v>EPA
United States      Industrial Environmental Research  EPA-600/7-80-015d
Environmental Protection  Laboratory         January 1980
Agency        Research Triangle Park NC 27711
Experimental/  -
Engineering Support
for EPA's FBC Program:
Final Report
Volume IV.  Engineering
Studies

Interagency
Energy/Environment
R&D  Program Report

-------
                  RESEARCH REPORTING SERIES


Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional  grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:

    1.  Environmental Health Effects Research

    2.  Environmental Protection Technology

    3.  Ecological Research

    4.  Environmental Monitoring

    5.  Socioeconomic Environmental Studies

    6.  Scientific and Technical Assessment Reports (STAR)

    7.  Interagency  Energy-Environment Research and Development

    8.  "Special" Reports

    9.  Miscellaneous Reports

This report has been  assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded  under the  17-agency  Federal Energy/Environment  Research  and
Development Program. These studies relate to EPA's mission to protect the public
health  and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations  include analy-
ses of  the transport of energy-related pollutants and  their health and ecological
effects; assessments  of,  and development of, control technologies for energy
systems; and integrated assessments of a wide range of energy-related environ-
mental issues.
                       EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for  publication. Approval does not signify that the contents  necessarily reflect
the  views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.

This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.

-------
                                    EPA-600/7-80-015d

                                          January 1980
Experimental/Engineering Support
        for EPA's  FBC  Program:
               Final Report -
   Volume IV.  Engineering Studies
                        by

              J.R. Hamm, D.F. Ciliberti, R.W. Wolfe,
                 R.A. Newby, and D.L Keairns

           Westinghouse Research and Development Center
                   1310 Beulah Road
                Pittsburgh, Pennsylvania 15235
                  Contract No. 68-02-2132
                Program Element No. INE825
              EPA Project Officer: D. Bruce Henschel

            Industrial Environmental Research Laboratory
          Office of Environmental Engineering and Technology
               Research Triangle Park, NC 27711
                     Prepared for

            U.S. ENVIRONMENTAL PROTECTION AGENCY
              Office of Research and Development
                  Washington, DC 20460

-------
                                PREFACE

     The Westinghouse R&D Center Is carrying out a program to provide
experimental and engineering support for the development of fluidized-
bed combustion (FBC) systems under contract to the Industrial Environ-
mental Research Laboratory (IERL), U. S. Environmental Protection Agency
(EPA), at Research Triangle Park, NC.  The contract scope includes atmo-
spheric (AFBC) and pressurized (PFBC) 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., sorption kinetics, regeneration, attrition, mod-
eling), alternative sulfur sorbents, nitrogen oxide (NOX) emission, par-
ticulate emission and" control, trace element emission and control, spent
sorbent and ash disposal, and systems evaluation (e.g., impact of new
source performance standards (NSPS) on FBC system design and cost).

     This report contains the results of work defined and completed
under technical directives issued by the EPA project officer.  Work on
these tasks was performed from January 1976 to January 1979 and is docu-
mented in the following EPA contract reports:
     •  The present report, which presents the results of four
        technical directives on systems evaluation
     •  Report on an engineering assessment of intimate coal/
        sorbent mixtures for S02 control in FBC applications which
        is reported in our 1978 EPA report, EPA-600/7-78-0051
     •  Report on the "Effect of SC>2 Emission Requirements on
        Fluidized-Bed Combustion Systems:  Preliminary Technical/
        Economic Assessment," issued in August 1978 (EPA-600/7-78-
        163, NTIS PB 286 871/7ST).2
                                    iii

-------
Work on the other tasks performed under this contract is reported
in:

•  Experimental/Engineering Support for EPA's FBC Program:
   Final Report Volume 1, Sulfur Oxide Control, EPA-
   600/7-80-015a, January 1980
•  Experimental/Engineering Support for EPA's FBC Program:
   Final Report Volume II, Particulate, Nitrogen Oxide, and
   Trace Element Control, EPA-600/7-80-015b, January 1980
•  Experimental/Engineering Support for EPA's FBC Program:
   Final Report Volume III, Solid Residue Study, EPA-
   600/7-80-015c, January 1980
•  Alternatives to Calcium-Based S02 Sorbents for Fluidized-
   Bed Combustion:  Conceptual Evaluation,  EPA-600/7-78-005,
   January 1978
•  Regeneration of Calcium-Based S02 Sorbents for Fluidized-
   Bed Combustion:  Engineering Evaluation, EPA-600/7-78-039,
   NTIS PB 218-317,  March 1978
•  Disposal of Solid Residue from Fluidized-Bed Combustion:
   Engineering and Laboratory Studies, EPA-600/7-78-049 (NTIS
   PB 283-082), issued in March 1978,  which presented the
   results of work performed from January 1976 to January
   1977
•  Evaluation of Trace Element Release from Fluidized-Bed
   Combustion Systems, EPA-600/7-78-050, NTIS PB 281-321,
   March 1978.
                               iv

-------
                                ABSTRACT

     Engineering studies addressing several aspects of fluidized-bed
combustion (FBC) system design and performance are reported.   An evalua-
tion on the impact of SC>2 emission requirements on FBC system perfor-
mance and cost is reviewed (EPA-600/7-78-163).  Stringent SC>2 emission
requirements can be satisfied economically if proper selection of design
and operating parameters is made.  Another study on the feasibility of
feeding coal/sorbent mixtures to FBC units is also reviewed (EPA-
600/7-78-005).  Critical data gaps exist for this concept.  Moreover,
general economic feasibility would not be expected.  An alternative  S02
control concept for pressurized  fluidized-bed combustion  (PFBC),  that  is,
pressurized scrubbing of the products of combustion with water,  is evaluated.
The concept is not  economically  competitive  because of  the requirement for
recuperative heating and reduced plant efficiency.  A potential  reduc-
tion in solid waste is realized with the concept, but„the SC>2 control
efficiency may be limited.

     An evaluation of PFBC examining the technical and economic  trade-
offs between the level of particulate control achieved and the frequency
of gas-turbine blade replacement is described.  The evaluation incorpor-
ates models of PFBC particulate  carry-over,  particulate control  device
efficiency, and turbine erosion.  Also, an indirect air-cooled PFBC  con-
cept is evaluated and compared with other PFBC concepts.  The indirect
air-cooled concept provides significant particulate control advantages
over the adiabatic coiabustor PFBC concept, while resulting in about  a
4 percent lower plant efficiency and a 1 percent higher cost of
electricity.

-------
                          TABLE OF CONTENTS
                                                                  Page
1.  INTRODUCTION                                                    1
2.  CONCLUSIONS                                                     2
    Effect of Emission Requirements on FBC Systems                  2
    Intimate Coal/Sorbent Mixtures for S02 Control                  2
    Feasibility Evaluation of Fluidized-Bed Combustion Using
       Pressurized-Water Scrubbing                                  2
    Particulate Control Trade-off for PFBC Systems                  3
    Indirect Air-Cooled Fluidized-Bed Combustion Concept
       Systems Evaluation                                           3
3.  RECOMMENDATIONS                                                 4
4.  SULFUR OXIDE CONTROL                                            6
5.  INTIMATE COAL/SORBENT MIXTURES FOR S02 CONTROL IN
    FLUIDIZED-BED COMBUSTION                                       12
6.  FEASIBILITY EVALUATION OF FLUIDIZED-BED COMBUSTION
    USING PRESSURIZED-WATER SCRUBBING                              14
    Introduction                                                   14
    Concept and Process Options                                    14
    Selection of Base Case Design Concept                          18
    Plant Basis                                                    20
    Material and Energy Balances                                   21
    Equipment Design                                               24
    Capital Investment                                             26
    Cost of Electricity                                            31
    Environmental Comparison                                       31
    Conclusions                                                    35
7.  PARTICULATE CONTROL TRADE-OFF FOR PFBC SYSTEMS                 37
    Overview                                                       37
    Background                                                     38
    Estimation of Particle Loading/Size Distribution               39
    Estimating Gas Turbine Impact                                  48
    Results                                                        59
                                    vi

-------
                     TABLE OF CONTENTS (Continued)
                                                                   Page

 8.  INDIRECT AIR-COOLED PRESSURIZED FLUIDIZED-BED COMBUSTION
     CONCEPT SYSTEMS EVALUATION                                    63
     Introduction                                                  63
     Background                                                    63
     Results of Study                                              70
     Particulate Control/Gas Turbine Expander Erosion
     Considerations                                                79
     Environmental Considerations                                  84
     Conclusions                                                   87
 9.  REFERENCES                                                    89
APPENDIX
 A.  GRADE EFFICIENCIES FOR PARTICULATE REMOVAL EQUIPMENT          92
 B.  PARTICLE SIZE DISTRIBUTION AT PERTINENT STATIONS IN
     PARTICULATE REMOVAL SUBSYSTEM FOR ALTERNATIVE CASE I          96
                                    vii

-------
                            LIST OF FIGURES

                                                                   Page
 1.  Concept and Process Options                                    15
 2.  Material and Energy Balances                                   22
 3.  Diagram of Particulate Removal System                          39
 4.  Particulate Removal Equipment Arrangement                      39
 5.  Grade Efficiency of Primary Cyclone Separator                  42
 6.  Grade Efficiency Curve for Tan Jet                             42
 7.  Schematic of Tan Jet                                           43
 8.  Granular-Bed Filter Module                                     43
 9.  Various Filter Performance Assumed for Final Cleanup
     Stage                                                          45
10.  Particulate Concentration at Outlet of First-Stage
     Granular-Bed Filter                                            48
11.  Particulate Concentration at Outlet of Second-Stage
     Granular-Bed Filter                                            48
12.  Projected Outlet Size Distribution Based on Rexnord
     Commercial Unit (Dolomite Particles)                            49
13.  Projected Outlet Based on Rexnord Commercial Unit (Ash
     Particles)                                                     49
14.  Projected Outlet Based on Rexnord Commercial Unit (Char
     Particles)                                                     50
15.  Projected Granular-Bed Filter Outlet Size Distribution
     Based on Westinghouse Bench-Scale Experiments (Dolomite
     Particles)                                                     50
16.  Projected Granular-Bed Filter Outlet Based on Westinghouse
     Bench-Scale Experiments (Ash Particles)                         51
17.  Projected Granular-Bed Filter Outlet Based on Westinghouse
     Bench-Scale Experiments (Char Particles)                       51
18.  Projected Granular-Bed Filter Outlet Size Distribution
     Based on Conventional Fabric-Filter Unit Performance
     (Dolomite Particles)                                            52
19.  Projected Granular-Bed Filter Outlet Based on
     Conventional Fabric-Filter Unit Performance
     (Ash Particles)                                                 52
                                   viii

-------
                      LIST OF FIGURES (Continued)
20.  Projected Granular-Bed Filter Outlet Based on
     Conventional Fabric-Filter Unit Performance
     (Char Particles)                                               53

21.  Blade Leading Edge Erosion Rates                               53

22.  Projected Turbine Life for a Particulate Removal
     System with Two Stages of Granular-Bed Filters                 57

23.  Projected Turbine Life for a Particulate Removal
     System with One Stage of Granular-Bed Filters                  57

24.  Projected Turbine Life for a Particulate Removal
     System with One Stage of Granular-Bed Filters                  58

25.  Cost of Electricity Increments due to Turbine Blade
     Replacement Using Granular-Bed Filters (performance
     based on granular-bed filter efficiency)                       61

26.  Cost of Electricity Increments due to Turbine Blade
     Replacement Using Granular-Bed Filters (performance
     based on fabric-filter efficiency)                             61

27.  Combined-Cycle  System Utilizing Fluidized-Bed Combustion
     with Indirect Heating of Part of the Working Fluid
     (no CBC)                                                       66

28.  Combined-Cycle  System Utilizing Fluidized-Bed Combustion
     with Indirect Heating of Part of the Working Fluid
     (Alternative Case I - CBC)                                     68

29.  Combined-Cycle  System Utilizing Fluidized-Bed Combustion
     with Indirect Heating of Part of the Working Fluid
     (Alternative Case II - no CBC)                                 70

30.  Plot Plan of Single Gas Turbine Module for Base Case           72

31.  Plot Plan of Single Gas Turbine Module for Alternative
     Case I                                                         72

32.  Plot Plan of Single Gas Turbine Module for Alternative
     Case II                                                        73

33.  Summary of Particulate Loading and Size Distribution  for
     Base Case                                                      82

34.  Summary of Particulate Loading and Size Distribution  for
     Alternative  Case  I                                             83
35.  Summary of Particulate Loading and Size Distribution  for
     Alternative  Case  II                                            85
                                      ix

-------
                             LIST OF TABLES
 1.  Material and Energy Balances                                   23
 2.  Gas-Cleaning Auxiliaries                                       24
 3.  Process Equipment                                              25
 4.  Capital Investment for Pressurized Water Scrubbing             27
 5.  PFBC Boiler Plant Equipment Costs                              28
 6.  PFBC Power Plant Cost Breakdown                                29
 7.  Comparison of Cost of Electricity                              32
 8.  Environmental Comparison                                       34
 9.  Projected Particulate Concentration Levels                     46
10.  Particulate Emission Levels                                    47
11.  Thermal History of Particles Entering a 0.05-in. Thick
     Boundary Layer                                                 55
12.  Summary of Plant Performance                                   71
13.  Summary of Plant Design Configurations                         74
14.  Capital Cost Estimate for Base Case                            75
15.  Capital Cost Estimate for Alternative Case I                   76
16.  Capital Cost Estimate for Alternative Case II                  77
17.  Specific Cost Comparison                                       78
18.  Cost of Electricity Summary                                    78
19.  Expander Inlet Particle Loading                                81
20.  Environmental Comparison of the Configuration                  86

-------
                              NOMENCLATURE

AFBC - atmospheric-pressure fluidized-bed combustion
Ca/S - calcium-to-sulfur ratio
 CBC - carbon burnup cell
  CF - capacity factor
 COE - cost of electricity
 DOE - Department of Energy
EGAS - Energy Conversion Alternatives Study
 EPA - Environmental Protection Agency
ERDA - Energy Research and Development Agency
 FBC - fluidized-bed combustion
 GBF - granular-bed filter
 HHY -
HRSG - heat recovery steam generator
IERL - Industrial Environmental Research Laboratory
 ISO - International Standards Organization
 NOX - nitrogen oxide
NSPS - New Source Performance Standard
 O&M - operating and maintenance
PFBC - pressurized fluidized-bed combustion
 SOX - sulfur oxide
 TDC - total direct costs
 TGA - therraogravimetrie analysis
                                     xi

-------
                               ACKNOWLEDGMENT

          We want to express our high regard for and acknowledge the
contribution of Mr. D. B. Henschel who served as the EPA project officer.
Mr. P. P. Turner and Mr. R. P. Hangebrauck, Industrial Environmental
Research Laboratory, EPA, are acknowledged for their continuing contri-
butions through discussions and support of the program.
          We gratefully acknowledge the contributions of the following
Westinghouse personnel:  K. D. Weeks for his assistance in evaluating
the air-cooled PFBC cycle and Dr.  D. H. Archer,  Manager, Chemical Engineering
Research, for his program consultation and continued support.
                                   xii

-------
                            1.   INTRODUCTION

     This volume documents five systems evaluation tasks that were per-
formed during the contract as technical directives from the EPA proiect
officer.  One study, the effect of sulfur dioxide (SO ) emission require-
ments on fluidized-bed combustion (FBC) systems, was issued as a sepa-
rate report in 1978 (EPA-600/7-78-163).  Another study, an engineering
assessment of intimate coal/sorbent mixtures for S02 control by FBC
applications, was also issued in 1978 (EPA-600/7-78-005).   Results from
three studies, the feasibility evaluation of pressurized-water scrubbing
for S02 emission control with PFBC (completed in 1977), a particulate
control/turbine life trade-off study for PFBC systems (completed in
1977), and an evaluation of indirect air-cooled PFBC concepts (completed
in 1978) were not previously issued as separate contract reports.

     A summary of the sulfur oxide (SOX) control report is presented in
Section 4, the intimate coal/sorbent mixture study is summarized in Sec-
tion 5, and results of the remaining three studies are reported in Sec-
tions 6 through 8.

-------
                            2.  CONCLUSIONS

     The primary conclusions from the five studies follow.

EFFECT OF EMISSION REQUIREMENTS ON FBC SYSTEMS
     •  AFBC and PFBC systems can economically meet the New Source
        Performance Standards (NSPS) for utility power plants:
        90 percent sulfur removal, 12.9 ng/J (0.03 Ib/MBtu) par-
        ticulate emission, and 258 ng/J (0.6 Ib/MBtu) nitrogen
        oxide (NOX) emission.
     •  The selection of FBC design and operating parameters to
        minimize the sorbent feed requirement is critical for
        realizing economical systems.

INTIMATE COAL/SORBENT MIXTURES FOR S02 CONTROL
     •  Sufficient technical data on the performance of intimate
        coal/sorbent mixtures, such as pellets consisting of coal
        and limestone powders, do not exist to project FBC perfor-
        mance reliably.
     •  The performance of intimate coal/sorbent mixtures in FBC
        is unlikely to be economically competitive with conven-
        tional FBC concepts.

FEASIBILITY EVALUATION OF FLUIDIZED-BED COMBUSTION USING PRESSURIZED-
WATER SCRUBBING

     •  A PFBC plant using the cold pressurized-water scrubbing
        concept for S02 control is not economically competitive
        with calcium-based PFBC or conventional steam power plants
        with stack-gas cleaning.

-------
PARTICULATE CONTROL TRADE-OFF FOR PFBC SYSTEMS
     •  A methodology has been developed and is available for
        evaluating trade-offs between fluid-bed corabustor,  gas-
        cleaning, and turbine design and operating parameters.
INDIRECT AIR-COOLED FLUIDIZED-BED COMBUSTION CONCEPT SYSTEMS EVALUATION

     •  Indirect air-cooled PFBC concepts will have
        - Lower performance (higher heat rates) than PFBC boiler
          concepts
        - Similar performance with PFBC adiabatic combustor
          concepts.
     •  The cost of electricity for indirect air-cooled PFBC con-
        cepts is essentially the same as for PFBC adiabatic cora-
        bustor concepts.
     •  Indirect air-cooled AFBC concepts will have lower perfor-
        mance (higher heat rates) than indirect air-cooled PFBC
        concepts.  The AFBC concept provides for turbine reliabil-
        ity using a clean gas.
     •  Environmental emissions standards can be met with all
        indirect air-cooled FBC concepts.

-------
                          3.   RECOMMENDATIONS

     The primary recommendations from work carried out under the techni-
cal directives are presented here,  followed by recommendations for
extended systems studies to evaluate and guide the development of eco-
nomical FBC systems operating within environmental constraints.

     •  Investigate the ability of  AFBC and PFBC processes to
        achieve more stringent emission standards,  with specific
        focus on the relationship of performance to combustor
        design and operating parameters.
     •  Evaluate and develop  advanced FBC sulfur removal concepts
        (e.g., sorbent pretreatment, sorbent regeneration,  alter-
        native regenerable sorbents).  This will be particularly
        important when solids procurement or disposal represents a
        constraint.
     •  Develop understanding of NOX minimization alternatives,
        perform system evaluation to select economic options,  and
        demonstrate capability.
     •  Carry out experimental test programs to obtain performance
        data on particulate control equipment applicable to AFBC
        and FFBC systems.   The primary need is an understanding  of
        high-temperature,  high-pressure particulate control equip-
        ment performance - e.g.,  cyclones,  granular-bed filters,
        fabric filters,  and other advanced filter and cyclonic
        concepts.   Data will  be important for process constraints
        (e.g., turbine erosion/deposition)  and environmental con-
        straints (leading  and potential fine particle emission
        criteria).

-------
•  Carry out experimental test programs to obtain understand-
   ing of turbine tolerance - e.g.,  erosiveness of particu-
   late, response characteristics of turbine materials,  effect
   of turbine design and operating parameters.
•  Extend and apply methodology for evaluating  fluid-bed
   combustor/gas-cleaning/turbine design and operating param-
   eter trade-offs to identify optimal fluidized-bed combus-
   tor systems for given application and environmental
   requirements.
Additional FBC systems studies are recommended  to

•  Assess the effects of potential NSPS on industrial FBC
   systems to aid EPA in developing standards.
•  Project and evaluate the environmental performance of FBC
   system designs currently proposed by Department of Energy
   (DOE) contractors or commercial vendors to understand the
   status of these designs.
•  Evaluate the impact of variable coal sulfur content and
   variable sorbent properties on the control of S02 emis-
   sions from FBC systems in order to quantify the effect of
   variable properties on sorbent consumption and system
   economics.
•  Evaluate FBC unit start-up and turndown techniques with
   respect to environmental performance in order to identify
   superior techniques and performance sensitivity.
•  Assess the technical/environmental performance of alterna-
   tive FBC operating regimes (e.g., turbulent fluidization,
   circulating fluidized bed, fast  fluidization, multtsolids
   systems) to understand their potential and limitations.

-------
                        4.  SULFUR OXIDE CONTROL

     Westinghouse evaluated the impact of up to 90 percent sulfur
removal on the capital and energy costs of conventional dense-phase,
fluid-bed, AFBC and PFBC power plants in a previous report as part of
this contract.  A brief summary of that report is presented here. The
full study is presented in EPA-600/7-78-163.
     Two levels of emissions standards were considered:

     •  The current (1978) EPA NSPS for large coal-fired boilers:
        SOX, 516 ng/J (1.2 Ib S02 MBtu); particulates, 43.0 ng/J
        (0.1 Ib/MBtu); and NOX, 301 ng/J (0.7 Ib N02 MBtu)
     •  A set of more stringent degrees of control:  SOX> 90 per-
        cent removal of coal sulfur content;  particulates,
        12.9 ng/J (0.03 Ib/MBtu); and NOX, 258 ng/J
        (0.6 Ib/MBtu).

These levels were selected for the study because they represent one set
of values considered during the planned revision of the NSPS for utility
boilers.

     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  Ca/S 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 SOX emission standard and for 90 percent sulfur

-------
removal FBC is potentially cost competitive with conventional coal-fired
power plants using lime-slurry scrubbing.   The competitiveness of FBC
depends upon proper selection of fluid-bed combustor operating condi-
tions—i.e., sufficiently long gas residence in the bed (sufficiently
low gas velocity and sufficiently deep beds) and sufficiently small sor-
bent particle size.  This selection of variables will result in a larger
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 one should emphasize maximization
of fluid-bed combustor performance rather than minimization of the com-
bustor cost through compact design.  The corabustor 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 depen-
dent on the combustor performance.

     The Ca/S 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 relative to FBC power
plant cost and performance for high-sulfur eastern coals (2 to 5 wt %
sulfur).  The Ca/S ratio has a dramatic impact on the FBC power plant
thermal efficiency, capital investment, and cost of electricity.  An
increased Ca/S ratio, if required  for lower SOX emissions, results in
increased auxiliary power consumption for solids handling and signifi-
cant 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 by using a kinetic model for SOX capture
Westinghouse had developed.  This model—using rate constants measured
in laboratory thermogravimetric analysis (TGA) equipment and confirmed
where possible by 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 corabustor operating/
design conditions.

     While the cost and performance of several subsystems in the FBC
power plants are uncertain (for example,  solids feeding and particulate
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/ per-
formance uncertainties will probably be small relative to the uncertain-
ties 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 an
increased Ca/S 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 NOX for FBC
power plants indicate that the more stringent emission requirements con-
sidered here of 12.9 ng/J (0.03 Ib/MBtu)  and 285 ng/J (0.6 Ib/MBtu),
respectively, are economically feasible and of lower cost impact than
the more stringent SOX requirement.   Conventional fabric-filter (bag-
house) techniques should permit achievement of this requirement, depend-
ing on particle size and future environmental standards.  We expect PFBC
plants 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 lower than
258 ng/J (0.6 Ib/MBtu) 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
combustor design and operating conditions are selected.   Development
programs should focus on developing large-scale information on the rela-
tionship between corabustor operating conditions and FBC plant emissions,
while engineering evaluation should assess FBC pollution control
capabilities.

     The detailed conclusions and recommendations developed in this
report are as follows:

     •  On the basis of available information, the more stringent
        emission requirements considered in this study (SOX, 90%
        sulfur removal; particulates, 12.9 ng/J (0.03 Ib/MBtu);
        NOX, 258 ng/J (0.6 Ib N02/MBtu) should be economically
        achievable  for both AFBC and PFBC  power plants.
     •  The  proper  selection of fluid-bed  corabustor design  and
        operating conditions is critical to the economical  reali-
        zation of these environmental goals.  The gas residence  in
        the  bed, in particular  (as  determined by gas velocity and
        bed  height),  should be  sufficiently long, and sorbent par-
        ticle size  should  be sufficiently  small.  In this assess-
        ment residence of  0.67  to 2.0 s  (gas velocities of  1.5 to
        1.8  m/s) and  particle sizes averaging 500  um appeared to
        offer effective SOX removal performance, although  these
        conditions  are not necessarily optimal.
     •  The  high  level of  SOX emission control considered has a
        greater  Impact on  the FBC power  plant  energy cost  than do

-------
    the  revised particulate and NOX standards considered.  The
    most critical process parameter with respect to FBC power
    plant cost and performance Is the Ca/S ratio.
•   The  fluid-bed combustor cost does not depend strongly on
    changes In design and operating conditions.  The fluid-bed
    corabustor should be designed to minimize the cost of plant
    energy rather than cost of the combustor.  For example,
    low - rather than high - fluldlzatlon velocities will
    probably result in lower FBC power plant energy cost.
•   Particulate control to levels as low as 12.9 ng/J
    (0.03 Ib/MBtu) should be economically achievable for AFBC
    using commercially available techniques.  Baghouses seem
    most suitable for this duty.  No testing of any type of
    final-stage particle control device on an AFBC unit, how-
    ever, has yet been conducted.
•   Particulate control to levels below 12.9 ng/J (0.03 lb/
   MBtu) 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 258 ng/J (0.6 lb N02/MBtu) requirement
   considered in this evaluation.   No direct control tech-
   niques for NOX have been clearly demonstrated on
   fluidized-bed combustors to date,  although several options
   are under study.
•  The greatest FBC power plant uncertainties presently
   involve  reliability questions - e.g.,  solids feeding,  par-
   ticulate control (especially for PFBC),  material erosion/
   corrosion/deposition,  and process control.   The impact of
   emission standards averaging time basis  and system reli-
   ability  has not been evaluated.
                               10

-------
•  AFBC and PFBC development programs should focus on more
   stringent emission standards and their relation to combus-
   tor 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.
                               11

-------
           5.  INTIMATE COAL/SORBENT MIXTURES FOR SC>2 CONTROL
                      IN FLUIDIZED-BED COMBUSTION
     Westinghouse performed a conceptual evaluation of the use of inti-
mate coal/sorbent mixtures (e.g. , pellets consisting of powdered coal
and limestone) as part of this contract.  This evaluation has been
reported previously, in EPA-600/7-78-005, and a brief summary of that
report is  included here.
     The study was carried out to investigate the technical and environ-
mental feasibility and economic potential of "intimate coal/sorbent mix-
tures" when used in an FBC system for power generation.   Various classes
of intimate coal/sorbent mixtures were first qualitatively screened for
feasibility on the basis of their probable performance assessment.
Intimate coal/sorbent mixtures selected as potentially feasible in the
initial screening were then subjected to an engineering assessment of
technical and environmental performance.  Areas such as SOX and NOX con-
trol, trace metal and particulate control, solid waste and plant effi-
ciency, and design factors for the fluidized-bed combustor were consid-
ered.  Because no actual performance or kinetic data exist for the inti-
mate coal/sorbent mixtures,  only potential performance could be
addressed and problem areas identified.

     Economic potential was examined by using optimistic performance
assumptions for the intimate coal/sorbent mixture.   Process alternatives
for the preparation of the mixtures were identified and  cost projections
for the preparation systems  were generated.
                                    12

-------
The major conclusions reached are as follows:
•  The only technically feasible intimate coal/sorbent mix-
   ture that could be identified for the current fluidized-
   bed combustion design concept is the consolidated coal/
   sorbent particle concept.
•  Attrition of the consolidated particle is the most criti-
   cal factor influencing the performance and feasibility of
   the concept.  Modifications to the corabustor design would
   probably be required in order to apply the consolidated
   particle concept.
•  The performance (technical and environmental) cannot be
   estimated without initiating a test program.  The overall
   technical and environmental performance of the consoli-
   dated particle concept could conceivably by worse than or
   better than the conventional fluid-bed combustor, but it
   is highly unlikely that any significant improvement in
   performance is to be realized.
•  Except under very extreme conditions, the consolidated
   particle concept will not be economically competitive with
   conventional FBC concepts.
•  Washing the pulverized coal during  consolidated  particle
   preparation could reduce  trace elements, ash, sulfur, and
   the sorbent requirement.  The economics of  this  option
   have  not been  investigated.
•  The most attractive  consolidated coal/sorbent particle
   from  the standpoint  of technical and  environmental  impact
   would utilize  a  binder to maintain  the  coal-ash  and sor-
   bent  particles  in discrete, consolidated particles  follow-
   ing combustion.  A  binder that will effect  this  behavior
   has not  been  identified.
                                13

-------
         6.  FEASIBILITY EVALUATION OF FLUIDIZED-BED COMBUSTION
                   USING PRESSURIZED-WATER SCRUBBING
INTRODUCTION

     The national and private development efforts for fluidtzed-bed com-
bustion are based on SOX absorption by calcium-baaed sorbents (limestone
or dolomite) at high temperatures.  Both AFBC and PFBC concepts are
being pursued with either regenerative or once-through sorbent opera-
tion.  We believe that once-through sorbent operation represents only
the  first-generation of FBC systems, but even with sorbent regeneration,
if it is eventually realized, FBC will produce significant quantities of
dry, granular, sulfated limestone or dolomite that must be disposed of
or utilized in an environmentally satisfactory manner.
     We have evaluated an alternative FBC concept that may be applicable
to PFBC.  This concept is a cold gas-cleaning scheme that uses water to
scrub the pressurized combustion products without additives for control-
ling S02-  The potential advantage of this alternative is the reduction
of solid-waste emissions.
     A feasibility study has been conducted to better define the concept
and to estimate its cost and performance.  Approximate material and
energy balances,  conceptual equipment designs, and process economic
estimates were performed in order to determine concept problem areas and
process economic and environmental feasibility.
CONCEPT AND PROCESS OPTIONS

     The basic PFBC concept with pressurized water scrubbing is shown in
Figure 1 with various process options indicated.   An understanding of
these options Is Important if one is to select the best process to be
evaluated.
                                    14

-------
   To
Gas Turbine
                Absorber
                Gas (Temperature)
         Combustion
         Products
      Part icu late
      Control
  (Type,  Number
   of  Stages)
    Coal
                                                                                     Dwg. 1687B33
                                       Tail Gas         «
                                       (Recycle, Exhaust)
 Pressurized
 Combust or
 (Pressure,
Temperature,
Ash or Inert
Bed)
   Particulate
   Control
(Type, Number
 of Stages)
        Air
        (excess air rate)
                                                                             Sulfur or*
                                                                             Acid
                                                                                  Sulfur or
                                                                                  Sulfuric Acid
                                                                                  Plant
                                                                                         Stripper
                                                                                         Gas
                                                         Makeup
                                                         Water
Pressurized
   Absorber

    (Type
   internal
    heat
  transfer,
   temper-
   ature)
                                                                    Pressure  ^k
                                                                    Reduction-*]
                                                                                                        Stripping Gas
                                                                                                        (Air,  Steam,
                                                                                                         Stack Gas)
                                                                                  Filter Cake
                                  Figure  1 -  Concept  and Process Options

-------
     Coal is combusted with air in the pressurized fluidized-bed corabus-
tor.  The bed consists of either coal ash or an inert bed material such
as alumina.  The bed temperature (760-1040°C) and pressure (620-
1600 kPa) are important process variables relating to the corabustor per-
formance and the cycle efficiency.  The excess air rate, also, is a
critical process variable since it defines the quantity of gas that must
be handled by the pressurized water scrubbing system.  The excess air
may range from 10 to 100 percent for fluidized-bed boilers and may be
about 300 percent for an adiabatic fluidized-bed corabustor (no heat
transfer surface in the bed).  The corabustor fluidization velocity and
heat transfer rates are assumed to be very similar to those of the
calcium-based corabustor,  as are the attrition and elutriation rates,
although they could be lower with proper selection of the inert bed
material.

     High-temperature particulate removal equipment (cyclones, filters)
could be situated so as to operate before the combustion products are
cooled,  and/or low-temperature removal equipment (filters, scrubbers,
electrostatic precipitators) could be placed to operate after the cool-
ing step.  Captured bed material (coal ash, inert material) could be
recycled to the corabustor or removed from the system.
     Cooling and reheating the combustion products would be a critical
step.   A recuperator or a convection-type steam generator followed by a
recuperator would cool the combustion products to a level suitable for
the absorber (<150°C) and would reheat the absorber gas to a temperature
resulting in an economical,  combined-power cycle.   Various types of
recuperators could be used:   a shell-and-tube heat exchanger constructed
from high-alloy tube materials (bare or finned) and designed for high
thermal expansion conditions,  the more conventional,  stove-type or
packed-bed-type heat exchanger requiring cyclic heating and cooling of
parallel vessels containing refractory material (packed-bed or checker
                                    16

-------
structure) with gas flow controlled by high-temperature valves,  or a
circulating pebble-bed heat exchanger requiring continuous circulation
of a refractory heat transport medium between parallel vessels.

     Conventional countercurrent absorber and stripper towers would be
used to remove the SOX from the combustion products at pressure and to
generate at atmospheric pressure an SC>2 gas suitable for elemental sul-
fur or sulfuric acid (l^SO^) recovery.  Packed columns or a plate-type
design could be applied with proper construction for the highly corro-
sive environment.  An internal heating or cooling surface could be
placed in the columns to control the column temperatures.  Mist elimina-
tors might be required in order to protect downstream equipment from
corrosion.

     The stripping gas could be either air, steam, or stack gas.  Each
would have advantages in terms of power requirements, oxygen content,
and capital investment.

     Elemental sulfur or H2S04 could be recovered from the stripper gas.
The composition of the stripper gas is critical to this  step.  A commer-
cial sulfur recovery process such as Allied Chemical's could be applied.
The Allied Chemical process requires  the use of a clean  fuel, such as
methane (City), for S(>2 reduction.  Alternatively, a  developmental pro-
cess such as  the Foster Wheeler RESOX Process, which uses coal as a
reductant, could be applied.  The tail gas from the  sulfur plant could
be exhausted  or  recycled to the absorber.

     The  circulating solution system  requires  heat exchange, cooling and
heating with  conventional  devices in  order to  control  the absorber and
stripper  temperatures.  In addition to a pump  to  circulate the solution
a means of pressure reduction such as a pressure  reduction valve or a
power  recovery turbine would be required,  since the  absorber is operated
at elevated pressure and the stripper is at  low pressure.
                                     17

-------
     Partlculate material trapped In the absorber must be removed In
order to maintain the absorber performance.  Various commercial devices
that will permit the filtration of a side stream of the circulating
solution are available.
     Makeup water would be fed to the system to account for filter cake
losses and evaporation losses.

SELECTION OF BASE CASE DESIGN CONCEPT

     A selection of a base design concept for the PFBC with pressurized
water scrubbing from the options presented in the previous section has
been made.   We have judged,  on the basis of preliminary considerations,
that the selected base-concept would probably be the most successful of
all of the concepts presented.

     A previous cycle study  for a PFBC concept that used low-temperature
venture scrubbing as an alternative to high-temperature particulate con-
trol was applied to reach the following conclusions:^

     •  A cornbustor temperature of about 927°C (1700°F), resulting
        in a combustion product temperature of about 871°C (1600°F)
        to the recuperator,  arid a corabustor pressure of about
        1034 kPa (150 psia)  are suitable combustor operating con-
        ditions for this concept.
     •  A recuperator effectiveness of at least 0.86 (resulting in
        a turbine inlet temperature of about 760°C (1400°F)) and an
        excess air rate of no more than 20 percent should be used
        for economic feasibility.   These conditions will yield a
        plant heat rate of about 10,000 kJ/kWh (9,500 Btu/kWh),
        including the boiler efficiency increase due to the elimi-
        nation of sorbent calcination energy losses and energy
        losses in the gas cleaning system.   Using a steam generator
        prior to using the recuperator or using the high excess  air
        fluidized-bed boiler or adiabatic combustor will not be
        economically feasible with this concept.
                                    18

-------
     Other  equipment  and  process  selections are as  follows:
     •  An  inert  ceramic  bed  (alumina)  in  the  corabustor because it
        should result in  superior combustor performance in  terms of
        particle  elutriation  and  potential ash fusion
     •  Two stages  of high-temperature  particulate  removal  equip-
        ment (cyclones) located directly after the  corabustor.  The
        first stage would recycle coarse material (alumina  and car-
        bon) to the combustor.  The second stage  would remove fines
        from the  corabustor products (coal  ash  and alumina)  in order
        to  protect  the recuperator from erosion and deposition.
        These fines would be  removed from  the  system.  No low-
        temperature particulate control equipment would be  used
        before water  scrubbing, and we  assumed that the absorber
        and stripper  could tolerate a relatively  high  particulate
        content.

     Because the recuperator  is a critical process  component, both the
shell-and-tube recuperator and the cyclic  stove  recuperator have  been
evaluated.   The packed bed concepts were  not  considered  because  of the
possibility of particle elutxiation and plugging.  An  effectiveness  of
0.90 was selected with a turbine  inlet temperature of  788°C (1450°F).

     Valve-tray columns were  selected for  the absorber and stripper to
improve performance under conditions of high  particulate content  and to
permit simplified periodic maintenance of  the columns.  Plastic  lining
was specified to protect against corrosion.   Preliminary calculations
indicate that internal heating or cooling would not be required  in the
columns.

     Stack  gas would be used for stripping rather  than steam or air.
Steam consumes a large quantity of power and  results  in a large water
loss.  Stack gas contains a lower oxygen content than does air and
                                    19

-------
results In less reductant consumption for sulfur recovery.  Preliminary
cleaning of the stack gas would be required in order to protect the
blower.
     Elemental sulfur would be recovered by using the Allied Chemical
Process with City as the reductant.  Developing technologies such as the
Foster-Wheeler RESOX process have not yet been demonstrated and may not
be very efficient or capable of producing commercial-grade sulfur.  The
sulfur recovery process tail gas would be exhausted on the basis of an
assumed 90 percent sulfur recovery efficiency.  On the basis of economic
projections, an S02 content of at least 4 mole % would be required in
the stripper off-gas.

     A hydraulic turbine would be used for power recovery from the cir-
culating scrubber solution.   The solution would be cooled by a cooling-
water exchange and heated by clean fuel (heating oil) combustion.   Low-
grade steam was considered for heating the solution, but the steam
requirements exceeded the availability in the plant.

     A typical design philosophy for large fluidized-bed combustion
plants calls for modular design with four parallel combustors in a
600 MWe power plant.   This philosophy has been followed in this design
evaluation that specifies parallel gas-cleaning trains.
PLANT BASIS

     The following plant basis was selected for the evaluation:

     •  594 MWe power plant  net output (635 MWe conventional PFBC)
        power plant net output
     •  Four boiler modules
     •  17.5 percent excess  air in primary combustors
     •  4 wt % sulfur coal with 10 wt % ash and a heating value of
        30 x 106 J/kg (13,000 Btu/lb)
     •  S02 emission controlled to 0.5 kg S02/GJ (1.2 Ib S02/
            Btu),  equivalent to about 81 percent sulfur removal
                                    20

-------
     •  Single sulfur recovery plant with 90 percent sulfur
        recovery efficiency
     •  Absorber 89.5 percent efficient in removing SOX.

This basis provides direct comparison with previous PFBC designs using
calcium-based, high-temperature gas cleaning.
MATERIAL AND ENERGY BALANCES
     Material and energy balances were performed for the base case
described and are summarized in Figure 2 and Table 1.
     An iterative approach was used for the absorber and stripper system
material and energy balances in order to provide reasonably near optimum
designs for these columns.  The minimum absorber operating temperature
possible [based on normal cooling water temperatures of 27-30°C (80-
85°F), and considering the absorber inlet gas temperature of 121°C] is
about 38°C (100°F).  This temperature was selected for the design in
order to yield the most efficient SOX absorption.  An operating tempera-
ture of 66°C  (150°F) was  selected for the stripper on the basis of max-
imizing the S(>2 concentration  in the stripper gas and minimizing evapo-
rative water  losses.  The maximum SOX content of the stripper gas for
this process  operated with realistic temperature conditions is about
6  mole %.  A  value of 5 mole % was  selected for the design in order to
give reasonable column dimensions.

     Energy balances around  the absorber and stripper indicate  that the
solution circulation rate would be  so great (361,725 kg-moles/hr) that
heat of absorption effects,  heat of evaporation effects, and sensible
heats  of entering gas streams  would be  negligible  and the  columns would
operate isothermally.

     A small  amount  of C02 would also  be  absorbed  from  the combustion
products  and  released into  the stripper gas (about 180  kg-moles/hr).
The particulate content  of  the circulating  solution was assumed to build
                                     21

-------
                                                                                   Dwg. 1 68783'*
         To Turbine
to
N5
        Coal
                      ©
           Recuperator

           AAA/Vr-n
                  I
                       Particulate
                       Control
Fines
                  8
                    Air
 Sulfur
Recovery
 Process
                                     Figure  2  - Material and Energy Balances

-------
          Table  1

MATERIAL AND ENERGY  BALANCES
   (594 MWe Power Plant)
Dwg .1712 SkO

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
IS
19
20
Stream
Combustion
products
Combustion
products
Absorber
gas
Absorber
gas
Stripper
stack gas
Stripper
air
Stripper
gas
Sulfur
Tail gas
Makeup water
Filter cake
Solution
Solution
Solution
Solution
Solution
Solution
Solution
Filtrate
Filter
solution
Flow Rate,
kg-moles/hr
89,255
89,255
88,847
88,847
3.454
3,454
4,543
204
-5,260
1,200
2, 722 kg/ hr
361,725
361, 725
361,725
361,725
361,725
361.725
333,773
27,952
27, 952
Composition, mole %
15% C02, 0. 284% S02
15%C02, 0.284%S02
15%C02. 0.030%S02. 0.5%H20
15%C02, 0.030%S02
4%02,C02.H20,N2
4%02> C02. H20, N2
5%S02, 16%C02, 15%H20, 3%02,
61% N2
Commercial grade sulfur
0.43%S02
H20
95 wt%particulate, 5 wt% acid solution
0. 004% S02 in water ( 0. 5 wt% solids)
0. 004% S02 in water ( 0. 5 wt% solids)
0. 004% S02 in water ( 0. 5 wt% solids)
0. 004% S02 in water ( 0. 5 wt% solids)
0.067%S02in water (0.5 wt% solids)
0. 067% S02 in water ( 0. 5 wt% solids)
0. 067% S02 in water ( 0. 5 wt% solids)
0.067%S02(Owt%solids)
0. 06 7 S02 in water ( 0. 5 wt% solids)
Temperature,
871(1600)
121(250)
38(100)
788(1450)
121 (250)
177 ( 350)
66(150)
38(100)
38(100)
27 (SO)
38(100)
35(95)
43(110)
66(150)
66(150)
68(155)
60 (140)
38(100)
38(100)
38(100)
Pressure,
kPa (psia)
1034(150)
1014 ( 147)
965(140)
952(138)
103(15)
172(25)
110(16)
103(15)
103(15)
965(140)
103(15)
965(140)
1000(145)
1296(188)
172 ( 25)
110(16)
117(17)
310(45)
310(45)
1034 ( 150)
               23

-------
up to about 0.5 wt % of partlculate material.  This would require that
8 percent of the circulating solution be continuously filtered in order
to maintain a steady particulate level in the solution.   The effects of
particulate levels of this order on the absorber, stripper,  hydraulic
turbine, pump, and heat exchangers requires further investigation.
     For the 4 wt % sulfur coal and the 90% sulfur recovery efficiency
assumed for the sulfur plant, the overall process sulfur removal
efficiency would be 80.5%, if we assume an absorber efficiency of
89.5 percent.  Coals with a higher sulfur content would require higher
sulfur removal efficiencies and greater solution circulations rates.
Coals with less sulfur would result in less S02 in the stripper gas.
     The auxiliaries (power,  fuel,  water) required for the pressurized
water scrubbing process are listed in Table 2.  Methane was used as the
reductant in the sulfur recovery process and fuel oil for the circulat-
ing solution heater.

                                Table 2

                        GAS-CLEANING AUXILIARIES
     Makeup Water                       365 £/min (96 gal/rain)
     Cooling Water                   72,000 A/min (19,000 gal/min)
     Methane                          5,200 m3/hr (183,200 scf/hr)
     Fuel Oil                         5,000 kg/hr (11,000 Ib/hr)
     Power                            3,870 kW
          Pump                        2,190 kW
          Blower                      2,520 kW
          Turbine                      -840 kW

EQUIPMENT DESIGN

     The major equipment items for the gas-cleaning system are described
in Table 3.   Following the philosophy of maximum use of  shop fabrica-
tion,  we used a modular design which resulted in a single recuperator

-------
                                                                           Table 3
                                                                                                              Dwg. 1712Bl*l
tsJ
Equipment
Absorber
Stripper
Recuperator
Number per
Combustor
3
3
1
Description
Plastic-lined vertical shell column, 17 m (12ft) diameter, 32m (105ft) tall, 40 valve trays,
0. 76 m ( 1 5 ft) tray spacing, 1400 kPa (200 psia) design pressure, with mist eliminators
Plastic-lined vertical shell column, 4m (12ft) diameter. 37 m (120ft) tall, 76 valve trays,
0. 46 m ( 1. 5 ft) tray spacing. 345 kPa (50 psia) design pressure, with mist eliminators
Shell-and-tube horizontal heat exchanger, floating-head design, 4 m (13ft) diameter, 23m (75ft) long;
         Alternative Recuperator
 Circulating Solution              3
 Filter
 Circulating Solution              3
 Turbine
 Circulating Solution              3
 Pump
 Circulating Solution Heat          3
 Exchanger

Circulating Solution              3
Cooler

Circulation  Solution              3
Heater

Stripper Stack-Gas Blower         3
 Stack-Gas Recycle System          1
 8,000. 2.5cm din)  OD finned tubes with 0.089cm (0.035 in) wall thickness,  15m (50ft) tube length,
 finned-tube-area-to-base-tube-area ratio =10, Inconel or Incoloy construction,  1400 kPa (200 psia)  .
 design pressure

 Two parallel vessels with internal refractory checker structure connected by high-temperature valves
 (4 per vessel); each vessel 6.7 m (22ft) in diameter and 46 m (150 ft) long; each vessel containing

 2.9 x 10 kg (6.5 x 10  Ib) of refractory checker structure;  1-hr cycle time assumed

 Continuous-pressure drum filter, handles 42.000 kg (93.000 Ib) of solution/hr, 29 m  (310 ft )
 filter area,  1000 kPa (150 psia)  inlet pressure, 690 kPa (100 psi) pressure drop

 Hydraulic turbine,  handles 7,600 4/min (2.000gal/min). recovers 67 kW (90 HP)


Centrifugal pump,  handles 7,6001/min (2.000gal/min). consumes  182 kW (244HP)


 Shell-and-tube heat exchanger,  heat duty of 1.4 x 10 W (4.78x10  Btu/hr),  tube surface of

1800m2 (19,000 ft2)

Shell-and-tube heat exchanger,  heat duty of 5.26 x 106 W (1.79 x 107 Btu/hr). tube surface of
                                                   560m
     2 (6,000 ft2)
                                                   Shell-and-tube heat exchanger,  oil-fired, heatduryof 1.4x10  W (4.8x10  Btu/hr), tube

                                                   surface of 1860 m2 (20,000 ft2)

                                                   Centrifugal unit, stack-gas rate of 11.000 kg/hr (24,200 Ig/hr), 216 kW (290 HP)  motor power

                                                   Baghouse, screw conveyor, airlock, lockhopper, valves, and booster fan-,  570.000 j/min (20,000acfm)
                                                   stack gas.

-------
(4 boilers per 594 MWe plant) and three parallel absorber/stripper gas
cleaning trains per fluidized-bed boiler.  A single sulfur recovery
plant was used for the 594 MWe power plant.
     The absorber and stripper columns were designed by following design
techniques and recommendations presented In the literature.4~6  For sim-
plicity we applied design relationships for dilute gas mixtures that
assumed the validity of Henry's Law.  These assumptions should be excel-
lent for the absorber and reasonable for the stripper.  Henry's Law con-
stants of 40 and 80 were assumed for the absorber and the stripper,
respectively.
     The greatest uncertainty In design concerns the recuperators.
Designs for two types of recuperators (shell-and-tube and refractory
stove) were developed and are described.
     The remaining items are essentially conventional devices modified
for the corrosion protection required and for the partlculate content of
the acid solution they must handle.  The large number of modules
required for the process indicates potential economic limitations.
CAPITAL INVESTMENT
     Equipment and total gas-cleaning process capital investments were
estimated on the basis of the descriptions in Table 3.  The results of
these estimates are presented in Table 4 and based on mid-1977 dollars.
Again, the greatest uncertainty surrounds the recuperator costs.  We
estimate that the total direct cost for the pressurtzed-water scrubbing
system would be $172/kW with the shell-and-tube and $144/kW with the
refractory stove recuperator design.
     A breakdown of the boiler plant equipment costs for the dolomite-
based PFBC plant and the water-scrubber-based PFBC plant is given in
Table 5.  A breakdown of the total power plant cost Is given In Table 6.
Both plants have the same coal-feed rate, Identical combustor designs,
                                    26

-------
                                  Table 4
          CAPITAL INVESTMENT FOR PRESSURIZED WATER SCRUBBING8
                                                                   Owg.77l8A30
Equipment
Purchased Equipment,
$x!06
Cost of Installed Equipment,
*x!06
Absorbers 2. 9 8. 9
Strippers 2.2 6.6
Recuperators
      Shell-and-tube                     21. 6
      Refractory stove                    16.8
Solution Filters                           1. 5
Solution Turbines                         0.7
Solution Pumps                           0. 3
Solution Heat
      Exchangers                        3.3
Solution Coolers                          1. 2
Solution Heaters                          0. 6
Stack Gas B lowers & Recycle System           1. 2
Sulfur Recovery
      Plant
TOTAL DIRECT COST
'Basis: mid-1977 dollars; 594 MW  plant
                          G
                49.8
                31.2
                 3.4
                 1.8
                 0.7

                 7.6
                 2.8
                 1.3
                 2.2

                17.0
102. 2 (shell-and-tube),  83. 6 (refractory stove)
and  identical combustion product  flow rates.  The  dolomite-based PFBC
plant produces 635 MWe of electrical  energy, but the water-scrubber-
based PFBC plant produces 594 MWe of  electrical energy because of  lower
plant efficiency.  Costs have been  taken from previous Westinghouse PFBC
cost studies and updated to include particulate cleaning equipment and
escalation.^>7

      We estimate that  the total power investment for the PFBC with
calcium-based, high-temperature gas cleaning would be $423/kW.   This
                                       27

-------
                                Table 5
                   PFBC BOILER PLANT EQUIPMENT COSTS
                                   Dolomite-Based
          Equipment	|    System, $/kW
Water-Scrubber
Based Systems,
 	$/kWa
Steam Generator                         20.89

Draft System
     Partlculate removal                45.78
     Draft flues and ducts               2.39
     Piping                              3.99
     Stack and foundation                0.68

Coal- and Sorbent-Handllng
and Feeding Equipment                   21.54
Ash- and Dust-Handling Systems           2.23
Stack-Gas Cleaning System                	
Instrumentation and Controls             4.47
Miscellaneous Equipment                  1.36
                                       103.33

Net Plant Electrical Output           635 MWe
     22.33


     19.66
      2.55
      4.27
      0.73

     13.03
      1.23
172.05 (140.74)
      4.78
      1.45

242.08 (210.77)
    594 MWa
aSystem with refractory stove recuperator is in parentheses;  system with
 shell-and-tube is shown to its left.
cost is based on mid-1977 dollars,  635 MW plant capacity,  once-through
operation with dolomite,  17.5 percent excess air,  and three stages of
particulate control equipment (final-stage,  granular-bed filter).

     The PFBC with pressurized-water scrubbing for S02 control would
cost about $639/kW with the shell-and-tube recuperator and $594/kW
with the refractory stove recuperator.  This estimate is based on  mid-
1977 dollars, 594 MW plant capacity, 17.5% excess  air, two stages  of
particulate control equipment (high-temperature cyclones), a combustor
                                    28

-------
                               Table 6
                   PFBC POWER PLANT COST BREAKDOWN
Item
Land and Land Rights
Structures and Improvement
Boiler Plant Equipment
Gas Turbine-Generator Equipment
Steam Turbine Generator Equipment
Electric Plant Equipment
Misc. Plant Equipment
Undistributed Costs
Other Plant Costs
Subtotal
Normal Contingency
Subtotal
Escalation
Subtotal
Interest during Construction
General Items and Engineering
Limestone-Based
System, $/kW
1.63
28.21
103.33
21.33
63.62
22.93
5.13
40.86
4.19
291.23
17.47
308.70
57.88
366.58
48.13
7.93
Water Scrubber
Based-System,
$/kWa
1.74
30.16
242.00 (210.77)
22.80
68.01
24.51
5.48
43.68
4.48
442.94 (411.63)
26.58 (24.70)
469.52 (436.33)
88.03 (81.81)
557.55 (518.14)
73.20 (68.03)
7.93
     TOTAL CAPITAL COST
422.64
638.68 (594.10)
aSystem with refractory stove recuperator is in parentheses;  system with
 shell-and-tube is shown to its left.
                                    29

-------
cost identical with the high-temperature, gas-cleaning case, and a
reduction of $10/kW to account for the elimination of dolomite-handling
equipment.

     A conventional, coal-fired steam power plant with limestone scrub-
bing for S(>2 control would probably cost from $500 to 570/kW.

     The alternative PFBC system using venturi scrubbing for particulate
control is estimated to cost between $459 and 502/kW based on 17.5 per-
cent excess air, two stages of high-temperature particulate removal
equipment, and mid-1977 dollars (Reference 3, Appendix A).

     The option of applying the RESOX process for sulfur recovery to the
PFBC with water-scrubbing for S02 control in place of the commercially
available Allied Chemical process would probably increase the capital
investment further because of the low sulfur-recovery efficiency
expected with a 5 percent SC>2 gas.  The basic RESOX plant would cost
about the same as the Allied Chemical process, but the tail-gas cleaning
plant for the RESOX process (Beavon process, for example) could easily
cost another $20 to 30/kW based on a sulfur recovery efficiency of 50 to
60 percent.

     An estimate of the most optimistic case for the pressurized-water
scrubbing concept for PFBC was also developed.  If the minimum modular
design is used (a single gas-cleaning train per coinbustor module) with
no increase in plant construction time, the refractory stove recuperator
design, a coal-fired solution heater, and a RESOX sulfur recovery plant
(assumed to have 90 percent sulfur-recovery efficiency), the total power
plant capital investment would be reduced to $577/kW instead of the more
realistic case of $59l/kW to 635/kW.

     In these capital cost estimates  we have assumed off-site disposal
of waste solids.  In the dolomite-based PFBC system, the bed overflow
and collected fly ash would be conveyed dry to on-site storage silos.  A
similar system for collected fly ash would be used in the water-
scrubber-based PFBC system, where no  accumulation of coal ash in the
                                    30

-------
inert combustor bed was assumed.   The filter cake is also disposed of
off site, handled with slurry techniques similar to those used on FGD
sludge.  Disposal cost is counted as an operating cost,  accumulated
within the cost of electricity.

COST OF ELECTRICITY

     The costs of electricity generated by PFBC with calcium-based S02
control and with pressurized-water scrubbing S02 control are developed
and compared in Table 7.  The energy cost of a conventional coal-fired
power plant with limestone scrubbing is also shown.3.7  The basis on
which these costs are derived is listed in the table.

     The energy cost associated with the pressurized-water scrubbing
concept  is projected to be 3.8 to 5.0 mills/kWh greater than the
calcium-based fluidized-bed combustion power plant energy cost and
2.7 mills/kWh greater to 0.4 less than a conventional power plant energy
cost.  For the most optimistic case previously defined the total energy
cost would be 23.7 mills/kWh.

     The cost of disposing of waste  solids and liquids does not  con-
tribute  significantly  to  the cost of electricity  for  the  disposal costs
assumed.  If higher costs should occur  in  the future  (say >$20/Mg) the
water-scrubber PFBC concept  could result in  competitive  costs  of
electricity.

     Also,  should  the  cost of sorbent  increase  significantly  (say  to
>$20/Mg) then  the  water-scrubber PFBC  concept could provide economic
incentive for  development.

ENVIRONMENTAL  COMPARISON

      The environmental performance  of  calcium-based PFBC and  of PFBC
with pressurized-water scrubbing for S02  control are compared in
Table 8. The concepts are  expected to be  comparable with respect to
                                     31

-------
OJ
                                                                       Table 7

                                                     COMPARISON  OF COST OF ELECTRICITY3
                                                                                                       I7HB93
Item
PFBC: Calcium-Based
SO.Control
PFBC: Pressurized -Water
Scrubbing S02 Control
Conventional Power
Plant: Limestone Scrubbing
Capital Investment, $/kW
Energy Cost,  mills/kWh
  Capital charges
  O&M
  Fuel (coal)
  Sorbent
  Auxiliary fuel
  Makeup water
  Cooling water
  Solid/ liquid waste disposal
     TOTAL
  a
                                                            423

                                                              10.4
                                                               1.1
                                                               6.8
                                                               1.5
                                                              0.6
                                                              20.4
Shell-and-Tube
Recuperator
639
15.6
1.7
7.2
0.6
0.1
<0. 1
0.2
Refractory Stove
Recuperator
594
14.5
1.6
7.2
0.6
0. 1
<0. 1
0.2
                            Basis:  Capital charges 15% of capital investment per year
                                   Capacity factor 70%
                                   0 & M 2. 36% of capital investment per year
                                   Sorbent (dolomite and limestone) at$10/Mg
                                   Ca/ S of 2. 0 for fluid-bed combustion,  1
                                   Coal at $0. 80/GJ($0.76/106Btu)
                                   Cooling water at 0. 5*/1031 (24« /M gal)
                                   Process water at 5tf 1031 (20(1 M gal)
                                   Methane and fuel oil at $ IIGJ ($ 0. 95/106 Btu)
                                   Dry solid disposal $4/Mg
                                   Sludge disposal at $ 10/Mg
25.4
24.2
  500-570

12. 2 -13. 9
 1.5-  1.7
    7.2
    0.6

   <0.1

    1.2
22. 7 - 24. 6

-------
NOX, particulates, and heat rejection.  The high-temperature, calcium-
based, gas-cleaning process, however, has the greater potential for SC>2
emissions lower than the current standard of 0.5 kg/GJ (1.2 Ib/lO^ Btu).
The PFBC water-scrubber concept is probably limited to sulfur removal
efficiency less than 90 percent because of limited S02 solubility in
water and the limited efficiency of sulfur recovery in commercial and
developmental sulfur recovery processes.  NOX emissions from the inert-
bed combustor could conceivably be less than or greater than the
calcium-based combustor NOX emissions; factors such as catalytic effects
from calcium compounds or alumina particles and the Influence of high-
versus-low combustion gas SOX content may affect the formation/decompo-
sition of NOX in the combustor.  Differences in particulate emissions
between the two cases are also possible because of the elutriation of
sorbent fines from the corabustor in the dolomite-based PFBC.  Particu-
late standards should be achievable with both concepts with properly
selected equipment.

     The solid wastes associated with the high-temperature, calcium-
based, gas-cleaning process would be larger in mass than those for the
pressurized-water-scrubbing concept by a factor of 2.4 if all forms are
considered, by a factor of 6 to 20 if the waste sorbent is compared to
the filter cake and attrlted alumina only.  The difficulty of handling
the waste solids, however, and the environmental impact of these wastes
would not necessarily be directly proportional to mass, and further pro-
cessing of the filter cake waste would be required.  The environmental
impact and exact nature of the filter cake material is unknown, but we
expect that this material could be handled by methods applied for cor-
rosive wastes in the chemical industry, with neutralization being poten-
tially acceptable.

     With respect to waste liquids, makeup water consumption, clean fuel
consumption, and the plant heat rate  the high-temperature gas-cleaning
process appears superior to the pressurized-water-scrubbing  concept for
                                    33

-------
                                                                 Table  8

                                          ENVIRONMENTAL COMPARISON  (600  MWe  Power Plant)
                                                                                                        Dwg. 171!892
CO
-p-

S02. kg/GJ(lb/106
6
N0x, kg/ GJ ( Ib/ 10
Partial late, kg/GJ(
PFBC with C<
so2<
Btu) <0. 5«
Btu) <0. 3«
ilcium-Based PFBC with Pressurized -Water-
Control Scrubbing S0? Control
1.2) 0.5(1.2)
0.7) Probably <0. 3«0. 7)
lb/106Btu) <0. 04KO. 1) <0.04«0. 1)
Heat Rejection. % less than . ~ 15
-14
  conventional plant
Waste Liquids                                           None

Waste Solids, Mg/hr/MW {fraction of coal feed mass)
    Total                                               0. 142 (0. 458)
    Coal ash                                           0. 031 (0.10)
    Sorbent                                            0.111(0.358)
    Sulfur                                             None
    Others                                             None

Makeup Water Con sumption, J/min                       None
Clean Fuel Consumption
    Methane, 103
-------
S02 control.  The amount of process water circulated in the water-
scrubber concept is about ten times as much as in a conventional plant
with limestone scrubbing.
     For the most optimistic case the consumption of clean fuels would
be reduced to zero, significantly more coal would be consumed (~7%) and
the emission of particulates and S02 would be expected to increase
slightly.  The solid waste generation would increase by about
0.003 Mg/hr/MW due to increased coal ash.
     The high energy consumption of the PFBC water-scrubber concept is
of particular concern.
CONCLUSIONS
     The PFBC power plant utilizing the cold gas-cleaning (water scrub-
bing) concept for S02 control is not economically competitive with
calcium-based PFBC or conventional steam power plants unless the cost of
sorbents and/or waste solids disposal is substantially increased above
the costs assumed in this study.  There appears to be no alternative
that could significantly improve the cold gas-cleaning concept eco-
nomics.  The recuperative heat exchanger is technically an item of great
uncertainty and may limit the concept feasibility.
     Environmentally, the pressurized-water-scrubbing concept could
eliminate the massive amount of sulfated dolomite waste generated by
PFBC, but the nature of  the environmental effect of the waste filter
cake produced in the process is uncertain.  The energy conversion effi-
ciency of the water-scrubber concept is very poor, and the sulfur
removal  efficiency associated with the concept is limited to less than
90 percent.
                                    35

-------
          7.  PARTICULATE CONTROL TRADE-OFF FOR PFBC SYSTEMS

OVERVIEW
     Work is reported in Volume 2 (EPA-600/7-80-015b) that provides per-
spective on determining the impact of emission requirements, fluidized-
bed combustor design and operating conditions, and turbine performance
constraints on PFBC particulate control requirements and plant econom-
ics.  An EPA technical directive was performed in 1978 to assess the
effect of final-stage filter performance and the effect of filter stag-
ing on particulate loading and size distribution emitted and the result-
ing implications on turbine life.  This work provided a basis for the
subsequent analysis presented in Volume 2.

     The results from the technical directive study are reported since
they illustrate the methodology used to project turbine blade life and
electrical energy costs as a function of different particulate loadings
to the turbine using the Westinghouse turbine erosion and particle pro-
file models.  The projections of the particulate loadings and size dis-
tributions used here were based on methods employed prior to the devel-
opment of the particle profile model described in detail in Section 5 of
Volume 2.  The model originally used for particle profile projections
was not capable of including the effects of a recycle cyclone.   The PFBC
configurations discussed in this analysis, therefore, are limited to the
carbon burnup cell (CBC) concept for high carbon utilization.  The tur-
bine erosion model has similarly been extended with additional  under-
standing and experimental determination of the important model
parameters.

     This work is important in that it documents the early analysis
and represents the basis for developing the tools that can provide per-
spective on important trade-offs and permit the design of reliable PFBC
plants that operate within environmental constraints.

                                    36

-------
BACKGROUND

     A substantial amount of work has been performed at the Westinghouse
R&D Center to analyze the flow trajectories and erosion effects of par-
ticulates entrained in the flow stream of gas turbines.  Reference 8
predicts quantitative turbine erosion rates for a typical 65 MW utility
gas turbine expander by combining three major complementary analyses.
These are the calculation of the inviscid dimensional flow stream
through the blading; the calculation of the trajectories of particles
entrained in the blading flow stream; and the calculation of the erosive
effects of those particulates whose calculated trajectories result in an
impact with the blade surfaces.

     We must emphasize that the analysis in Reference 8 did not include
the effect of particulate deflection and velocity reduction due to
profile-boundary layers on the blading surfaces.  Note also that the
author cautioned that the erosion model used was based on data available
in the literature, which at best is sketchy.

     On the basis of the observations of actual erosion patterns on
experimental coal-burning gas turbines,9,10 we have concluded that sec-
ondary flow phenomena have a substantial effect on the trajectories of
small (i.e., <10 Mm) particles.  Accordingly, an analysis was carried
out**- that includes the effects of viscous boundary layers on the tra-
jectory of small particles.  The results of this analysis confirmed
qualitatively the tendency of the boundary layer flows to concentrate
the particulates in certain regions of the gas turbine.  Because of the
complexity of the problem, however, quantitative erosion rates were not
calculated.

     Another phenomenon that Is of interest, particularly for very high-
temperature turbine applications, is the temperature reduction of the
particle as it penetrates the boundary layer next to a cooled blade sur-
face.  To determine the temperature history of a particle as it passes
through the boundary layer one must also calculate the trajectory veloc-
ity history.

                                    37

-------
 ESTIMATION  OF PARTICLE LOADING/SIZE  DISTRIBUTION
     The works  reported above are related  to  the erosion caused by par-
 ticles after they enter the  turbine.  One  must, of course, determine the
 concentration and size distribution  of  the particles that enter the tur-
 bine.  These parameters are  determined  by  the partlculate removal equip-
 ment Installed  in the hot gas stream between  the turbine and the pres-
 surized fluidlzed-bed boiler in which are  generated both the hot gas
 stream and  the  entrained partlculates.   (Essentially the same particu-
 late removal equipment would be used If  the hot gas stream source were
 an adlabatic fluldized-bed combustor or  a  fluldized-bed gasifler.)  The
 performance of  the partlculate removal  equipment depends upon the con-
 centration, density, and size distribution of the entering particles,
 and the detailed design of the equipment would depend upon these fac-
 tors.  The  general approach, however, is to remove the bulk of the large
 particles in the early stage(s) and  remove the fines in the later
 stage(s).   The  performance and cost  of a particulate removal system in a
 plant with  a pressurized fluidlzed-bed boiler have been described in
 some detail.12

     Figure 3 is a flow diagram of the separation equipment arrangement
 designed for the present study.  Figure 4  shows a typical arrangement of
 the particulate removal equipment and the  piping connecting the pres-
 surized fluidized-bed boiler to the  gas  turbine.  As Figure 3 indicates,
 the effluent from the pressurized fluidized-bed boiler first enters a
 cyclone separator,  a relatively inexpensive component that can handle
 high particulate loadings  and has a much higher removal efficiency for
 the larger particles than for the smaller.

     The particles collected by the primary cyclone separator contain a
 large portion of char (carbon), which is fed to a CBC to recover the
chemical heat of combustion.   Additional combustion air is fed to the
CBC to complete the reaction.  The volume of flow through the CBC is
small compared to that of  the main flow stream,  so the exit stream is
                                    38

-------
                                                                      Dwg. 6407A85
From PFBB
               Cyclone
               Separator
                                       Cyclone
                                       Separator
                                  Carbon
                                  Burnup
                                    Cell
\
                  Air
            T
                                                       Particulates
                                           Particulates
                                      Air
                               First-Stage
                                Granular-
                                Bed Filter
                                                                                Particulates
Second-Stage
  Granular-
  Bed Filter
                                 Particulates
               Figure  3 - Diagram  of Particulate Removal System
                   Gas Turbine Generator
                                                     Cyclones
                                                   Carbon
                                                   Burnup Cell

                              Pressurized f^\-^f^\ Pressurized
                              Fluidized-  I     /Sv    Jpluidized-
                              Bed Boiler  \	X   \—/ Bed Boiler
                                B                     A

            Figure 4  - Particulate Removal Equipment Arrangement1^
                                            39

-------
passed through a small cyclone separator before being introduced into
the main flow.  This cyclone is designed to accommodate the very high
particulate loadings leaving the CBC and is equipped with secondary
inlet air jets that prevent fouling of the unit and provide angular
momentum for the particle separation process.

     The combined flow stream then enters the granular-bed filter, which
is capable of high removal efficiency for particles smaller than 10 vim.
Two stages of granular-bed filters are shown in Figure 3.  Since the
granular-bed filters are by far the most expensive component of the par-
ticulate removal system, this study will investigate whether the second
stage is economically justified.
     An analysis made specifically for the present study has been used
to calculate the particulate concentration and size distribution at each
location within the particuate removal system.  The analysis considers
the sorbent, the ash,  and the char separately, since each of these con-
tituents has a different effective density,  an important factor in the
efficiency of a particulate removal component.
     The design conditions of the fluidized-bed boiler used to calculate
the rate and size of the elutriated particles is as follows:

     Superficial  bed velocity   1.52 m/s (5  ft/s)
     Bed depth                  2.4 m (8 ft)
     Pressure                   1013 kPa (10 atm)
     Bed temperature            1010°C (1850°F)
     Sorbent                    Dolomite
     Sorbent feed size           3.2 x 0 mm (1/8 x  0 in)
     Coal feed size             6.4 x 0 mm (1/4 x  0 in)
     Excess  air                 Primary bed,  6%;  CBC 36%
     Ca/S atom ratio            1.5:1
   Carry-over from the  bed was estimated by  calculating  the size of the
particle whose terminal velocity was equal  to the  superficial velocity
in the combustor.   All  of the feed material  below  this size was assumed
                                    40

-------
to be elutriated.   Efforts to include sorbent attrition in the calcula-
tion involved the use of an average attrition rate that was a function
of bed velocity and residence time.  A more detailed account of the cal-
culations involved in computing the carry-over characteristics is pre-
sented elsewhere.^

     The performance of the primary conventional cyclone is shown in
Figure 5 as a function of species density.  These estimates were made on
the basis of a commercial vendor's estimates of performance at pressure
and temperature.  The cyclone pressure drop is estimated at 165 cm wg at
an inlet velocity of 17 m/s.

     The performance of the CBC cyclone (Figure 6) is similarly based on
the manufacturer's estimates for the elevated temperature and pressure.
The specific device considered was a Donaldson Tan Jetv^y cyclone oper-
ating with a primary flow pressure drop of about 180 cm wg.  This device
also employs a high-pressure, secondary flow of clean gas to impart the
rotary motion to the dust-laden gases entering the device.  A schematic
of this device is presented in Figure 7.

     The granular-bed filter concept used to establish the costs of the
final filter system is basically the same Ducon filter that had been the
basis of cost estimates for the previously published ECAS^2 report.
Figure 8 presents a schematic of the filter system.  In the Westinghouse
design relatively few, but large (7-8 m dia), pressure vessels were used
to effect a cost savings over a large number of small modules.

     Although the performance of cyclone  separators is known to a  rea-
sonable degree of accuracy, the performance of the granular-bed filter
- because it is still in an early  stage of development  — is poorly def-
ined.  In the long range granular  bed filters may achieve a removal
efficiency equal to that currently achieved by conventional low-
temperature fabric filters.  At present,  Westinghouse has bench-test
results of a granular-bed filter whose performance  is not as good  as
                                     41

-------
                                                     Curve 694490-A
                                                 Specific Gravity
                                                  2.8 1.35.8
0.01
 0.1
                                     D Ipm)

     Figure  5  - Grade Efficiency  of  Primary  Cyclone Separator

                                     Specific Gravity
                                      2.88 .88.64                     ^'
   10


   30

F  90

-  70


   90


   98
   99


 99.9

 99.99
               Tan Jet
                   TM
               i   ill
                                                 T= 927 °C I1700°F)
                                                 P= 1013kPa(10Atm)
                                 i    ill
                                                    i    iii
  .10                  1.0                10.0                100
                                     D . Mm
                                      P
           Figure 6 - Grade Efficiency Curve for Tan Jet
                                       42

-------
                                                  6214816;
                 1 Clean Gas Out
Secondary
  Gas
                 Dust
-Dirty Gas In
                                   -Body
                                 Tangential Gas
                                 Jet Assembly  out
                              Gas Jets
     Figure  7  -  Schematic  of  Tan  Jet
                                             Cwg. 6Z24A1Z
  Catalyst- <
  Laden
  Oust
         ,	I
                    Filter Element Internals
            Clean Gas
          Operating Cycle
                                    Cleaning Cycle
  Figure  8  - Granular-Bed  Filter  Module
                             43

-------
 that  of  a  fabric  filter.   As  a  third  benchmark,  results  are  available
 from  the EPA^ on an  existing commercial  unit  (Rexnord)  whose  perform-
 ance  is  substantially inferior  to  that  of fabric filters.

      In  an attempt  to bracket the  actual  situation,  the  calculations
 presented  here have been based  on  the grade efficiencies of  each of the
 three cases mentioned above.  The  actual  grade efficiencies  used for the
 fabric filter and the Westinghouse granular-bed  filter performance are
 pre-sented in Figure  9 along  with  the Rexnord data.

     Using these  three efficiency  criteria, we have  calculated  the con-
 centration of particulate  at  the outlets  of the  first and second stages
 of granular-bed filters, and  these are  shown in  Tables 9 and 10.  A more
 graphic  illustration  of these performances is shown  in Figures  10 and
 11.  Figures 12 through 20 show the size  distribution of the particles
 leaving  these filters.  Similar information at the discharge of the
 cyclone  separators has not been included  since preliminary analysis
 showed that the turbine erosion rates corresponding  to these particulate
 loadings were so  high that a  system not including granular bed  filters
 would be impractical.

 ESTIMATION OF GAS TURBINE IMPACT

     On  the basis of  the information available in all of the above ref-
 erences, a relatively simple method was conceived with which to calcu-
 late the erosion  rate of gas  turbine blading as a function of particle
 distribution and concentration.   A parametric, quantitative assessment
 of power generation cost penalties has been determined from those data
 on erosion rate, practical wear limits, and the cost of blading replace-
ment (compared to the cost of particulate removal equipment as a func-
 tion of its removal efficiency).

     Reference 8 shows that maximum erosion occurs at the leading and
 trailing edges of rotors and at the trailing edge of stators.  Although
                                    44

-------
                                                    Curve 694489-A
                                                      \ I  I I I U
              1 - Fabric Filter, Ref. W
              2 -® Granular-Bed Filter Data
              3 - Rexnord Granular-Bed
                 Filter Ref. 13
                                                                99
                           0.1               1.0
                             Particle Diameter,
             Figure 9 - Various Filter Performances Assumed
                        for Final Cleanup Stage
10.0
these rates are  roughly  comparable in magnitude, the calculations
neglect the effect  of  blade profile boundary layers.   Since  the  boundary
layer thicknesses are  larger at the trailing edge than at  the  leading
edge, the actual erosion rates will be reduced a greater amount  at the
trailing edge.   This  is  particularly true for the particulates that pass
through a granular-bed filter since they have a large  proportion of very
small particles  that  are readily deflected and slowed  by the boundary
layer velocity profile.   This general reasoning is  supported by  the
experimental evidence^,  which shows the maximum erosion at the leading
edge of the rotor blades.   On the basis of these considerations, we have
made calculations for this analysis only for the rotor leading edge.

     The calculation  procedure is essentially a stepwise  integration of
the erosion of each particle size over the range of particle sizes
entering the gas turbine for each of the three  types of particulates.
                                      45

-------
                                             Table 9

                           PROJECTED PARTICULATE CONCENTRATION LEVELS
Projected Based
Upon
Location,
Outlet of
Dolomite Concentration,
g/sm3 (gr/scf)
Ash Concentration,
g/sm3 (gr/scf)
Char Concentration,
g/sm3 (gr/scf)
Rexnord Granular-
Bed Filter
Rexnord Granular-
Bed Filter
J) Bench Tests
Granular Bed Filter
J) Bench Tests
Granular Bed Filter
Conventional
Fabric Filter*
Conventional
Fabric Filter*
First GBF
Second GBF
First GBF
Second GBF
First GBF
Second GBF
0.1326
(0.05795)
0.0866
(0.03783)
0.00787
(0.00344)
0.000214
(0.0000935)
0.00103
(0.000449)
0.000000
(0.000000)
0.2179
(0.09518)
0.1243
(0.0543)
0.0138
(0.00601)
0.000290
(0.000127)
0.00203
(0.000886)
0.000000
(0.000000)
0.0398
(0.01737)
0.0244
(0.01065)
0.00240
(0.00105)
0.0000559
(0.0000244)
0.000353
(0.000154)
0.000000
(0.000000)
*Particle concentrations indicated are
 a conventional low-temperature fabric
based upon assumptions
filter.
that GBFs perform as effectively as

-------
          Table 10




PARTICULATE EMISSION LEVELS
Performance No. of (
Characteristics Stage:
Rexnord 1
Rexnord 2
(w) Bench Tests 1
(W) Bench Tests 2
Fabric Filter 1
Fabric Filter 2
Total Particulates Emitted
3BF
5 g/snH (g/scf)
(0.1705)
0.390
(0.1028)
0.235
(0.0105)
0.0240
(0.00025)
0.000572
(0.0015)
0.00343
nil
g/MJ
(0
0
(0
0
(0
0
(0
0
(0
0

(lb/106 Btu)
.137)
.136
.191)
.0820
.0195)
.00838
.000465)
.000200
.00279)
.00120
nil

-------
oo
                0.229
               (a 10)


                0.206
               
-------
                                        Curve 713968-A
                                                                                                     Curve 713969-A
    0.01
   0.1

   0.5
    1
    2
    5
    10
a   20
i/t
|   40

1   60
     99 -
    99.9 -
   99.99
      0.2
                                 I    I  I  I  II ITJ

                              Dolomite Particles
                 Exit of Second GBF
                   I   i   I
          0.4   0.6    0.8 1     2
                    Particle Diameter, t
3 4 5 678 10
  0.01

   0.1

   0.5
    1
    2
    5
    10
a   20
*/>
f   40

I   60
»
£   80
    90


    99


  99.9


 99.99
                                          ~nn—n	'—'—'  MINI
                                                  Ash Particles
                                                                                       Exit of First GBF
                                                                               I
                                                                             Exit of Second GBF
                                                                                     I  I
                                                         I   ill
          0.2   0.4  0.6   1     2   3  4 5 678 10
                    Particle Diameter, urn
Figure 12  - Projected  Outlet Size  Distribution
              Based on Rexnord Commercial  Unit
              (Dolomite  Particles)
                          Figure 13  - Projected  Outlet  Based  on
                                        Rexnord Commercial Unit
                                        (Ash Particles)

-------
                                        Curve 713967-A
 0.01


  0.1
  0.2


    1
    2

    5

   10
   40
c  60
s
£  80
   90




   99


  99.9


 99.99
                    I
Char Particles
                               Exit of First GBF
                 Exit of Second GBF
            0.2     0.4  0.6   1.0     2   4
                     Particle Diameter, urn
                 6   8  10
                                             0.1

                                             0.5 -
                                               i
                                               2-
                                               5
                                              10

                                            S 20
                                            e
                                            | 40
                                            3

                                            I"
                                            £ 80

                                              90
                                              99



                                             99.9


                                            99.99
                                                                             Curve 713973-fl
         I I  I I I |      I


         Dolomite Particles
                                                                                       Exit of First GBF
               Exit of Second GBF
    I	I
               I
                         i  i  i i  i i
0.2    0.4  0.6 0.8 1     2
          Particle Size, u m
6  8 10
  Figure 14 -  Projected Outlet Based on  Rexnord
                Commercial Unit
                               Figure 15 -  Projected Granular-Bed Filter Outlet
                                             Size Distribution Based on
                                             Westinghouse  Bench-Scale
                                             Experiments (Dolomite Particles)

-------
                                         Curve 713966-A
   0.01
    0.1

    0.5
      1
      2
      5
     10
     20
 12
 o>
 •o
 0)
 Q.
c   60
8
    80

    90
    99


   99.9

  99.99
                   T —i  -T  |    -l    r

                        Ash Particles



                    Exit of First GBF


                            Exit of Second GBF
                                              i I  i 11
                       I	i
i   ill
            .2     .4  .6 .8 1      2
                       Particle Size, M m
                                       3  456 8 10

 0.01

  0.1

  0.5
    1
    2
    5
   10

J   20

!   40
)
   60

-   80

   90


   99


 99.9 -
                                                                                                    Curve 713965-A
                                                            99.99
                                 I   I   I  I  I  I I I  I

                                           Char Particles
T I  i  i i i l
                                                -Exit of First GBF


                                                    -Exit of Second GBF
                                                           l	i
                               0.2    0.4  0.6 0.8 1      2
                                          Particle Size. iam
34   6  8  10
Figure 16 - Projected Granular-Bed  Filter
             Outlet  Based on Westinghouse
             Bench-Scale Experiments
             (Ash Parties)
                                                               Figure  17 - Projected  Granular-Bed Filter
                                                                            Outlet Based on Westinghouse
                                                                            Bench-Scale Experiments
                                                                            (Char Partcles)

-------
                                               Curve 713971-A
         0.01
                                                                                                        Curve 713972-A
Ul
 0.0 -

 0.5
   1
   2
   5
  10

i  2°
I
I  40

:  60
»
!  80


  95


  99


 99.9
         99.99
                                 1     2   3
                            Particle Diameter, \i m
4 5 6 7 8 10
         Figure 18 - Projected Granular-Bed  Filter
                      Outlet Size Distribution
                      Based on Conventional Fabric-
                      Filter Unit Performance
                      (Dolomite Particles)
                                                                    a 01

                                                                     0.1
                                                                     0.5
                                                                      1
                                                                      5
                                                                     10
                                                                      20

                                                                     40

                                                                      60
                                                                  0)
                                                                  a.
  90
  95
  99


 99.9


99.99
                                I    1  I  I  I I I I |      I    I  I  I  I  M |

                                           Ash Particles
                                I    I   i i  i  i i 11
                                                         i  i  i
                                               1      2   3 4  5 6  8  10
                                         Particle Diameter, \im

                          Figure  19 - Projected Granular-Bed Filter
                                       Outlet Based on Conventional
                                       Fabric Filter Unit
                                       Performance  (Ash  Particles)

-------
                                                     Curve 713970-A
Ui
U)
           99.
                                     1     2   3  4  5 678 10
                               Particle Diameter, mn
                                                                           35x6
                                                                           (14)

                                                                           30x5
                                                                           (12)

                                                                        £  25x4
                                                                      CE
                                                                      3 ID
                                                                      il
                                                                      3> E
                                                                      ee —
20x3
 (8)

15x2
 (6)

10x2
 (4)

 5X1
 (2)
           '   I   '   I
                                                                                                               Curve 71 3t-64-A
                                         '   r
                         Particle Density

                             =2.5g/cm
                         op  = 1.5g/cm

                Particle Concentration = (0.00023g/sm
                                                                                           i   i   i   i   i   i   i    i   i   i
                                                                                                             10
                                           12
                   Particle Diameter, (D ),
                                  P
                                              14
        Figure  20 - Projected  Granular-Bed  Filter  Outlet
                      Based on Conventional Fabric-Filter
                      Unit Performance  (Char  Particles)
Figure 21 -  Blade  Leading  Edge Erosion
              Rates

-------
 1.  Choose a step size (e.g., 1 Mm).
 2.  Enter Figure 11 (or appropriate subsequent figure) and read off
     the percentage undersize at each side of the step.
 3.  Take the difference between these two values.  This gives the
     percentage of the particulate size at the mean of the step.
 4.  Multiply this percentage by the total particulate concentration
     entering the turbine.
 5.  Enter Figure 21 (taken from Reference 8) and read from the
     curve the metal recession (for 10,000 hours of operation) at
     the rotor leading edge for the particular particle size of this
     integration step.  (Note the density parameter.  The erosion
     rate reduces slightly for lower density particles since they
     deviate less from the gas streamlines.  Note also that the ero-
     sion of Figure 2L is for a given particle concentration.)
 6.  Multiply the erosion read from the curve by the ratio of the
     actual concentration (from Step 4)  to the concentration for
     which Figure 21 has been calculated.
 7.  From Table 11,  which traces the trajectory of a given size par-
     ticle in a specified boundary layer thickness,  read the velo-
     city of the particle at impact with the blade surface.
 8.  Calculate the square of the ratio  of the velocity at impact to
     the velocity entering the boundary layer.
 9.  As mentioned in Reference 8,  the erosive effect of a particle
     is proportional to the square of the impact velocity.   There-
     fore,  multiply  the recession rate  calculated in Step 6 by the
     factor calculated in Step 8.
10.  Repeat Steps 1  through 9 over the  complete range of particle
     sizes.
11.  Add the erosion rates of all the particle sizes.
12.  Based on the estimate of the Westinghouse Gas Turbine Division
     that the maximum allowable amount  of erosion of a turbine rotor
     blade would be  254 mm (0.100 in),  multiply 10,000 hours by the
                                54

-------
                                                     Table 11

                               THERMAL HISTORY  OF  PARTICLES ENTERING A 0.005 INCH
                             THICK  BOUNDARY  LAYER  WITH 10 DEC.  INCIDENCE,  AT Rl-L.E.
                    GAS VEL. = 1200 FT/S,  TO =  2060.DEC. R,  TW = 1460.DEC.R, THER.EMISS.
0.9
.94115-26
.11111-08
.22222-06
.33333-06
.44444-06
.55556-06
.66667-06
. 00000
.32000+00
.63997+00
.95989+00
.12797+01
.15995+01
.19191+01
.99154+00
.94377+00
.88793+00
.83243+00
.77733+00
.72263+00
.66827+00
.11980+04
.11862+04
.11716+04
.11568+04
.11411+04
.11245+04
.11071+04
.15990+04
.15931+04
.15359+04
.15784+04
.15705+04
.15623+04
.16635+04
.12000+04
.12000+04
.11998+04
.11996+04
.11992+04
.11988+04
.11932+04
-.21156+03
-.21013+03
-.20672+03
-.20732+03
-.20592+03
-.20453+03
-.20315+03
.15994+04
.15994+04
.15993+04
.15992+04
.15990+04
.15988+04
.15986+04
.10401+02
.10393+02
.10430+02
.10523+02
.10679+02
. 1 0903+02
.11205+02
.11773+05
.11741+05
.11713+05
.11696+05
.11691+05
. 1 1 699+05
.11721+05
.49687+05
.33002+04
.15458+04
.99565+03
.72610+03
.56570+03
.45902+03
ijl

-------
          ratio  of  254 mm  (0.100  in)  to  the  total  recession  calculated  in
          Step 11.  This gives  the  time  in hours required  to erode  away
          254 mm (0.100 in).
     13.   Repeat the  above procedure  for each  of the  particulate
          constituents —  i.e., dolomite, ash,  and char.
 RESULTS
     The  boundary  layer near the leading edge  of  a rotor  blade is  very
 thin, of  course, since it has had  little distance in which  to develop.
 The  thickness is a function of the Reynolds number and can  be estimated
 from simple flat plate theory.  Because of  the rapid acceleration  of the
 main stream flow around the leading  edge of a  blade, the  boundary  layer
 growth tends to be retarded in this  region compared  to a  flat plate.   In
 Figure 144 of Reference 15, an experimental determination of boundary
 layer thickness is given  for an airfoil at approximately  the right
 Reynolds  number and  thickness chord  ratio.  The thickness shown in the
 figure near the leading edge is approximately  127  urn (0.005 in).   Accor-
 dingly, calculations of the erosive  life of the blading have been  made
 for  boundary layer thicknesses of  0.0,  127, 254,  and 508  pm (0.0,  0.005,
 0.010, and 0.020 in) to determine  sensitivity  to  this critical
 parameter.
     Another parameter of importance, which is not well defined at  this
 point, is the erosiveness of the particulates.  On the basis of evidence
available from the literature,  Reference 8 assumed that the erosiveness
of the ash and dolomite particles was 1/25 of  that of silcon carbide
 (SiC) particles.
     Figures 22 through 24 show the  results of the calculation for  the
 three levels of performance of the granular bed filters.  Blade life,  in
hours, is plotted against boundary layer thickness with parameters of
erosiveness.  The  range of erosiveness chosen  is  from twice to half what
was  indicated in Reference 8.
                                    56

-------
                                                                                                         Curve 689722-A
    20,000

    18,000

    16,000

    14,000

£  12,000
g
•=_  10,000
CD
3   8000

     6000

     4000

     2000
                                          Curve 6897Z1-A
Erosiveness Compared
to SiC Particles    .
       0 0  (2)  (4)  (6)  (8)  (10) (12) (14) (16) (18) (20) (22)
            51   102  152  203  254  305  356  406 457  508 559
                  Boundary Layer Thickness, \i m (mils)
 Figure  22 -  Projected Turbine Life for  a
                Participate  Removal  System  with
                Two Stages of Granular-Bed
                Filters (performance  based  on
                Rexnord, commercial  unit)
    24,000

    22,000

    20,000

    18,000

    16,000

e  HOOO

1  12,000
oT
2  10,000

     8000

     6000

     4000r-

     2000
                                                     Erosiveness Compared
                                                     to SiC Particles
                                                                                                             1/50,
                                          0   (2)  (4)  (6)  (8)  (10) (12) (14) (16)  (18)  (20)  (22)
                                               1  102  152  203  254  305 356 406  457  508  559
                                                   Boundary Layer Thickness, urn (mils)

                                     Figure 23 -  Projected Turbine  Life for  a
                                                   Particulate Removal System  with
                                                   One  Stage of Granular-Bed
                                                   Filters (performance  based  on
                                                   Westinghouse bench-scale
                                                   experiments)

-------
      Figure 22 shows  the life of  the  turbine with  two  stages of granular
bed  filters, each with a performance  equal to that of  the Rexnord com-
mercial unit.  As the figure indicates,  for a boundary layer thickness
in the  expected range of 0 - 127  ym (0  to 5 mils), the life of the  tur-
bine  is very short — no more than six  months, even assuming an optimis-
tic  level  of erosivity.   One stage of Rexnord filter was  found to give
inadequate life.

      Figure 23 shows  the life of  the  turbine with  one  stage of granular
bed  filters with a performance equal  to  that of the Westinghouse bench-
test  unit.   As the figure indicates,  this level of performance increases
the  turbine life  appreciably, although  it is still much shorter than is
usual for  utility equipment.

     Not shown on a figure is the life  resulting from  the use of two
stages  of  granular-bed filters with the Westinghouse measured
                  30,000
                  28,000
                  26,000
                  24,000
                  22,000
                  20,000
                  18,000
                e
                § 16,000
                § 14,000
                  12,000
                  10,000
                   8000
                   6000
                   4000
                   2000
                     0
                                        1/25
               Erosiveness Compared
               to SiC Particles
i
   i
              i
                          i
                                                         Curve 689733-A
                      0  12) (4)  (6)  (8) 110) 112) (14) (16) (18) 120) 122)
                        51  102  152  203 254 305 356 406 457 508 559
                              Boundary Layer Thickness, pm (mils)
  Figure 24 - Projected Turbine  Life for a Particulate Removal  System
              with One Stage of  Granular-Bed Filters (performance
              based on conventional  fabric filter  efficiency)
                                      58

-------
performance.   This Is because the partlculate concentration is so low
that even if  a zero boundary layer thickness is assumed, the turbine
life is 19 years.   In other words, the erosion rate is negligible.

     Figure 24 shows the life of the turbine with one stage of granular-
bed filters with a performance equal to that of a conventional  low-
temperature fabric filter.  The turbine life here is appreciably
increased over that of the previous two cases.  The curves are rela-
tively steep, especially so near the zero boundary layer thickness, but
with the erosiveness that Reference 8 assumed, the turbine life is
approximately two years.  With two stages of granular-bed filters, how-
ever, the erosion rate is essentially zero since the particulate  concen-
tration is zero when calculated to six decimal places.

     If we assume that replacing the turbine blading would be con-
sidered an operating and maintenance (O&M), expense, we  can calculate
it  in the form of a  cost of electricity (COE)  in units  of mills/
kWh.  Based on information  from the Westinghouse Gas Turbine Division,
the installed cost of a complete change of  turbine blading is approxi-
mately $2 million  per turbine.  (No charge  is  included  for plant  down
time.  We have assumed for  purposes of this  study  that  blade  changes
would be made during normal maintenance periods and  not as a  result of
forced outages.)   As shown  in the  EGAS report,H the  power output of a
pressurized  fluidized-bed power  plant with  two W501  gas turbines  is
679,000 kW.   Thus,  the  equation for the cost of electricity in mills/kWh
due to a  blading  change  in  terms  of the time in hours between blading
changes  is as follows:

       COE »  $2 x  106/(turbine-change)  (2  turbines)  (1000 mills/$)
               CF   kW rating  (679>°00 ™  rating)  (A hr/change)

 Note that as the  capacity factor (CF)  goes  down,  the kW output  goes down
 so the COE goes  up.   On the other hand, as  the CF goes down,  the coal
                                     59

-------
 input and  resultant particulate loading goes down so the life, A, goes
 up.  Since  the life was calculated on the basis of full load, the
 sensible approach is to take CF = 1.0.  Thus, the equation becomes:
With  this equation the lifetimes shown in Figures 22 through 24 can be
converted to COE.  This result is shown in Figures 25 and 26 for two of
the three assumed granular bed filter performances.  The calculation was
not made for the Rexnord performance since the turbine lives would be
too short to have practical application in a utility power plant.  The
COE is plotted versus the boundary layer thickness, with parameters of
erosiveness for one stage of granular-bed filters.  The A COE associated
with  a two-stage, granular-bed system has been calculated and is also
shown in Figures 25 and 26.

     Calculations have been made to estimate the cost of granular-bed
filters for the gas flow conditions of this power plant.  Each module
handles a volume flow rate of 15.72 m3/s (33,300 acfm) , which results in
a filter pressure vessel diameter of 7.6 m (25 ft), if we assume a
design face velocity of 15.2 m (50 ft/mi n) , a face area per filter ele-
ment of 0.37 m^ (4 ft^), and a plan area per element of 0.26 m^
(2.8 ft^) (which includes the open flow area around each element).  The
cost of a 7.6 m (25 ft)  diameter, granular-bed filter module for opera-
tion at 982°C (1800°F) and 1013 kPa (10 atm)  has been estimated on the
following basis :

    Cost base                    - mid-1975 $
    Field labor                  - 51% of direct installation costs
    Professional services        - 10%
    Escalation                   - constant dollars
    Interest during construction - 10%
    Construction time            - 5 years
                                    60

-------
                     \    I    I	1	1	1	1
                                  Erosiveness Compared
                                  to SiC Particles
            ^ Cost Due to
              Second Stage
                                                                                                       Curve 689720-A
          (2)  (4)  (6)   (8) (10) (12) (14) (16)  (18) (20)  (22)
           1   102  152   203 254 305 356 406  457  508  559
                Boundary Layer Thickness, urn (mils)

Figure 25  - Cost  of Electricity Increments due
              to Turbine  Blade Replacement  Using
              Granular-Bed Filters (performance
              based on  granular-bed  efficiency)
       3.0

       2.8

       2.6

       2.4

       2.2

       2.0

       "
       1.6
                                                                 o
                                                                ~
   o 0.8

      0.6

      0.4

      0.2
                                                                           I    I    I    I
                                                                                  ^ Cost Due to
                                                                                -*— Second Stage
                                     Erosiveness Compared
                                     to SiC Particles
                                                                          I
         0   (2)  (4)  (6)  (8)  (10)  (12)  (14)  (16)  (18)  (20) (22)
             51   102  152  203  254  305  356  406  457  508 559
                  Boundary Layer Thickness, M m (mils)

Figure 26  - Cost  of Electricity Increments due
              to Turbine  Blade Replacement Using
              Granular-Bed Filters (performance
              based on  fabric-filter  efficiency)

-------
     The resulting cost for a 7.6 m (25 ft) single-stage, granular-bed
filter module is $1,850,000.

     The equation for the cost of electricity due to capital expense is

COE - (yearly cost of capital, $/$-yr) (capital cost. $) (1000 mills/$)
              (capacity factor) (8760 hrs/yr) (plant power, kW)

The incremental COE due to the incremental cost of two stages compared
to one stage is calculated to be 1.08 mills/kWh.

CONCLUSIONS

     1.  This analysis provides a basis for evaluating the economic
         advantage of improving filter performance rather than replacing
         turbine blades.
     2.  More accurate experimental information on particulate erosive-
         ness and staged-bed filter performance is required before a
         definitive comparison can be made.
     3.  Although the cost of staged filters is substantial, the differ-
         ence between this cost and the cost of blade replacement in the
         case of least frequent blade replacement (Figure 25) is only
         about one half mill/kWh,  which indicates that rather large
         capital costs can be tolerated for efficient filter systems.

RECOMMENDATIONS

     1.   Accelerate development of granular-bed filters and alternative
         filter concepts  designed  for gas  turbine applications.
     2.   Continue experimental studies to  investigate particulate ero-
         sion rates of gas turbine blading.
                                   62

-------
      8.   INDIRECT AIR-COOLED PRESSURIZED FLUIDIZED-BED COMBUSTION
                       CONCEPT SYSTEMS EVALUATION
INTRODUCTION
     A gas turbine cycle combustor using PFBC with indirect heating of
pressurized air in immersed tubes is being investigated by Curtiss
Wright under DOE funding.1°  We compared the performance and cost of
energy of a combined-cycle plant using this configuration and those of a
combined-cycle plant using an adiabatic fluidized-bed combustor.  We
considered two configurations of the partially indirectly heated system:
one with a CBC and one without.  In each case we selected the amount of
excess air that would give an overall carbon loss equivalent to 1 per-
cent of the energy in the coal.

     Since environmental concerns are of primary importance  to EPA, this
study included an assessment of  the effect of the partially  indirectly
heated concept on pollutant emissions (particulates,  SOX, NOX, products
of  incomplete combustion, and solid wastes) compared  to  the  adiabatic
PFBC System.

     This  technoeconomic study  was  carried out  in late  1976  and  early
1977 and  reflects  the FBC technology  and  the  economic situation  that
obtained  in that  time period.   No attempt has been  made  to update  the
results.

BACKGROUND

      The products  of combustion from PFBC of  coal  are passed through a
gas turbine expander.   The cost of  the particulate removal equipment
 required to clean the combustion products well enough to avoid problems
 with erosion of, corrosion of,  and deposition  on the expander parts  is
 expected to be significant.   This is particularly true of gas turbines
                                     63

-------
 with an adiabatic  fluidized-bed  combustor,  where  air equivalence  ratios
 of  approximately 3 prevail  (200% excess  air),  since  the  cost  of the  par-
 ticulate removal equipment  is  roughly  proportionate  to the volume  of  the
 combustion  products and  the ratio of combustion products to coal  is
 high.

      In the partially  indirectly heated  gas turbine  combustor concept
 the volume  of  the  combustion products  that  must be cleaned is reduced
 substantially  by using the  minimum amount of combustion  air and heating
 the balance of the gas turbine airflow indirectly with tubes  submerged
 in  the  fluidized bed.  This indirectly heated  air is  then mixed with  the
 combustion  products after they have been cleaned, and this mixed  stream
 constitutes the flow to  the gas  turbine  expander.  There is a trade-off,
 therefore,  between the cost of the particulate removal equipment and  the
 cost of  the heat transfer surface required  for indirectly heating  the
 air that bypasses  the  combustor.
 Description of Systems Evaluated
      The coal-fired  power systems that were evaluated and compared in
 this study  are:
     •   Base Case  -  A  combined-cycle system with an adiabatic
         fluidized-bed  combustor and in situ desulfurization
     •   Alternative  Case I  - Partially indirect heating  with  a CBC
     •   Alternative  Case II - Partially  indirect heating  without a
         CBC.

     Performance calculations were made  for each of these configurations
with Ohio Pittsburgh No.  8  seam coal with 3 percent moisture.

     The gas turbine design conditions were as follows:
     Ambient air conditions - International Standards Organization (ISO)
     Compressor airflow - 345 kg/s
                                    64

-------
     Compressor pressure  ratio  - 10
     Compressor isentropic efficiency  - 0.853 (polytropic efficiency
      W 0.89)

     Expander isentropic  efficiency -  0.927 (polytropic efficiency
      W 0.90)
     Temperature drop due to heat transfer between combustion products
       and combustion air - 8°C.

     The fluidized-bed combustor design conditions common to all cases
were as follows :
     Primary bed temperature - 1010°C  (value used in EGAS17)
     Coal size - <6.4 mm
     Dolomite size - <4.8 mm
     Ca/S atom ratio - 1.5
     Superficial velocity -^1.5 m/s
     Maximum bed depth - 4.6 m
     Pressure loas (including particulate  removal equipment) - 7.5%.

     The design conditions for  the heat recovery steam generator  (HRSG)
were as  follows :
     Type - unfired
     Pinch - 22.2°C
     Offset - 2.8°C.

     The Base  Case was previously  treated  in Reference  18 in a somewhat
different configuration.   The  configuration  used in  this study is shown
in Figure 27.   A Ducon cyclone separator was used for  the first stage of
particulate  removal,  a Ducon granular-bed  filter for the second stage.
Grade  efficiency plots for these components  are given  in Appendix A.
The fluidized-bed  design conditions specific to  the  Base Case were as
follows:
     Excess  air -  237 percent
     Superficial velocity - 1.5 m/s
     Bed depth - 2.0 m
                                     65

-------
                      7
                   Back-flushing
                   .,
                   Air
                  RuMized-
                    Bed
                  Combustor
                  Spent
                  Sorbent
5. )681M5

 Granular-Bed
l—,  Filter
                      Generator
                                     Heat Recovery
                                     Steam Generator
                                            Steam Turbine
                                            Generator Set
        Figure 27 - Combined-Cycle System Utilizing Fluidized-Bed
                     Combustion (Base Case)



      Combustion losses

         Incomplete combustion     -  0.4

         Losses  to  atmosphere       -  0.8

         Sensible heat in solids    -  1.25

         Desulfurization reactions  -  Q.i

                              Total     2.55 percent

      Overall combustion efficiency  97.45 percent.


With  allowances for  thermal losses from the ductwork,  the temperature of

the gas at the gas turbine  expander  inlet was 996°C, and the gas  turbine

expander cooling airflow was 8.3 percent.
                                      66

-------
     The Alternative Case I configuration is shown in Figure 28.   A
Ducon cyclone separator was used for the primary bed effluent,  a  Tan Jet
cyclone separator for the CBC effluent, and a Ducon granular-bed  filter
for the final separator.  Grade efficiency plots for these separators
are given in Appendix A.  The fluidized-bed design conditions specific
to Alternative Case I are as follows:
     Primary bed
        Excess air - 0 percent
        Bed depth - 4.6 m
        Superficial velocity - 1 m/s*
        Carbon losses - 10 percent of equivalent energy in coal
     CBC bed
        Excess air - 0 percent
        Bed depth - 4.6 m
        Superficial velocity - 0.85 m/s
        Carbon losses - 10 percent of  input
        Temperature - 1010°C

     Combustion  losses
        Incomplete  combustion
        Losses to atmosphere
        Sensible heat in solids
        Desulfurization reactions  •
                             Total
     Overall  combustion efficiency  - 96.85 percent.

     The  effectiveness  of  the  submerged heat exchangers  in the primary
 and  CBC beds  for indirectly  heating  part of the gas turbine working
 fluid  were assumed  to  be 85  percent.   This assumption gives an air
 outlet temperature  of  907°C  from both the primary and CBC bed  heat
 exchangers.   After  the  combustion products and the indirectly  heated air
 *Superficial velocity reduced below 1.5 m/s nominal design value to
  satisfy limits on maximum bed depth.
                                     67

-------
                                                                    Dwa. 1701B60
                     Primary
                     Bed
                     Cyclone \i
     Fluidized Bed
     Combustor
                                                                   Back-flushing
                                                                   Air
                                                  Granular -
                                                  Bed
                                                  Filter
Coal&
Sorbent
                                Ash-Char.
                                & Sorbent
                                Fines
                                                                     Steam Turbine/
                                                                     Generator Set
Booster
Compressor
Generator
                                                         Heat
                                                       Recovery
                                                         Steam
                                                       Generator
        Figure 28  -  Combined-Cycle  System Utilizing  Fluidized-Bed
                      Combustion with Indirect  Heating of Part  of
                      the Working Fluid (Alternative Case I - CBC)
                                         68

-------
have been mixed, and radiation losses and heat transferred between the
hot products and the combustion air have been allowed for, the gas tur-
bine expander inlet temperature is only 928°C.

     The Alternative Case II configuration is shown in Figure 29.  This
configuration is obviously considerably less complicated than that with
the CBC shown in Figure 28.  A Ducon cyclone was used for the first-
stage separator, a Ducon granular-bed filter for the second stage.
Grade efficiency plots for these separators are given in Appendix A.
The fluidized-bed design conditions specific to Alternative Case II were
as follows:

     Excess air - 60 percent
     Superficial velocity - 1.5 m/s
     Bed depth  - 3.3 m
     Combustion losses
         Incomplete  combustion      -   1.00 percent
         Losses  to atmosphere       -   0.80
         Sensible heat in  solids    -   1.25
         Desulfurization  reactions  -   0.10
                            Total      3.15
     Overall  combustion  efficiency -  96.85  percent.

     The effectiveness of the  heat transfer surface  submerged in the bed
 was again assumed  to be  85 percent,  giving  an air  temperature out of the
 heat exchanger  of  907°C.   After the  combustion products have been mixed
 with the indirectly heated air, and  allowances for losses to the atmo-
 sphere and heat transferred from the hot combustion products to the com-
 bustion air have been made, the temperature at the gas turbine expander
 inlet is 942°C.

      Alternative Case I  is probably not a practical configuration
 because of the likelihood of severe corrosion of the immersed air heater
 tubes in a bed with zero excess air.  It does, however, represent one
 boundary of the design spectrum for partially indirectly heated systems
 (the other boundary being the Base Case).
                                     69

-------
                                                     Dug. 7719A58
                          Back-flushing
                          Air
                      Fluidized-Bed
                      Combustor

                     Spent
                     Sorbent
                  Coa1& I
                  Sorbent I
                                         Granular-Bed
                                         —,  Filter
                  Booster
                  Compressor
                                       Steam Generator
                                             Steam Turbine/,
                                             Generator Set
    Figure 29 -
Combined-Cycle System Utilizing  Fluidized-Bed Combustion
with Indirect  Heating of Part of  the  Working Fluid
(Alternative Case II - no CBC)
 RESULTS OF  STUDY


      A summary  of the plant performance for the cases studied is given
 in Table 12.


      All three  of these  configurations  have heat rates that are  appre-

 ciably better than that  for a conventional  steam plant with FGD  (i.e.

 10,475 kJ/kWhl?)  with the  admittedly optlmistic bed teraperature  Qf

 1010°C.   If  limitations  on bed temperature  per  se and/or hot gas duct

 temperature  require the  bed temperature to  be reduced significantly, the

 heat  rates  for all configurations will be increased uniformly.   If lim-

 itations on  the maximum metal  temperature of the  in-bed  air tubes

 require the  teraperature of  the  indirectly heated  air to  be reduced sig-

nificantly,  the  heat rates  of  the partially indirectly heated  configu-

rations will increase  relative  to that  for the Base  (adiabatic) Case.
                                      70

-------
                                Table  12
                     SUMMARY  OF  PLANT  PERFORMANCES


Case

Gas Turbine
Output,
MW/G.T.
Module
Steam
Turbine
Output,
MW/G.T.
Module
Total
Electrical
Output,
MW/G.T.
Module
Coal Feed
Rate,
ton/hr/
G.T.
Module


Heat Rate,
kJ/kWh
 ?ase
 Adiabatic)
Alternative 1
Alternative II
73.8

66.3
67.8
34.1

27.4
28.7
107.9

 93.7
 96.5
36.2

32.7
33.4
9148

9587
9504
     Plot plans of a single gas turbine module for the Base Case, Alter-
native Case I, and Alternative Case II are shown in Figure 30 through
32.  Table 13 summarizes the plant design configurations for plants with
a nominal capacity of 400 MW.

     Estimates of the capital cost of nominally 400 MW plants for each
of the three configurations were made on the basis of the following
assumptions:
     Cost base - 1st quarter of 1976
     Construction time - 4 years
     Indirect construction costs - 13.5 percent of the total direct
     costs*
     Professional services - 10% of the direct plus indirect costs
     Contingency -  10 percent
     Escalation rate - 6-1/2 percent
     Interest during construction - 10 percent
     Expenditure rate - S curve supplied by NASA for use  in  the
     EGAS studyl7
 *Equivalent  to ~50 percent of direct installation costs.
                                    71

-------
                                                                   Dwp. 7681 13
   Compressor   FBC
             o
Air
                                         Granular-Bed
                                             Filters
                                                                            Generator
    Figure 30 -  Plot Plan of Single Gas  Turbine  Module  for Base  Case
 Compressor      FBC
Granular-Bed
   Filters
                                                                 0 ".  'S31A1?
                                                                        Generator
                                                             Expander
              Carbon
              Burnup
              Cell
         Figure 31 -  Plot Plan of  Single Gas  Turbine Module  for
                       Alternative Case 1
                                        72

-------
                                                             Dwr. 7681A14
                                    Granular-Bed
                                       Filters
Compressor
FBC
                            Cyclones
Oi
                                        <>
                                                       Expander  ,
                                                                Generator
         Figure 32  -  Plot Plan of Single Gas Turbine Module for
                     Alternative Case  II
     Cost escimates for equipment manufactured by Westinghouse, such as
the gas turbines,  heat recovery  steam generators, and steam turbines,
were obtained from cost correlations  supplied by the pertinent Westing-
house divisions during EGAS.1''   Cost  estimates for high-temperature par-
ticulate removal equipment were  based on information obtained from
equipment suppliers during EGAS.  Cost estimates for FBC equipment,
solids feeding equipment, and in-bed  heat transfer surface were made
using procedures that originated in the Evaluation of the Fluidized Bed
Combustion Processes^ and were  used  in EGAS.

     Summaries of the capital cost estimates for  the  Base Case, Alter-
native Case I, and Alternative Case II are given  in Tables  14  through
16.  The only costs that vary significantly with  system configuration
are  those for the FBC modules and the gas-cleaning equipment.   While  the
variations in the cost of these two components are opposite (e.g.,  the
                                    73

-------
                                 Table  13

                  SUMMARY OF PLANT DESIGN  CONFIGURATIONS
Configuration
Case
Base
Alternative I
Alternative II
 Capacity, MW                   431.6
 No.  of  Gas Turbines              4

 Combustion Modules
  Number                         8
  Diameter, m                    3.8
  Height, m                     25.6
  Beds/module                    4
  Bed depth, m                   1.98

 CBC  Modules
  Number                         —
  Diameter, m                    —
  Height, m
  Beds/module                    —
  Bed depth, m

 Ist-Stage Separator Modules
  Volumetric flow, am-Vs        29.3
  Number                        16
  Diameter, m                    0.76
  Height, m                      3.0

 CBC  Separator Modules
  Volumetric flow, am^/s
  Number
  Diameter, m
  Height, m

 2nd-Stage Separator Modules
  Volumetric flow, am-Vs        14.6
  Number                        32
  Diameter, m                    7.77
  Height, m                      8.84

No.  of HRSG Modules              4

No.  of Steam T-G Modules         1
374.8
  4
  8
  3.65
 39
  4
  4.57
  4
  3.65
 10.97
  1
  4.57
 16.1
  8
  1.0
  4.0
  3.21
  4
  2.6
  3.05
 17.9
  8
  8.23
  9.3

  4

  1
386.0
  4
  8
  3.65
 41
  5
  3.35
 25.3
  8
  1.3
  5.25
 12.7
 16
  7.01
  8.23

  4

  1
                                    74

-------
                                Table 14
                  CAPITAL COST ESTIMATE FOR BASE CASE
                            (4 gas turbines)
                             Item                                 106 $
1.00 Land and Land Rights                                          4.600
2.00 Structures & Improvements (on-site waste disposal)           13.845
3.00 Heat Rejection System                                         3.605
4.00 Material Handling and Storage                                16.445
5.00 Energy Conversion
     PFBC                                                          2.381
     Combustion air piping  v                                      0.827
     Transport air subsystem                                       1.546
     Gas cleaning                                                 41.341
     Refractory-lined pipe                                         1.608
     Refractory- and metal-lined pipe                              2.368
     Gas turbine/generator                                        36.542
     Steam turbine/generator                                       7.606
     HRSG                                                          6.366
                                            Subtotal              100.585
6.00 Auxiliary Mechanical Equipment                                5.037
7.00 Auxiliary Electrical Equipment                                9.614
8.00 Station  Transmission Equipment                                2.403
     Total direct costs                                           156.134
     Indirect construction costs  (13.5% of  total  direct  costs)    21.078
                                            Subtotal              177.212
     Professional services (10% of direct and  indirect costs)     17.721
                                            Subtotal              194.933
     Contingency (10%  of above)                                    19.493
                                            Subtotal              214.426
     Escalation during construction  (6 1/2-4)  (15.8% of  above)     33.879
     Interest during  construction (10-4)  (21.4% of above)          45.887
     Total  capitalization                                         294.193
                                     75

-------
                                Table 15

              CAPITAL COST ESTIMATE FOR ALTERNATIVE CASE I

                             Item                                 106
1.00 Land and Land Rights                                          4.144

2.00 Structures and Improvements (on-site waste disposal)         12.473

3.00 Heat Rejection System                                         3.108

4.00 Material Handling and Storage                                14.815

5.00 Energy Conversion

     PFBC                                                         12.614
     Combustion air piping                                         0.836
     Transport air subsystem                                       1.452
     Gas cleaning                                                 13.408
     Refractory-lined pipe                                         0.926
     Refractory- and metal-lined pipe                              3.032
     Gas turbine/generator                                        36.540
     Steam turbine/generator                                       6.400
     HRSG                                                          5.357

                                            Subtotal              80.565

6.00 Auxiliary Mechanical Equipment                                4.342

7.00 Auxiliary Electrical Equipment                                8.288

8.00 Station/Transmission Equipment                                2.072

     Total direct costs                                          129.807

10.0 Indirect construction costs (13.5% of total direct costs)     17.524

                                            Subtotal             147.331

11.0 Professional services (10% of direct and indirect costs)     14.733_

                                            Subtotal             162.064

12.0 Contingency (10% of above)                                   16.206
                                            Subtotal             178.370

13.0 Escalation during construction (6 1/2-4) (15.8%)             28.183

14.0 Interest during construction (10-4)  (21.4%)                  38.171

     Total capitalization                                        244.724
                                    76

-------
                                Table 16

              CAPITAL COST ESTIMATE FOR ALTERNATIVE CASE I

                             Item                                 106 $

1.00 Land and Land Rights                                          A.A55

2.00 Structures and Improvements (on-site waste disposal)         13.A08

3.00 Heat Rejection System                                         3.372

4.00 Material Handling and Storage                                15.926

5.00 Energy Conversion

     PFBC                                                         10.09A
     Combustion air piping                                         0.836
     Transport air subsystem                                       1.517
     Gas cleaning                                                 18.583
     Refractory-lined pipe                                         0.688
     Refractory- and metal-lined pipe                              3.200
     Gas turbine/generator                                        36.SAO
     Steam turbine/generator                                       6.AOO
     HRSG                                                          5.608

                                            Subtotal              83.A66

6.00 Auxiliary Mechanical Equipment                                A.711

7.00 Auxiliary Electrical Equipment                                8.992

8.00 Station/Transmission Equipment                                2.2A8

     Total direct  costs                                           136.578

10.0 Indirect  construction  costs  (13.5% of  total  direct costs)     18.A38

                                            Subtotal              155.016

11.0 Professional  services  (10% of direct and indirect  costs)      15.501

                                            Subtotal              170.517

12.0 Contingency (10% of above)                                   17.052
                                             Subtotal              187.569

 13.0 Escalation during construction (6 1/2-A) (15.8%)             29.636

 1A.O Interest during construction (10-A) (21.A%)                   A0.1AO

     Total capitalization                                        257.3A5
                                     77

-------
adiabatic system has the highest gas cleaning cost and the lowest com-
bustion cost), they are not equal, and the specific costs of the the
three configurations vary significantly, as shown in Table 17.  The
adiabatic system is estimated to have the highest specific capital cost
and Alternative I the lowest.
                                Table 17
                        SPECIFIC COST COMPARISON
               Case
          Base
          Alternative I
          Alternative II
Specific Cost-$/kW
       682
       653
       667
     The cost of electricity was calculated for the three cases studied
on the basis of the following assumptions:

     Life of plant - 30 years
     Annual charge - 18 percent
     Capacity factor - 65 percent
     Fuel cost (1st quarter of 1976) - $0.95/GJ
     Fuel escalation rate - 5%/year
     O&M (including dolomite) -2.5 mills/kWh

     Table 18 summarizes the energy costs for the three cases.

                                Table 18

                 COST OF ELECTRICITY SUMMARY, mills/kWh
Item
Capital
Fuel
O&M
Total
Base
21.6
21.5
2.5
45.6
Alternative I
20.6
22.6
2.5
45.7
Alternative II
21.0
22.3
2.5
45.7
                                    78

-------
This table shows that the variations in specific capital costs of the
three systems are nearly balanced by the variations in fuel consumption
and that the spread in cost of electricity among the three cases is
about 0.2 percent.  Since the uncertainties in estimating the cost of
the cost-variable components (the fluidized-bed combustor modules and
the gas cleaning equipment) are considered large as compared to the
spread in cost of energy, no significance can be attributed to this
spread in cost of energy.

     The in-bed heat transfer surfaces used in this study were plain
tubes placed in horizontal array.  No attempt was made to evaluate the
finned/finned-tubing concept that was proposed for the partially indi-
rectly heated cycle by Curtiss Wright.

PARTICULATE CONTROL/GAS TURBINE EXPANDER EROSION CONSIDERATIONS

     The requirements for particulate removal from the combustion  prod-
ucts of PFBC for  economical gas turbine expander life are predicted  to
be  in excess of  those for meeting emission limits.21  For that  reason
this assessment  of the merit of the  partially indirectly heated  concept
emphasizes  the  control of gas turbine expander erosion and  deposition
rather  than particulate  emissions.

     Erosion of  and  deposition on gas turbine expander parts  are func-
tions of  the concentration,  size distribution, and physical properties
of  the  particles entrained  in the working  fluid.   (They are also func-
tions of  the size and design of  the gas  turbine  expander, but considera-
tion of the latter aspect  is outside the scope  of  this  study.)

     The  particulate in the products of  combustion from fluidized beds
with in situ desulfurization has  three  components:   ash from the coal,
fines  from the  desulfurization  sorbent,  and  unburned carbon.

     The ash from FBC of coal consists  of friable  platelets that have an
erosivity substantially less than the fused  cenospheres generated in the
combustion of  pulverized coal.   All of the ash in the coal is assumed to
                                     79

-------
 be entrained in  the products of combustion, either as free ash or asso-
 ciated with unburned carbon.  Actually, small quantities of ash are car-
 ried out with  the coarse spent sorbent that is removed directly from the
 fluidized bed.   The size distribution of  the free ash is based on data
 given in Reference 22.

     For this  study we have assumed desulfurization using once-through
 dolomite.  The dolomite feed is single screened so it contains a sig-
 nificant amount  of fines that are elutriated from the bed almost immedi-
 ately after the  dolomite is injected into it.  The amount of excess sor-
 bent is small  (50%) and the bed volume is large, so the residence time
 of the coarse dolomite is long (~10 hr).  As a result, there is a sig-
 nificant amount  of attrition and decrepitation of the coarse dolomite
 and subsequent elutriation of the sorbent fines generated thereby.  The
 size distribution of the attrited and decrepitated sorbent fines was
 also based on data given in Reference 22.

     The quantity of unburned carbon (char) entrained in the combustion
 products from a  fluidized bed is primarily a function of bed temperature
 and the amount of excess air.  Bed depth, feed particle size and distri-
 bution, and superficial velocity also are factors.  Fluidized-bed com-
 bustion efficiency is based on data given in Reference 23.  The size
 distribution of  the char is assumed to be the same as that for the coal
 feed for those particles having diameters smaller than the particle
whose terminal velocity is equal to the bed superficial velocity.  The
 char composition is assumed to be that of devolatilized high-volatility
bituminous coal.

     The concentration and size distribution of particles in the dis-
 charge of the particulate removal equipment are based on separate calcu-
 lations for each of the three constituents.  This individual calculation
 is necessary because of the differences in the density of ash, spent
 sorbent, and char, which have a significant effect on the performance of
centrifugal separation equipment.
                                    80

-------
     Appendix B contains detailed information on the concentration and
size distribution of ash, sorbent,  and char particles at various stages
in the particulate removal subsystems for Alternative Case I.   Summaries
of this information are given in Figures 32 through 34 for the Base
Case, Alternative Case I, and Alternative Case II.   Table 19 compares
the particulate loadings at the gas turbine expander inlet for each of
these cases.

                                Table 19

                    EXPANDER INLET PARTICLE LOADING
Loading, g/sm^
Sorbent
Ash
Char

Base
0.0050
0.0030
0.00004
0.00804
Alternative I
0.0033
0.0031
0.00073
0.00713
Alternative II
0.0038
0.0025
0.00006
0.00636
     This table shows that the amount of char is less than 1 percent of
the total particulate for the Base Case and for Alternative Case II and
about 10 percent for Alternative Case I, which has the CBC.  Since there
is a good possibility that the fine carbon particles in the products
of combustion will be oxidized after the bypass air is added, the
amount of char shown for Alternative Case I is considered to be of no
significance.

     The concentrations of ash in the alternative cases are within
about 15 percent of the value for the Base Case so the variance is
insignificant.

     The only variation considered to be significant is that in the sor-
bent concentration, with the values for the alternative cases being 25
to 35 percent lower than that for the Base Case.  Since the sorbent is
the most erosive constituent of  the particulate,  this difference might
                                     81

-------
                   0.01
oo
to
                      1 -
                •Z    10 -
                I
                     90 -
                     99 -
                   99.99
 Participate loading at Expander Inlet
Sorbent ttOOSOg/sm3    (a 0022 gr/scfl
       a0030g/sm3    (0.0013gr/scf)
       0.00004 g/sm3   (0.00002 gr/scft
       0.00804 g/sm3   (0.00352 gr/scf)
    °  Char
    o  Ash
    a  Sorbent
                                            10
                                   10"
                                                                     L   I  I  I
                                                   1st Stage
                                                   Separator
                                                   Inlet
                             Particulate Loading at 1st*
                             Stage Separator Inlet
                              Sorbent  & 17 g/sm3  (2.26 gr/scf)
                              Ash     3.02 g/sm3  (1.32 gr/scf I   -
                              Coal     0.14q/sm3  (a 06 gr/scf I
                                      8.33g/sm3  (3.64 gr/scf)
                                 i    i  i r	i	i    i  i
101
102
10"
                                                                                 Particle Size-urn

                           Figure  33  -  Summary  of  Particulate Loading and  Size Distribution for  Base Case

-------
                                                                                                                                 Curve 68951 5-B
00
u>
                 0.01
             o>
                   10
                  50
90
                  99 -
               99.99
            "I	1—TTT
            a  Char
            o  Ash
            A  Sorbent
                Partlculate Loading
                at Ejqpander Inlet
      Sorbent 0.00332 g/sm3   (0.00148 gr/scf)
   -  Ash     0.00309 g/sm'   
-------
 be  expected  to  significantly affect  the life of the gas  turbine expander
 vanes  and  blades.  Examination of  the size distribution  plots for the
 sorbent  particles  entering  the expander inlet for  the three cases (Fig-
 ures 33  through 35) shows,  however,  that  100 percent of  the particles
 are smaller  than 10 pm and  50 percent are smaller  than 2 ym.  A recent
 survey^ of  turbine manufacturers  indicated that 0.009 to 0.045 g/sm^ of
 particles  larger than 10 vm is tolerable.  We conclude,  therefore, that
 none of  the  cases  would have an  erosion problem if, in fact, a granular-
 bed filter or other device  having  the performance  assumed becomes a com-
 mercial  reality.

     If  the  performance assumed  for  the granular-bed filter cannot be
 achieved in  a practical separation device with the feed  sorbent size
 distribution used  here, an  excessive amount of particle  larger than
 10  ym may  be present at the expander inlet.  Since a substantial frac-
 tion of  the  sorbent elutriated from  the beds are fines present in the
 sorbent feed, the  use of double-screened sorbent would be expected to
 reduce substantially the amount  of sorbent in the  combustion products
 going to the gas turbine expander.  This suggests  that there may be a
 trade-off  between  the cost of double-screened sorbent and the cost of
 replacing  gas turbine expander parts.  Consideration must be given, how-
 ever, to the effect of the use of double-screened  sorbent on the effec-
 tiveness of  its desulfurization.

ENVIRONMENTAL CONSIDERATIONS

     We have estimated the emission performance of the three configura-
 tions using available process models and emission data for PFBC.   Using
a Ca/S atom feed ratio of 1.5 for all three configurations,  based on an
average activity dolomite such as Tymochtee, we determined the sulfur
 removal efficiency.  In addition, we estimated the NOX,  CO,  and particu-
late emissions and projected the  resulting solid waste product rate.
These estimates are shown in Table 20.

     Table 20 indicates that the  three configurations would satisfy all
of  the current NSPS for coal-fired boilers :   all S02 emission of
                                    84

-------
                                                                                                                             Curve 690376-6
00
                                 Particulate Loading
                                   at Expander Inlet:
                           Sorbent  0.00375g/sm3   (0.00164 gr/scf)
                           Ash      0.00247 g/sm3   (0.00108 gr/scf)
                           Char     0.00006 g/sm3   (0.00003 gr/scf)
                                   0.00630 g/sm3   (0.00275 gr/scf)
              Ist-Stage
            Cyclone Inlet
                                 0 Char
                                   Ash
                                   Sorbent
                                                                                                  Sorbent  & 874 g/sm
(3.877 gr/scf)
(Z 581 gr/scf)
(0.301 gr/scf)
                 99.99
                                         10
                                                                                  10*
                                                                            Particle Size - \m
   103
                      Figure  35  -  Summary of PartiCulate  Loading and  Size  Distribution for Alternative  Case II

-------
                                Table 20
             ENVIRONMENTAL COMPARISON OF THE CONFIGURATIONS
Configuration
Base Case

Alternative
Case I
Alternative
Case II
S02,
ng/J
(Ib/MBtu)
284
(0.66)
116
(0.27)
116
(0.27)
NO
ng/J
(Ib/MBtu)
215
(0.5)
86
(0.2)
150
(0.35)
CO,
p£m
50

300
100

Particulate,
ng/J
(Ib/MBtu)
7.3
(0.017)
7.7
(0.018)
6.5
(0.015)
Solid kg/kg
Waste, kg/kg
(Ib/lb coal)
0.38
(0.38)
0.38
(0.38)
0.38
(0.38)
516 ng/J (1.2 Ib/MBtu), an NOX emission of 301 ng/J (0.7 Ib/MBtu), and a
particulate emission of 43 ng/J (0.1 Ib/MBtu).  The base configuration
would achieve 90 percent sulfur removal at the selected Ca/S feed ratio
of 1.5, while the two options would achieve 96 percent sulfur removal
because of significantly longer gas residence in the fluidized-bed com-
bustor.  Sulfur losses from the CBC have been accounted for in the esti-
mate for Alternative Case I.  The desulfurization efficiency for the
Base Case could be increased at a modest cost, however, by increasing
the bed depth.

     The oxides of nitrogen would vary considerably among the configura-
tions because of the variation in excess air levels.  The low excess air
level in Alternative Case I would result in a low NOX emission but a
relatively large CO emission.  Unburned hydrocarbon emissions may also
be significant in Alternative Case I, but little information is avail-
able to make such a projection.

     Particulate emissions would be comparable for the three configura-
tions and much less than the environmental limit if the particulate con-
trol system selected for turbine protection were to be used.  The solid
                                    86

-------
waste produced by the three configurations would be similar in rate of
production and in physical/chemical properties.   Small differences could
exist in the particle size distributions of the waste materials, but the
resulting environmental factors (e.g.,  leaching characteristics) would
be expected to be very similar.  The solid waste production could be
reduced slightly in the alternative cases by operating at the smaller
Ca/S feed ratios required to yield the 90 percent sulfur removal effi-
ciency of the base configuration, or a Ca/S ratio of about 1.2 for
Alternative Case 1 and 1.3 for Alternative Case II.  The solid waste
would thus be reduced to 0.34 and 0.35 kg/kg coal, respectively.  This
reduction in ratio could also result in a reduction in particulate emis-
sions for the alternative cases.

     The emissions from the three configurations would also be  sensitive
to the properties of the coal and sorbent selected for operation.  The
coal ash and sulfur content directly affect the fluidized-bed combustor
control requirements, while dolomites vary significantly in sulfur
removal activity, attrition resistance, and trace  element content.

CONCLUSIONS

     As a result of  this  evaluation  of  partially indirect heating  of  the
working fluid for a gas turbine  with a  pressurized fluidized-bed  com-
bustor we conclude that:

     •  With the projected granular-bed filter  performance, gas
        turbine  expander  erosion problems are not  anticipated  for
        either  the base (adiabatic)  configuration  or  the  two  par-
        tially  indirectly heated configurations.   If,  however,  the
        particulate  removal  performance projected  herein  cannot be
        attained on  a  commercial basis, both  of the  partially
        indirectly heated alternatives  would  have  a  potential for
        significantly  larger expander life because of lower sor-
        bent fines concentration in the gas entering the  expander.
                                     87

-------
•  With the replacement cost of gas turbine expander parts
   assumed to be uniform, the estimated costs of electricity
   for the three configurations considered are essentially
   equal.  The variations in capital costs among the three
   cases would be balanced by variations in heat rate*
•  Acceptable environmental performance is indicated for all
   three configurations.  The partially indirectly heated
   alternatives indicate a potential for environmental per-
   formance significantly better than that of the Base Case.
•  All three of the configurations considered have a poten-
   tial for heat rates appreciably better than that of a
   conventional coal-fired steam plant with flue gas
   desulfurization.
                              88

-------
                           9.   REFERENCES

1.   Newby,  R.  A.,  and D.  L.  Keairns,  Alternatives to Calcium-Based 862
    Sorbents for Fluidized-Bed Combustion:  Conceptual Evaluation,
    report  to EPA, Westinghouse Research and Development Center,
    Pittsburgh, PA, January 1978, EPA-600/7-78-005.
2.   Newby,  R.  A.,  N. H.  Ulerich, E. P. O'Neill, D. F. Ciliberti, and
    D. L. Keairns, Effect of S02 Emission Requirements on Fluidized-Bed
    Combustion Systems;  Preliminary Technical/Economic Assessment.
    Report  to EPA, Westinghouse Research and Development Center,
    Pittsburgh, PA, August 1978, EPA-600/7-78-163, NTIS PB 286 811/
    7 ST.
3.   Keairns, D. L., et al. Fluidized Bed Combustion Process
    Evaluation - Phase 11 - Pressurized Fluidized Bed Coal Combustion
    Development, Report to EPA, Westinghouse Research Laboratories
    Pittsburgh, PA, September 1975, EPA-650/2-75-027c, NTIS PB 246-116.
4.   Treybal, R. E., Mass-Transfer Operations, New York:  McGraw-Hill
    Book Co., Inc.; 1955.
5.   Riesenfeld, F.  C., A. L. Kohl, Gas Purification, Second Ed.,
    Houston:  Gulf  Publishing Co.; 1974.
6.   Perry, R. H.,  C. H. Chilton, Chemical Engineer's Handbook, 5th Ed.,
    New York:  McGraw-Hill  Book  Co.;  1973.
7.  Hoke, R. C., Emissions  from  Pressurized Fluidized Coal Combustion.
    Proceedings of  the 4th  International  Conference  on Fluidized  Bed
    Combustion, December  1975, McLean, VA:  The  Mitre Corporation;
    1976; and  Monthly Reports  to EPA.
8.  Advanced Coal  Gasification  System for Electric  Power Generation,
    Quarterly  Progress Reports  for ERDA,  Westinghouse Electric  Corpora-
    tion, Pittsburgh, PA,  October  1975 and  January  1976, Contract No.
    E(49-18)-1514.
                                    89

-------
 9.  Smith, J., et al., Bureau of Mines Coal-Fired Gas Turbine Project,
     RI 6920, U. S. Department of Interior, Bureau of Mines, 1967.
10.  Atkin, M. L., and G. A. Duke, The Operation of a Modified Ruston
     Hornsby Gas Turbine on a N.S.W. Bituminous Coal, Aeronautical
     Research Laboratories Report 133, Department of Supply, Australian
     Defense Scientific Service, 1971.
11.  Ulke, A., and W. T. Rouleau, The Effect of Secondary Flows on Tur-
     bine Blade Erosion, ASME Paper 76-GT-74.
12.  Beecher, D. T., et al., Energy Conversion Alternatives Study -
     Westinghouse Phase II Final Report, Vol. Ill, Summary and
     Advanced Steam Plant with Pressurized Fluidized-Bed Boiler,
     November 1, 1976, Contract NAS 3-19407.
13.  McCain, J. D. , Evaluation of Rexnord Gravel Bed Filter, Report to
     EPA, EPA-600/2-76-164, NTIS PB 255 095.
14.  Spagnola, H.,  Operating Experience and Performance at the Sunbury
     Baghouse, Symposium on Particulate Control in Energy Processes,
     San Francisco, CA, May 11-13, 1976.
15.  Goldstein, S., Modern Developments in Fluid Dynamics, Volume II,
     New York:  Oxford University Press; 1938.
16.  Moskowitz, S., and G.  Weth, Pressurized Fluidized Bed Pilot Plant
     for Production of Electric Power using High Sulfur Coal.  Proceed-
     ings of the 12th Intersociety Energy Conversion Engineering Confer-
     ence, Washington, DC,  August 28 - September 2, 1977, pp. 696-703.
17.  Beecher, D. T., et al., Energy Conversion Alternatives Study -
     Westinghouse Phase II  Final Report, Vol. I; Summary and Combined
     Gas-Steam Turbine Plant with Integrated Low-Btu Gasifier,  Contract
     NAS 3-19407, November  1,  1976.
18.  Keairns, D. L., D. H.  Archer, R.  A. Newby, E.  P. O'Neill,
     E.  J. Vidt, Evaluation of the Fluidized Bed Combustion Process,
     Vol.  I.   Report to EPA,  Westinghouse Research Laboratories, Pitts-
     burgh,  PA, December 1973, EPA-650/2-73-048a,  NTIS PB 231-162.
19.  Coal-Fired Power Plan:  Capital Cost Estimates,  Report to  EPRI,
     Bechtel  Power  Corporation;  January 1977, EPRI AF-342.
                                    90

-------
20.   Archer,  D.  H.,  D.  Lt  Keairns,  J.  R.  Hamm,  et al.,  Evaluation of the
     Fluidized Bed  Combustion Process,  Vol.  I,  II, and  III.   Report to
     the Office of  Air Programs, EPA,  Westinghouse Research Labora-
     tories,  Pittsburgh,  PA; December 1971,  NTIS PB 211-494, 212-116,
     AND 213-152.
21.   Menguturk,  M.,  and E. F. Sverdrup, Tolerance of a Large Electric
     Utility Turbine to Erosion Damage by Coal Gas Ash Particles, ASTM
     Symposium of Erosion, Vail, CO, October 24-26, 1977.
22.   Merrick, D. , and J.  Highley, Particle Size Reduction and Elutrta-
     tion in a Fluidized Bed Process, Recent Advances in Pollution Con-
     trol, AIChE Symposium Series No. 137, Vol. 70; p.  366-78.
23.   Hake, R.C., et al.,  Studies of the Pressurized Fluidized-Bed
     Combustion  Process, Report to EPA, Exxon Research and  Engineering  Co.,
     Linden, NJ, September  1977, EPA-600/7-77-107.
24.   CFCC Development Program - Advanced Clean-up Hardware Performance
     Guidelines for Commercial Plant (Task 4.1.2), March 1978,
     FE-2357-37.
                                    91

-------
                     APPENDIX A




GRADE EFFICIENCIES FOR PARTICIPATE REMOVAL EQUIPMENT
                          92

-------
                                                              Dug. 6429A58

                                                             Final Stage
Coal

Sorb
      Primary
        Bed
                           V
                  Oucon
                  Primary
                  Cyclone
                                                                GBF
 Ash and Spent
-"-* Sorbent
        Air
 C
 B
 C


T
Air
                                                  I Tan Jet

                                                  ICBC Cyclone
Ash&
Attrited Sortent
            Figure A-l -  Schematic  of Particle Control System

                                                         Curve 694490-A

                                                    Specific Gravity

                      a 01
                       0.1 -
                                      181.35.88
                     i


                     1

                     I
                      99.9
                      99.99
        Figure  A-2 -  Grade Efficiency of  Primary  Cyclone Separator
                                           93

-------
                                                                                  Curve 687623-B
0.01

 0.1


   1



  10


  30

  30

  70


  90


  98
  99


99.9
                                            Specific Gravity
                                             2.88 .88.64
99.99
                Tan Jet
                      TM
                                                          T= 9Z7°C(1700°F)
                                                          P= 1013kPa(10Atm)
                                                               i    ill
                                                                                      i    i   r
                                                                                         i     i  i
.10
                            1.0
                                                    10.0
100
                                           Dp.
            Figure A-3  - Grade Efficiency Curve for  Tan Jet

-------
                                       Curve 687622-A
       0.01
        o.it-
         10


         30
     *
     d   50
     o

     1   70


         90
          98
          99


        99.9
        99.99
0.10
                       1	1  IT
   i   i  I
       LO
Particle Size, pm
                                                   10.0
Figure A-4 - Grade Efficiency  of Granular-Bed  Filter
                               95

-------
                               APPENDIX B

          PARTICLE SIZE DISTRIBUTION AT PERTINENT STATIONS IN
         PARTICULATE REMOVAL SUBSYSTEM FOR  ALTERNATIVE CASE I
     Analyses of particle size distributions and concentration were made
at each pertinent station in the particulate removal subsystem for each
case studied.  Plots of the size distribution at each station for
Alternative Case I are included in this appendix.  Figure B-l gives the
size distribution plots for sorbent particles, Figure B-2 gives those
for ash particles, and Figure B-3 gives those for char particles.

     The concentrations of sorbent, ash, and char particles at each of
the pertinent stations are given in Table B-l.

                               Table B-l
Solids Concentrations, g/sm^
Station So
rbent Ash
Primary Bed Exit 10.35 10.92
1st Stage Separator Exit 0.55 0.69
Carbon Burnup
CBC Separator
GBF Inlet
GBF Exit
Cell Exit 9.52 8.86
Exit 1
.72 1.31
0.67 0.76
0.011 0.0085
GT Expander Inlet 0.0034 0.0026
Char
8.38
0.21
1.08
0.021
0.20
0.0025
0.0007
Total
29.64
1.44
19.46
3.04
1.63
0.022
0.0067
                                   96

-------
                                                                                                                         Curve 689518-B
                    0.01
VO
                .
                £
                      10
                     50
                     90
                     99
                   99.99
—i—r—r~i	1	1—r—r
       Case I - Sorbent
 o  Feed to Primary Bed
 a  To ht-Stage Separator
 o  From Ist-Stage Separator
 ?  To CBC
 o  From CBC to CBC Separator
 &  From CBC Separator
 a  To Granular Bed Filter
 o  From Granular Bed  Filter
                                         10
                                                                          Particle Size,  urn

                         Figure B-l - Size  Distributions of Sorbent  Particles  for Alternative  Case  I

-------
                                                                                                                        Curve 689516-B
VO
oo
                     0.01
                 .s-    10
                 2    50
                 I
                 •E
                 i    »
                      99
                    99.99
—i—r—r-|	1	
       Case I - Ash
 o To Ist-Stage Separator
 a From Ist-Stage Separator
 o From C8C Cyclone
 a To Granular Bed Filter
 o From Granular Bed Filter
              ~i~ n
                                                                                        I     I    I  I  I      I     I    I  I  1
                                           «-l
                                                              10
                                                 10'
                                           Particle Size, Mm
10
io3
                                                                                                                                       Tr-
                                                                                                   I   1  I
                           Figure B-2 -  Size  Distributions  of  Ash Particles for  Alternative Case I

-------
                                                                                                                           Curve 689517-6
                     0.01
VO
                 .2*
                 o>
                      10
                      50
                      90
                      99 -
                    99.99
   I   I   I  I      I      I
      Case I - Char
o  Coal Feed to Primary Beds
D  To Ist-Stage Separator
o  From Ist-Stage Separator
a  To CBC
o  From CBC to CBC Separator
o  From CBC Separator
h  To Granular Bed Filter
»  From Granular Bed Filter
                                          10
                                                10
                                          Particle Size,
                            Figure  B-3 - Size  Distributions  of  Char Particles for  Alternative  Case I

-------
                                TECHNICAL REPORT DATA
                          (Please read Instructions on the reverse before completing)
 1. REPORT NO.
 EPA-600/7-80-015d
                           2.
                                                      3. RECIPIENT'S ACCESSION NO.
      AND SUBTITLE Experimental/Engineering Support for
 EPA's FBC Program: Final Report
 Volume 4.  Engineering Studies
            S. REPORT DATE
              January 1980
            6. PERFORMING ORGANIZATION CODE
 7. AUTHOR(S)
 J.R.Hamm, D. F.Ciliberti, R.W.Wolfe,
  R.A.Newby, and D. L.Keairns
                                                      8. PERFORMING ORGANIZATION REPORT NO.
 9. PERFORMING ORGANIZATION NAME AND ADDRESS
                                                      10. PROGRAM ELEMENT NO,
 Westinghouse Research and Development Center
 1310 Beulah Road
 Pittsburgh,  Pennsylvania 15235
            INE825
            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 REPORT AND PER
            Final; 12/75 - 12/78
                                                                       ERIOD COVERED
            14. SPONSORING AGENCY CODE
              EPA/600/13
 15. SUPPLEMENTARY NOTES JERL-RTP project officer is D. Bruce Henschel, Mail Drop 61,
 919/541-2825. EPA-600/7-78-163 also relates to this work.
 16. ABSTRACT The report gives results of engineering studies addressing several aspects
 of fluidized-bed combustion (FBC) system design and performance, as applied to
 coal.  It reviews an evaluation of the impact of SO2 emission requirements on FBC
 system performance and cost. Stringent SO2 emission requirements can be satis-
 fied economically if design and operating parameters are properly selected. An
 alternative SO2 control concept for pressurized FBC (PFBC), pressurized scrub-
 bing of the products of combustion with water,  is evaluated. The concept is not eco-
 nomically competitive because of reduced plant efficiency and the need for recuper-
 ative heating. A potential reduction in solid waste is realized with the concept, but
 the SO2 control efficiency may be limited. An evaluation of PFBC, examining the
 technical and economic trade-offs between the level  of particulate control achieved
 and the frequency of gas-turbine blade  replacement, is described. The evaluation
 incorporates models of PFBC particulate carry-over, particulate control device
 efficiency, and turbine erosion.  Also,  an indirect air-cooled PFBC concept is com-
 pared with other PFBC concepts. The indirect  air-cooled concept provides signifi-
 cant particulate control advantages over the adiabatic combustor PFBC concept,
 resulting in about 4% lower plant efficiency and 1% higher cost of electricity.
 7.
                             KEY WORDS AND DOCUMENT ANALYSIS
                DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COS AT I Field/Group
 Pollution           Dust
 Combustion         Gas Turbines
 Fluidized Bed Processing
 Coal
 Sulfur Oxides
 Scrubbers
Pollution Control
Stationary Sources
Particulate
13B       11G
21B       13G
13H,07A
21D
07B
131
 8. DISTRIBUTION STATEMENT
 Release to Public
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES

   111
20. SECURITY CLASS (Thispage)
Unclassified
                        22. PRICE
EPA Form 2220-1 (9-73)
                                        100

-------