screen analysis was analyzed with a Coulter Counter in an
attempt to obtain a better estimate of the fines size distribution. These
results are shown in Figures D3 and D4.
Note that the equivalent spherical diameter (d ) has been used in
..,.,. . r . . , ..-...,... s
these plots and that this can differ significantly from d obtained
...• r ...... . . . .6. 3 avg
by the screen analysis. The average particle diameter and the equivalent
spherical diameter are related by the following equation:
d = d (n<|>) ,
avg s
where n is the specific surface of the particle, and $ is the sphericity
of the particle. . For-the limestone it-was possible to obtain r\ as a,.
function of d by measuring the pressure drop through a bed of par-
ticles of size d , and of known bed voidage. Ergun's equation was
then used to obtain the product of d :
+1.75
The pressure drop data are shown in Figure D5. Having obtained n(d ),
and assuming to be similar to sharp sand ( = 0.6), the size distribu-
tion obtained as d with the screens could be converted and plotted as
avg F
the spherical equivalent diameter (d ). Although this analysis was
S
carried out as carefully as possible, it should be noted that these data
apply strictly only to the stone studied here and are not generally
applicable to all FBC sorbents.
142
-------
0.01
10.000
Figure Dl - Coal and char size distributions (PFBC)
0.01
10. IKK)
Figure D2 - Coal and char size distributions (AFBC)
143
-------
I I I III 1 1—I I I I 111 1 1 I I I II 1 1—I—I I/Ill
0.10
10. on
Figure- D3'- Sorbent size distributions (PFBC)
i—i i i i 111 1—i—i ii i i i 11 1—y—i i i 11
10.000
Figure D4 - Sorbent size distribution (AFBC)
144
-------
Curve 687625-B
100 X 10
100
E
u
I
CT>
Q_
10.
(40-200)Mesh
35-40
1.0
J L
1
J L
1.0
10.0
V(cm/s)
100
Figure D5 - Pressure drop vs. flow for different
size fractions of limestone
145
-------
Elutriation of Ash from the Bed
It has been assumed in this work that all the ash in the coal is
elutriated, which implies that the ash elutriation rate per weight of
coal fed (W /W ) is equal to x , and the percent ash of the coal:
3. C Si
W /W = x
a c a
The size distribution of the ash was estimated from work done by the
U. K. National Coal Board (NCB), and reported by Merrick and Highley.
A Coulter Counter analysis of the fines from an Exxon miniplant (run 34)
indicated a minimum particle size of about 0.1 ym. The. size distribution
constructed by these means for both the pressurized and atmospheric cases
is shown in Figure D6.
Elutriation of Char from the Bed
To estimate the amount of char lost from the bed it was assumed
that all of the combustion efficiency loss was due to lost char.
D2
Exxon has produced a correlation of combustion efficiency with excess
air and bed temperature. This allo.ws, • the- weight of char- lost (W ,0 per
weight of coal fed to be obtained by the following heat balance:
(1 - n , )W H = W H
comb c c ch ch
or
W , /W = (1 - n , )(H ,H ) ,
ch c comb c/ ch
where H and H , are the heating values of the coal and char respectively,
c ch
and combustion efficiency n , is obtained from the correlation.
comb
To estimate the size distribution of the elutriated char it was
assumed that the char particles maintained the same size as the parent
3
coal particle but experienced an apparent density decrease to 0.88 g/cm
due to loss of volatiles. The diameter of the char particle whose
terminal velocity corresponds to the superficial velocity in the bed was
calculated using the data of Pettyjohn and Christiansen. It was then
146
-------
I I I I I I I I
99.9
99.996
1000
Figure D6 - Size of ash elutriated from bed for both PFBC and AFBC
-------
assumed that the size distribution of the coal feed below the maximum
diameter elutriated typified the elutriated char size distribution.
Clearly, the fine fractions of the material will be elutriated faster,
which will bias the distribution to the finer sizes; these finer char
particles, however, are more apt to be burned, which would bias the
distribution toward the coarser material. The estimate adopted here
lies between these extremes.
Elutriation of Sorbent from the Bed
Sorbent elutriated from the bed can come from two sources, elutria-
tion of fines (W ) in the feed material (W ,.) and attrition of the
se sf
material in the bed with subsequent elutriation (W ). We assumed that
sa
all of the feed material with a diameter smaller than that whose terminal
velocity corresponds to the superficial velocity in the bed was elutriated
immediately. If y is the cumulative fraction of the sorbent feed less
m
than this critical diameter, the loading from elutriated feed is simply:
W /W = y
se sf m
The size distribution of this material is obtained from the feed
distribution.
It is somewhat more difficult to estimate loadings and size distri-
butions for the attrited material, since the experimental data are
limited and have not been assimilated to develop predictive capability.
Preliminary estimates from Exxon's miniplant and Argonne's pressurized
unit indicate that roughly 4 percent/hr of the bed weight is a reasonable
estimate of attrition rates. Some experimental work is currently being
D4
conducted at Westinghouse Research and Development Center that suggests
that the attrition rate can be adequately described by an expression of
the form
rate = K(U - U Jt~n
s mf
148
-------
for the range of superficial velocities being considered here. Prelimin-
ary results indicate that the constants n and K have values of about
-•:-• _4
n = 0.7 and K = 3 x 10 for a good attrition resistant sorbent. An
average attrition rate can be calculated by integrating over the resi-
dence time (T) for the particles.
K(Us-Umf)t~0'7dt
or
X(T) = 0.001(U - U ,)t~0'7 .
s mr
The loading of attrited material (W ) can then be expressed by the
S3.
following
(W /W ) = (1 - v 'iTxCr")
\ w / w _/ ~- \± y / i A l l / .
sa sf m
Combining the attrited material with the immediately elutriated material
gives the total sorbent loading to the flue gas (W ) as
st
W W + W
st se sa ,1 \ ~f \
W7 = W , = ym + (1 - ym)Tx(T) '
sf sf
The residence time for the solids can be estimated by the use of the
following expression:
(288) (T) (D) (1 - V ) (p b) [(W /W ) <|> + (1 -
T = (EF) (P) (MW) (U ) (W _/W ) (R) (W 7W~
s st sti st su c
where
T = bed temp., °C
D = bed depth, cm
V = void fraction, including tubes
3
p' • = bulk density sorbent, 1.38 g/cm
W /W = air to coal ratio - stoichiometric, 10.413
a c
= air equivalence ratio, 1.20
3. •
x , = weight fraction of ash in coal, 0.107
ash
EF = expansion factor, 1.4
149
-------
P = pressure IcPa, 101.3, 1013
MW = molecular weight of flue gas
U = superficial velocity, cm/s
s
,(W f /W , ) . = weight of sorbent to weight of sulfur, 5.75
SI SU S C
R = calcium to sulfur ratio; 2, 5, 10, 20
(W . /W ) = weight fraction sulfur in coal, 0.043.
SU O
The pressurized case has a residence time that varied from about 2.8 to
0.28 hr when calculated in this manner; depending upon the sorbent feed
rate; while for thfe atmospheric case t va'ried from 6;8 to 0.68 for the
different values of R;
The attrited material size distribution was estimated using Westirig-
fj'4 • • •
house experimental data as a guide. The size distribution of the
combined attrited and elutriated sorbent is shown in Figures D3 and D4.
We observed that for the size distributions assumed, the distribution
fines created by attrition closely resembled the size distribution
of the fedd elutriated from the bed. This allowed one size distribution
to fee use'd for all values of R; since the relative amounts of material
attritfed arid fines in the feed had little impact on the combined size
distribution.
To obtain the mass of flue gas per gram of coal fed (W /W ) , a
fg c
mass balance around the fluidized bed was employed, assuming the following
flows :
coal + air + sorbent = flue gas + char loss + ash
+ spent sorbent + half-calcined sorbent.
If R is taken as the calcium to sulfur atomic ratio used, and £. is the
percent sulfur removal (assumed to be 0.90), the overall stoichiometry
can be written as
R(M C03 • CaC03) + (SC^) + 1/2 £(02) + (R - O MgO •
+ (R + O
150
-------
This allows the mass balance to be completed. Letting
M = molecular weight sulfur = 32
O
>L = molecular weight dolomite = 184
= molecular weight half calcined = 140
M = molecular weight sulfided = 186
bD
XS = excess air
the mass of flue gas per unit weight of coal fed is:
(W, /W ) = 1 + (W /W ) _ (1 + XS) - (1 - n , ) (H /H , )
fg c a c st comb c ch
x
- * . + rr^- [KM,, - (R -
ash M D
s
It is now possible to estimate the total loading of solids from the bed
as well as the relative amounts of ash, char, and sorbent:
W_ „ i x + (1 - n . ) (H /H , ) + R(W ,/W ) x (y + (1 - y ) x)
total _ ash comb c ch sf su st su m m
W,. (W^ /W )
fg fg c
Table Dl summarizes the results of the application of these equations
for both the pressurized and atmospheric cases.
Although it is difficult to make comparisons between different
systems, the projected overhead loading of 0.395 g solids/g coal fed for
a calcium to sulfur ratio of 2 seems reasonable, as Exxon has reported
a range from 0.13 to 0.21, and ANL has reported numbers from about
0.1 to 0.5 for the pressurized case.
For the atmospheric case, total solids loadings of between 0.5 and
0.85 g solids/g coal fed have been reported at superficial velocities
near the 3.05 m/s (10 ft/s) assumed here, which compares well with the
value of 0.535 projected. Data on the size distributions of the solids
carry-over from the combustor are scarce, but even less data on the
distribution of the fines exist. Figure D7 shows the combined solids
size distribution projected by this analysis for calcium/sulfur = 2 for
151
-------
Curve 6972<*2-A
o>
IM
O>
0.1
1.0
10.0 100.0
Diameter (urn)
1000.0
Figure D7 - Total solids from combustor (PFBC)
Curve 6972^3-A
0.1
1.0
1D.O
Diameter
100.0
1000.0
Figure D8 - Total solids from combustor (AFBC)
152
-------
PFBC. Included is a range of distributions reported by Exxon. The
manner in which the data are extrapolated into the fines region has a '
great impact on the performance of the particulate removal train. As
mentioned before, particle size analysis done on Exxon fly ash indicated
that the minimum particle size was about 0.1 ym. The total solids
size distribution from the AFBC is shown in Figure D8. As can be observed,
the projected distribution agrees rather well with the experimental data
D8
reported by Pope, Evans and Robbins.
Table Dl
EFFLUENT LOADINGS FROM BED
Pressurized Case
R (Calcium/
Sulfur
2
5
10
20
2
5
10
20
g Sorb/
g Coal
0.216
0.472
0.863
1.597
0.356
0.792
1.466
2.749
g ASh/
g Coal
0.107
0.107
0.107
0.107
Atmospheric
0.107
0.107
0.107
0.107
-------
Coal
Sorb
Primary
Bed
Air
Ash and Spent
Sorbent
Primary
Cyclone
C
B
C
Air
Dwg. 6429A58
Final Stage
GBF(PFBC)
ESP (AFBC)
or
Fab. Flit (AFBC),
Tan Jet
CBC Cyclone
Figure D9 - Schematic of particle control system
-------
Primary Cyclone
The primary cyclones used for this study were of a fairly conven-
tional design operating at an.inlet velocity of about 24 m/s (80 ft/s).
3
Each unit has a capacity of about 27,184 m /hr and a pressure drop of
about 21 kPa (3 psi). A plot of particle penetration versus particle
size for the various particulate densities is shown in Figure DIG. These
data are based on manufacturers' projected performance at pressure and
temperature.
The projected performance of this cyclone on the individual species
of particulates carried out of the bed is summarized in terms of overall
efficiencies in Table D2.
Table D2
PRIMARY CYCLONE EFFICIENCIES
PFBC AFBC
Ash 0.925 0.926
Char 0.994 0.998
Sorbent 0.989 0.991
The estimated overall efficiency for all particulate matter has been
calculated for the various calcium to sulfur ratios considered and is
presented in Table D3. Figures Dll and D12 show the projected size dis-
tribution of the combined solids leaving the primary cyclone for a cal-
cium to sulfur ratio of 2.
Table D3
OVERALL EFFICIENCIES OF PRIMARY CYCLONE
Calcium/Sulfur PFBC AFBC
2 0.972 0.977
5 0.979 0.982
10 0.983 0.985
20 0.984 0.987
155
-------
d.oi
di
20
50
00
Hi—1
CD
c~
:cu
o,
95
99
99.9
99.99
Curve 694490-A
Specific Gravity
2.8 1.35.88
1.0
10
100
Figure D-10 - Grade efficiency for primary cyclone
156
-------
Curve 69721IS-B
0.01
0.1
1
5
OJ
I "
?. 50
Calcium/Sulfur = 2
95
99
99.9
99.99
0.10
1.0
10.0
100.0
D (pm)
P
Figure Dll - Combined solids leaving primary cyclone (PFBC)
Curve 6972*46-9
Figure D12 - Combined solids leaving primary cyclone (AFBC)
157
-------
Carbon .Burhup Cell
All of the solids caught by the primary cyclone are considered to
be fed to the carbon burnup cell (CBC) to maintain high overall combus-
tion efficiencies. The superficial velocity in this bed is somewhat
lower than the primary - i.7 m/s (5.5 ft/s) versus 2.1 m/s (7 ft/s) for
the pressurized case and 2.7 m/s (9 ft/s) versus 3.1 m/s (10 ft/s) for
the atmospheric case. The temperature of the carbon burnup cell has
been taken to be 1093°C (2000°F) for both cases; The materials carried
over from the carbon burhiip cell wer£ characterized in the same way as
was the material from the primary bed except that no attrition of the
sbrbent was accounted for; It has beeh noted that the irate of attrition
decrease's with time as the sharp corners are rounded off•, so it is
probably not a bad estimate to ignore attrition in the carbon burnup
cell. Figures Dl3 and DlA show the projected size distribution of the
combined solids leaving the carbon burnup cell for a calcium to sulfur
ratio of 2;
Carbon . Burhujj Cell .Cyclone .
Since the amount of carbon fed to the CBC is small relative to the
other solids, the amount of air fed to the CBC is also relatively small.
3
The grain loading coming from the CBC is thus rather high (200 to 400 g/m )
[100-200 g/scf], so a nohfbuling type of cyclone such as the Tan Jet by
Donaldson has been considered for use at this point. Donaldson has pro-
vided estimated performance at elevated temperature curves which are
shown in Figure D15. The overall Tan Jet efficiency is projected to be
very high since all the particulates that pass into it have already been
captured once by a standard cyclone and are, thus, fairly coarse. The
cleaned gas from the CBC is then merged with the gas cleaned by the pri-
mary cyclone and is fed to the third and final stage of fine particulate
removal. Tables D4 and D5 present the estimated efficiencies for the Tan
Jet on the individual species and overall solids removal efficiencies as a
function of calcium to sulfur ratio. Figure D16 shows the projected
combined solids size distribution leaving the Tan Jet, and Figures D17
158
-------
Curve 6972
-------
.0.01
0.1
1
Curue 687E23-B
3 30
r
70
90
98
99
99.9
99.99
~i 1 1—F
Tan }e\
TA1
_L
0.10
1.0
10.0
D (pml
100
'PFBC
Figure D15 - Grade efficiency for fan jet
_V , r
Curve 6972"<9-B
Calcium/Sulfur =2
AFBC
10.0
-Ar-
1.0
0 (pm)
10.0
100
Figure D16 - Combined solids leaving the CBC cyclone
160
-------
Curve 6972^1-A
O>
o
£
"c
99,99 *
0.01
10.0
Diameter (urn)
100.0
1000.0
Figure D17 - Combined solids to final control device (PFBC)
Curve 6972Wt-A
0>
tsi
c
OJ
OJ
0.
0.1
1.0
100.0
1000.0
10.0
Diameter (\un)
Pigure D18 - Combined solids to final control device (AFBC)
161
-------
Table D4
PROJECTED EFFICIENCIES OF TAN JET FOR EACH SPECIES
Species PFBC AFBC
Sorbent
Char
Ash
0.9968
0.9995
0.9769
0.9975
0.9990
0.9769
0.9896
0.9929
0.9943
0.9954
0.9930
0.9952
0.9961
0.9967
Table D5
PROJECTED OVERALL EFFICIENCIES FOR THE TAN JET
Calcium/Sulfur PFBC AFBC
2
5
10
20
and D18 show the size'.distribution of the merged streams from the pri-
mary collector and the Tan Jet. This stream is then fed to the final
stage of particle removal equipment.
Fine Particulate Removal
It is in the choice of the fine particulate removal that the signi-
ficant differences between the gas cleaning systems arise. Two major
choices for the final stage of particulate removal appear to exist for
the atmospheric-pressure case. These are: fabric filtration and electro-
static precipitation, and there is some degree of uncertainty in the use
of either of these devices. The precise nature of the electrical proper-
ties of the ash, sorbent, and char from the fluidized-bed combustion
process are not well known, so migration velocities and possible reentrain-
ment problems in electrostatic precipitators can only be estimated. In
view of .the considerable problems experienced in trying to precipitate
low-sulfur Western coals in conventional boilers, these considerations
may be of some consequence in the fluidized-bed case. For the purposes
162
-------
D9
of this work, data reported by J. D. McCain have been used. These
data were obtained by measurements on a cold-side pilot-scale ESP oper-
ating on a coal-fired boiler burning low-sulfur coal. It has been assumed
that the different components of the particulate are collected with equal
efficiency since the grade efficiency curve used already indicates the
problems of reentrainment and gas bypassing by the lower-than-expected
efficiencies in the larger particle sizes. The hot-side precipitator
D9
data from McCain do not indicate this problem with the larger particle
sizes at only-modestly larger collection area to flow ratios. The lower
efficiency cold-side data were adopted as a conservative estimate of
D9
performance. The grade efficiency assumed for the ESP from McCain is
shown in Figure D19.
The other alternative for the atmospheric case is the use of a fabric
filter system. There may be some uncertainty associated with the erro-
siveness of the particulate, and hence with bag life, but it seems likely
that operating conditions can be chosen to avoid serious maintenance
problems with a baghouse system. There is probably less uncertainty
about the performance of the filter system, compared to the ESP, since
the physical nature of the particles is unlikely to have much impact on
the capture efficiency of the filter. Baghouses, however, do have poten-
tial problems with start-up and blinding, which necessitate careful
operation. The grade efficiency used in this study has been taken from
data reported by H. Spagnola and is shown in Figure D20. These data
were collected from a large baghouse facility operating on power plant
flue gases.
The final stage of cleanup for the pressurized system is probably
the least well-defined piece of equipment, since hot pressurized particu-
late removal is not yet state-of-the-art practice. The device that has.
been used as the basis of this study is a granular bed filter of the
Ducon Corporation design. Since preliminary operation of a 850 m /hr
scale granular bed filter unit at the Exxon miniplant has been plagued
by various mechanical problems, data from bench-scale experiments con-
ducted by Westinghouse have been used rather than those from the
163
-------
Curve 694492-A
99,9'8
-. 99,9
f; 99.8
I 99.5
o. 99
gp 98
I 95
§ 90
.o
'•6
-2
60
30
f = t600C
SCA=67 fi$(m3/s)
Current Density =
15,6nA/cm2
Low-Sulfur Coal
0,1
0,5 1.0
Particle Diairieter, (prii'j
5,0 10.0
Figure' Dl9 - Grade efficiency for ESP (Ke'f-,D'9 j
Curve S94489'-A
99.99
99.9
99
95
* 90
c
o>
'o
£ 50
c
0
"o
3 10
5
1
0.1
0.01
0.
T r — i i i i M | 1 — T" i i i i 1 1 -|. 1 — i — i i MM
: ^^^_cr^ :
_
— -
_
_ —
-
—
_
_ ~
Filtration Velocity 1 cm'/s
T = 163°C j
AP = 0'.75kPa
- Reverse Air Cleaning -
I . 1 III - 1 II Mil 1 . 1 1 1 M 1 1
01 0.1 1.0 10
U.U1
0.1
1
5
10
c"
_o
'50 |
c
-------
miniplant. The grade efficiency used in this study is shown in Fig-
ure D21. It should be noted that these data were obtained on a small
experimental unit and are more optimistic than those realized in larger
units. Confirmation of this performance in larger units is required.
The entire hot gas fine particulate removal picture may change if
the preliminary work done on rigid ceramic filters reported by Ciliberti
can be developed to commercial scale. These filters may have some prob-
lems with cleaning if a sticky ash is encountered. They appear, however,
to have a large cost advantage over granular bed filters as well as
significantly higher collection efficiencies in the submicron range.
Ceramic bag filters are also promising, as indicated by work reported
, A D13
by Acurex.
Results
The projected effectiveness of each control system for all three of
the particulate species is shown in Table D6 where the fractional pene-
tration for the overall system is presented.
Table D6
OVERALL FRACTIONAL PENETRATION FOR EACH SPECIES
.012
Species
Sorbent
Char
Ash
PFBC
GBF
0.000263
0.000030
0.001114
AFBC
Fab. Filt.
0.000030
0.000018
0.000161
ESP
0.000313
0.000189
0.00125
These species penetrations can be weighed to give the projected
overall particulate penetrations that are present in Table D7. As can
be noted from the individual species size distribution curves of Fig-
ures Dl through D4 and D6, all of the particulates emitted from the final
stage device for all cases have essentially the same size distribution,
with the .effluent being generally log-normal and running from 0.1 pm to
about 10 to 15 urn, with a mass median diameter from 0.5 to 1 ym. Since
165
-------
a-.
0.01
0.1
1
10
s*
•£. 30
1
c 50
o
1 70
o>
Q-
90
98
99
99.9
99.99
0. ]
Curve 687622-A
1 1 1 1 I 1 1 1 1"
_^/ -
_
"~ -
~~ -
~ -
—
1 111 III
0 1. 0 10.
D (urn)
en
.c.
Curve 694491-A
- Ill No. 6 As Fired 3% Moisture
10.7% Ash
4.3'% Sulfur
Figure D21 - Grade efficiency for
granular bed filter
6 8 10 12 14 16 18 20
Calcium/Sulfur Ratio
Figure D22 - Projected particulate emitted
vs. calcium to sulfur ratio
-------
the amount of sorbent fines produced by the feed and attrition vary with
the calcium to sulfur ratio used, the data are presented with R as a
parameter. These penetrations are projected by applying the appropriate
grade efficiency curve to the mass loadings and size distribution shown
in Figures D17 and D18.
Table D7
OVERALL SYSTEM PARTICULATE FRACTIONAL PENETRATION
R(Ca/S)
2
5
10
20
PFBC
GBF
0.000451
0.000382
0.000334
0.000305
AFBC
Fab. Filt.
0.000055
0.000044 .
0.000038
0.000034
ESP
0.000485
0.000408
0.000369
0.000344
Since the total loading to the system increases as R increases,
the total mass escaping the system increases even though the penetration
decreases slightly with increasing R as indicated in Table D7. This
effect is shown in Figure D22, which is a plot of the total weight of
particulate escaping the system per unit energy of coal fed.
For AFBC, the electrostatic precipitator is projected to control
the emission of particulate material to below the particle emission
requirement of 12.9 ng/J (0.03 Ib/MBtu) for calcium to sulfur ratios
less than five. The fabric filter system is projected to control the
particulate emissions to levels less than 12.9 ng/J (0.03 Ib/MBtu) for
all calcium to sulfur ratios considered. With the investment and oper-
ating and maintenance costs of electrostatic precipitators and baghouses
not being sufficiently different to have a significant impact on the
overall cost of electricity, the baghouse system is recommended.
Particulate control with PFBC may be different from AFBC in the
sense that the process constraints on particulates may supersede the
167
-------
environmental requirement of 12.9 ng/J (0.03 Ib/MBtu), depending on
size distribution. The granular bed filter is projected to meet the
emission control target for calcium to sulfur ratios less than 1-2.
The projections of sorbent requirements in this report for 90 percent
SCL removal ranged from a calcium to sulfur ratio of 2.9 to 7.0 for AFBC,
and from 1.7 to 4.5 for PFBC. It is, thus, apparent that, according to
these projections of particle control performance, the targets of 90 per-
cent SO- removal and 12;9 rig (0.03 Ibj particulate/J (MBtu) should be
a'ctiievabie simuitanedusiy; As emphasized earlier, the estimates of par-
ticle control device performance used here are largely projections that
must be confirmed on operating fluidized-bed combustion units.
The impact of coal sulfur-content, coal ash-content, and coal heat-
ing value are important in determining the particulate emission, but their
irifiue'rice should not change the conclusions already reached. The inter-
cept of the curves in Figure D13 at a calcium to sulfur ratio of 0 should
increase approximately linearly with an increase in the ratio of the
coal ash-eontierit over the coal Heating value. The slope of the curves
will increase approximately linearly with the quantity of coal sulfur-
cdriterit over the coal heating value. As an example, a 50 percent
increase in the coal sulfur-content and coal ash-content from the base
values in Figure D13 (4.3 percent arid 10.7 percent, respectively) would
result in a particulate emission for the fabric filter system of less
than 12.9 ng/J (0.03 Ib/MBtu) for the range of calcium-to-sulfur ratios
considered and would result in a particulate emission less than 12.9 ng/J
(0.03 Ib/MBtu) for granular bed filters at calcium to sulfur ratios less
than 3 and for electrostatic precipitators at calcium to sulfur ratios
less than about 1.5.
REFERENCES
D.I Merrick, D., and U. Highley, "Particle Size Reduction and Elutriation
in a Fluidized Bed Process," AIChE Symposium Series 137, Vol. 70,
1974.
168
-------
D2. Studies of the Pressurized Fluidized-Bed Coal Combustion Process,
Office of Research and Development, EPA, Exxon Research and Engineer-
ing Co., Linden, NJ, September 1977, EPA-600/7-77-107, NTIS
PB 272-722.
D3. Pettyjohn, E. S., and E. B. Christiansen, "Effect of Particle Shape
on Free-Settling Rates of Isometric Particles," Chem. Eng. Prog.,
44(2), 1948.
D4. Vaux, W. G., Work to be reported by Westinghouse under contract to
EPA, contract 68-02-2132.
D5. Hoke, R. C., et al., "Studies of the Pressurized Fluidized-Bed Coal
Combustion Process," report to EPA, Exxon Research and Engineering
Co., Linden, NJ, September 1976, EPA-600/7-76-011.
D6. Vogel, G. J., "Recent ANL Bench-Scale Pressurized-Fluidized Bed
Studies," presented at the 4th International Conference on Fluidized
Bed Combustion, McLean, VA, December 1975.
D7. "Summary Evaluation of Atmospheric Pressure Fluidized-Bed Combustion
Applied to Electric Utility Large Steam Generators," Vol. I: Final
Report and Vol. II: Appendices, report to Electric Power Research
Institute, The Babcock and Wilcox Co., EPRI FP-308, October 1976.
D8. Aulisio, C. , et al., "Results of Recent Test Program Related to
AFB Combustion Efficiency," presented at the 5th International
Conference on Fluidized Bed Combustion, Washington, DC, December 1977.
D9. McCain, J. D., "Results of Field Measurements of Industrial Particle
Sources and Electrostatic Precipitator Performance," J. Air Pollution
Control, 25(2); February 1975: 117.
D10. Spagnola, H., "Operating Experience and Performance at the Sunbury
Baghouse," Symposium on Particulate Control in Energy Processes,
San Francisco, CA, May 11-13, 1976.
169
-------
Dll. Ciliberti, D., D. C. Realms, D. H. Archer, "Particulate Control
for Pressurized Fluidized-Bed Combustion Processes," presented at
the 5th International Conference on Fluidized Bed Combustion,
Washington, DC, December 1977.
D12. Ciliberti, D., "High-Temperature Particle Control with Ceramic
Filters," report to EPA, Westinghouse Research and Development
Center, Pittsburgh, PA, October 1977, EPA-600/2-77-207.
D13. Shackleton, M. A. and D. C. Drehmel, "Barrier Filtration for HTHP
Particulate Control," presented at Symposium on the Transfer and
Utilization of Particulate Control Technology, Denver, CO, July 1978.
170
-------
APPENDIX E
CONVENTIONAL POWER PLANT DESIGN AND COST BASIS
The cost of electricity for a 796 MW conventional pulverized-coal
power plant with lime-slurry wet-scrubbing for sulfur oxide control has
been projected at two levels of control: 1.2 Ib SO?/MBtu (516 ng/J)
(83% sulfur removal efficiency for a 4 wt % sulfur coal) and 90 percent
sulfur removal efficiency. The basis for these projections is a design
study performed by General Electric/Bechtel and described and expanded
F1
upon by TVA.
The GE and TVA study used a basis consistent with that applied for
conceptual AFBC and PFBC power plant design (the basis applied in the
EGAS program) and is thus valuable for comparison purposes.
The General Electric conventional plant design is for a 3.9 wt %
sulfur coal, meeting a standard of 90 percent sulfur removal for a
4.5 wt % sulfur coal. The conventional plant has a stack gas temperature
of 175°F (79°C) (50°F [28°C] of reheat) and operates with 33.8 percent
plant thermal efficiency. The lime-slurry scrubber system was on-site
calcination of limestone and has a 5-year slurry disposal capacity and
a 5.5-year plant construction time. The wet lime absorber parameters
are summarized in Table El. The General Electric/Bechtel costs are
used directly for the case of a 4 wt % sulfur coal with 90 percent sulfur
removal efficiency except for two changes suggested by TVA: pond size
was increased to a 30-year capacity; and the stack cost was increased to
reduce stack corrosion.
The unmodified General Electric/Bechtel conventional plant cost is
$771.3/kW. TVA recommends an increased stack cost of $4.33 x 10 and
increased disposal cost of $22.92 x 10 , resulting in a total plant
investment $805.6/kW. The cost of electricity is
171
-------
Capital charges 25.46 mills/kWh
O&M 3.74
Fuel 10.10
Total 39.30 mills/kWh
The cost of limestonej with a calcium-to-sulfur ratio of 1.10 and a
delivered cost of $5/ton ($5.51/Mg), is included in the O&M cost at
0:374 mills/kWh.
TVA modified the General Electric/Bechtel design fbr the conventional
power plant for the case of 83 percent SOx removal with a 3.9 wt % sulfur
coal, using' the premises listed in Table E2y with a 30-^ear dispbsal
capacity arid bn-site limestone calcination.
The TVA-modified conventional plant electrical capacity is 798.2 MWe.
the plant capital investment is $782.3/kW, and the cost of electricity is,
with" limestone at ($5/51/Mg).
Capital charges 24.73 mills/kWh
6&M 3 -. 74
Fufel lb;10
Total 38.57 mills/kWh
Both of the General Electric/Bechtel and TVA conventional plants are
assumed to be capable of satisfying the more stringent NO and particulate
X E2
emission levels considered in this report with little cost impact.
REFERENCES
El. Utility Boiler Design/Cost Comparison: Fluidized-Bed Combustion
versus Flue Gas Desulfurization. Report to EPA, TVA, November 1977,
EPA-600/7-77-126.
E2. Effects of Alternative New Source Performance Standards on Flue Gas
Desulfurization System; Report to EPA, PEDCo Environmental, Inc.,
March 1978, EPA-600/7-78-033, NTIS PB 279 080.
172
-------
Table El
WET LIME ABSORBER SYSTEM PARAMETERS, CONVENTIONAL
FURNACE-STEAM CYCLE
(Basis: 90% SOX removal for 4.5% sulfur coal -
General Electric/Bechtel Design)
S02 Absorbers (6)
No. of Stages
Superficial Gas Velocity
Total Pressure Drop
Liquid/Gas Ratio
Presaturation Sprays
Mist Eliminator Wash Sprays
Lime: S02 Stoic. Ratio
Absorber Hold Tank Residence Time
Recycle Slurry Solids
Lime Makeup Slurry Solids
Spent Slurry Pond Solids
TCA-Type
3 (6 in. of Spheres/Stage)
8 ft/s
9 in. H20
72 gal/Mscf
2.5 gal/Mscf
2 gpm/ft2
110%
5 min
10% wt
20% wt
40% wt
Table E2
WET LIME ABSORBER PARAMETERS, CONVENTIONAL
FURNACE-STEAM CYCLE
(Basis: 83% SOX removal for 3.9% sulfur
coal - TVA design)
S02 Absorbers (4)
No. of Stages
Superficial Gas Velocity
Total Pressure Drop
Liquid/Gas Ratio
Presaturation Spray
Lime/S02 Absorbed Stoichiometry
Absorber Hold Tank Residence Time
Recycle Slurry Solids
Lime Makeup Slurry Solids
Spent Slurry Pond Solids
Calcination Off-Gas meets S02
emission requirements
3 (TCA-type)
12.5 ft/s
6 in. H20 in the scrubber
50 gal/Mscf
2.5 gal/Mscf
1.05
10 min
8% wt
20% wt
40% wt
173
-------
TECHNICAL REPORT DATA
(Please read Inunctions on the reverse before completing)
1. REPORT NO.
EPA-600/7-78-163
2.
3. RECIPIENT'S ACCESSION NO.
4.T.TLE AND SUBTITLE Effect of SQ2 Emission Requirements
on Fluidized-bed Combustion Systems: Preliminary
Technical/Economic Assessment
6. REPORT DATE
August 1978
6. PERFORMING ORGANIZATION CODE
7.AUTHOR.S) R.A.Newby, N.H.Ulerich, E.P.O'Neill,
D. F. Ciliberti, and D. L. Keairns
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Westinghouse Research and Development Center
1310 Beulah Road
Pittsburgh, Pennsylvania 15235
10. PROGRAM ELEMENT NO.
E HE 62 3 A
11. CONTRACT/GRANT NO.
68-02-2132
12. SPONSORING AGENCY NAME AND ADDRESS
EPA,,Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF R.EPORT AND PERIOD COVERED
Topical; 8/77-1/78
14. SPONSORING AGENCY CODE
EPA/600/13
15. SUPPLEMENTARY NOTES JERL-RTP project officer is D. Bruce Henschel, Mail Drop 61,
919/541-2825.
i6. ABSTRACT The report gives results of a preliminary technical/economic evaluation to
project the impact of SO2 control requirements (up to 90% control) on the capital and
energy costs of atmospheric-pressure and pressurized fluidized-bed combustion
(AFBC and PFBC) power plants. Ability of AFBC and PFBC to reduce emissions of
particulates and NOx is also considered. Performance and economic projections are
presented, both for the current New Source Performance Standards and for more
stringent controls; the AFBC and PFBC projections are compared with equivalent pro-
jections for conventional boilers with flue gas desulfurization. The projections show
that AFBC and PFBC plants can achieve SO2 control up to at least 90%, and still
remain economically competitive with conventional boilers using scrubbers. However,
FBC plant design and operating parameters are critical to the achievement of high
levels of SO2 control at competitive energy costs. In particular, increased gas resi-
dence time in the bed, and reduced sorbent particle size, are important in effective
attainment of high levels of SO2 control. The projections of FBC performance at 90%
SO2 removal must be confirmed experimentally on FBC units large enough to be
representative of commercial-scale combustors.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
C. COSATI l-'icld/Gloup
Pollution Nitrogen Oxides
Fluidized Bed Processing Boilers
Sulfur Dioxide Flue Gases
Capitalized Costs Desulfurization
Electric Power Plants
Dust
Pollution Control
Stationary Sources
Energy Requirements
Particulates
Flue Gas Desulfurization
13B
13H,07A 13A
07B 21B
14A,05A 07D
10B
11G
13. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (This Report!
Unclassified
21. NO. OF PAGES
192
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
174
------- |