Unrtad States
Environmental Protection
Agency
Municipal Environmental Research
Laboratory
Cincinnati OH 45268
EPA-600/2-80-095
Augubt 1980
Research and Development
A Field Test Using
Coal:dRDF Blends in
Spreader Stoker-Fired
Boilers

-------
                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 ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution-sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia  22161.

-------
                                      EPA-600/2-80-095
                                      August 1980
    A FIELD TEST USING COAL:dRDF BLENDS
     IN SPREADER STOKER-FIRED BOILERS
                    by

             Gerald H. Degler
              H. Gregory Rigo
      Boyd T. Riley, Jr. (Consultant)
      Systems Technology Corporation
            Xenia, Ohio  45385
          Contract No. 68-03-2426
              Project Officer

             Carlton C. Wiles
Solid and Hazardous Waste Research Division
Municipal Environmental Research Laboratory
          Cincinnati, Ohio  45268
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
    OFFICE OF RESEARCH AND DEVELOPMENT
   U.S. ENVIRONMENTAL PROTECTION AGENCY
          CINCINNATI, OHIO  45268

-------
                                  DISCLAIMER
     This report has been reviewed by the Municipal Environmental Research
Laboratory, U.S. Environmental Protection Agency, and approved for publi-
cation.  Approval does not signify that the contents necessarily reflect the
views and policies of the U.S. Environmental Protection Agency, nor does
mention of trade names or commercial products constitute endorsement or
recommendations for use.
                                      11

-------
                                  FOREWORD
     The U.S. Environmental Protection Agency was created because of increasing
public and government concern about the dangers of pollution to the health and
welfare of the American people.  Noxious air, foul water, and spoiled land are
tragic testimonies to the deterioration of our natural environment.  The
complexity of that environment and the interplay of its components require a
concentrated and integrated attack on the problem.

     Research and development is that necessary first step in problem solution;
it involves defining the problem, measuring its Impact, and searching for solu-
tions.  The Municipal Environmental Research Laboratory develops new and
improved technology and systems to prevent, treat, and manage wastewater and
solid and hazardous waste pollutant discharges from municipal and community
sources, to preserve and treat public drinking water supplies, and to minimize
the adverse economic, social, health, and aesthetic effects of pollution.  This
publication is one of the products of that research and provides a most vital
communications link between the researcher and the user community.

     In recognition of the fact that more than 50 percent of the roughly
42,000 industrial boilers in the U.S. are coal-fired boilers, the Environmental
Protection Agency undertook a project to investigate the technical and
environmental implications of using densified (pellet form) refuse derived
fuel as a substitute for stoker coal.  This report presents the results from
co-firing 258.5 Mg (285 tons) of dRDF when firing at various volumetric blend
ratios of coalrdRDF,  i.e., 1:1, 1:2, and 0:1.

     The investigation specifically addresses the performance of the fuel
handling and feeding system, the boiler, and the resulting emissions from
each blend firing.   Since a spreader stoker-fired 7.6 kg/sec (60,000 Ib/hr)
boiler was co-fired for 230 hours (132 hours continuously) without major
difficulty, the results are sufficiently encouraging to suggest a larger term
demonstration of co-firing coal and dRDF.
                                       Francis T. Mayo, Director
                                       Municipal Environmental Research
                                         Laboratory
                                     iii

-------
                                    ABSTRACT

      This study program was initiated with the overall objective being to
 characterize and demonstrate the technical, economical, and environmental
 feasibility of combusting densified forms of refuse derived fuels (dRDF)
 blended with coal in spreader stoker-fired boilers.

      The testing was conducted at the Maryland Correctional Institute Power
 House located in Hagerstown, Maryland.  A total of 258.5 Mg (285 tons) of
 pelletized 1/2-inch-diameter by 3/4-inch-long dRDF was co-fired with coal in
 spreader stoker boilers rated at 7.6 and 9.9 kg/sec (60,000 and 75,000 Ib/hr)
 of 1034 kPa (150 psig) saturated steam.

      The field tests were designed to investigate (1)  the material handling
 characteristics of dRDF,  i.e., storage in warehouses,  drop boxes, and open
 slabs;  conveying; and feeding out of bunkers; (2) boiler performance, i.e.,
 boiler  efficiency, spreader limitations, grate speeds, underfire and overfire
 air requirements, steam production, flame impingement, slagging, fouling,
 clinkering, and combustion gas analysis; and (3) environmental performance,
 i.e., size, mass, opacity, and resistivity of particulates; gaseous (SOX,
 NO ,  Cl, F, He) emissions; and trace organic and inorganic emissions.
   X

      With the steam demand limiting the  test boiler to a 30-55 percent load,
 the 258.5 Mg (285 tons) of dRDF were satisfactorily co-fired with coal for
 230 hours (132 hours continuously).   The results indicate that coal:dRDF
 blends  up to 1:2 can be handled and burned in conventional spreader stoker-
 fired boilers without major equipment modification.  The fuel  blends  were
 handled satisfactorily, although some pellet deterioration (due to excessive
 handling and rain damage)  caused much dusting and slightly impeded the pellet
 flow.   After adjustments  of the air controls, the spreader-feeders,  and the
 grate pulse interval,  the  blends generally burned as well as coal alone.
 Moreover,  as more dRDF was substituted for coal,  the flame volume increased,
 the opacity decreased,  the fly ash carbon burnout improved,  and the turndown
 ratio of  boiler operation  increased.   Relative to the  particulate emissions
 from coal-only  firing,  the emissions  from the blend firing decreased  slightly
 in  mass  flux, dropped  significantly  in particulate  size  and stack opacity,
 and had  resistivities within  the range for satisfactory  electrostatic precipi-
 tator performance.  Also as dRDF substitution increased,  chlorine and trace
metals  (specifically Pb, Sb,  Br,  and Mn)  increased, and  SOX decreased
correspondingly.

     This report was  submitted  in fulfillment  of  Contract No.  68-03-2426  by
Systems Technology Corporation  under the  sponsorship of  the  U.S.  Environmental
Protection Agency.  This report  covers the period June 29,  1976  to December 30,
1977.
                                       IV

-------
                                  CONTENTS

Foreword	      ill
Abstract	       iv
Figures	     viii
Tables	      xii
Acknowledgment	     xiii

     1.  Introduction  	        1
               Feasibility of burning dRDF	        1
               Types of refuse fuels	        1
               Previous dRDF test programs 	        2
               Current dRDF test programs	        2
               Site selection	        2
               Test program outline	        4
     2.  Summary and Conclusions 	        5
               Introduction  	        5
               Test objective	        5
               Site selection	        5
               Test design	        5
               Test results	        5
                    Material handling  	        5
                         Pellet storage  	        5
                         Pellet feeding  	        7
                         Pellet properties   	        7
                    Boiler performance 	        7
                         Spreader-feeder performance 	        7
                         Combustion of dRDF	        8
                         Fouling 	        8
                         Clinkering  	        8
                         Corrosion 	        8
                         Boiler operation  	        9
                         Ash handling	        9
                         Mass and energy balance	        9
                    Environmental performance  	       10
                         Data normalization	       10
                         Particulate emissions 	       10
                         Gaseous emissions 	 ....       11
                         Trace organic and
                           inorganic emissions 	       11
               Conclusions	       12
                    Fuel handling system	       12
                    Boiler performance 	       12
                    Environmental performance  	       12
                    Summary	       12

-------
3.  Material Handling   	       13
          Fuel mechanical properties	 .       13
               Coal	       13
               Pelletized dRDF	       15
          Storage of dRDF	       15
               Open containers	       18
               Warehouse	       18
               Remote slab	       18
               On-site slab	       20
          Fuel handling system	       20
               Description	       20
               Operation	       22
               Performance	       22
               Alternative fuel blending method 	       23
               Summary	       24
4.  Boiler Performance  	       25
          Boiler description  	       25
          Boiler conditions	       25
          Fuel properties	       27
               Coal properties	       27
               dRDF properties	       30
               Blend properties	       30
               Conclusions	       32
          Fuel handling and response in
            boiler systems  	       32
               Fuel distribution	       35
                    System description  ....  	       35
                    Cold flow test	       36
                    Hot flow test    	       40
                    Normal boiler operation  	       43
               Ash handling	       44
                    System description  	       44
                    Grate	       45
                    Bottom ash	       45
                    Ash silo	       49
                    Reinjection and  collector
                      fly-ash  flows  	       49
               Air and  gas handling	       53
                    System description  	       53
                    Underfire  air setting  	       56
                    Overfire air setting	       56
                    Induced  draft fan	       60
               Furnace  performance   	       61
                    Heat release rate	       61
                    Flue gas temperature	       61
                    Fouling, gas, and  wastage	       61
                    Firing phenomenon  	  ....       72
               Boiler controls   	       72
              Mass  and energy balance	       77
              Low-load performance  	       82
         Summary	       82
                                vi

-------
     5.  Environmental Performance 	      84
               Introduction  	      84
               Field sampling setup	      84
               Test chronology and procedures	      84
                    Opacity	      89
                    Particle mass flux	      89
                    Size distribution	      91
                    Fly ash resistivity	      91
                    S02	      94
                    Oxides of nitrogen	      95
                    Halogens	      95
                    Oxygen	      95
                    Hydrocarbons 	      95
                    Trace organic emissions  	      95
                    Trace inorganic emissions  	      95
               Data analysis and normalization 	      97
                    Data analysis	      97
                    Data normalization	      98
               Particulate emission test results 	     101
                    Opacity	     101
                    Particulate concentration  	     102
                    Size distribution	     106
                    Fly ash resistivity	     106
                    Overall ESP performance  	     106
               Gaseous emissions test results  	     110
                    S02	     110
                    Oxides of nitrogen	.' . .  .     113
                    Halogens	     113
                    Hydrocarbons 	     113
               Trace compound emissions test results 	     120
                    Trace organic emissions  	     120
                    Trace inorganic emissions  	     121
                    Summary	     126

References	     128
Appendices

     A.  Emissions, fuel, and ash data summaries	     129
     B.  Summary sheets for ASME abbreviated efficiency
           tests and Boiler No. 1 and 2 specifications  	     138
     C.  Procedure for estimating stack velocity 	     148
     D.  Cascade impactor data	     151
     E.  Discussion of Monsanto1 s ESP Test Data	     157
     F.  Heavy metals emissions data summaries 	     163
     G.  Physical and chemical characterization of
           dRDF/coal fly ash	     173
     H.  Preceding coal:dRDF studies 	     207
                                  vii

-------
                                    FIGURES
Number

  1       Comparison of the size distributions for the
            December and Hay coals	    14

  2       Characteristic length distributions for pellets
            burned in March	    16

  3       Characteristic length distributions for pellets
            burned in May	    16

  4       Comparison of arithmetic (interpolated)
            and field-measured blend bulk densities  	    17

  5       Comparison of pellet length distributions for
            different types of storage 	    19

  6       Effect of storage duration and method on pellet
            moisture content 	    19

  7       Comparison of a deteriorated pellet (left)  and
            a well-formed pellet (right) 	    20

  8       Front- and side- view drawings of  the
            temporary fuel handling  system 	    21

  9       Cross  section of Boiler  No.  2	    26

10       Coal and  dRDF size distributions compared
            with recommended size  spectra	    29

11       Cross  section of Hoffman Combustion Engineering
            spreader-feeder 	    35

12        Spreader-feeder  injecting a  dRDF:coal blend
            into  furnace	    37

13       Uniform distribution of  coal and dRDF pellets
            near the  furnace rear wall	    38

14       Isolation of  grate dRDF  to determine  its
            spread density   	    39
                                    viii

-------
                              FIGURES (continued)

Number

 15       Ash reinjection and overfire air ports in
            rear wall of Boiler No. 1	    42

 16       Effect of blend and load on grate pulse interval
            or relative grate speed  	    46

 17       Sieve analysis of bottom ash samples for coal,
            blend, and dRDF firings	    47

 18       Drawing of a typical ash collection drain tube  to
            monitor relative ash flow in collector and reinjector  .  .    50

 19       Ash flows in reinjector hopper drain tube for
            coal, blend, and dRDF firings	    51

 20       Ash flows in collector hopper drain tube for
            coal, blend, and dRDF firings	    52

 21       Carbon content of reinjector ash for coal,
            blend, and dRDF firings	    53

 22       Reinjector ash size distributions for coal,
            blend, and dRDF firings	    54

 23       Collector ash size distributions for coal,
            blend, and dRDF firings	    55

 24       Carbon dioxide levels in furnace vs. time as
            determined with a water-cooled probe 	    57

 25       Relationship of furnace excess air level with
            blend and load	;	    58

 26       Grate seal leakage forcing flames toward center
            of furnace during a 1:1 blend firing in Boiler No. 2 ...    59

 27       Heat release rates per unit grate area for
            coal, blend, and dRDF firings	    62

 28       Heat release rates for coal, blend, and
            dRDF firings	    63

 29       Effects of blend and load on flue gas temperature	    64

 30       Variations in ash fusion hemispheric temperatures
            under reducing atmospheric conditions during
            December runs	    65
                                      IX

-------
                              FIGURES (continued)

Number                                                                   Page

 31       Variations in ash fusion hemispheric temperatures
            under reducing atmospheric conditions during March runs  .  .    66

 32       Variations in ash fusion hemispheric temperatures
            under reducing atmospheric conditions during May runs   ...    67

 33       Effects of blend and load on flue gas temperature
            before and after a test	    68

 34       Effects of dRDF and load on flue gas temperature
            before and after a test	    69

 35       Drawing of typical clamp-on corrosion test  shield  	    70

 36       Furnace flames viewed at 3.3 m (10 ft) above  the
            grate during blend and dRDF firings	    73

 37       View from top tube hatch in Boiler No. 2 to show
            firing with a 1:1 blend	    74

 38       Effects of blend and load on flame temperature
            measured with an optical pyrometer ........  	    75

 39       Pressure chart recordings for coal,  blend,  and
            dRDF  firings	    76

 40       Effects of blend and load on carbon  content of
            bottom ash   	    78

 41       Effects of blend and load on carbon  content of
            collector ash	    79

 42       Effects of blend and load on carbon  content of
            fly ash	    80

 43       Effects of blend and load on input/output efficiency  	    82

 44       View of MCI power  plant  showing  stack sampling
            shed  and temporary fuel  handling system at right	    85

 45        Layout  of  Boilers  No. 1,  2,  and  3  with sampling
            locations  indicated  	  	    87

46        Schematic  of  EPA Method  5  sampling train setup  	    90

47       Typical dust  loading of MRI  cascade  impactor
            stages during  a  1:1 blend  firing 	    92

-------
                              FIGURES (continued)

Number                                                                   Page

 48       Schematic of MRI cascade impactor sampling
            train setup	    93

 49       Schematic of WAHLCO resistivity probe assembly 	    94

 50       Schematic of Battelle Tenex sampling train setup 	    96

 51       Graphic representation of probably similar (A & B)  and
            potentially dissimilar (A & C) regression lines
            through data set	    99

 52       Effects of blend and load on stack opacity	103

 53       Effects of blend and load on particulate mass
            emission rate	104

 54       Effect of blend on color of stock aerosol	105

 55       Typical MRI cascade impactor results for blend firing  ....   107

 56       Average size distribution for coal, blend, and
            dRDF firings during March tests  	   108

 57       Average size distribution for coal, blend,
            and dRDF firings during May tests	108

 58       Effects of blend and load on aerosol resistivity	109

 59       Cell configuration in the portable ESP	Ill

 60       Effects of blend and load on sulfur dioxide emissions  ....   112

 61       Effects of blend and load on nitrogen oxide emissions  ....   114

 62       Effects of blend and excess air on nitrogen oxide
            emissions	116

 63       Effects of blend and load on chlorine emissions	117

 64       Effects of blend and load on fluorine emissions	118

 65       Effects of blend and load on hydrocarbon emissions	119
                                      XI

-------
                                    TABLES

Number                                                                  Page

  1       Summary of Previous Co-Firing Tests  	     3

  2       Percentage of Pellets in Hand-Sorted Samples 	    23

  3       Average Properties of Coal on Both an As-Received and
            a Moisture-Free and Ash-Free Basis 	    28

  4       Average Properties of dRDF on Both an As-Received
            and a Moisture-Free and Ash-Free Basis 	    31

  5       As-Fired Properties for Blends in March Tests  	    33

  6       As-Fired Properties for Blends in May Tests  	    34

  7       Area Density of Pellets Removed from the Cold
            Flow Test	    40

  8       Metal Wastage Rate Data for Eight Specimens  	    71

  9       Heat Balance Summary Based on As-Received Fuel 	    77

 10       Ash Mass Balance	    81

 11       Chronological Listing of Test Conditions 	    88

 12       Fuel Elemental Composition Normalization Factors
            for Adjusting Emissions to  a Standard Fuel	   101

 13       Effect of Blend on Aerosol Resistivity    	   110

 14       Relationship of NOX Concentration and Excess Air
            Percentage for Coal,  Blend,  and dRDF  Firings  	   115

 15        POM Concentrations  for  Coal and  Blend Firings	   120

 16        Trace  Metal  Concentrations Found in Coal and dRDF Fuel  .  .  .   122

 17        Average  Heavy Metal  Emissions  in Ash from
            Blend  Firing Tests	   124

 18        Blend  Heavy  Metal  to  Coal-Only Heavy Metal  Ratios
            in Ash Samples	   125

                                     xii

-------
                                 ACKNOWLEDGMENT

     On behalf of Systems Technology Corporation, the authors gratefully
acknowledge the direction and cooperation of the EPA Project Officer,
Mr. Carlton C. Wiles of the Municipal Environmental Research Laboratory,
Cincinnati, and the timely and valuable support of Mr. Robert Olexsey of the
Industrial Environmental Research Laboratory, Cincinnati.

     The authors are also grateful to Mr. James Farrell of the State of
Maryland Department of General Services for his assistance and comments
during the testing at the Maryland Correctional Institute in Hagerstown,
Maryland.  In addition, special thanks are given to Mr. Lou Baltozer and his
powerhouse staff for their generous assistance that was essential to the
successful completion of the test program, and to Mr. Argo Kraus, retired
Detroit Stoker Field Engineer, for his resolution of testing difficulties
during the initial firing of the coalrdRDF blends.
                                      Xlll

-------
                                    SECTION  1

                                  INTRODUCTION


 FEASIBILITY  OF  BURNING DENSIFIED  REFUSE DERIVED FUEL  (dRDF)

      During  the first half  of  the 1970's, several  short-duration, coal-dRDF
 test  burns had  indicated  that  it  may be feasible to use dRDF as a substitute
 for stoker coal.  However,  while  these tests have  provoked the interest of
 the resource recovery community,  they have  failed  to  answer, the questions
 most  critical to determining this feasibility:

      1.   Can dRDF be burned within existing environmental constraints?

      2.   Does  dRDF burning have  any detrimental effects on a boiler system
          or its performance?

      3.   Is dRDF an economical substitute  for coal?

 The present  study was designed to explore the answers to the.boiler perfor-
 mance questions  (Section  4) and the environmental  questions (Section 5).
 The economics of producing  dRDF was addressed by the National Center for
 Resource Recovery, Inc. (NCRR), under Grant Number 804150.

 TYPES OF REFUSE FUELS

     Beginning  in the 1950's in Europe and  in the  late 1960's in the United
 States, the  technical community has had an  increasing interest in the fuel
 value of urban  solid waste.  As a result of this interest and the impetus
 caused by the energy crisis of 1973, four basic types of solid waste fuels
 have been developed:  (1) unsorted urban refuse, (2) fluff refuse derived
 fuel  (fluff  RDF), (3) powdered refuse derived fuel (powdered RDF), and
 (4) densified refuse derived fuel.  Unsorted urban refuse, the oldest type,
 is thermally processed in mass-burning incinerators.  This type of facility
 usually excludes such bulky objects as applicances, rolled carpets, and
 furniture.   Fluff RDF is produced by shredding mixed urban refuse and passing
 the milled material through a series of material separation steps to remove
many of the noncombustibles.  Powdered RDF  is usually produced by an acid-
 embrittling and hot-milling drying operation.  Densified RDF is produced from
 either fluff or powdered RDF with  equipment such as pelletizers, brickquetters
and cubetters.   The dRDF is intended for plants which generally burn lump-
 sized coal,  such as industrial or  institutional stoker-fired plants, rather
 than plants which burn pulverized coal.

-------
 PREVIOUS dRDF TEST PROGRAMS

      Perhaps the first summary of the early programs to investigate the
 feasibility of burning dRDF with coal was a report by R. T. Stirrup, Fellow
 of the Institute of Public Cleansing and Director of Public Cleansing, City
 of Southford, England.  Published in 1965, this report covers applied research
 in England and Europe during the 1956-1960 period.  Specifically, it describes
 programs which prepared briquettes out of mixed refuse and co-fired the
 briquettes with coal to generate steam in short-term tests.  One of these
 programs generated 3 pounds of steam for each pound of briquettes burned.

      Since the early 1970's, several similar short-duration programs were
 conducted in the United States.   Table 1 lists these programs,  and Appendix H
 details the results of each.

 CURRENT dRDF TEST PROGRAM

      Since most of the previous  programs had test firings  lasting less than
 12 hours,  the EPA contracted with Systems Technology Corporation (SYSTECH)
 to conduct a comprehensive technical and environmental test program to
 determine  the feasibility of co-firing dRDF and coal in spreader stoker-fired
 boilers.

 SITE  SELECTION

      As an integral part  of the  Environmental  Protection Agency  (EPA)  planning
 for the test program,  the EPA awarded a grant  to NCRR in Washington,  D.C.,  to
 produce at least  907 Mg  (1000 tons)  of dRDF.   Consequently,  to keep costs
 within  budget limits,  the principal  requirement in selecting a boiler  plant
 for the test program was  a site  within a reasonable trucking distance  of
 Washington,  D.C.

      A  second requirement  was that the site have a spreader  stoker  boiler
 which would  be representative of many similar  stoker-fired boilers.   In
 addition,  it should have  a. variable  grate speed,  an adequate fuel storage
 capacity,  a  feeding system,  and  other facilities readily adaptable  to  the
 testing requirements.  A  third requirement  was  a boiler  plant with  management
 sufficiently interested and  cooperative  to  ensure the  successful  performance
 of  the  test  program.

     Accordingly, four boiler plants  within a  241-km (150-mile)  radius of
Washington, D.C., were established as  candidate  sites.   After SYSTECH
engineers visited and evaluated each  plant,  the  Maryland Correctional
 Institute  (MCI) Boiler House  in Hagerstown, Maryland, was selected  as  the
testing site.  This plant met the three  requirements as  follows:

     1.    The MCI plant had  three Erie City  Iron Works boilers rated at
          3.1, 7.6, and 9.9 kg/sec (2.5,000,  60,000,  and  75,000 Ib/hr)  of
          1034-kPa (150-psi) saturated steam.  The  steam generation capacity
          was sufficient to ensure continued plant  operation if a boiler
          should go off-line because of malfunctions due to  dRDF burning.

-------
                      TABLE 1.   SUMMARY OF  PREVIOUS  CO-FIRING  TESTS

1.

2.


3.


It.


5.


6.


7.


8.


9.

*10.


*11.

*12.

Location of test
Fort Wayne, Ind.
Municipal Power Plant
Sunbury Steam Electric
Station, Pennsylvania
Power & Light
Piqua, Ohio,
Municipal Power
Plant
Wright-Patterson
Air Force Base

Eugene Water &
Electric Board

University of
Wisconsin
Oshkosh, Wisconsin
Appleton Diversive


Menasha Paperboard
Mill

Chanute Air Force Base

Waupun, Wisconsin


Green Bay, Wisconsin

Stockertown,
Pennsylvania
Test sponsor
(Producer)
National Recycling
Center
(Elo & Rhodes)


Black-Clawson
Fibreclaim, Inc.

Air Force
Black-Clawson
Fibreclaim, Inc.
Sandwell
International, Inc.
(Vista)
Wisconsin Solid Waste
Recycling Authority
(Vista)
Wisconsin Solid Waste
Recycling Authority
(Grumman)
Wisconsin Solid Waste
Recycling Authority
(Grumman)
U.S. Army CERL
(Vista)
Wisconsin Solid Waste
Recycling Authority

Ft. Howard Paper
(Grumman)
Hercules Cement
(Vista)
Date
of test
1972

1975


1975


1975


1974


1976


1976


1976


1975

1976


1976

1975

Type of Vol blend
dRDF coal: dRDF
cubette 3:1
1 1/2" x 1 1/2" x 2"
5/8" pellets


3/8" pellets 1:1


3/8" pellets 1:1
1:2

3/8" pellets


1 1/8" pellets 1:1, 1:3,
and 0 : 1

3/4" pellets


3/4" pellets 3:2


1 1/8" pellets 1:1
0:1
3/4" pellets 20%, 30%, and
40% by heating
value
3/4" pellets 1:3
1:2
1 1/8" & 5/8"
pellets
Amt dRDF Test
fired duration
36 Mg

73 Mg 2 days


20 Mg 7 hr


36 Mg 34 hr
6 hr

19 Mg 1 1/2 hr


19 Mg


36 Mg 8 hr


19 Mg


136 Mg

19 Mg


36 Mg

182 Mg 7 days

* Appendix H does not  include a discussion on these tests.

-------
      2.    The  plant  could  accommodate  all  of the  test  equipment and proce-
           dures.   All  flows  in  and out of  the boiler were  readily  accessible.
           The  coal silo  system  could be easily bypassed  to permit  installing
           a  temporary  coal-dRDF fuel handling system.

      3.    The  plant  management  expressed sufficient interest and willingness
           to cooperate in  the test program.

TEST  PROGRAM OUTLINE

      The test  program  consisted  of four separate  field tests:  (1) co-firing
coal  and 20.9  Mg  (23 tons) of dRDF in  a series  of short  runs during
December 1976,  (2) a coal  base  test in January  1977, (3) co-firing coal and
106.1 Mg (117  tons)  of dRDF  in  a series  of longer duration tests during
March 1977,  and (4)  co-firing coal and 127.9 Mg (141 tons) of dRDF also in a
series of  longer duration  tests  coupled  with  electrostatic precipitator (ESP)
evaluations  during May 1977.

     Throughout each test, SYSTECH engineers monitored and evaluated the fuel
handling system, the boiler  performance, and  the stack emissions.

     While Section 2 presents a  summary  and conclusions  from the program,
Sections 3,  4, and 5 describe each phase and aspect of the test program and
evaluate the results for each.

-------
                                   SECTION 2

                            SUMMARY AND CONCLUSIONS
 INTRODUCTION

     Densified refuse derived fuel may be considered one of the more market-
 able products recovered from municipal solid waste.  When densified in the
 form of pellets, cubettes, or briquettes, it may be handled, transported, and
 fed separately or blended with coal and burned in existing-stoker-fired
 boilers without major equipment modification.

     Over the past few years, several limited tests have tentatively confirmed
 that dRDF is a viable coal substitute.  While these tests produced positive
 results, boiler monitoring and emission tests were performed only in a few
 instances.

     As a. result of these encouraging tests, the Environmental Protection
Agency sponsored two parallel efforts:  one to determine the economics of
preparing dRDF and the second to assess the technical and environmental
implications when the fuel is used as a coal substitute.  This report presents
 the technical and environmental evaluation of co-firing tests conducted at
the Maryland Correctional Institute power plant in Hagerstown, Maryland.  The
dRDF used in these test were pellets prepared by the National Center for
Resource Recovery under a research grant.

TEST OBJECTIVE

     The objective of the study was to determine, characterize, and demon-
strate the technical and environmental feasibility of combusting dRDF with
coal in spreader stoker-fired boilers.  The tests were to be conducted in a
stoker-fired boiler which would have a rating between 3.1 and 25.1 kg/sec
 (25,000 and 200,000 Ib/hr) of steam and would be within 241 km (150 miles) of
NCRR in Washington, D.C.  The study was to specifically address fuel handling,
boiler performance, and environmental effects when dRDF pellets, cubettes,
and briquettes were fired with coal.

SITE SELECTION

     After all the candidate boiler plants within 241 km (150 miles) of
Washington, D.C., were surveyed, the MCI plant was selected because it met
the above described criteria most satisfactorily.  This plant had three Erie
City Iron Works boilers, one each rated at 3.1, 7.6 and 9.9 kg/sec (25,000,
60,000, and 75,000 Ib/hr) of 1034 kPa (150 psi) saturated steam.

-------
  TEST DESIGN

      The  test was  designed  to  combust 258.5 Mg  (285 tons) of dRDF during
  236 hours of firing various blend ratios of coal:dRDF.  These tests were
  conducted in a series of burns with volumetric  coal:dRDF ratios of 1:1, 1:2,
  and 0:1 and with test durations ranging from 20 minutes to 132 hours.  The
  series of coal:dRDF tests were preceded and followed by a coal-only test with
  duplicate conditions.  Also, because of required plant steam demand, all of
  the tests were conducted at only 30 to 55 percent of boiler design capacity.
  The initial tests were designed to ensure that dRDF could be safely burned
 without jeopardizing the boiler's capability of meeting the steam demand.
 These tests included monitoring the performance of the spreaders while
  introducing dRDF into an unfired boiler and a series of short-duration burns
 to determine the combustion properties and the boiler performance.  Subsequent
 field tests involved a study of (1) the material handling characteristics of
 dRDF,  i.e., storing, conveying, feeding out of bunkers, etc.; (2)  boiler
 performance, i.e.,  grate speeds, underfire and overfire air requirements,
 steam production,  spreader limitations,  boiler efficiency,  flame impingement,
 slagging,   fouling,  clinkering, combustion gas analysis, etc.; and  (3) environ-
 mental performance, i.e., particulates,  gaseous emissions,  and trace organic
 and inorganic emissions.   Since only pelletized dRDF was available,  testing
 with cubettes and  briquettes was not conducted.

 TEST RESULTS

 Material Handling

     Throughout  the field testing,  259 Mg  (285  tons)  of dRDF were  received,
 stored,  and conveyed to  the  boiler  without  major difficulty or malfunction of
 the fuel handling system.  Difficulties  were  limited  to dusting  and  pellets
 hanging  up in feed  hoppers.

 Pellet Storage—
     At  successive  periods,  the pellets  were  stored in  20-cubic-yard  open
 containers, in a warehouse  (uncovered),  and on an outdoor concrete slab
 (tarpaulin covered).

     Twenty-Cubic-Yard Containers—Since the pellets received during  the
 winter tended to steam, they eventually  froze into a solid mass.   Minimal
 rodding, however, broke up the  frozen pellets, and subsequent handling  further
 restored the individual pellet  integrity without significant  degradation  to
 the pellet.

     Warehouse—Approximately 125 Mg  (140 tons)  of pellets were  stored  in an
unheated warehouse  for 2 months.  With the exception of mild  odors and  some
fungus growth, this storage  proved to be the most effective  in maintaining
pellet integrity over extended  periods of storage time.  Since the depth  of
the piles  was limited to 1.8 m  (6 ft), increases  in temperature  due to
composting  effects were negligible, and the pile  temperature stabilized at
60°C (140°F).

-------
      Open Slab—The  pellets  stored  in  the warehouse were  subsequently moved
 to  an outdoor storage  area.   The  pellets were stored  in 1.8-m  (6-ft) piles
 on  an outdoor slab and covered with a  tarpaulin.  Moisture accumulation under
 the tarpaulin caused pellets at the top of the piles  to deteriorate and cake.
 Also,  some pellets sustained minor  damage, i.e., swelling and  roughened
 edges, because  of water infiltration onto the slab.

 Pellet Feeding—
      The  pellets were  conveyed to the  boiler feed hopper  by a  temporary fuel
 blending  and handling  system.  The  coal and dRDF were volumetrically blended
 in  the various  ratios  by separately feeding coal and  pellets from two hoppers
 to  a  common bucket elevator  which subsequently conveyed both coal and pellets
 to  a  weigh lorry.  The fuels were volumetrically blended  by filling the feed
 conveyors  to capacity  (level with the  conveyor flights) and operating the
 coal  and  pellet  feed conveyors at speeds commensurate with the desired blend
 ratio, i.e., a  1:2 coal:dRDF blend  would require the dRDF conveyor to run at
 twice the  speed  of the coal  conveyor.  Although this feeding system generally
 worked well, it  had  some difficulties with deteriorated pellets.  As the
 amount of  fines  increased (due to excessive handling), the pellets would not
 flow  from  the feed hoppers without  rodding.  These fines  also caused consid-
 erable dusting  throughout the plant.  This dusting was subsequently controlled
 by  installing a  steam  jet at the  conveyor transfer point.

 Pellet Properties—
     The 1/2- x  3/4-inch pellets  had an average bulk density of 425 kg/m3
 (26.5 lb/ft3) and ranged from 400 to 466 kg/m3 (25 to 29  Ib/ft3).  The
 material density for intact  pellets ranged from 1.22 to 1.34.x 1Q3 kg/m3
 (76 to 84  lb/ft3) while that for  deteriorated pellets averaged
 0.98 x 103 kg/m3 (61 lb/ft3).  The  as-received properties were 12.10 to
 15.12 MJ/kg (5200 to 6500 Btu/lb),  20  to 29 percent ash,  9 to 10 percent
 fixed carbon, 12 to  13 percent moisture, 50 to 57 percent volatiles, and
 1142°C to  1152°C (2088° to 2105°F)  hemispheric reducing fusion temperatures.
NCRR projected  that  further  processing of the shredded refuse to remove glass
 and other  inerts could produce a  pellet with a heat content of 19.1 MJ/kg
 (8200 Btu/lb) and an ash of  10 to 12 percent.

 Boiler Performance

 Spreader-Feeder  Performance—
     In a  cold  flow  run (furnace  not fired) to test the fuel distribution of
 the Hoffman Combustion Engineering  spreader-feeders, two  different sized
pellets were distributed onto the grate:  1/2 x 3/4 and 1x2  (diameter x
average length  in inches) pellets.  Because the Hoffman spreader throat has a
maximum size restriction of  1 1/2 inch, the 1-inch-diameter by 2-inch-long
pellets tended  to hang up and slug-feed the furnace.  However, the
 1/2-inch-diameter by 3/4-inch-long  pellets generally were handled and fed well
with the larger  pellets traveling to the rear of the grate and the fines
 falling close to the spreader.  During the initial combustion  tests with
 100 percent pellets, the spreader had  to be adjusted  to decrease the pellet
 trajectory in the furnace by approximately 0.3 m (12  in.).  In addition, the
maximum steam load that the  boiler  could carry was 6.8 kg/sec  (54,000 Ib/hr)
or  70 percent of design capacity.   This derating is the direct result of
volumetric limitations of the spreader feeder.

-------
 Combustion of dRDF—
      While operating  at  these partial boiler capacities, the combustion of
 the various blends of coal:dRDF was generally as good as the combustion of
 coal only.  However,  when  the dRDF substitution was increased, the flame
 length,  intensity, and volume of  the fireball increased correspondingly. As
 the intensity of the  fireball grew, the flame temperature, measured about
 1.5 meters above the  center of the grate, also increased from 1200°C (21928F)
 for 100  percent coal  firing to 1240°C (2264°F) for 100 percent dRDF firing.

      When test firing  the  1:1 blend and 100 percent dRDF, the fireball was
 kept well away from the rear wall of the furnace by adjusting the overfire
 air.  Once these jets were adjusted for minimum smoke and maximum efficiency
 for coal-only burning, they continued to meet the mixing and wall protection
 requirements when burning blends and 100 percent pellets.  As viewed from the
 side of the furnace when firing both pellets and blends, the bed was well
 burned out by the time it approached the front ash pit.   The flame pattern
 above the grate indicated that the fuel bed was maintaining proper porosity
 with minimum clinkering or agglomeration.   This operation was achieved when
 burning a double screened,  high ash fusion temperature 1370°C (2498°F)  coal.
 With little attempt to optimize the system,  a 10 to 12 percent carbon dioxide
 content in the flue gas at the boiler outlet was readily obtained.

 Fouling—
      Inspection of  the furnace interior after the tests  revealed that a light
 coating of ash had  accumulated on the tubes.  Also, an interim boiler inspec-
 tion revealed  that  one-third  of the rear wall of the boiler was covered with
 slag.   This  slagging was  subsequently eliminated when a  spreader was adjusted
 to  prevent pellet impingement  on the  rear  wall.   Subsequent inspections of
 the boiler after being on-line for 8  days  revealed  that  the slag had sloughed
 off.

 Clinkering—
     During  the initial tests  frequent  clinkering occurred  on the grate when
 firing  a  1:1 blend.  This clinkering  was subsequently attributed to  a low
 hemispheric fusion  temperature,  1204°C  (2200°F),  of the  coal.   When  the
 coal was  changed to one having  a  higher  ash  fusion  temperature,  1373°C
 (2500°F),  the  clinkering  stopped.  While coal with  low fusion temperatures
 clinkered, the 100 percent  pellets, which  had a  low fusion  temperature  of
 1151°C  (2103°F), did not  clinker.  This  observation is valid  within  the
 constraints of the test conditions, i.e.,  a  4-hour  test  burn  at  a boiler
 capacity  of 30 percent  of rated design capacity and 100  to  130  percent  excess
 air.

 Corrosion—
     Eight clamp-on corrosion test specimens were installed on  the supply
 tubes of  the rear screen wall 1.5 m (5 ft) above  the  fuel bed.  After
 478 hours of exposure  to various blend and coal-only  firings, normal wastage
 (less than 5 mils per  year) was evident  on all specimens  except  the
 1018 specimen.  This test specimen, which had extremely  high  metal wastage,
was mounted in the area where the heavy  slagging  occurred because of  the
maladjusted spreader.

-------
Boiler Operation—
     Air Flow Controllers—During periods of load shedding, the fuel bed was
more susceptible to clinkering when coal:dRDF blends were fired.  The
clinkering was eliminated by biasing the underfire air control to supply
approximately 70 percent excess air to the fuel bed.  On the basis of these
results, boilers which are tight (minimum air leaks) should be capable of
satisfactorily burning coal:dRDF blends with 50 percent excess air.

     Oscillating Grate Dwell-Shake—Throughout the test, the duration and
amplitude of the grate shake pulse was adjusted to advance the fire line at
the rear of the boiler approximately 15.2 cm (6 in.) per excitation.  In all
advances, the pulse frequency was the principal controlling variable.  At
40 percent load, the frequency of the pulse decreased from 11 minutes for
100 percent coal to 3 minutes for 100 percent pellets.  When firing a blend,
the pulse duration tended to increase because the bulk density of the blend
ash was less than that of the coal ash.

Ash Handling—
     Bottom Ash—The sieve analysis of. bottom ash samples taken during blend
firings indicated that conventional pneumatic ash handling systems should be
able to handle the bottom ash from blend firings as well as they do the
bottom ash from firing coal-only.  On a few occasions fire occurred in the
bottom ash hopper during blend firing.  Rodding of the clinkers in the ash
hopper revealed that the ash had a taffy-like consistency.  Under similar
conditions, when firing coal only, the bottom ash was much easier to break up
by rodding.

     The bottom ash removal system malfunctioned only during 100 percent
pellet firing.  The bottom ash was so fine that it would not de-entrain
properly in the cyclone.  The particles, which had been wetted by the steam
in the vacuum ejector, passed through the cyclone and eventually plugged the
ejector.

     Dust Collector Ash—As dRDF was substituted for coal, the fly ash
particles became finer.  The size of the particles in the dust collector
ranged from 200 micrometers for 100 percent coal firing to 90 micrometers
(sizes at the 50th percentile) for 100 percent pellet firing.  Also, the
carbon content of the fly ash decreased significantly with increasing dRDF
substitution.  The primary factor contributing to this occurrence was the low
fixed carbon content of dRDF (12 to 18 percent) compared to coal which had
65 to 85 percent fixed carbon.

Mass and Energy Balance—
     Mass Balance—The mass balance indicated that an unusually large amount
of the fuel ash had accumulated in the collectors.  Subsequent analysis of
the collector fly ash revealed that the high collector ash weights were due
to the presence of 50 to 70 percent carbon in the collector ash.  Also, since
90 percent of the particles exiting the boiler were greater than 50 micro-
meters in diameter, these large particles were removed by the cyclone.  The
carbon content of the bottom ash varied from 2 to 10 percent, and the carbon
content of the stack fly ash (not captured by the cyclone) was 30 to
40 percent.  The analysis of the stack fly ash as a function of blend revealed
that its carbon content decreased as the dRDF substitution increased.

-------
       Efficiencies—During the testing  the boiler  efficiencies were extremely
  low,  namely 55  to 60 percent.   When the  boilers were  installed in 1963, they
  produced a boiler efficiency  of 79  percent at  an  excess air of 34 percent and
  rated design capacity.   These low efficiencies were primarily due to the low
  boiler loads (less than 30 percent  of  rating), high excess air (80 to
  115 percent), and extremely high losses  of combustibles in the refuse (up to
  25 percent).  The analysis of  the results indicated that the coal-only and
  blend firing efficiencies had  no discernable differences.  However, this
  observation may be unique to  the boiler  installation  at MCI since the large
  amount of  unburned combustibles  removed  by the collectors is certainly an
  anomally to expected  boiler performance.

  Environmental Performance

  Data  Normalization—
       Since  the co-firing  tests spanned a  6-month period, the properties of
  the coal and dRDF  burned  in the  successive  tests varied widely.   The boiler
  excess air  in the  tests also varied considerably.   To eliminate the effects
  of these variables, all the emissions data were corrected to 50 percent
  excess air  and then normalized to a reference coal and dRDF composition.   All
  the co-firing emissions data were then statistically compared with a coal-
 only baseline plot of emissions concentration versus boiler load.   If the co-
 firing emissions data fell outside the 90 percent  confidence limits for the
 coal-only emissions data, they were considered to  be significantly different.

 Particulate Emissions—
     Mass Concentration—The particulate mass concentration (grams/standard
 cubic  meter, g/scm) in the 1:1 and 1:2 blend firings was slightly  less  than
 in the coal-only firing.  However, the reductions  were not  significant  at the
 90 percent confidence level.  The mass flux at  a 40 percent boiler load  for
 1:1 and 1:2 blend  firings averaged 0.45 g/scm corrected to  12  percent C0a.
 The coal fired during these tests was a nominal size of 1 1/4  * 1/4  inch  with
 a maximum of 30  percent  passing through a nominal  1/4-inch  screen.

     Particulate Size—As more dRDF  was substituted for coal,  the  particulate
 diameter decreased.  In  the May tests,  the diameters for the  coal-only  firings
 were 3 micrometers, and  those  for the dRDF-only firings were  0.8 micrometer
 (at the 50  percentile point).

     Particulate Resistivity—Because of  the unusually high carbon content  in
 the fly ash  during the coal-only firing,  the resistivity was generally less
 than 106 ohm-cm.   As  dRDF  was  substituted for coal,  the carbon burnout in the
 fly ash improved,  and  the  resistivity increased to  2 x 10*" ohm-cm for the
 1:1 blend firing.

     Electrostatic  Precipitator Performance—A  mobile  5-cell electrostatic
 precipitator (owned by the  EPA  Industrial Environmental Research Laboratory
 in Research  Triangle Park,  North  Carolina)  was  evaluated to determine the
 effect  of dRDF on ESP  performance.   The evaluation, however, could not be
 conclusive because  of  short  circuiting within the ESP.  This short circuiting
developed when the  dielectric blocks from which the electrodes were suspended
became  coated with  high carbon  content aerosol.  Appendix E discusses the
test results.

                                      10

-------
      Opacity—As  dRDF was  substituted for coal, the overall opacity of the
 plume reduced significantly.  At 40 percent boiler load, the opacity for
 coal-only  firing  was 16 percent (based on a 1.22-m (4-ft) diameter stack).
 At  the same boiler load and excess air, the opacity was only 10 percent for
 dRDF-only  firing.

 Gaseous Emissions—
      SO2—Since the dRDF had a sulfur content of 0.4 percent, the S02 emissions
 reduced with increasing dRDF substitution.  The decrease was particularly
 significant for the 1:2 and 0:1 (100 percent dRDF) blend firings.  At
 40  percent boiler load and the same excess air levels, the S02 dropped from
 1300  ppm for coal-only firing to 250 ppm for dRDF-only firing.  This reduction
 in  S02  follows exactly the reduction in sulfur content of the fuel (see
 Table 6 in Section 4).

      NOX—There were no significant changes in NOX as dRDF was substituted
 for coal.  At 40  percent boiler load and the same excess air levels, the NOX
 concentrations ranged from 200 to 350 ppm with either fuel.

      Chlorine—As dRDF was substituted for coal, the chlorine in the emissions
 increased  from 60 ppm for coal-only firing to 650 ppm for dRDF-only firing.
 There appeared to be no appreciable change in chlorine concentrations as the
 load  changed from 20 to 50 percent of design capacity.

      Fluorine—Fluorine concentrations also increased with increasing dRDF
 substitution.  However, the concentrations were very low, e.g., 8 ppm for
 coal-only  firing and 12 ppm for dRDF-only firing at a 40 percent boiler load
 and constant excess air conditions.

      Hydrocarbons—There were no significant changes in hydrocarbon emissions
when  substituting dRDF for coal.  At a 40 percent boiler load, the total
hydrocarbons ranged from 10 to 25 ppm.  As the boiler load increased, the
hydrocarbon concentrations decreased significantly.  This reduction is
probably attributable to the improved conbustion conditions at higher
boiler  loads.

Trace Organic and Inorganic Emissions—
      Organic Emissions—The overall emissions of polycyclic compounds for
coal-only  and blend firings were well below the threshold limits proposed by
the National Academy of Science.  Typical measured values were:  543 ng/m3
for anthracene/phenanthrene, 100 ng/m3 for methyl anthracene, and 137 ng/m3
 for fluoranthene  (all at 1:1 blend firing).

      Inorganic Emissions—The analysis of the fly ash for trace metals
revealed that relative to coal-only firing the blend firing enriched some
metals but reduced others.  For example, when firing a blend of 1:2 coal:dRDF,
 the amount of lead in the stack particulates was 8217 ug/m3.  This compares to
a lead  concentration of 230 yg/m^ for coal-only firing.  While dRDF was the
main  contributor of Br, Mn, Pb, and Sb, coal was the primary source of As,
Ni, and V.
                                     11

-------
      Several elements, particularly As, Ga, Na, and Sb, tended to concentrate
 in small particles.  In addition, as the dKDF substitution increased, both
 the solubility of the fly ash and the amount of small-size particulates in
 the respiratory range increased.  Consequently, each of these effects pose
 potential hazards from (1) respiration of heavy metals associated with aerosols
 and (2) leaching of high levels of heavy metals in landfills.

 CONCLUSIONS

 Fuel Handling System

      Pelletized refuse can be stored, handled,  and blended with coal in
 conventional material handling equipment provided the pellet structural
 quality is maintained.   However, if pellets deteriorate because of excessive
 handling and exposure to rain,  they will hang up in bunkers and will generate
 considerable dust as they are conveyed throughout the plant.

 Boiler Performance

      Boiler performance was evaluated at reduced load conditions,  i.e.,  30 to
 55 percent of design capacity.   At these boiler loads,  some minor  operational
 difficulties were encountered with slagging and clinkering throughout the
 testing.   These difficulties were controlled by making  simple adjustments.
 Typical adjustments  included (1) biasing the air controls  to  higher  excess
 undergrate air  levels to  prevent clinkering in  the fuel bed during load  shed,
 (2) properly adjusting  the spreader  feeders to  prevent  dRDF impingement  on
 the side  and rear walls  of the  furnace,  and (3)  setting the grate  dwell  and
 pulse  intervals to compensate for the reduced ash bulk  densities when the
 blends  were fired. The boiler operation  was restricted  only when the spreaders
 and ash handling  system became  capacity  limited during  dRDF-only firing.   The
 increasing  dRDF substitution resulted in (1)  improved carbon  burnout in  the
 fly ash,  (2) decreased plume opacity,  and  (3) improved  low-load performance
 (more than  a 4:1  turndown  without excessive smoking).

 Environmental Performance

     Compared to  the particulate emissions  from coal-only  firing,  the emissions
 from the blend  firing decreased  slightly in particulate concentration, dropped
 significantly in particulate size and  stack opacity,  and had  resistivities
within  the  range for satisfactory ESP performance.  Of  the gaseous emissions,
 SOx decreased and  chlorine increased, both  significantly.   Analysis  of the
 trace inorganic elements in  the  fly ash  when dRDF was fired revealed that
concentrations of Pb, Cd, Mn, Zn,  and Sb were significantly higher than  other
elements.  Since the solubility  of the fly  ash  increased with increasing dRDF
substitution, landfilling  the dRDF residue  could result in hazardous levels
of heavy metals in the leachate.

Summary

     While the test  was limited  to firing at reduced boiler loads, the
preliminary results  from these field tests  indicate that coal  and  dRDF can  be
co-fired at volumetric coal:dRDF ratios  up  to 1:2 with  only minor  adjustments
to the boiler and fuel handling  systems.  Subsequent testing  should  address
the long-term effects of corrosion and erosion on boiler tubes.

                                      12

-------
                                   SECTION 3

                               MATERIAL HANDLING
     The primary factors affecting the flow of a solid fuel through a material
handling system are the following fuel properties:  (1) size distribution,
(2) moisture content (inherent and free), (3) bulk and particle density,
(4) bulk compressibility, and (5) configuration and roughness.  Previous
tests at Chanute Air Force Base and the Pennsylvania Power & Light Company
revealed that excessively handled dRDF becomes fluffy with increased fines
which then promote high angles of repose and bridging in bunkers and hoppers.
Since the handling and storage of the dRDF pellets at MCI required five
separate fill/dump operations, similar dRDF deterioration and bridging were
anticipated.

     Therefore, to ensure that the plant would continually meet its steam load
requirement and not have to be shut down because of blended fuel bridging in
the bunker, the dRDF and coal blending and handling system was installed
independently of the main coal feed system.   Since the stoker spreader
feeders might disperse the dRDF and coal blend differently than coal alone,
the stoker spreader operation was monitored to characterize feeding behavior
and dispersion of fuel on the grate for coal, a blend of coal and pellets,
and pellets alone in a cold, idle furnace.

FUEL MECHANICAL PROPERTIES

     Since the size and density of both the coal and the dRDF directly
influence the performance of the fuel handling system, tests were conducted
to quantify these properties.  The following paragraphs discuss these pro-
perties.

Coal

     Hoffman Combustion Engineering recommends that the coal fed to its
spreader stoker have a nominal size of 1 1/4 x 1/4 inch with a maximum of
40 percent passing through a nominal 1/4-inch screen.  A sieve analysis of
the December coal showed that the distribution was 100 percent less than
3/4 inch with 70 percent passing through a 1/4-inch screen.  Test firing of
this coal showed that the high fraction of fines impeded proper plant
operation,  i.e., the fines caused furnace pressure pulsations as they entered
the furnace, plugged the grate and air ports (larger size fines), and over-
loaded the fly ash system (smaller size fines).  In view of the poor boiler
system performance with the December coal, a double screened stoker coal had
to be acquired before meaningful co-firing tests could be started.


                                     13

-------
     Accordingly, arrangements  were made with the Delta Coal Company (the
supplier for MCI) to  provide  specification coal for the March and May tests.
Although not standard fuel  for  the Hagerstown plant,  the coal acquired was
Swickley seam coal, a commercially available spreader stoker coal.  The coal
was obtained from the supply  prepared  for an industrial spreader stoker plant
about 64 km (40 miles)  from MCI.

     The coal sieve analysis  in Figure 1 compares the coal for the December
and May field tests.   The May size distribution is also representative of the
coal for the March field tests.   The January coal had a size distribution of
100 percent less than 1/2 inch  with 70 percent passing through a 1/4-inch
screen with an occasional 76-mm (3-in.)  piece.   The March and May coal was
double screened, 1 x  0  Stoker coal with 30 percent passing through a 1/4-inch
screen.   The bulk density of  the  March and May coal averaged 777 kg/m3
(48.5 lb/ft3), and the  material density ranged between 1.35 and
1.43 x 103 kg/m3.
                                l-ROSLIN—flAMMLER EXPONENT
                                                   a DECEMBER
                                                   A MAY(AVG.)
                                     0»   05 075 •
                                   PARTICLE SZE D. WOCS
          Figure 1.   Comparison of the size distributions for  the
                     December and May coals.
                                      14

-------
Pelletized dRDF

     The determination of the optimum pellet size was based primarily on
obtaining a pellet that would fall in the middle of the recommended coal
particle size range.  The rationale for such a size was (1) to prevent the
pellets from segregating from the coal during handling, (2) to ensure that
the pellets would flow through the unmodified spreader without jamming, and
(3) to spread the pellets onto the grate in a pattern similar to that of
coal.  Cold flow tests, i.e., feeding blends of coalrdRDF into the furnace
without combustion or air flow, were conducted on nominal 1/2-inch-diameter
by 3/4-inch-length and 1-inch-diameter by 2-inch-length pellets.

     The test results indicated that the 1/2-inch-diameter pellets flowed
satisfactorily through the material handling equipment and the spreader
feeders.  The 1-inch-diameter pellets tended to hang up in the spreader
feeders and to slug feed the furnace.  Since NCRR had a die available to
produce the 1/2-inch-diameter pellets and they performed satisfactorily
throughout the cold flow tests, only the 1/2- x 3/4-inch pellets were used in
the combustion tests. Throughout the combustion tests, the pellets were
sampled from the furnace hopper for length analysis.  Figures 2 and 3
illustrate the length distribution data.

     The March test pellets were stored in a warehouse and had a lesser
deterioration and data spread than the May test pellets which were stored on
an open slab.  These discrepancies could be attributed to differences in the
storage conditions and/or the original pellet characteristics.  The pellet
bulk density averaged 425 kg/m3 (26.5 lb/ft3) and ranged from 400 to 466 kg/m3
(25 to 29 lb/ft3).  Bulk density was determined by filling and weighing a
1-ft3 container and then subtracting the container tare weight.  The material
density for intact pellets ranged from 1.22 to 1.34 x 103 kg/m3 while that
for deteriorated pellets averaged 0.98 x 103 kg/m3.  Material density was
determined by weighing a pellet and then determining the amount of volume
displaced by the pellet when immersed in a liquid.

     Figure 4 shows the bulk density of various coalrdRDF blends.  The fact
that the measured bulk density of the blend was higher than the arithmetic
bulk density can be attributed to the dRDF particles filling the interstices
in the coal and vice versa.

STORAGE OF dRDF

     Throughout the pellet storage SYSTECH engineers were able to avoid
spontaneous combustion by following the storage procedures for lignite.
These procedures required keeping the storage period to a minimum and limiting
the pellet piles to maximum depths of 1.8 to 2.4 m (6 to 8 ft).

     Each of the field tests required 91 to 125 Mg (100 to 140 tons) of dRDF.
Since the supply from NCRR was generally 11.8 to 14.5 Mg/wk (13 to 16 tons/wk),
the deliveries had to be accumulated for 8 to 10 weeks before each test could
be conducted.  The pellets were transported from the NCRR test facility in
Washington, D.C., to the Hagerstown, Maryland, plant in tarpaulin-covered,
20-yd3, open roll-on containers.  With deliveries from December 1976 through
May 1977, the pellets acquired totaled 255 Mg (281 tons).
                                     15

-------
                        PABTXXE SOE D. INCHES
 Figure 2.
Characteristic length distributions for
pellets  burned in March.
                                         a MAY 10 1:2
                                         • MAY 11 1:2
                                         t MAY 13
                                         o MAY 13 1:1
                                         A MAY 12 1:1
                                         • MAY 10 1:0
                       PWTKXE SZE 0, HOtS
Figure  3.   Characteristic length distributions for
            pellets  burned in May.
                          16

-------
                                             MEASURED DENSITY
                                        \  \     /- ARITHMETIC DENSITY
                                           V
                                            \
                                 % d-RDF ( by volume)

           Figure 4.  Comparison of arithmetic  (interpolated)  and
                      field-measured blend bulk densities.
     Since the slab beside the  fuel handling  system at MCI was too small to
accommodate the pellet accumulations,  the  pellets  were stored as follows:
During December 1976 and January  1977,  the pellets were kept in the roll-on
containers until they were removed for burning.  During February and March of
1977, the pellets were stored in  an unheated  warehouse within a residential
community.  With the advent of  warmer  weather and  the possibility of offensive
odors reaching the nearby homes,  the pellets  were  subsequently moved to an
open slab about a half mile from  the MCI power plant and stored under a
tarpaulin cover during April and  May of 1977.  The following sections discuss
each of the storage conditions  and their effects on the pellets.
                                      17

-------
 Open Containers

      When the pellets were stored in the roll-on containers during December
 1976 and January 1977, they occasionally froze near the edges.  Since the
 initially received pellets tended to steam, evaporating moisture trapped
 under the tarpaulin eventually froze the pellets into a solid mass.  However,
 minimal rodding broke the mass into blocks which could flow freely.  Subse-
 quent movement broke up the blocks into individual pellets.  Problems were
 also encountered with trash remaining in the containers from previous garbage
 loads.   The contaminating trash included such materials as cans, cardboard
 boxes,  and wood and metal pieces which had to be removed from the fuel to
 prevent their jamming the conveyors and/or the fuel spreader.  While this
 problem was minimal (only 10 percent of the containers had such waste),  it
 does demonstrate that care must be exercised in this area.

 Warehouse

      When the pellets were stored in the warehouse during February and
 March,  they were dumped from the containers and pushed by a front-end loader
 into 1.8-m (6-ft)  deep piles in the warehouse.   In addition to mild odors,
 fungus  growths appeared on the peaks of the piles or wherever there was  a
 moisture vent.   There did not appear to be any rodent or insect damage to the
 stored  pellets.  After the stored pellets were removed,  the warehouse was
 easily  cleaned.

      Figure  5 indicates the extent of the pellet length reduction due to the
 warehouse storage.   However,  Figure 6 indicates that this reduction cannot  be
 attributed  to a  moisture  loss.   A reasonable cause for the reduction could
 be  the  additional  pellet  handling and/or differences in the NCRR production
 procedures.

 Remote  Slab
                                                                 »
     When the pellets  were stored during April  and May on the open  concrete
 slab at  the MCI  power  plant,  they were  dumped  from roll-on containers and
 then pushed by a front-end loader into  1.8- to  2.1-m (6-  to 7-ft) deep piles
 on  the  slab.  Although a  plastic  tarpaulin  protected each pile from the
 weather,  moisture accumulation  under  the tarpaulin caused pellet  deteriora-
 tion and  caking  (or capping)  on the  tops of the piles.  A similar capping
 occurred  on the  piles  stored  in the warehouse.   This caking consisted of a
 5-  to 10-cm (2-  to 4-inch)  thick  layer  in which the mechanical integrity of
 the individual pellets was greatly reduced.  In addition,  because of  poor
 slab drainage, run-off water  infiltrated some of  the piles and deteriorated
 pellets on the slab surface.  Such pellets  swelled,  and their initially
 smooth sides  became rough.  Figure 7  compares a deteriorated pellet with a
 good one.  However, the deteriorated  pellets were relatively few, and they
were still usable although their  rough  sides impeded their flow out of
 storage bins.  Rodding was  required to  assist the flow of these pellets  from
 the storage bin.
                                      18

-------
2 8
a
I
LU
JT8
i
£
OJ O_
                                     3/23 — STORED UNDER TARP
                                          ^   STORED IN WAREHOUSE
                                      3/29 V— & HANDLED 2 EXTRA
                                      3/311   TIMES
                                        ^
                                         1"
                NOMINAL PELLET LENGTH
 Figure 5.  Comparison of pellet  length distributions for
            different types of  storage.
                            INCREASING AGE

o_



K
o
UJ
CO
I °-
UJ
cr
i-
CO
O


liN.v/nc«oiwva «V3C



JANUARY PELLETS
X
r >





















i
















	 • 	 m^*
APRIL PELLETS
jl 	
/ ^\


IT T1










>


T




. _











I










i
                                              DIFFICULTY W/FLOW •
                                              PELLETS GOT WET
                                              FINES VERY HIGH
       3/23 3/29     5/9  5/10 5/11 5/12 5/f3 5/14
                 DAY OF BURN

Figure 6.  Effect  of storage duration and method on pellet
           moisture content.
                            19

-------
          Figure 7.  Comparison of a deteriorated pellet (left) and
                     a well-formed pellet  (right).
 On-Site Slab

      Some of the 35 Mg (40 tons) of dRDF remaining from the January and March
 tests and stored in the warehouse were transferred to the slab beside the
 fuel handling system at MCI for subsequent use during the May tests.   These
 pellets were stored on the slab in a 2.4-m (8-ft)  deep pile.   The pile was
 periodically monitored with a thermocouple imbedded 1.5 m (5  ft)  down from
 the top of the pile.   After the pile temperature rose to 60°C (140°F),  it
 dropped as the pile dried  out to ambient temperature by the time  of the May
 test.

 FUEL HANDLING SYSTEM

 Description

      Coal  is  delivered  to  the Hagerstown plant  in  trucks which dump through a
 grizzly grate  onto  a  drag  chain feeder  to the bucket  elevator inlet.   The
 bucket  elevator delivers the  coal  to  a  square concrete  silo.   A "weigh lorry"
 on rails over  the firing aisle and  located  above the  stoker feeder  hoppers
 transports  the coal from the  silo  to  the  stoker feed  hopper and then weighs
 and dumps  the  coal.

     As noted  previously,  since  the blended dRDF and  coal  fuel  could have
bridged  the silo and  thereby  caused a plant shutdown, a  temporary fuel
blending and handling system was installed  to bypass  the  silo  during the
field tests.
                                      20

-------
     As shown in Figure 8, the system included a canopy-covered  slab with
storage space for about 27 Mg (30 tons) of both coal and pellets.   The
conveying equipment consisted of two 8-yd3, pyramid-bottom bins  that fed the
fuel into cleated, pin-pan, Z-belt conveyors which  in  turn emptied the  fuel
into the feed hopper of a bucket elevator.  Orbital vibrators were installed
on the coal and pellet bins to facilitate free fuel movement.  The bucket
elevator had a straight back with the other three sides inclined at 45  degrees
to feed the fuel into a 30-degree, 0.2-m (8-in.) square chute.

     The starters on the Z-belt conveyors and the bucket elevator were  inter-
locked.  The activation of a single start button began the blend feed.  The
feed from the bins could be stopped from a station  next to the weigh lorry or
from outside the building next to the fuel bins.  Normally the fuel feed was
stopped and the elevator allowed to empty before a  complete  shutdown.

     The coal and dRDF were blended to  the various  coalrdRDF ratios by
changing drive pulleys to vary the speed of one of  the Z-belt conveyors.   The
volume of fuel loaded per foot of conveyor was maintained by scraping  the
fuel load level with the top of the conveyor flights.
                                                                      SWIVEL CHUTE
                           Z-BELT CONVEYOR (TYP)
                    FRONT VIEW
                                                      SIDE VIEW
              Figure 8.  Front- and side-view drawings of  the
                         temporary fuel handling system.
                                      21

-------
      The temporary blending and handling system could transfer 2450 kg
 (5400 Ib) of a 1:1 coal and dRDF blend into the weigh lorry in approximately
 30 minutes or at a rate of 1.39 kg/hr (3.1 Ib/sec).  The limiting factor on
 this rate was the speed of the Z-belt conveyors.

 Operation

      Two crews operated the fuel handling system.   One crew,  consisting of a
 truck driver and a front-end loader operator,  retrieved the dRDF from storage
 and delivered it to the test site.  The second crew consisted of two men at
 the test site, one was a front-end loader operator and the other was a
 helper.   This crew filled the coal and dRDF bins after each loading to the
 weigh lorry.

 Performance

      In  general,  the temporary fuel blending and handling system performed
 well throughout the test program.   The mixture of  1 x 0 coal  with less than
 30 percent  fines  and 1/2-inch-diameter x 3/4-inch-long pellets fed well and
 required less hoeing in the stoker feed hopper than coal alone.   During the
 January  and March field tests,  the blend fed much  easier than coal alone.
 However, near the end of the May field tests,  the  pellets would  "rat-hole"
 (assume  a high angle of repose)  in the 8-yd3 pyramid feed bins unless they
 were rodded periodically.   The "rat-holing" was due primarily to pellets
 whose sides had become fluted because of water damage.   Consequently,  such
 pellets  tended to interlock and  bridge.

      Lengths of pipe or chunks  of  blacktop  (picked up by the  front-end
 loader)  intermingled with  the pellets and jammed the Z-belt conveyors once or
 twice each  8-hr shift.   While thorough cleaning of the transportation equip-
 ment and careful  operation of the  front-end loaders would have prevented the
 inclusion of these materials,  a  grizzly  grate  and  a magnet, such as  used in
 a  coal plant,  would  have been a  more practical means of  preventing the
 introduction of such refuse into the fuel flow.

      At  the  outset of  the  field  tests, dust released during coal and dRDF
 fuel transfer from the  bucket  elevator  chute to the weigh lorry  was  excessive.
 While a  shroud over  the chute  initially  contained  the dust, the  subsequent
 motion of the bucket elevator blew out the  loose dust.   The dust had a lint-
 like consistency  and settled  throughout  the plant.   Consequently,  a  hood was
 installed over  the weigh lorry and coupled  to  what  proved to  be  an inadequate
 exhaust  fan.   Although  this  installation  provided  some relief, the dust  was
 still excessive until a steam jet  was  installed under the hood.   This  steam
 jet  wetted the dust  particles and  adequately suppressed  the dust from
 spreading throughout the plant.

     To  quantify  the blending system performance,  five fuel samples  were
 taken from the feed  trough  (which  filled  the lorry)  on each of two represen-
 tative field  test days.  The samples, approximately  4.5 kg (10  Ib)  each,  were
 hand-sorted  into  coal and pellets,  and the  two sorts  for  each  sample were
weighed.   The  remaining dust  (primarily dRDF)  from  the hand sorts  was  weighed
with the pellets.  The  consistency of the weight percentage of the samples

                                      22

-------
evidenced the repeatability and homogeneity of the coal:dKDF blending.   The
relative standard deviation of the pellet fraction was less than 5 percent in
these samples.  Table 2 summarizes the sampling results.
            TABLE 2.  PERCENTAGE OF PELLETS IN HAND-SORTED SAMPLES
               Blend
Time of Sample
% Pellets (by weight)
9 May 1:1
1:1
1:1
1:1
1:1
10 May 1:2
1:2
1:2
1:2
1:2
10:15
11:30
12:50
2:00
3:15
9:15
10:10
12:30
1:35
-
0.54
0.46
0.44
0.47
0.53
0.66
0.63
0.65
0.62
0.64
Alternative Fuel Blending Method

     Although the fuel blending and handling system functioned as designed,
some preliminary tests had to be performed before the drives on the Z-belt
conveyors could be set at the proper speed ratios.  For these initial tests,
the operator of the front-end loader alternately loaded coal and dRDF into
the feed hopper of the bucket elevator.  When the hopper was full, the bucket
elevator was started, and the gate to the feed chute was opened.  The layered
fuel flowed satisfactorily from the hopper.  Moreover, samples of the blend
taken from the weigh lorry were consistently mixed.

     Consequently, the alternate coal and dRDF layering may be considered as
effective in blending the two fuels as the proportioning conveyor mixing
system.
                                      23

-------
Summary

     Except for flow problems experienced with deteriorated pellets, the
field tests demonstrated that conventional equipment can adequately blend
coal and dRDF in various coalrdRDF ratios and can handle both the coal-only
and the blended fuel.  Such equipment, however, may require provisions to
suppress dusting at conveyor transfer points.
                                     24

-------
                                   SECTION 4

                               BOILER PERFORMANCE
BOILER DESCRIPTION

     The MCI boiler plant in Hagerstown consists of three 1034-kPa (150-psig)
Erie City Iron Works boilers.  Their design steam ratings are 9.9, 7.6, and
3.2 kg/sec (78,500, 60,000, and 25,000 Ib/hr). Figure 9 shows a cross section
of a typical boiler.  Each unit is equipped with a Hoffman Combustion
Engineering "Firerite" spreader-feeder with an appropriate number of spreader-
feeders to distribute the lump fuel in the furnace.  The large coal pieces
that do not burn in suspension are consumed on the surface of the front ash
discharge vibrating grates.

     The Erie City Iron Works boilers have tube-and-tile furnaces.  The
waterwalls are composed of nominal 8.26-cm (3 1/4-in.) diameter tubes that
are partially covered by refractory to approximately 2.4 m (8 ft) above the
grate surface.  The gases exit from the furnace passing through a two-drum,
vertically baffled boiler bank consisting of rows of in-line 5.7-cm
(2 1/4-in.) diameter tubes arranged in two gas passes.  The boiler unit flue
gases pass through a decantation two-stage multiclone collector.  The fly ash
captured in the first-stage collector is injected into the furnace to complete
combustion of the fly char, and the fly ash in the second-stage collector is
pneumatically transported to disposal.  The cleaned gases are induced through
a centrifugal fan and exhausted to a breeching (common to all boilers) and
then to the stack.

BOILER CONDITIONS

     Before testing the boilers, all the associated instruments, such as the
steam flow meters and the pressure and temperature gauges, were calibrated by
Johnson Controls.  The boiler settings, grates, and grate seals were visually
inspected to determine their general condition and to seal obvious leaks.
Refractory cement was applied at various locations on the boiler setting to
seal leaks.  In addition, the spreaders were adjusted for proper distribution
on MCI normal coal.

     Most of the December through March tests were performed in Boiler No. 1,
the 9.9 kg/sec (78,500 Ib/hr) boiler.  Boiler No. 2 was not used for the
initial testing because it was the only boiler equipped with both electric
and steam-driven facilities and had the capability of cold starting.  Conse-
quently, most of the preparatory effort was devoted to putting Boiler No. 1
in good operating condition.  However, the automatic control systems for
                                     25

-------
Figure 9.  Cross section of Boiler No.  2.
                    26

-------
Boiler No. 1 were in such poor condition that the boiler had to be operated
in the manual mode much of the time.  The boiler tubes, grate, and refractory
were in generally good condition, but there was some grate seal leakage.

     Since the restriction on Boiler No. 2 was eventually lifted and because
tests had to be scheduled according to the extremely limited availability of
an EPA mobile electrostatic precipitator which could be tied into only one
boiler, Boiler No. 2 was selected for the May tests.

     During the start-up of the May test, air leakage around the grate seals
produced air levels higher than those normally existing in boilers of this
type.  The leakage was detected when Orsat analyses were performed on combus-
tion gases sampled at various cross-sectional scans and heights above the
grate.

     The following two sections on the boiler performance when substituting
dRDF for coal are as follows:  the first examines the properties of the coal
and dRDF and their blends, and the second covers how the fuels responded in
the boiler subsystems.

FUEL PROPERTIES

     The following discussion on the properties of coal, dRDF, and coal:dRDF
blends is preparatory to discussing their handling and burning in the field
test program.

Coal Properties

     The coal used during the four field tests (December, January, March, and
May) came from different mines.  This variation in the coal supply was the
result of the procurement procedures of the State of Maryland and the need to
use a specification coal during actual test runs.

     During December the plant burned a coal which had a low ash fusion
temperature, high heating value, and 3 percent sulfur.  In January, the coal
(supplied from the Pittsburg/Swickley seam) had a high ash fusion temperature.
However, the fines were excessive with 85 percent passing a 6.3-mm (1/4-in.)
sieve.  In March and May, the coal met the ash fusion, heating value, and
size constraints of the stoker manufacturer.  Laboratory analyses of the coal
samples taken throughout the test program are presented in Table 3.  Exami-
nation of these coal properties reveals that the coal burned in March was
different from that fired in May.

     During each daily test 2.2- to 4.5-kg (5- to 10-lb) coal samples were
taken periodically for subsequent analysis.  At the end of the day the
samples were mixed, and the composite was divided by sectioning.  The final
samples were placed in two separate containers for shipment to laboratories.
First, a 300- to 500-gram sample was placed in a 1/2-liter rigid polyethlyne
jar with a vapor proof lid and sealed.  This sample was sent to the SYSTECH
laboratories for moisture determinations.  Second, samples were placed in
4-mil polyethylene bags and taped shut.  Subsamples from these macro samples
were sent to Broeman Laboratories and Commercial Testing and Engineering

                                      27

-------
     TABLE 3.   AVERAGE PROPERTIES OF COAL ON BOTH AN AS-RECEIVED AND
               A MOISTURE-FREE AND ASH-FREE BASIS

As Received
% Moisture
% Ash
% Volatile
7, Fixed Carbon
Btu/lb
MJ/kg
Dry Basis
% C
7, H
% N2
% Cl
% S
% Ash
% 02
Btu/lb
MJ/kg
Fusion
Initial
1st Softening
2nd Softening
Fluid
Mineral Analysis
Phos. Pent Ox.
Silica
Ferric Ox.
Alumina
Titania
Sodium Ox.
Potasium Ox.
Lime
Magnesia
Sulfur Tri Ox.
Undetermined
December
Average

2.12
10.78
29.42
57.68
13,471
31.33

77.15
4.77
1.26
.26
3.57
11.01
1.98



1128°C
1192-C
1249°C
1371°C

.48
35.43
34.94
22.39
.56
.25
.99
1.63
.28
1.23
1.82
January
Average

7.03
14.30
16.16
62.50
11,797
27.44

70.50
3.80
1.57
	
1.80
15.40
6.93



1332CC
1368°C
1379°C
1414°C

.92
43.50
21.00
20.70
1.42
2.70
2.37
.38
.58
.83
...
March
Average

3.78
10.23
22.43
63.55
12,959
30.14

74.15
4.38
1.59
.11
1.72
10.63
7.42



1274°C
1308° C
1335°C
1371°C

.41
52.02
12. 74
25.64
.70
.47
1.87
2.18
.36
1.66
1.95
May
Average

1.27
21.95
22.55
54.23
11,706
27.23

67.40
4.33
1.35
.05
1.22
22.23
3.42



1482+° C
1482+° C
1482+° C
1482+° C

.33
59.62
5.80
27.43
.90
.31
2.32
.54
1.15
.32
1.32
December January March May
Average Average Average Average

COAL MOISTURE AND ASH FREE

33.78 20.60 26.14 29
66.22 79.40 73.84 70



86.70 83.34 82.98 86
5.36 4.49 4.90 5
1.42 1.85 1.77 1
.29 	 .13
4.01 2.12 1.93 1

2.23 8.19 8.38 4
15,446 14,996 15,069 15,
35.93 34.88 35.05 35

NOTE: Hemispheric (Second Softening)
temperatures in excess of 1204
(2200'F) are preferred.
















.13
.87



.73
.56
.72
.07
.43

.35
246
.46


°C














Corporation for fuel property analyses.  These analyses included standard
ASTM ultimate and proximate determinations, ash chemistries, and reducing
atmosphere fusion temperatures.  The analyses were ultimately expanded to
include a determination of chlorine.  Table 3 lists the averages of the test
coal properties.  The individual determinations from which the averages were
derived are contained in Appendix A.

     Part of each macro sample was sized by using a Tyler portable sieve
shaker and a standard set of ASTM coal-sizing sieves.  The sieves used were
nominal 1-, 3/4-, 1/2-, 3/8-, and 1/4-inch screens.  Figure 10 shows the size
distributions of the coal samples overlayed on a probability display of the
recommended spreader stoker coal distribution.

     The bulk density of the coal was determined by loosely filling a
0.03-m3 (1-ft3) container with coal, without its being agitated or tamped,
and by weighing the container on a 100-lb capacity platform scale.  The bulk
                                     28

-------
                              SIEVE ANALYSIS
                              l-ROSLIN—RAMMLER EXPONENT
o MARCH AVERAGE d-RDF
A MARCH AVERAGE COAL
• MAY AVERAGE d-RDF
A MAY AVERAGE COAL

                                 PARTICLE SIZE D, INCHES
              Figure 10.   Coal and dRDF size distributions compared
                          with recommended size  spectra.
density was  determined by dividing the weight of  the  material by the volume
of the container.   The density of the coal particles  was  determined by
weighing  lumps  of  coal; measuring the amount of liquid  displaced by each
piece when it was  immersed in a water-filled, 500-mSL, graduated cylinder;
and dividing the piece weight by the total volume of  coal.

     Because the use of a fuel may be limited by  its  heat density (the amount
of energy per unit volume),  the amount of air required  to burn the fuel, or
the ability  to  introduce the fuel into the furnace, there is a particular
interest  in  the heat density (MJ/m3) of the fuel  and  the  theoretical mass
(kg) of air  required to burn a unit mass (kg) or  unit energy value (MJ) of
fuel.  The data obtained from the laboratory analyses of  the fuel were
substituted  in  Equation 4.1  to determine the theoretical  air requirements:
      TA = 11.53C + 34.34(H2-02/8) + 4.29S   kgair/kgfuel
                        (4.1)
where C, H, 0,  and  S  are the respective weight fractions  of  the element in
the fuel ultimate analysis.
                                      29

-------
      The air requirement can also be expressed in terms of the mass per heat
 content equivalent of fuel by using the available fuel properties information:

                           (kg,  ,/kg .  ) x 106
                kg .  /MJ •    fue*   air -                        (4.2)
                   .


      The volumetric  heat rate (megajoules per cubic meter)  is the parameter
 which determines the amount of energy that the volumetric feeders can intro-
 duce into the furnace.   This parameter is particularly important  in a retrofit
 application to determine the maximum amounts of dKDF which  can be substituted
 for coal and still have sufficient feeder capacity.


                      =  J/kg x kg/m
                            106
 dRDF Properties

      Procedures similar to those for the coal  sampling  were  used  to  collect
 dRDF samples.   Each time the dRDF was sampled,  one  sample was  sealed to
 permit  moisture determination at the SYSTECH laboratories and  another was
 immediately  processed  for size and composition.

      Table 4 lists  the averages of the dRDF  properties  including  ultimate and
 proximate analyses,  chlorine,  reducing atmosphere ash fusion temperatures,
 and  ash chemistries.   The table also includes  the same  information recalcu-
 lated on a moisture-free and an ash-free basis.  The individual determinations
 from which the  averages were derived are contained  in Appendix A.

      The theoretical air requirements (expressed as kilograms  of  air per
 megajoule) and  the volumetric  heat rate (megajoules per cubic  meter)  were
 computed by  substituting the laboratory data in Equations 4.1  through 4.3.

 Blend Properties

      The theoretical combustion properties of  coal:dRDF blends can be mathe-
 matically computed once the  heating value, the bulk density, the  mass, and
 the  heat or  volumetric  blend ratio of the components are known.

      Throughout the current  study  the mixture of coal and dRDF was given a
 volumetric ratio designation.   For example, a  1:1 blend would  be  an  admixture
 of 1 m3  of coal with 1  m3  of dRDF.

      It  is important, also,  to  know both  the weight and the  heat  mixture of
 the blended  fuels in order to  compare the test results with  predictions made
by others as well as to  size the fuel handling system, to estimate the ash
handling requirements,  and to  determine  the best feed rate of  blended fuel
entering the boiler.
                                     30

-------
TABLE 4.  AVERAGE PROPERTIES OF dRDF ON BOTH AN AS-RECEIVED AND
          A MOISTURE-FREE AND ASH-FREE BASIS

As Received
% Moisture
% Ash
'/. Volatile
7. Fixed Carbon
Btu/lb
MJ/kg
Dry Basis
% C
% H
7. N2
% Cl
% S
Z Ash
X 02
Btu/lb
MJ/kg
Fusion
Initial
1st Softening
2nd Softening
Fluid
Mineral Analysis
Phos. Pent. Ox.
Silica
Ferric Ox.
Alumina
Titania
Sodium Ox.
Potasium Ox.
Lime
Magnesium
Sulfur Tri. Ox.
Undetermined
December
Average

13.40
19.97
56.54
10.10
6488
15.09

43.98
5.29
.35
.40
.40
23.19
30.80
March
Average

12.62
24.41
54.08
8.89
5534
12.87

39.17
4.47
.39
.45
.26
27.97
27.30
May
Average

12.22
28.75
49.27
9.76
5266
12.25

35.63
4.54
.85
.36
.28
33.02
25.33
December March May
Average Average Average

dRDF MOISTURE AND ASH FREE

85.04 85.80 83
14.97 14.21 16



54.29 54.42 53
6.53 6.20 6
.43 .54 1
.49 .62
.53 .36

38.02 ' 37.87 37



.38
.62



.36
.75
.25
.54
.42

.68
9785 8772 8956


1103CC
1142CC
1191°C
1246° C

.87
55.52
2,27
13.45
.66
6.82
1.30
10.75
1.14
6.03
1.19


1116°C
1151nC
1179°C
1213'C

.73
71.58
2.89
4.43
.99
5.66
.53
7.50
1.12
1.22
1.87


10963C
1152=C
1163°C
1218°C

.65
63.65
2.64
8.39
.69
7.53
.91
9.74
1.59
3.20
1.00
22.76 20.40 20

MOTE: Hemispheric (Second
Softening) temperatures
excess of 1204*0 (2200°
are preferred.












.83


in
F)













                            31

-------
       The  properties  of  the  fuel  for  the various blends used in March and May
 are  listed  in Tables 5  and  6.

       The  weight  fraction of dRDF determined by physically separating the coal
 from the  dRDF and weighing  each  component had a relative standard deviation
 of less than 5 percent.  Hence,  the  blending system was producing a uniform
 blend.  The hand-sorted weight ratios are consistently higher than the
 calculated weight ratios, probably because all fines not identified as coal
 were  considered  dRDF.  Since the coal contained between 15 and 30 percent
 fines, the dRDF  fraction was biased by considering all unidentified fines as
 dRDF.

       It would normally be expected that a linear interpolation should be
 applicable for blends when  the density of coal and the density of dRDF on a
 volumetric basis are used to determine the relative density of the two
 fuels.  The fact that the fuel densities are uniformly higher than the
 interpolated densities is understandable when both coal and dRDF are con-
 sidered as ensembles of solid particles which have different void sizes.
 When two solid fuels are admixed, the relative void between the two fuels
 frequently becomes smaller because some of the coal particles will fill voids
 in the dRDF and some of the dRDF particles will fill voids in the coal
 structure.  Consequently,  the bulk density of an admixture would likely
 be higher  than the density determined by straight  interpolation.   This error,
 however,  is small and was  determined  to be less than 3 percent for all
 blends.

 Conclusions

      In  the current  tests,  the  substitution of  dRDF for coal increased the
 ash and  decreased the volumetric  heat density of  the blend.   However,  the
 air:fuel ratio  (kilograms  air per megajoule equivalent)  actually declined as
 dRDF  was substituted  for coal.   If  the coal:dRDF blends  can be fired  at the
 same  excess air  level as coal alone,  the  forced and induced draft  fans
 should be  adequate, assuming there  is little  efficiency  deterioration.  The
 decrease in heat  content for a  given  fuel  volume could cause the  furnace  fuel
 feeders  to become limiting.   Increases in  ash content  could overload  ash
 handling equipment or air pollution control equipment.

 FUEL  HANDLING AND RESPONSE IN BOILER  SYSTEMS

      The three boilers at the Hagerstown plant are  similar  in  configuration
 although different in capacity.   Each boiler can be  divided  into five  sub-
 systems:   fuel feeding distribution,  furnace, controls, air  and gas handling,
 and ash handling.  The fuel  distribution system moves  the fuel from outside
 the boiler  into the furnace  where it  is combusted and  converted  to heat
 energy and ash.   The  ash handling system removes the solid  residue from the
 furnace and air pollution control equipment.  The air  and gas handling  system
moves  air  to the  furnace for combustion, draws the  combustion products  through
 the heat transfer sections,  and exhausts the gases  to  the atmosphere.   The
 furnace and boiler system removes heat from the combustion  products and
converts it into  steam.  The control  system coordinates the  activities  of the
fuel and air handling systems in  response to energy requirements of the
furnace system as coupled with plant  demand.

                                      32

-------
       TABLE 5.   AS-FIRED PROPERTIES FOR BLENDS IN MARCH TESTS*
                                             AS FIRED
     VOLUMETRIC BLEND
1:0
1:1
1:2
0:1
PARAMETER
Btu/#
MJ/kg
Moisture
Volatiles
Fixed Carbon
Ash
C
H
N
0 <
s
Cl
%dRDF
by Volume
by Weight
by Heat

12714
29.5
4.9
30.4
54.2
10.5
71.5
4.7
1.3
5.6
1.5
.06

0
0
0

9180
21.3
7.4
44.9
29.8
17.9
51.3
4.6
.8
16.8
.8
.24

50
35
20

8706
20.2
7.7
46.8
26.5
18.9
48.6
4.6
.8
18.3
.7
.3

67
52
36

6034
14.0
9,6
57.8
8.1
24.5
33.4
4.5
.4
26.8
.2
.4

100
100
100
*Unless noted, all values are weight percent on a wet basis.
                                 33

-------
      TABLE  6.   AS-FIRED  PROPERTIES  FOR BLENDS  IN MAY  TESTS*
 VOLUMETRIC BLEND

 PARAMETER

 Moisture
 Volatiles
 Fixed  Carbon
 Ash
 Btu/lb
 MJ/kg
 Carbon
 Hydrogen
 Oxygen
 Nitrogen
 Sulfur
 Chlorine

 FUSION TEMP. -C

 Init.  Def.
 1st Soft.
 2nd Soft.
 Fluid

 THEORETICAL AIR
   1:0
1274
1308
1335
1371
                                         AS FIRED
1:1
1:3
22.6
54.2
22.0
11706
27.2
66.5
4.3
3.4
1.3
1.2
.05
6.6
31.7
38.4
23.2
8988
20.9
54.1
4.1
9.8
1.1
.86
.15
1:2
0:1
7.9
38.3
29.1
24.7
8382
19.5
47.3
4.1
14.2
.9
.66
.21
16.6
48.6
9.0
25.9
5130
11.9
30.9
3.8
21.8
.6
.23
.33
                  1116
                  1151
                  1179
                  1213
kg/kg fuel
kg/MJ in fuel
% Weight Rate dRFD
% Heat Rate dRFD
9.04
.351
0
0
7.26
.347
35
20
6.27
.332
52
37
3.93
.331
100
100
*Unless otherwise noted, all values are weight percent on a wet basis
                                  34

-------
Fuel Distribution

System Description—
     Each of the three boilers  is  equipped with Hoffman Combustion Engineering
Type A "Firerite" underthrow spreader-feeders.   Figure 11 is a cross section
of a typical spreader-feeder mounted  on the  stoker front plate of the boiler.
The solid fuel is placed  in  the coal  hopper  and flows by gravity to a recipro-
cating feed plate which advances the  lump fuel  over the distributor blades.
As the reciprocating feed plate moves forward,  the lump fuel falls into the
rotor and is carried between the rotor and the  rotor casing through 220 degrees
of arc before being thrown into the furnace.

     Adjustments on the spreader include varying the rotor speed to change
the length of the particle trajectory and varying the position of the rotor
casing (circular tray) to alter the arc of the  particle trajectory.  Slowing
the rotor shortens the particle throw.  When the rotor casing is moved
forward  (toward the furnace), the  fuel particles are thrown into a higher
trajectory.  Although the two adjustments jointly determine the longitudinal
fuel pattern, rotor speeds are  adjusted grossly while the casings are adjusted
finely so that the fuel introduced by each spreader-feeder lands at approxi-
mately the same distance  from the  back furnace  wall.
                        FEEDER BODY
                   DISTRIBUTOR BLADE
                      FACE PLATE
                  WATER GASKET-,
                                                COAL HOPPER
FEED PLATE
                ARCH
              PATH OF COAL
                           END OF COAL SPREAD
                               TOP OF GRATE
       Figure 11.  Cross  section  of  Hoffman Combustion Engineering
                   spreader-feeder.
                                      35

-------
 Cold Flow Test—
      The  cold flow test  (feeding  fuel  to  an  idle boiler with no fire) was
 designed  to demonstrate  that  coal:dRDF blends could be successfully handled
 and distributed into  the furnace.  The first goal of the cold flow test was
 to operate and verify the performance  of  the new fuel conveying system that
 had been  installed  to bypass  the  existing coal silo.  This system conveyed
 dKDF and  coalrdRDF  blends into the coal weigh lorry which in turn supplied
 the stoker coal hoppers  of the three boilers.  Initially the operation of
 the conveyor system was  unacceptable because the coal and pellet conveyors
 ran too slowly.  Subsequently this problem was easily corrected.

      To test the ability to feed pellets,  two basic pellet sizes were fed
 into the  furnace:   nominal 1-inch-diameter by 2-inch-length pellets were used
 for the first experiment, and nominal 1/2-inch-diameter by 3/4-inch-length
 pellets were used for the second.   In each experiment the test sample was
 44 kg (97 Ib).   In the first experiment,  with the reciprocating feed plate
 adjusted for maximum stroke, 4 1/2 minutes were required to feed the nominal
 1-inch-diameter by 2-inch-length pellets,  an an average rate of about
 9.7 kg/min (21.3 Ib/min).  During  this experiment unusual feeder noise
 indicated some  pellet breakage was occurring.  The noise was attributed
 primarily to the rotor casing (circular tray) clearance limitations.   Although
 there were no difficulties  during  this test,  it was concluded that  if the
 1-inch-diameter  by 2-inch-length pellets were simultaneously fed to all four
 stoker spreader-feeders  (which are driven  by  a single motor and drive shaft),
 they  could cause shear pin  failure.

      In the second experiment, with the feeder adjustment  similar to  that
 used  during normal coal firing,  the 1/2-inch-diameter by  3/4-inch-length
 pellets were fed into  the furnace  in about 3  minutes, an  average rate of
 about  11.3 kg/min  (25  Ib/min).  During this experiment  no  unusual noise or
 other  events were  noted.

      In both cold  flow experiments the fuel distribution patterns (both
 lateral and  longitudinal) were excellent,  and the  pellets were  generally fed
 to the  desired furnace locations.   Figure  12  shows  fuel being fed through the
 spreader.  Fine material  accumulated  at the bottom  of the  spreader  opening
 because  the  feed fan air  jet was turned off during  the  test  to minimize
 dusting  inside the boiler.   Figure 13 shows coal and pellets  spread on top of
 each other with the fuels rather uniformly dispersed.  The  spreader performed
 as intended in that the larger pellets  with the greater mass  traveled
 to the rear of the furnace with the  fines  falling closer to  the  spreader.

     The area densities of the 1/2-inch-diameter by 3/4-inch-length pellets
 spread into the furnace were determined as follows:  A rough  square of  the
 fuel on the grate  was  isolated with a flat shovel;  see Figure 14.  The
 isolated pile of fuel  was then collected and weighed.  The  average density
across the grate section  was determined by dividing the weight of the  collected
pellets by the area  covered by the pile.  For each of four  distances  from the
back wall, Table 7  lists  the spread density and condition of  the pellets.
                                      36

-------
Figure 12.  Spreader-feeder injecting a dRDF:coal blend into furnace.
                                  37

-------
00
             Figure 13.  Uniform distribution of coal and dRDF pellets near the furnace rear wall.

-------
Figure 14.  Isolation of grate dRDF to determine its spread density,

-------
        TABLE 7.  AREA DENSITY OF PELLETS REMOVED FROM THE COLD FLOW TEST
      Distance (m)            Spread density
     from back wall               (kg/m2)                 Pellet Condition
1.
1.
1.
2.
07
52
83
90
(42
(60
(72
(114
inches)
inches)
inches)
inches)
29
12
12
21
.5
.9
.0
.3
(6.
(2.
(2.
(4.
06
65
46
35
Ib/ft2)
lb/ft2)
Ib/ft2)
lb/ft2)
whole
whole
whole
large number
fines
of
      The cold flow test indicated that  the 1/2-inch-diameter  by  3/4-inch-
 length pellets produced a spread density which was  high near  the front and
 rear walls  and low in the middle of the grate.   Approximately twice  as much
 fuel was located at the rear wall than  in the middle  of the grate.   Subsequent
 testing during the hot flow test confirmed that the pellet throw was too long.

 Hot  Flow Test—
      It was decided to fire 100 percent pellets for a short period of time to
 determine their impact on furnace performance.   A total of 945.7 kg  (2085 Ib)
 of 1/2-inch-diameter dRDF pellets were  weighed  in the lorry.   When the stoker
 coal hopper of the No.  2 boiler was empty,  the  pellets in the lorry  were
 discharged  into the hopper.   The steam  pressure was 1055 kPa  (153 psi),
 and  the steam flow was about 3.8 kg/sec (30,000 Ib/hr).  The  cam on  each of
 the  Hoffman spreader-feeders was adjusted to  the maximum feed stoke.  The
 spreader rotor rpm was  60,  the  same as  that used for  coal.  After the pellets
 were fed into the  furnace,  it was immediately obvious that the fuel  trajectory
 was  too long,  causing fuel  and  flame to impinge on  the rear furnace  wall.
 The  trajectory was reduced  some 0.3 m (12 in.)  by slightly retracting the
 circular tray.   Fifteen minutes after starting  the  pellet feed,  the  steam
 pressure had  decreased  to 1027  kPa (149 psi).   This pressure  generally pre-
 vailed  for  the remaining 5  to 7 minutes of  the  test.   The pressure had dropped
 because the spreader-feeders were incapable of  supplying a sufficient quantity
 of fuel to  the furnace  to maintain the  steam  pressure.   The average  firing
 rate  for  the  three feeders  collectively was 48  kg/min (105 Ib/min) or about
 15.9  kg/min  (35  Ib/min)  for  each feeder.  While  the*steam pressure decreased
 during  the  20-minute  firing  with a 0:1  blend, the steam flow  remained about
 3.8 kg/sec  (30,000  Ib/hr) which was  50  percent  of boiler rating.  This
 dropping  steam pressure  indicated that  the boiler system could not carry the
 50 percent  load when  firing  100 percent  pellets.

      Following  the  successful 0:1 test,   a 1:1 blend test was  fired.   During
 this  test there was frequent  clinkering  on the  grate.   One cause  of  this
clinkering is  that when  a fuel  with  too  low art  ash  fusion temperature is
placed  on the  grate and  agitated,  hot burning particles with molten edges
roll against and under each  other  as  the grate  is vibrated.   The  burning
particles, which are now within  the  ash  bed, heat the  ash so that the softened
ash sticks together, restricting  cooling air flow,  and begins  to  form a skin
clinker layer across the bed  surface.  This skin clinker becomes  progressively

                                      40

-------
less permeable than the open portions of the ash and fuel bed; consequently,
less air is supplied to the fuel/ash bed prompting progressively deeper
clinkering.  As new fuel is fed into the furnace and falls on top of the skin
clinker, the airflow is retarded still more, and the clinkering condition is
further aggravated, and smoking becomes severe.

     The ash fusion temperature analysis of the coal revealed that the
hemispheric reducing ash fusion temperature was 1191°C (2176°F).  Since
reducing hemispheric ash fusion temperatures in excess of 1204°C (2200°F) is
preferred for proper combustion conditions, it is not surprising that
clinkering was encountered.  Since continuation of this clinkering during
blend firing would not permit evaluation of the effect of pellets, a different
coal had to be procured for subsequent tests.

     Even though the pellets have an ash fusion temperature less than that of
coal, the furnace operated satisfactorily when the 100 percent pellets were
burned.  Assuming that the ASME definition for lignitic type ash (defined as
ash in which the sum of CaO + MgO is greater than Fe203)i applies to dRDF ash
(which has a CaO + MgO to Fe203 ratio of 4), then dRDF burns similarly as a
lignite coal.  Therefore the boiler design rules for lignite, rather than
those for bituminous coal, more nearly apply when burning dRDF.

     During January the testing switched from Boiler No. 2, which is a
7.6 kg/sec (60,000-lb/hr) boiler with three spreader-feeders, to Boiler
No. 1, which is a 9.9 kg/sec (78,500-lb/hr) boiler equipped with four
spreader-feeders.  In the January tests the boiler was fired with a 1:1 blend
for 4 hours and with 100 percent pellets for 2 hours.  During the 1:1 firing,
the overfire air and fly ash reinjection air pressures were adjusted to
produce the desired flame mixing and low smoke opacity.  Further discussion
on these adjustments are provided later in the text.  The stack plume was
relatively clear during both the 1:1 and 0:1 blend firings.  While firing
with the 1:1 blend, there was no reoccurrence of the clinkering which was
experienced during the December tests.  This improved performance was due to
the coal having a hemispheric fusion temperature of about 1373°C (2500°F).
The December coal fusion temperature was 1204°C (2200°F).

     When the blend ratio was changed to 100 percent dRDF, the furnace
remained clear of smoke streamers for the first 20 to 30 minutes of firing,
and there was no significant smoking.  Subsequently two rows of ash were
observed on the fuel bed in line with the fly carbon reinjection nozzles.
Figure 15 shows the location of three of the fly carbon reinjection nozzles
at the back of the furnace.  Each of the two rows was in line with the double
fly ash reinjection ports (one reinjector for the dust caught in the boiler
bank passes, the other for the coarse stage multiclone).  This ash layer
phenomenon might be alleviated by spacing the fly ash reinjection nozzles
more uniformly across the rear wall of the furnace or increasing the number
of reinjection lines.  The reinjection ash rows moved well through the furnace
      Winegartner, E. C.  Coal Fouling and Slagging Parameters.  American
Society of Mechanical Engineers, 1974.


                                      41

-------
Figure 15.  Ash reinjection and overfire air ports in rear wall of Boiler No. 1.

-------
as the grate vibrated, and they remained sufficiently porous for airflow.
There was also a base of burned-out ash under the ash rows.

     The furnace volume appeared more than adequate for the combustion
taking place, although a maximum of 6.8 kg/sec (54,000 Ib/hr) of steam could
be generated when firing 100 percent pellets.  This unmodified vibrating-
grate, stoker-fired boiler supported 70 percent of the nominal rating for
2 hours when firing 100 percent pellets.  The magnitude of this derating is
the amount predicted by the volumetric limitations of 1.47 m3/hr (52 ft^/hr)
for the feeder.  The forced draft fan capacity was sufficient to meet the
underfire air requirements at this rating.

     When a 1:1 blend was fired, there was no difficulty in maintaining the
1034 kPa (150 psig) boiler steam pressure.  However, when 100 percent pellets
were introduced, the steam pressure fell to about 1000 kPa (145 psig) while
supplying 6.8 kg/sec (54,000 Ib/hr) steam flow.  This pressure drop was due
to volumetric feeding limitations of the spreader-feeders.  After the initial
loss in pressure, the boiler operated at this reduced pressure for the rest of
the test.

     The principal change required during the 1:1 and 0:1 blend tests was to
reduce the spreader rotor speed so that the throw of the pellets would be
approximately 0.15 m (6 in.) less than that for 100 percent coal.  This
adjustment was necessary to prevent furnace rear wall fuel impingement.

     During the combustion tests with 1:1 blend and 100 percent dRDF, the
fireball was kept well away from the rear and front walls of the furnace by
adjusting the overfire air.  Once these jets were adjusted for minimum smoke
and maximum turbulence efficiency for coal-only burning, they continued to
meet the mixing and wall protection requirements when burning blends and 100
percent pellets.  As viewed from the side of the furnace when firing both
pellets and blends, the bed was well burned out by the time it approached the
front ash pit.  The flame pattern above the grate indicated that the fuel bed
was maintaining proper porosity and that the combustion was good.  With
little attempt to optimize the system, a 10 to 12 percent carbon dioxide
content in the flue gas at the boiler outlet was readily obtained.

Normal Boiler Operation—
     The cold and hot flow tests showed that dRDF could be properly distributed
into the furnace.  The blend also had a distribution pattern on the furnace
grate which was similar to that of coal.  This finding is not particularly
surprising since the size distributions and material densities (not bulk
densities) of the coal and pellets were similar.  Therefore, with the same
velocity and angle of injection into the furnace, pellets and lumps of coal
with equal weight would be expected to travel approximately the same distance.

     There was a severe slag accumulation on the rear furnace wall in line
with Spreader-feeder No. 1 during the May testing.  To stop the slagging, the
throw of all pellets was reduced approximately 15.24 cm (6 in.), and the
circular tray in Spreader No. 1 was retracted slightly to reduce the arc of
the pellet trajectories.  After this adjustment the pellets still carried


                                      43

-------
 the same distance from the feeder and landed at the same place on the grate.
 However, the 15.24-cm (6-in.) throw reduction eliminated the slag accumulation
 on the furnace wall and reduced smoking.  In December,  January,  and March,
 the material impinging on the side walls was not excessive.   In May, however,
 Spreader No.l had to be adjusted because of its throwing too far and spraying
 the left side wall of the furnace.  Careful measurement of the spreader
 showed that the circular tray had skewed left.   Once the tray was properly
 aligned, the fuel impingement on the left side  wall was eliminated.

      During January and March, Boiler No. 2 had a recurring  problem of
 clinkering on the left side when burning coal.   Reports received after the
 field testing stated that this clinkering had been eliminated by readjusting
 the spreader circular tray.

      Proper adjustment of the spreader-feeders  is critical to the successful
 combustion of coal:dRDF blends.   During part of the March tests, clinkering
 on the grate in front of Spreader No.  2 recurred, but the rest of the fuel
 bed remained free burning.  Clinkers formed on top of the piled burnt-out  ash
 at the front and moved out  of the furnace with  some difficulty.   The rear wall
 of the furnace remained clear of slag throughout these  tests.   During an
 unexpected furnace outage caused by a control loop failure (the  dRDF in the
 furnace was not related to  this  failure), the furnace spreaders  were inspected.
 While Spreaders No.  1, 3, and 4  were clear of pellets,  Spreader  No.  2 had a
 heavy accumulation of partially  pyrolyzed pellets in the feed throat.   Careful
 measurement of the tray position from the inside of the boiler indicated  that
 the pellets were being thrown at too high an angle out  of the spreader so
 that they  ricocheted off the refractory feed throat and accumulated  at a
 point in the furnace approximately two-thirds of the way back.   This malad-
 justment was solved  by retracting the  circular  tray approximately 3.18 mm
 (1/8 in.).   The furnace  was then brought back on line from a cold start with
 a  1:2 blend and run  continuously at loads of 3.2 kg/sec to 4.2 kg/sec
 (25,000 to  33,000 Ib/hr)  for 48  hours  with no further clinkering.

 Ash Handling

 System Description—
      Each of the  three boilers at  the  MCI plant  is equipped  with a dry
 pneumatic ash handling system.   As  the solid fuel burns on the vibrating
 grate,  the  grate  is  periodically pulsed  to  advance the  ash to  the front of
 the  boiler  where  it  falls into a refractory-lined ash pit.   The  bottom ash is
manually hoed from the ash  pit into the  pneumatic ash handling system.  After
 the  ash  is  conveyed  to a  cyclone separator where it  is  de-entrained  from  the
 carrier air,  it  falls  into  an ash  storage silo.   The  vacuum  source for the
pneumatic system  is  a  steam ejector.

     Ashes  are also  collected from  under  the  boiler  tube  bank  and  from the
particulate  collectors.   The ash from  the tube tank and primary  cyclone is
reinjected  into the  furnace  through nozzles  in the rear of the furnace.   The
secondary multiclone collector ash  is  pneumatically transported  to the
de-entrainment cyclone and  ash silo.
                                      44

-------
     To discuss the findings on relative boiler ash flows when substituting
dRDF for coal, the following aspects of the boiler ash handling system are
addressed:  grate, bottom ash, ash silo, and reinjection and collector
fly-ash flows.

Grate —
     As the fuel is spread into the furnace, portions of it burn in suspen-
sion, and the remainder falls to the grate.  The primary combustion air is
introduced into the furnace through the grate supporting the burning fuel and
then into the furnace.  The grate is maintained with a constant 5-cm (2-in.)
ash covering by periodically vibrating it to advance the accumulated ash.  In
order to achieve a steady-state condition, as much ash must be removed as is
added to the furnace with the fuel.  The two principal means of control of
the ash bed depth are the dwell period between grate pulses and the duration
of the grate pulse.  In addition to frequency, the amplitude of the agitation
can also be adjusted to account for differences in ash bulk density.

     Figure 16 shows the dwell between grate pulses for various blends as a
function of boiler load.  When the load increased, the frequency of pulses
was increased to maintain the same depth of bed on the grate.  Also, as the
dRDF substitution ratio increased, the frequency and/or duration of pulse was
increased to maintain the uniform bed depth.  In all instances, pulse
frequency was the principal controlling variable.  The duration of pulse was
adjusted to cause the fire line at the rear of the boiler to advance approxi-
mately 6 inches per excitation.  The amplitude of the pulse was also adjusted
to help clear the rear fire line from the furnace wall.  Minimal adjustment
to the amplitude was required.  When firing a blend, the pulse duration
tended to increase because the bulk density of blend bottom ash was less than
that of the coal bottom ash.  The outliers shown in Figure 16 were so indi-
cated because the ash bed depth and fire line were not representative of
normal boiler operation.

Bottom Ash—
     As the ash was shaken from the grate into the ash pit, it was allowed to
accumulate for approximately 8 hours.  After this period the ash doors were
opened, and the ash was hoed into the boiler's pneumatic ash handling system.
When the grate was free of clinkers, the ash which accumulated in the hopper
was free flowing and easily handled.  Figure 17 shows the results of manual
sieving of various bottom ash samples and indicates that the coal and dRDF
ashes are equally well handled by a conventional pneumatic ash handling
system.

     Because the spreader paddles were worn, a significant amount of fuel
"dribbled" onto the front of the grate and subsequently into the ash pit.
Since this fuel had not ignited on the grate, mixing the unignited coal with
the hot bottom ash often resulted in fires in the ash pit.  When the ash pit
had fire, the blend ash tended to be more plastic than coal-only ash and had
a taffy-like consistency.  The coal ash under similar conditions was much
easier to break up.

     The ash flow from the various blends was monitored by first cleaning the
ash pit completely.  Then, after the ash was allowed to accumulate in the pit

                                      45

-------
cc
3
_J
OJ
Q
    12 -
CO
LU

Z  8-
LU
*
<
I
CO
LU
4 -
                            AD
           OUTLIERS
            O.i
                0.2     0.3     0.4      0.5
                                                   BLEND
                                                    I --0
                                                    I •• I
                                                    I -• 2
                                                    0 I
0.6     0.7
             BOILER
               I 2
               o •
               A A
               O •
                *
0.8
                          £ - FRACTION OF RATING

  Figure 16.   Effect of blend  and load on grate pulse  interval
               or relative  grate  speed.
                                46

-------
BOTTOM ASH SIEVE ANALYSIS  • MAncH 21
                           o MARCH 22
         BLB4D 1:0            A MARCH 28
                           A MARCH 31
                           o APRIL 1
                                               BOTTOM ASH SIEVE ANALYSIS
                                                         BLEND 1:1            o  MARCH 23
                                                      n.(KMll«-«»««l.BE
-------
  BOTTOM ASH SIEVE ANALYSIS
             BLEND ! :0
        *- HQst.m-tuMMt.efl E
a MAY 4
• MAY 3
o MAY 18
A MAY 17
x MAYS
BOTTOM ASH SIEVE ANALYSIS
           BLEND 1:1
                                                                  vROSLIN—RAMMLEM EXPONENT
                                                        -M-
• MAYS
a MAY 12
o MAY 13
* MAY 13P.M.
                                                                     PARTKXE SZ£ 0.
BOTTOM ASH SIEVE ANALYSIS
            BLEND 1 2
                           BOTTOM ASH SIEVE ANALYSIS
                                     BLEND 01
                                                                  '•RCSUN-RAMWLEfl EXPONENT
                                          I
                                       "  i
                                        1 £

                                       - 0 *
               i5 37* S  'S
           P,V*TICL£ S2E 0- I
                                                                     PAflTTOf SCE 0, INCHES
                                Figure  17  (concluded)
                                             48

-------
for 1 hour, it was manually shoveled from the ash pit into tared 55-gallon
drums and weighed.  As would be expected, the quantity of ash increased with
increases in dRDF.

Ash Silo—
     The ash silo is used for dry ash storage.  The bottom ash and the fly
ash are both transferred to the silo by the pneumatic ash handling system.
The ash is withdrawn from the bottom of the silo by a rotary drum mixer where
the ash is blended with water for dust control before discharge into a truck.

     The only problem with this system occurred during 100 percent pellet
firing when the bottom ash was so fine that it would not de-entrain properly
in the cyclone.  Rather, the paper ash particles, which were wetted by the
steam in the ejector, carried through the cyclone and eventually plugged the
ejector.  Although the ejector was easily cleaned, continued maintenance of
the ejector will likely require developing some other means for keeping it
clear.

Reinjection and Collector Fly-Ash Flows—
     To measure the change in rate of fly-ash reinjection into the furnace
when changing from coal to dRDF, a secondary measurement technique was used.
Because the reinjection and collector dust streams were piped directly into
the plant's pneumatic ash handling system, it was impractical to isolate the
flow from one collector and weigh it as had been done for the bottom ash.
Consequently, the flow was measured by installing a drain tube near the
bottom of the collector and reinjector hoppers.  Figure 18 illustrates this
sampling technique.  As fly ash accumulates in the hoppers, it flows down the
sides.  A portion of that flow will fall into the drain tube.  The amount of
ash collected in the drain tube over a fixed period of time indicated the
relative fly-ash flow in the hopper.

     The fraction of ash falling to the ash pit was relatively independent of
the type of fuel being used (e.g., it did not matter if the fuel was coal or
dRDF), but it was strongly dependent on the ash content of the blend.
However, the weight of the carbon fly ash accumulating in the reinjector and
collector drain tubes (see Figures 19 and 20) did not appear to have any
correlation to the ash content of the fuel.  Since the dRDF had a higher ash
content than the coal, it was anticipated that the ash flow into the drain
tube would increase with increasing dRDF substitution.  This occurrence
however, was not indicated by the data.

     The lack of variation in the ash accumulation in the drain tube as a
function of fuel ash content was perhaps due to incomplete burning of carbon
or the difference between coal and dRDF carbon fly ash particle gas
de-entrainment characteristics.  The data shows, however, that combustion
improved as dRDF was substituted for coal.  Figure 21 shows the carbon content
of the reinjector ash as a function of the boiler load and coal:dRDF ratio.
While the data has much scatter, a general trend toward improved carbon
burnout with dRDF substitution is evident.  Consequently, since the decreasing
amount of carbon with increasing dRDF substitution tended to offset the
increasing ash content of the fuel, the drain tube ash flow measurements for
various coal:dRDF blends could be deceptive.

                                      49

-------
                                          . 2" SCH 40 PIPE - .31 m LONG
                                        PIPE CAP
Figure 18.   Drawing of a typical  ash collection drain tube  to
             monitor relative ash  flow in collector and reinjector.
                                   50

-------
  1000
   800
   600
m
m
   400
cc
a

cc
O
5  200

LJJ

CC
     0


     0.00
                                      • .  2
                                      * •  •   •

                                *o  V    *   '
                                • °  8   a°   8
               0.10       0.20       0.30      0.40       0.50


                                     FRACTION OF RATING
                                                                 BLEND


                                                                  1:0  —

                                                                  1:1  —


                                                                  1:2


                                                                  0:1  —
                                                                         BOILER

                                                                           I 2
0.60       Q70
0.80
       Figure 19.   Ash  flows into reinjector hopper drain tube  for

                    coal,  blend,  and dRDF firings.
                                          51

-------
z
5
O
   1000
    800
                                                BLEND

                                                 1:0

                                                 1:1

                                                 1:2

                                                 0:1
                                                                       BOILER

                                                                        I  2
<
01
CD

(-
Z
<
   600
   400
o


O  200
O
                            * »
      0

      QOO
0.10       020       0.30       0.40      050

                   FRACTION OF RATING
0.60
0.70
080
          Figure 20.   Ash flows  into collector hopper  drain tube  for

                       coal, blend,  and dRDF firings.
                                          52

-------
                            REINJECTOR  ASH
    100
    80
%C
    60
    40
    20
                             BLEND

                              I = 0
                              I = I
                              I = 2
                          .   OH
                                       BOILER
                                         I 2
                                         o •
                                         A A
                                         a •
                              A  •
•A
 o
                          •
                          a
0.00
0.10
0.20
0.30
      0.40
                                                0.50
0.60
0.70
0.80
                                 ~ FRACTION OF RATING
              Figure  21.   Carbon content of reinjector ash for coal,
                          blend, and dRDF firings.
     As dRDF is substituted  for coal,  the  ash  distribution becomes finer.
Figures 22 and 23 show that  the 1:1  and  1:2  size  distributions are similar
to, but finer than, the 1:0  size  distributions.   The ash from firing a 0:1
blend produced the highest amount of fines.  A possible explanation of this
phenomenon is that as more paper  is  burned,  there is an increased amount of
fine ash which is easily lofted by the underfire  air.

Air and Gas Handling

System Description—
     Each boiler has three separate  air  systems:   underfire air, overfire
air, and an induced draft fan.  The  underfire  airflow was controlled from the
fuel flow to the furnace.  The overfire  air  was adjusted manually to achieve
the most smoke-free operations.   The induced draft fan was controlled by a
draft sensor in the upper part of the furnace  to  maintain a specified
negative pressure in the firebox.

     The underfire air and induced draft fans  are equipped with both electric
motors and steam turbine drives.   The overfire air fans are driven by an
electric motor.
                                      53

-------
 REINJECTOR
MCI FLY ASH
 COAL (1:0)            MAY. 1977

            n-ROSLIM-MAMMLER E1POMENT
• 5/3
A 5/4
c 5/16
x 5/17
+ 5/17
                  :L£ SIZE O MICRONS
REINJECTOR
BLEND 1 2
              PARTICLE SIZE D MICPOMS
                                  REINJECTOR
                                  BLEND 1:1
MCI
MAY 1977
                                                                                             S
                                                    « 5 12
                                                    x 5'12
                                  REINJECTOR
                                  d-ROF |0 1)
                                   5 14  77
                                           MCI
                                           MAY 1977
• 5/10
x 5/10
           Figure  22.   Reinjector ash  size  distributions  for  coal,
                          blend,  and dRDF  firings.

-------
COLLECTOR
COAL (1.0)
        MCI FLY ASH
COLLECTOR
           MCI
           MAY. 1977
                                      COLLECTOR
  MCI
                                                            MAY, 1977
                                                 n-ROSLIN—HAMMLER EXPOMCNT
                                                    PADTICLE SIZE D. HICHONS
                                        5/12
                                        5/13
                                                 COLLECTOR
MCI FLY ASH
                                      d-ROF(Ol)               MAY. 1977
                                                 n ROSLIN-RAMMLEP. EXPONENT
                                                      ^u
                                                                            •t r
                                                                            K *
                                                    PARTICLE SIZE D. MICRONS
Figure  23.   Collector ash size distributions for coal,
              blend,  and  dRDF  firings.
                               55

-------
 Underfire Air Setting—
      When the underfire air fans were set for manual control to follow the
 steam flow, a clean ash bed was maintained; but when they were placed on
 automatic control, clinkering occurred.  The controls were designed so that
 the amount of fuel fed to the furnace is varied to maintain boiler pressure.
 In addition, the airflow is controlled proportionate to the fuel flow.
 Consequently, during a load shed, fuel on the grate could deliver an extra
 0.6 to 1.3 kg/sec (5,000 to 10,000 Ib/hr) of steam due to the lag time
 associated with this control mode.  As a result, the ash bed could change
 from oxidizing to reducing conditions which promote clinkering.   This
 condition occurred when the airflow followed the fuel feed rate rather than
 the amount of fuel in the furnace.  To prevent this problem during the May
 testing, the underfire was biased upward as though the boiler were carrying
 1.3 kg/sec (10,000 Ib/hr) more steam than it actually was.  When the controls
 were set for 50 percent excess air, rapid load sheds induced clinkering.
 After the air was biased upward to provide 20 percent more excess air,  the
 load sheds caused no problems.

      The need for this biasing was attributed to the appreciable amount of
 underfire air leakage around the grate.  When the boiler was inspected
 before testing,  the seals between the grate and the windbox appeared to be
 tight.   However,  when a water-cooled lance was used to probe the furnace
 during operation,  the excess air levels at 1 m (3 ft) and 3 m (10 ft)  above
 the center of the  grate were 30 and 50 percent,  respectively.   In the center
 of  the duct  leaving the boiler,  the excess air levels were about 120 percent.
 Figure 24 shows  the carbon dioxide level inside the furnace compared with
 that at  the  furnace outlet.   Figure 25 shows the excess air in the breeching.
 Figure 26 shows  the leaning of the flame toward the center of  the furnace
 which confirms  the hypothesis  of excess underfire air leakage.

      In  view of  the  above,  it  appears that a coal:dKDF blend can be  fired at
 the same settings  as coal only if adequate air is available to prevent
 clinkering during  load  sheds.   In addition,  excess  air levels  of 50  percent
 should be attainable in a tight  boiler since the excess air at the grate  in
 the test boiler with significant air leakage around the grate  was 30 percent.

      When firing 100 percent dRDF in March,  no problems were encountered  when
 the  air  control was  set on manual at an air  flow typical of that required
when burning  a specification stoker coal and supplying a desired steam  flow.
 In  May,  however, when pellets  were  first introduced,  ignition  problems
developed when the underfire air was automatically  controlled.   A malfunction
of  the air control system resulted  in a 600  percent excess  air condition  on
the  fuel  bed which nearly extinguished the flame.   Once the ignition problem
was  diagnosed and  the air controller was put  on  manual (it  could  not  be
biased enough to stay on automatic),  the fire  returned  to  normal,  and the
feeders  operated at  a low fraction  of  capacity with steam  pressure returning
to normal.

Overfire Air Setting—
     Combustion air is normally  provided through  the  grate  as  underfire air
so that the fuel is supplied with sufficient oxygen  to  ensure  its  complete
combustion.  However, because  of  the  different combustion  rates  of various

                                      56

-------
      10
                                            PROBE 6 FT
                                            ABOVE GRATE
                                            (CENTERED IN FURNACE)
%CO2  8
                        PROBE 10 FT
                        ABOVE GRATE
                        (1/s INTO FURNACE
                        FROM WALL)
                        BOILER LOAD - 30% OF DESIGN RATING
                                   BREACHING
                  o    o
                                    o    o
                11:00
12:00
13:00
  TIME
              Figure  24.  Carbon dioxide  levels in  furnace vs.  time as
                           determined with a water-cooled probe.
                                        57

-------
140
40
                        '•    • A 1   A«  AO  A     00
            O.I
      0.4       0.5

- FRACTION OF RATING
BLEND
 I =0
 I •• I
 I = 2
 0 = 1
                                   BOILER
                                     I  2
                                     o •
                                     A A
                                     a •
                                       *
                                                                DESIGN EXCESS AIR
                                                                       LEVELS
0.6
        0.7
      Figure  25.   Relationship of furnace excess  air level with
                   blend  and load.
                                   58

-------
10
                Figure 26.  Grate seal leakage forcing flames toward center of furnace
                            during a 1:1 blend firing in Boiler No. 2.

-------
  lumps of  coal and  the nommiformity of coal distribution onto the grate, some
  of the air  coming  up through the grate into the combustion zone is channeled,
  and it bypasses the areas where it is needed.  With such channeling, pockets
  of pyrolysis gas form above the grate.  If these pockets of hydrocarbons pass
  through the flame  zone without being exposed to adequate oxygen to complete
  combustion, smoking will occur.  To prevent smoking, the gases must be mixed
  so that they do not drift uniformly upward and out of the furnace without
  combustion.  These pyrolysis gases are mixed with overfire air jets.  The
 proper use of overfire air jets is particularly critical when a boiler is
 being operated at a low fraction of load.

      In order to find the proper setting for coal:dRDF operation,  the overfire
 jet pressure was increased in 12.7 mm (0.5 in.) of water increments starting
 at 12.7 mm (0.5 in.)  of water until the opacity meter showed  no reduction in
 smoke with increasing jet pressure.   Because of time constraints and the
 amount of fuel available, this procedure was iterated three times  for the
 front and rear rows of  overfire air jets for each blend during the first day
 of each test.   After  the minimum smoking settings were determined,  they were
 used  in the subsequent  tests.

      The differences  between the settings  for  coal-only and blend  firing were
 as follows:   When  firing blend,  the  air pressure applied to the front
 5.08  x 1.9 cm (2 x  3/4  in.)  rectangular overfire air jets  was reduced from
 178 to 114 mm (7 to 4.5  in.)  of  water  while the air pressure  applied to the
 rear  5.08  x  1.9  cm  rectangular  overfire air jets was increased from 38.1 mm
 (1 1/2 in.)  to approximately 76.2 mm  (3  in.) of water.   The pressure for the
 rear jets  was  increased  because  the high volatile content  of  the dRDF produced
 a large  fireball which enveloped more  of the furnace volume than the fireball
 from coal-only firing.   The  pressure for the front  jets  was decreased because
 the fuel was burning more in suspension  and further to  the  back of  the furnace.
 Consequently,  there was  only burnt-out  fuel in  the  forward  portions  of the
 grate  which  required less overfire air  than normally needed for  coal.

     Regardless  of  the blend  employed,  including 100 percent  pellets,  the
 amount of  overfire  air required  was the same.

 Induced Draft Fan—
     The induced draft fans were adequate  for all  the  tests performed.
However, the boilers were not fired above  70 percent of  rating on blends
because of limitations in steam  demand and  make-up  water equipment.   Attempts
 to operate at 100 percent  of boiler design  capacity by blowing steam were
aborted when it was determined that the make-up water  equipment  had  insuffi-
cient capacity to handle  the make-up water  flow.  Consequently,  it could  not
be determined whether the  existing induced  draft  fans have  sufficient  capacity
for firing various  blends up to  100 percent of  rating.

     A study of the fuel properties listed  in Table 6 reveals  that the
air:fuel ratio for  coal-only and coal:dRDF  blends is approximately the  same
with blends requiring slightly less air per megajoule heat  release than  coal.
This is particularly significant because if coal and blends can  be fired  at
the same excess air levels, the amount of gas passing through  the boiler  to
the collectors and  fans will be the same when the boiler efficiencies  are

                                      60

-------
equivalent.  However, as discussed subsequently, the efficiency of a boiler
firing dRDF is expected to be slightly lower than the efficiency of a boiler
firing coal because of the higher hydrogen content and bound moisture of the
blend.

     If the feeders do not limit the substitution ratio which can be used
without derating the boiler, then the fan capacity could become the limiting
factor.  The induced draft fan, rather than either the overfire air blower or
the underfire air fan, would most likely have an insufficient capacity.

Furnace Performance

     After the fuel has been mixed with air and ignited, it burns to release
the chemical energy in the fuel.  The heat is recovered in two different
sections of the boiler to produce steam.  First, some heat is transferred by
radiation to boiler tubes in the furnace walls.  Second, additional heat is
withdrawn from the gases as they pass through the convection section of the
boiler.  The overall performance of a furnace-boiler combination depends to a
large extent on the radiant heat transferred in the furnace and the removal
of sufficient heat from the combustion products so that the fly ash is solid
and not molten in the convection section.  The heat transfer characteristics
of the rest of the boiler are governed by the gas properties and the mass
flow rate.

Heat Release Rate—
     The design heat release rates for Boilers No. 1 and No. 2 were:
961 MJ/m3/hr (25,800 Btu/ft3/hr) and 616,000 MJ/m2/hr (543,000 Btu/ft2/hr)
for Boiler No. 1 and 1002 MJ/m2/hr (26,900 Btu/ft3/hr) and 662,000 MJ/m2/hr
(583,000 Btu/ft2/hr) for Boiler No. 2.  Figures 27 and 28 show the heat
release rates attained during the tests.  The maximum heat release rates were
low because of low steam demand requirements.  Even in the coldest time of
the winter, only 7.1 kg/sec (56,000 Ib/hr) of steam were required to meet the
heating needs of the Maryland Correctional Institute.  Consequently, the
tests do not show the effect of coal:dRDF blends on boiler performance at or
near design heat release rates.

Flue Gas Temperature—
     Of particular concern to a boiler operator is the temperature of the
gas leaving the boiler.  This temperature indicates the amount of potentially
available energy that is lost to the environment.  Figure 29 is a graph of
the flue gas temperatures in the furnace when firing coal-only and coal:dRDF
blends.  The exhaust gas temperature characteristics of Boilers No. 1 and
No. 2 differ.   The exhaust temperatures also differ when firing coal-only or
when firing coal:dRDF blends for the same boiler load.

Fouling, Slagging, and Wastage—
     Coal:dRDF blend firing resulted in occasional slagging, slight fouling,
and perhaps slightly higher than normal wastage.  Since these are areas of
major concern to potential fuel users, a detailed discussion is presented
although the total evaluation effort is not complete or conclusive because of
the relatively short duration of testing and the low boiler loads maintained
throughout the tests.

                                      61

-------
f-
m
    500,000
    400,000
   300,000
    200,000
    ! 00,000
                                                               BLEND
                                  BOILER
                                   I  2
                                   o •
                                   A A
                                   n •
                                     *
000
0.10
0.20
0.30
                                               0.40
                                              0.50
0.60
                               - FRACTION OF RATING
          Figure  27.   Heat  release  rates per  unit grate area  for
                       coal, blend,  and dRDF firings.
                                      62

-------
BOILER
25000



20000
15000
s—
\
13
H- 10000
m

5000
n
BLEND 1 2
| : 0 0 •
| : | A A
1 : 2 n •
• 0 : 1 *
o
A
* • ° A A
a
a
i i i i j i
    0.00     0.10       0.20      0.30      0.40      0.50     0.60





                         jL- FRACTION OF RATING







Figure  28.  Heat  release rates  for  coal, blend,  and  dRDF firings.
                                  63

-------
   "t O
   co in
   CM
   O O
   CD O
   CM ID
 LU
 ID O> O
 pi CM in
 < cj ri-
 ce
 LU
 Q.

 LU

 CO •* O
 < o o-
   CNJ rj-
   o o
   N. in
   i- CO
                                                           1 - MFG PROJECTION
     0.00     0.10      0.20     0.30      0.40     0.50
                                  •£ ~ FRACTION OF RATING
0.60
0.70
0.80
        Figure 29.  Effects of blend and  load  on  flue gas temperature.
      Since the boilers were operated at low loads  throughout  the  tests,
 furnace  temperatures were not high.  Even under these conditions,  the  boiler
 deficiencies were corrected with simple adjustments.  The  findings imply that
 a boiler owner should not anticipate immediate failures.   The long-term
 effects,  however, are as yet unknown.

      When pellets impinge on a wall, they tend to  stick and burn  because of
 the low  fusion characteristics of the dRDF.  For fuels with a higher ash
 fusion temperature,  the material rebounds and falls to the grate  where it is
 burned.   Severe slagging occurred during a portion of the May tests when
 one-third of the rear wall of the furnace opposite Spreader No. 1, which was
 maladjusted,  was covered with slag.  The remaining two-thirds of  this  wall
 remained  clear.   This slag was generally loose and could be easily removed.
 Fouling also accumulated rapidly on the leading rows of the convection section
 of the boiler in line with the maladjusted spreader.  The remaining two-
 thirds of  the convection section remained clean.   The fouling was  very loose,
was easily removed by rapping, and had a porous structure.  At the end of the
blend firing,  the/leading tube elememts had a velvet-like ash accumulation.
The same  type  of coating was found at the conclusion of the low-grade  coal
burning in December.   The potential hazard of this coating cannot  be assessed.
                                      64

-------
     After the spreader throw on Spreader No. 1 was adjusted, the accumulation
ceased.  By the time the boiler was brought off line 8 days later for inspec-
tion, the slag had sloughed off.  Inspection of the furnace interior revealed
that a glassy slag layer had accumulated on the lower portion of the side
wall where a grate clinker had contacted the wall.  This material was easily
removed.

     The reducing atmosphere hemispheric fusion temperatures for the ash
processed through the system for the December, March, and May tests are
graphed in Figures 30 through 32.  The low December ash fusion temperatures
for coal explain the clinkering observed.  The March and May results show
that blending coal and dRDF depresses the fusion temperature of the coal.
Interestingly, the dRDF-only bottom ash shows a higher fusion temperature
than coal.  If the glass fragments blow out of the furnace while the fuel  is
in suspension and leave only paper ash, a higher fusion temperature is
possible.
     O u_
     o  o
     ? O
     CO «
     t- CM
Q.

LU
     ™
       8
     0) O
     Tf O.
      CO Q
      O> O
      O O'
      »- eg
              71
                                                       EOT TOM ASH


                                                       COAL
              1:0
                           1:1
                              BLEND
    Figure 30.  Variations in ash fusion hemispheric  temperatures  under
                reducing atmospheric conditions during December  runs.
                                      65

-------
0.

01
   o u.
   o  o
   (Q O



   " CM
   o
   CD
   CM
   3 8
   CM PJ
   »- CM
   § 8_
   go o
     o'
     CM
                           COAL




                           BOTTOM  ASH




                           RE1NJECTOR ASH




                           DRDF
                1:0
  1:1


BLEND
                                                          1:2
    Figure 31.  Variations in ash fusion hemispheric  temperatures under

                reducing atmospheric  conditions during March runs.
                                     66

-------
    O
   CM  O
   00  O
   ^  O
   l~-  O
   ro  m'
   •*-  c\
H  .DR«,E
                                                               |g
-------
     In Figures  33  and  34 the flue gas temperature plots before and after the
boiler was  fired with blends show that there was some change in the character-
istics of the heat  transfer surface.   Because the "after" data lie outside
the confidence band for the "before"  data, the heat transfer surface had
deteriorated somewhat.   The blend ash was more insulating and/or more fouling
than the coal-only  ash.   The increase in exhaust gas temperature was greater
after 48 hours of blend firing in March than after 136 hours of blend firing
in May.  It should  be noted however that because different coals were used
for the two tests,  the  fouling may be related to the mixture properties and
not just to the  presence of dRDF.
       03
LU
       O
       LU
       CC
       H-
       cc
       LU
       CL

       LU
           LO O
           T- O
           m CD
           o o
           CD O
           "* 0
           O O
           OJ -t
   o> o
   rf O
   T- CO
           8-L
                                         • o  MARCH 31 & APR. 1
0.00     0.10     0.20    0.30     0.40
                ^ - FRACTION OF RATING
                                                    0.50
                                                     0.60
        Figure 33.   Effects of blend and  load  on flue gas temperature
                    before and after a  test.
                                      68

-------
            s> s-
            I? 8
            co co
            § 8
            CM m
        C3   -to
        MJ   O  O
        g   OJ  -t
        LL
        O
        UJ
        tr

        I
        £
        0.
        UJ
o>
2  8
               CNJ
             00 §
             co O
             CO
                                 00
                                                    	 •  MAY 4 & 5
                                                    	o  MAY 16 & 17
               0.0     0.10     0.20     0.30     0.40
                                £ —FRACTION OF RATING
                                          0.50
0.60
          Figure  34.   Effects of dRDF and load on flue gas temperature
                       before and after a test.
     During  the May tests,  eight corrosion test specimens were  installed  on
the downcomers in the rear  screen wall 1.52 m (5 ft) above the  fuel bed.
These specimens were clamp-on erosion shields similar to the one shown  in
Figure 35.   Since the shields were bolted in place before bringing the
boiler on line, they were exposed first to coal-only firing (1:0); then to
1:1, 1:2, and 1:1 blend  firings; next to 0:1 (100 percent pellets) firing;
and finally  to coal-only firing again.  At the end of 478 hours of exposure,
the specimens were removed  from the furnace and cleaned by the  procedures
described in Corrosion Engineering.2  The weight loss of the specimens was
converted to a wastage rate in mils per year by Equation 4.4.
     2Fontana, M. G.,  and M.  D.  Green.
N.Y.C., 1967.
                             Corrosion Engineering.  McGraw-Hill,
                                      69

-------
                                    25.4 cm (10 in)
                                           SHIELD FABRICATED  FROM
                                           111,304,309,316 S.S.
         Figure 35.  Drawing of typical  clamp-on corrosion test shield.
                                 MPY  =
534 W
 DAT
(4.4)
where W = weight loss, mg
      D = density of specimen, g/cm
                               ip
      A = area of specimen,  in.
      T = exposure time, hr

     The wastage rate was less than  127 micrometers  (5  mils)  per year for
carbon steel while the wastage (weight loss)  of  some  stainless steel specimens
(309 and 310) was not even detectable.

     One specimen of 1018 cold rolled steel was  in line with  the maladjusted
spreader.  It had a relatively high  wastage.   Since  the surface of this
material was exposed to burning material,  it  was in  a strong'reducing atmos-
phere.  A twin test specimen not  subjected to fuel impingement had a wastage
rate of only 76 micrometers  (3 mils) per year.   Table 8 lists the data used
to calculate the wastage rates.   While the data  provided by this corrosion
test provides some guidance, further corrosion testing  needs  to be carried
                                      70

-------
 TABLE 8.   METAL WASTAGE RATE DATA FOR  EIGHT SPECIMENS
                                        Specimen
No.
1
2
3
4
5
6
7
8
Material
1018
304
1018
304
309
310
309
310
Location
1
2
3
4
5
6
7
8
W
mg
19310
3620
2007
4111
n/d
50
240
n/d
ID
g/cm3
7.86
8.02
7.86
8.02
8.02
8.02
8.02
8.02
(&)
100.00
98.75
98.75
97.50
99.38
100.00
98.75
98.75
T
hr
478
478
478
478
478
478
478
478
Mils/y
2700*
5.00
3.00
6.00
n/d
0.07
0.34
n/d
mm/y
.69
.13
.08
.15
—
.002
.009
—
On.Line
 Off Line
               1400
               1200
              April 27, 1977
              May 17,  1977
                                                   Elapsed Time
                                                     478 Hours
 *Specimen in  line with  spreader  spraying fuel directly on the back
  furnace wall.
o
8
 12    16    20
NO. OF TUBES -
                                                             1.52 m
                                                             (5 ft.)
24    28    30
                              71

-------
 out over longer periods of time (6 months to 1 year) with the boiler operating
 at or near rated capacity.

 Firing Phenomenon—
      During the combustion of a series of different mixtures, the coal:dRDF
 blends seemed to perform as well in the furnace as coal.   The principal
 differences at equivalent loads were that as the dRDF substitution increased,
 the height, intensity, volume, and violence of the fireball increased
 correspondingly.  Moreover, because of the intermingling of paper platelets
 with the fuel, more sparklers or live pieces of glowing char were carried by
 the combustion products toward the heat transfer elements.   These sparklers
 increased as additional dRDF was introduced into the furnace.

      Figure 36, a set of photographs taken through a port in the side wall of
 the furnace 3.3 m (10 ft) above the grate, shows the extent to which the
 flame filled the furnace volume.  As seen in the progressive photographs,  as
 the dRDF subsitution ratio increased,  the number of sparklers and the height,
 intensity,  violence, and volume of the fireball increased.

      Figure 37, a photograph taken through the tube removal hatch in the top
 of the furnace, shows the fire distribution for a 1:1 blend.   As seen in this
 figure,  the furnace front has a relatively good burnout,  and the furnace rear
 has a square fire line.   The two zones of high-intensity  flame in the fire
 bed indicate that the fuel was spread  in two distinct waves:   one at the rear
 of the furnace and the other in the middle of the furnace.   This nonuniform
 spreading was  not detected during  the  cold flow test.   However,  it did not
 seem to  have any impact  on the degree  of burnout.   The lighter fire to the
 front left  was the result of starving  Spreader No.  2 to rectify a worn r6tor
 deficiency  which caused  an excessive left throw of  the fuel.

      The  radiant heat transfer characteristics of the fireball probably
 improved  as more dRDF was substituted  for coal.   Figure 38  shows flame
 temperature versus  load  as a function  of blend.   The flame  temperature was
 measured  by a  Leeds  & Northrup optical pyrometer focused  on the  middle of  the
 fireball  at the center of the furnace.   Since an optical  pyrometer measures
 the  product of emissivity and temperature,  an increased reading  would indicate
 that  the  furnace radiant  heat transfer characteristics have improved (assuming
 excess air  remains  constant).   Consequently,  the data indicates  that substi-
 tuting dRDF for coal  would likely  have a slightly beneficial  effect on the
 heat  absorbing capacity  of the furnace.

 Boiler Controls

      The  boiler  controls  in  the  Hagerstown  plant are  typical  of  those installed
 in heating  plants in  the  mid  1960's.   They  consist  of  a master  controller
which modulates  the  fuel  supply  in  response  to  changes  in boiler pressure.
As the steam demand on the  facility  increases,  the  pressure in  the  steam
distribution system decreases, and  the  amount  of fuel  introduced into the
furnace is  increased.  The underfire air  is  modulated  in  response  to fuel
flow.  In response to  load changes,  the  cam  connecting  the  underfire air to
the fuel flow  causes  the  air  to modulate  to  "optimum"  firing  conditions.
Consequently, when there  is a  load shed,  the  airflow  drops  while  there  is


                                      72

-------
                                              1:1  Blend  Firing
                                             1:2   Blend  Firing
                                             0:1  dRDF Firing
Figure 36.  Furnace flames viewed at  3.3 m (10 ft)  above the
           grate during blend and dRDF firings.

                              73

-------
                            BOILER REAR WALl
COMBUSTION
   GAS
 SAMPLING
  PROBE
     Figure  37.  View  from top tube hatch in Boiler No. 2
                show  firing with a 1:1 blend.
to
                                  74

-------
        co
    LU
    I
    LL
    U-
    o
    UJ
    cc

    I
          CM
        CD O
        *- O
        CD •*
        *- CM
 O
        •* O
        co o
        o> o
        o o
                                                    BLEND
                                                      1:0  -
                                                      1:1  -
                                                      1:2
                                                      0:1  -
                                                         BOILER

                                                          1  2
                                                          o  •

                                                          &  A

                                                          O  •

                                                            *
0.00    0.10      0.20       0.30      0.40
                   •£ — FRACTION OF RATING
                                                       0.50
                                                         0.60
       Figure  38.   Effects of blend and load on flame temperature
                    measured with an optical pyrometer.
still fuel in  the  furnace for the higher load level.  During such a load
change the fuel bed  has  reducing conditions.   Conversely, during a load gain
the air input  leads  the  fuel increase,  and the fuel bed has oxidizing con-
ditions.  This type  of control network can cause problems when the fuel on
the grate is a low fusion coal or a coal:dRDF blend.  During the co-firing
tests, some clinkering occurred during load sheds.  This clinkering was
overcome in the tests by biasing the underfire air control upward.  With a
combustion control system designed to maintain constant oxygen levels in the
flue gas, this clinkering should not occur.

     The control system  and feeders allowed the boiler to follow the load
without any discernable  difference in pressure fluctuations in the header
when firing coal only and a 1:1 blend.   Circular charts for the steam pressure
are shown in Figure  39.   The modulations in steam pressure were minor for
coal only and  a 1:1  blend;  however, when a 1:2 blend was fired, the feeders
were volume-limited  and,  as a result, had a lag such that the pressure
modulated 7 to 14  kPa (1  to 2 psi) in a sawtoothed pattern.
                                      75

-------
Figure 39.  Pressure chart recordings for coal, blend, and dRDF firings.
                                    76

-------
     When  100  percent pellets were  fired,  the feeders were the  rate-limiting
component.   The 9.9-kg/sec  (78,500-lb/hr)  boiler could maintain 7.1 kg/sec
(56,000 Ib/hr) of steam per hour with  100  percent pellets and manual airflow
control.   When the 7.6 kg/sec (60,000-lb/hr) boiler burned 100  percent pellets
while using  the automatic air/fuel  ratio controller, unburned fuel accumulated
at the base  of the rear wall, the entire fuel bed had a very sparse fire, and
the steam  pressure dropped 413 kPA  (30 psi).  This condition was due princi-
pally to high  excess air levels.  Because  of the control system limitations,
the air control could not be adequately biased downward; consequently, there
was a 600  percent excess air level  in  the  furnace.  The difficulty with the
boiler control was easily resolved  by  taking the underfire air  modulation
control off  automatic to allow an operator to manually control  the airflow
and to track the steam load rather  than the fuel feed rate.  With the air
control in the manual mode, steam pressure variations were about 14 kPa
(2 psi) with the air supply fixed for  12 percent C02 at the peak of a load
swing.

Mass and Energy Balance

     Table 9 summarizes the boiler  efficiency data which were calculated by
the AMSE Short Form loss method.  The  complete forms are included in
Appendix B.


           TABLE 9.  HEAT BALANCE SUMMARY BASED ON AS-RECEIVED FUEL
           BLEND                  1:0      1:1      1:2      0:1


           PARAMETER

           Fraction of Rating         .17      .33      .30      .19

           Excess Air (%)         104      82       99       113


           LOSSES

           Dry Gas               17.9     13.7     17.8     19.4

           Fuel Moisture             .1       .9      1.2      4.0

           H20 for H2 Combustion      4.0      5.1      5.4      8.1

           Combustibles in Refuse    18.3     25.3     16.6      3.0

           Radiation               3.7      1.8      1.8      3.7

           Unmeasured              1.5      1.5      1.5      1.5


           TOTAL                 45.5     48.3     44.1     39.7


           EFFICIENCY             54.5     51.7     55.9     60.3
                                        77

-------
          The  high carbon losses in the refuse are most  unusual and account for
    the  extremely low efficiencies (normal spreader-feeder  efficiencies are
    between  74  and 80 percent).  The carbon content of  the  refuse (bottom ash,
    collector ash,  and fly ash) for the various coal:dRDF blends  is shown in
    Figures  40,  41,  and 42.   The carbon content of the  three  ash  streams varied
    as follows:   (1)  bottom ash, 2 to 10 percent; (2) collector ash,  50 to
    90 percent;  and  (3) fly ash, 30 to 40 percent.
                            BOTTOM  ASH
%c
                                                          3LENC

                                                          I •• 0
                      BOILER
                        i  2
                                                          NOTE SiVPLES COLLECTED
                                                              FROM THE FRONT
                                                              OP THE GRATE
                                o
                                •
                                     A    A
                                      O
0.20
0.30
0.40
                                                    0.50
0.60
0.70
0.80
                                     - FRACTION OF RATING
     Figure  40.   Effects of blend and load  on  carbon content of bottom ash.
                                          78

-------
                             COLLECTOR  ASH
%C
    100
     80
     60
     40
     20
      000      0.10
                                      A

                                       A
BLEND


 i : 0
                                                   BOILER

                                                     I  2


                                                     o •

                                                     A A


                                                     a •
                                                                 SAMPLES COLLECTED
                                                                 ^ROV THE DRAI\ TUBE
0-20       0.30      0.40       0.50




           «£ - FRACTION OF RATING
0.60      0.70
              0.80
   Fieure  41.   Effects of  blend  and load  on carbon content of  collector ash.
                                           79

-------
                                    FLY  ASH
%C
    100
     80
     60
     40
    20
     0
0.00      0.10
                                                            BLEND
                                                             I --0
                                                             I •• \
                                                             I : 2
                                                             0 = I
            BOILER
              I 2
              o •
              A A
              D •
                                                            NOTE'- SAMPLES COLLECTED
                                                                  AT THE STACK
                                    I
                         020      0.30      0.40      0.50
                                  £- FRACTION OF RATING
0.60
0.70      0.80
     Figure 42.   Effects of blend and  load on carbon content  of fly  ash.
                                         80

-------
      The  mass balance for the various blends is presented in Table 10.  This
balance indicates that an abnormally high amount of ash was removed by the
collector.   Analysis of fly ash samples taken from the collector revealed
that  85 to  90 percent of the particles were larger than 50 micrometers.
Since collectors have much higher efficiencies for particulates in this size
range,  the  higher collector weights may be justified.  However, such an ash
weight  distribution is not typical of expected boiler performance.

                        TABLE 10.  ASH MASS BALANCE
Fuel
Blend Flow
Test Coal:dRDF kg/hr
=/
Ash
in
Fuel
Ash in
Fuel
kg/hr
Bottom Ash
kg/hr
Carbon With
Free Carbon
Fly Ash
kg/hr
Carbon With
Free Carbon
Collector Ash*
kg/hr
Carbon With
Free Carbon
May 4, 1977   1:0     872     21.9    191     82      89

May 13, 1977   1:1    1489     23.3    347    232     238

May 11, 1977   1:2    2035     23.4    476    324     341
 7.7

 6.8

10.2
104

110

145
219

369

300
 * The collector weight was determined by difference.
     In addition,  the  carbon content of the bottom ash in the various blends
varied little.  When the  dRDF substitution was increased, the fly ash burnout
was improved.

     Figure 41  indicates  that the carbon content of the collector ash dropped
significantly when the amount of  dRDF was increased in the 1:1 to the
1:2 blend firings.  Since this drop effectively offsets the wet flue gases
losses, the boiler efficiency did not change appreciably as more dRDF was
substituted for coal.   As a result, the boiler efficiency had only minor
differences over  the various coal:dRDF blend ratios tested.  This fact is
confirmed when  the input/ouput efficiency data are plotted as in Figure 43.
To determine the  input/output efficiency, the fuel in the feed trough was run
out and the steam integrator read.   After the feed trough was refilled with
weighed fuel, it  was again emptied  and the final integrator reading determined,
Dividing the fuel heat content into the heat content of the steam yields the
efficiency.  Even though  the steam meter was calibrated, the differences in
efficiency when comparing those calculated by the loss method with those
computed by the input/output method were significant enough to suggest a
constant multiplier error in the  steam meter.  Hence, efficiency values are
not provided in Figure 43.   The input/output efficiency data shown in
Figure 43 confirm the  heat loss calculations in that there is no distinguish-
able difference in the efficiencies when firing blends or coal.  This
conclusion is unique to the boiler  installation at MCI since the large amount
of ash and unburned carbon losses attributed to the collector is certainly an
anomally to expected boiler performance.
                                      81

-------
  INPUT/OUTPUT
     EFF.-%

   1 DIV= 10%
                                BOILER No. 2
               BLEND
                  1:0 •  MAY
                  1:0 o  MARCH
                  1:0 a  JANUARY
                  1:1 A  MAY
                  1:2 a  MAY
                  0:1 *  MAY
O       Q  Q
       Q
              L
              00      0.10     0.20     0.30     0.40     0.50

                                   •^  - FRACTION OF RATING
                    0.60
      Figure 43.   Effects of blend and load on input/output efficiency.
Low Load  Performance

     A positive  result  of  the tests was the substantially improved low-load
performance and  the  decreased plume opacity (indicative of better burnout)
when dRDF was  substituted  for coal.

     Normally  a  3:1  or  4:1 turndown ratio on spreader-feeders is considered
the practical  limit  to  avoid  severe smoking.  With the use of dRDF, this
turndown  ratio was increased.   Although of lesser importance for base-loaded
industrial plants, the  demonstration of the boiler's ability to operate at
extremely low  loads  is  particularly advantageous for heating plants and
institutional  facilities which must  support a very small summer base load yet
have sufficient  capacity to meet  severe winter heating requirements.

SUMMARY

     The  Hagerstown  experience has  increased the knowledge of blend behavior
in a spreader-feeder.   The fuel entered the furnace satisfactorily, burned
                                      82

-------
well, and met plant energy requirements.   The operative and control dis-
crepancies were all corrected by simple adjustments.  Some biasing of the air
controls was required to prevent slagging on the fuel bed during load sheds.
The only other limitations on the boiler operation occurred when the boiler
was operated on 100 percent dKDF.  During this test both the spreader and the
ash handling systems became capacity-limited.

     Proper adjustment of the spreader-feeders is critical to prevent slagging
and fouling.  Some slagging and fouling occurred (in excess of what would occur
when firing with coal only) on the walls slightly above the grates but was
readily removed.  The corrosion experiment resulted in wastage comparable to
what might be expected for coal-only firing.  This test was too short in
duration, however, to permit any definite conclusions on material wastage.

     While the boiler performance when firing coal:dRDF blends up to 1:2  (by
volume) was generally similar to that when firing coal only, final conclusions
must await further long-term demonstration testing in which boiler loads can
be established at rated capacity and satisfactory boiler operating character-
istics can be maintained.

     A follow-on demonstration test is currently scheduled to be carried out
in a spreader-feeder fired boiler with a rated capacity of 18.9 kg/sec
(150,000 Ib/hr) and superheat capability.
                                      83

-------
                                    SECTION 5

                            ENVIRONMENTAL PERFORMANCE
 INTRODUCTION

      If a boiler plant can comply with emission regulations  when firing  coal,
 the question arises whether it can still do so when substituting dRDF  for
 some of the coal.   The low sulfur content of the dRDF  could  degrade pre-
 cipitator performance which,  when coupled with the  high  ash  content of dRDF,
 could result in the emissions exceeding the regulation limits.   Also,  the
 high chlorine content of the  dRDF raises concern about the long-term corrosion
 effects on the boiler system.

      To answer this question, the current study assessed changes in emissions
 as  dRDF was substituted for coal.  Although the quantitative results for the
 coal:dRDF emissions are significant,  the principal  conclusions  are drawn by
 comparing the coal:dRDF blend emissions with coal-only emissions.  Therefore,
 base lines were established before and after each blend  run  by  duplicating
 all test conditions for coal-only firing.

      This section  discusses the  method of sampling  and data  analysis.  It
 also covers the impact of  substituting dRDF for coal on  particulate emissions,
 gaseous emissions,  and trace  compound  emissions in  the order given.

 FIELD  SAMPLING SETUP

      Since at least four 2-week  tests  were to  be conducted at the Hagerstown
 plant,  a weatherproofed test  shed enclosed in  sheet metal and readily
 accessible to the  stack on  two sides  (see  Figure 44) was  constructed for the
 environmental testing.   Placed 4.9  m  (16  ft)  above  the roof  line, the shed
 encompassed one  quarter of  the single  stack that served  all  three boilers.
 Four 0.1-m (4-in.)  half couplings were installed in the  stack to serve as
 sampling ports,  two on the  north side  and  the  other two  on the  west side,
with each pair arranged vertically  with  a  0.61-m (2-ft)  separation.  A stair-
well from the  plant  catwalk system  provided  access  to  the shed.  A trans-
missometer was  installed in another pair  of  ports spanning the  stack on the
north-south axis at  about 0.9 m  (3  ft)  above  the roof  line and  4.6 m (15 ft)
below the  other  sampling ports.

     One  port  in the stack  was used to  insert  a sintered steel  filter through
which stack  gas  samples were  extracted,  cooled  in a condenser/knockout box,
and piped  through heated umbilicals to  SYSTECH's  sample  analysis trailer.
This trailer housed  facilities for  wet chemistry  analyses.   In  the trailer,
the samples were distributed  through a manifold  to provide flue  gas to the

                                      84

-------
             Ill    ill
                  ^PHK.^-^
Figure 44.  View of MCI power plant  showing stack  sampling shed
           and temporary fuel handling system at  right.
                              85

-------
 gas analyzers.   A Theta Sensor,  Inc.,  trigas  analyzer  was  used  for  continuous
 oxygen,  nitrogen dioxide,  and sulfur dioxide  measurements.   A slip  steam was
 also distributed to an AID gas chromatograph  to  determine  both  total hydro-
 carbons  and the composition of hydrocarbons lighter  than C6.  The manifold
 was also used  to distribute calibration gases.

      Another sampling  port was used  to insert a  1/4-inch-diameter steel pipe
 into the center line of the boiler outlet.  The  pipe was connected  with
 copper tubing  to an Orsat  analyzer located on the boiler house  floor.  This
 sample system was used to  determine  the characteristics of  the  flue gas
 leaving  the boiler.  Figure 45 illustrates the overall arrangement  of the
 boilers,  breeching,  and locations where the various  ash, fuel,  and  flue gas
 samples  were taken.

 TEST CHRONOLOGY AND PROCEDURES

      Table  11 illustrates  the test chronology for the  entire  program.  The
 consumption of  dRDF throughout the program was 20.9  Mg (23  tons) in December,
 106.1 Mg  (117 tons)  in March,  and 127.9 Mg (141  tons)  in May  for a  grand
 total of  254.9  Mg  (281 tons).  During  the May test,  Boiler  No.  2 was con-
 tinuously fired with coal:dRDF blends  for 132 hours.   The  test  time breakdown
 for  the blend firings  was  58  hours for  the 1:1 blend,  53 hours  for  the
 1:2  blend,  and  29 hours  for  the 0:1  blend (100 percent dRDF).   The  table
 lists the fraction  of  the  boiler load  rating, the coal:dRDF ratio,  and the
 number of emission  data measurements for each test blend.   The  test program
was  designed so  that three  sets of emissions  data would be  acquired for each
 test blend.  While  good boiler testing  practices would dictate  that the
boiler be stabilized for 24 hours on each blend before collecting emissions
data, the limited supply of dRDF necessitated that the stabilization period
be limited  to overnight  (approximately  12 to  15 hours).  Each test blend was
subjected to the following battery of emission tests:

     3 each    -    EPA Method 5         -    Particulate mass flux, Cl,
                                             F,  S02, S03, and trace organic
                                             and inorganic  compounds

     3 each    -    Cascade Impactors   -    Particulate size distribution

     2 each    -    EPA Method 7        -    Nitrous oxides

     6 each    -    Orsat               -    C02, 02,  and CO
     8 each    -    Orsat               -    CO
                                               2
     4 each    -    AID Gas             -    Total hydrocarbons
                    Chromatograph

     4 each    -    Wahlco Probe        -    Resistivity

     1 each    -    Tedlar Bag          -    Record sample
                                     86

-------
00
                                                       NO it  SAMPLING PORTS FOR PARTICIPATE (FLUX
                                                             AND SIZ.E)SOX,NO,,02,CO CON-LINE AND
                                                             ORSATj HYDROCARBONS, CL,F, METALS AMD
                                                             RESISTIVITY ARE LOCATED IN THE STACK
                                                             26 FT. ABOVE "PHE BREECHING CENTER-
                                                             LINE.
                                      A3H SAMI'Lf 5<	


                              ' JJ    ID FANS
                               Figure 45.    Layout  of  Boilers No.  1, 2,  and  3 with  sampling
                                               locations  indicated.

-------
    TABLE  11.   CHRONOLOGICAL  LISTING OF  TEST CONDITIONS

Date
12/8/76
12/10
12/13
12/14
1/20/77
1/21
1/24
1/25
3/19
3/21
3/22
3/23
3/24
3/28
3/29
3/30
3/31
4/1
5/3
5/4
5/5
.w 5/10
''4 5/11
5/12
5/13
5/14
5/16
5/17

Boiler Load


Boiler No. Fraction of Rating Blend Data Replication
1
1
1
1
1
1
1
1
1
2
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
.45

.45
.43
.53
.47
.53
.50
.51
.40
.40
.43
.44
.38
.36
.26
.27
.35
.25
.20/.22
.31/.51
.36/.30
.36/.34
.347.28
.397.30
.277.26
.217.36
.17
1:1
1:0
1:1
1:1
1:0
1:0
1:0
1:0
1:0
1:0
1:0
1:1
1:1
1:0
1:2
1:2
1:0
1:0
1:0
1:0
1:0
1:2
1:2
1:1
1:1
0:1
1:0
1:0
1
2
2
2
2
1
1
1
2
2
1
2
1
1
2
1
2
1
1
2
2
2
2
2
2
2
2
2
Note:  Boiler No. 1  is rated at 9.9 kg/sec (78,500 Ib/hr) and
      Boiler No. 2  is rated at 7.6 kg/sec (60,000 lb.hr).
                                88

-------
     Continuous -   Theta Sensors, Inc. -    C02, S02> and N02
                    Trigas Meter

     Continuous -   Leads & Northrup    -    Opacity
                    T r ansmi s some t e r

     As Req'd   -   Draeger Tubes       -    CO, C02, and SOX

     The Battelle Tenex traps were only run three times throughout the test,
i.e., at 1:0, 1:1, and 1:2 test blends.  When the measured emissions produced
unusual results, the Tedlar bag sample was used to further clarify the data
through gas-chromatograph analysis.

     The Draeger tubes were used periodically throughout the program whenever
a quick, approximate concentration of a particular pollutant gas was desired.

     The following paragraphs detail the test procedures for collecting,
monitoring, and analyzing each emission.

Opacity

     Opacity was measured by a Leads & Northrup single-pass transmissometer
spanning the stack and calibrated with neutral density filters.  The trans-
missometer was calibrated with the neutral density filters by installing the
unit in a pipe section whose length was equivalent to the stack diameter.

Particle Mass Flux

     The particle mass flux was measured by an EPA Method 5 train, which is
schematically illustrated in Figure 46.  Because of the low gas flow rates,
nominal 12.7-mm (1/2-in.) nozzles were installed to produce nominal
0.0047-m3/sec (1-cfm) flow rates through Greenburg-Smith impingers.  The
3-m (10-ft) diameter stack was traversed from two sides by a single 3.7-m
(12-ft) stainless steel probe.

     The stack was sized for acceptable flow rates with all three boilers
operating simultaneously.  During testing, however, only a single boiler was
on line at part load.  Consequently, the velocity of the stack gases was less
than 2.1 m/sec (7 ft/sec) which is below the detection limit of an S-type
pitot tube.  Although sophisticated velocity monitoring equipment was con-
sidered, the large quantities of dust made its application inappropriate.  As
an alternative method for measuring the flow rates, the flue gas composition
at the stack, the boiler load and efficiency, and the ultimate analysis of
the fuel were used to calculate the velocity.  The mathematics relating these
parameters and the apparent pitot reading are presented in Appendix C.

     Because of the velocity conditions at the stack, the extent of aniso-
kineticism could not be determined precisely.  In any event, the error due to
anisokinetic sampling in these experiments is likely negligible because of
                                       89

-------
            ffn
'/,
PROBE
HEATED     FILTER HOLDER
AREA
                                                           THERMOMETER
                                                                          CHECK
                                                                          VALVE
             	i__	:_i_zi	
  REVERSE
  TYPE
   PITOT
   TUBE
1
                 PI TOT
                 MANOMETER
                                   IMPINGERS
                               THERMOMETERS
                                             ICE BATH
                ORIFICE
                                    BY-PASS
                                    VALVE
                                     VACUUM
                                     GAUGE
                                                              T
                                                        MAIN
                                                        VALVE
                                                                    VACUUM LINE
           MANOMETER
                DRY TEST METER     AIR TIGHT PUMP
         Figure 46.   Schematic of EPA Method 5  sampling train setup.

-------
 the  fine  particle  sizes  and  low gas velocities.3  As a general rule, when
 an aerosol  is  less than  5 micrometers  in diameter, there is no need for
 isokinetic  sampling.  At Hagerstown, approximately 65 percent of the par-
 ticulates were less than 5 micrometers.  In view of the large amount of less
 than 5-micrometer-diameter particulates present in the flue gas and the
 utilization of a calculated  stack velocity, the particulate emissions should
 be representative  of actual  plant operation.

 Size Distribution

      The  size  distribution of  the aerosol emitted from the Hagerstown plant
 was  monitored  in the stack downstream  of the multiclone collectors.  All
 measurements were  made with  an MRI cascade impactor which was operated
 according to the manufacturer's recommendations.  The method of forming a
 nonrebounding  substrate  for  the impactor plates was developed at SYSTECH
 according to other researchers' experience with the MRI cascade impactor.
 After the impactor plates were dipped  in benzene in which Apiezon-H grease
 was  dispersed,  they were baked overnight at 232°C (450°F).  The resulting
 coating was extremely uniform.  "Blank" test runs were performed in which the
 MRI  cascade impactor was inserted in the stack with a filter installed before
 the  impactor.   These tests confirmed that the coating on the impaction disks
 did  not come off on the  0-rings or during handling.  The substrate forming
 method, therefore,  proved to be an acceptable procedure.  Consistent particle
 size  distributions, as shown on the stages in Figure 47, further substantiate
 the  validity of this method.

      The  MRI cascade impactor  was always inserted 1.5 m (5 ft) in from the
 west  side wall.  After the impactor was heated, it was connected to the train
 shown in  Figure 48, checked, and inserted into the stack.  When the sampling
 was  completed,  the  impactor was removed from the stack and disassembled.  The
 impactor  plates were placed  in tared petri dishes.  After the plates were
 returned  to the laboratory for final weighing in a clean environment, the net
weight gain per stage was used to determine the cumulative mass distribution.
 The  characteristic  aerosol diameter of each stage was obtained from the
 factory-supplied calibration curves for the test conditions and unit density
 aerosols.

 Fly Ash Resistivity

      The  particle  resistivities were measured on site with a WAHLCO probe.  A
 sketch of this probe is  shown  in Figure 49.  Dust samples were cyclonically
 captured  from  the  stack  gases  in a collector cup.  The captured particles
 collectively became a resistor between electrodes A and B.  After a constant
voltage was applied to the electrodes and the current flow was measured, the
 resistivity was computed by substituting the voltage and current values in
     3Watson, H. H.  American Hygiene Association Quarterly, Volume 15
1954.  p. 21.
                                      91

-------
NJ
                Figure 47.  Typical  dust  loading of  MRI cascade impactor stages during 1:1 blend firing.

-------
                                                                                     CHECK
                                                                                     VALVE
CASCADE
IMPACTOR
HEAD
          GAS
          FLOW
/s
                                                                               VACUUM LINE
                       MANOMETER
             DRY TEST METER    AIR TIGHT PUMP
         Figure 48.   Schematic of MRI  cascade impactor  sampling train setup.

-------
                                         CYCLONE BODY
                                                   COLLECTING
                                                   CUP AND
                                                   ELECTRODE
  TO ASPIRATION
  ASSEMBLY
CAP WITH      STATIC
DISCHARGE     SHIELD
PIN AND PURGE
AIR FITTING
@ - ELECTRODE PIN

(S>- DISCHARGE PIN
            Figure  49.   Schematic of WAHLCO resistivity  probe  assembly.
Equation  5.1.   Particle resistivities  were continuously checked through  the
blend  and coal-only tests.
                        Resistivity  =  (voltage/current) L
             (5.1)
where L  =  constant
S02
     Sulfur  dioxide levels in the flue  gas  were determined by two different
techniques.   With one technique, S02 levels were continuously monitored by a
trigas meter manufactured by Theta  Sensor,  Inc.  This electro-chemical sensor
was calibrated  with standard S02 gases.   With the second technique, wet
chemistry determinations were made  by analyzing the sulfur level in the first
impinger of  the EPA Method 5 train.  A  10 percent hydrogen peroxide/water
solution in  the impinger was titrated with  barium perchlorate to yield a
measure of the  SOX in the flue gas.  Both samples were collected at the same
location; i.e.,  the stack.  When the results from the two techniques were
cross correlated,  the corresponding values  proved to be similar.  Since the
SOa concentration is an order of magnitude  less than the S02 concentration in
a flue gas stream,  this cross correlation was considered to be valid.
                                       94

-------
 Oxides of Nitrogen

      The emissions of oxides of nitrogen were determined by two different
 techniques.   In one technique,  the Theta Sensor trigas  monitor was  employed
 and in the second, an EPA Method 7,  a phenol  disulphonic test  method, was
 used.

 Halogens

      During  each particulate mass flux evaluation,  the  halogen emissions were
 simultaneously determined by replacing the water  in the first  impinger  in  the
 EPA Method 5 train with a 10 percent hydrogen peroxide  solution.  After the
 Method 5 testing was completed, the  first impinger  was  then analyzed for
 chlorine and fluorine by using  specific ion electrodes.

 Oxygen

      In addition to monitoring  SO and NO , the Theta Sensor trigas monitor
 had the capability of continously monitoring  oxygen.  All  three of  these
 parameters were continuously recorded on a strip  chart.   Standard calibration
 gases  were used to calibrate the instrument at the  beginning and end of each
 test day.  In addition,  Orsat readings were taken at the stack,  and the 02
 readings were cross correlated  with  the 02 readings on  the  Theta Sensor trigas
 monitor.   Identical readings verified the integrity of  the  sampling line.

 Hydrocarbons

      To determine the emissions of hydrocarbons from the Hagerstown plant,
 continuous gas  samples  extracted from the stack were passed through a sintered
 steel  filter, a condenser  knockout box,  and a heated umbilical  to the SYSTECH
 trailer where they were  analyzed in  the AID flame ionization detector-equipped,
 field-portable  gas chromatograph.  This instrument  was  operated  in the  total-
 izing  mode and  calibrated  with  a methane gas.   Consequently, the total  hydro-
 carbons are  expressed in terms  of a  methane equivalent.  The gas samples were
 fractionalized  with a molecular sieve column  capable of  distinguishing  between
various hydrocarbons  lighter  than C6.

Trace  Organic Emissions

     The trace  organic emissions  were collected by  a Battelle  Tenex sampler.
Figure  50  illustrates how  the sampler was connected after the  filter in the
EPA Method 5  train.   The sampler  was  maintained at  50°C  (122°F) by a recir-
culating water  bath.  After  the  samples  were  taken, the  probe washes, filters,
and Tenex  traps were  all preserved and  sent to  the  Battelle  Columbus Labora-
tories  for analysis.

Trace  Inorganic Emissions

     The inorganic  compounds  emitted  from a boiler  can be in either the
aerosol or the  gaseous phase.   To  quantify these emissions,  SYSTECH modified
the standard  EPA Method  5  train  so that  while  the train  used the normal
hardware,  the first  impinger was  loaded with a  10 percent hydrogen peroxide

                                      95

-------
                                                                          CHECK
                                                                          VALVE
                                               BATTELLE TENEX TRAP
  REVERSE
I  TYPE
1  PITOT
                                                                    VACUUM LINE
           MANOMETER       DRY TEST METER    AIR TIGHT PUMP

        Figure 50.   Schematic of Battelle  Tenex sampling train setup.

-------
 solution,  and the next  two impingers were  charged with a catalyzed ammonia
 persulfate reagent.   These impinger solutions were used to  ensure the capture
 of the gas phase metals.

      The probe washings,  filters,  and  impinger solutions were all analyzed in
 SYSTECH's  laboratories  by using  atomic absorption spectrometry with appro-
 priate detectors and  furnaces.   Each filter was cut in half, dryed, and
 desiccated.   The half to  be digested was then weighed for total particulate.
 The remaining half was  retained  for voucher or repeat analysis.  One to two ml
 of concentrated HN03  were added  to each filter and dried slowly on a heated
 steam table.   The filters were not dried completely to avoid losing volatile
 metals such as Pb.  Additional HN03 was then added to cover the sample.  The
 sample was refluxed until the digestion was complete.  The  reflux was removed
 from heat,  and after  the  sample  had cooled, concentrated HC1 was added until
 the sample was in solution.  After the sample was filtered, the filtrate was
 brought to a  known volume for atomic absorption analysis.   Filter blanks were
 analyzed along with the samples  to provide supporting data.

      Fly ash  samples  were dried, desiccated, and then halved' and quartered to
 obtain a representative 2-g sample.  The bottom ash samples were ground up to
 a  minimum  sieve size  in a Wiley  mill;  then the ground samples were halved and
 quartered,  and a 2-g  aliquot was taken for analysis.  These samples were
 digested by the same  procedure as previously described for  the filters.  With
 the high silica content,  it was  difficult to dissolve the entire sample.
 Consequently,  the digestion was  considered complete when the sample had a
 straw-like  color after a  minimum of 2  to 3 hours of refluxing.

      For the Hg  analysis,  a separate digestion was necessary.  Each sample
was weighed and  put into  a BOD bottle.  The procedure for analyzing Hg
 required a  persulfate digestion  as described in EPA Manual  of Methods for
 Chemical Analysis of  Water and Wastes.  All samples were analyzed in duplicate.

DATA ANALYSIS  AND NORMALIZATION

Data Analysis

     The data  management  and interpretation was complicated by the test
program and load  limitations; for example, only one boiler would be on line
at  a  time.   Since the single on-line boiler had to follow a modulating steam
 load,  the test matrix had  to include the steam load as a variable.  Conse-
quently, data  were taken  at various load points as the boiler met the varying
steam  needs.

     Since soot  formation,  flame temperature, boiler efficiency, hydrocarbon
emissions,  etc.,  are  all  functions of  the boiler firing rate, all data were
analyzed as a  function of  the boiler rating to minimize the effect of testing
in  two  different  size boilers.   Because of the limited number of data points
and the similarity of the  data taken on Boilers No. 1 and 2, all data were
combined.

     A method  of data interpretation different from usual techniques had to
be employed to  identify the effect of  substituting dRDF for different test

                                     97

-------
  coals with  varying  coal  chemistries.  The  selected method was based on the
  principle of  establishing  the range of likely values  for the reference
  parameter and then  applying  that  range to  the dependent parameters to ensure
  meaningful  results.4  Accordingly, with  this method a regression analysis was
  conducted with 90 percent  confidence intervals for the best fit lines based
  on the coal-only (1:0) data  set.  Figure 51 illustrates the principle.  After
  the data for  two different blends were independently regressed and the con-
  fidence interval of the  regression line  for the first data set (A) was
  established,  the two data  sets could not be called different if the regression
  line for the  second data set (B) was within this interval.  However, if the
  regression  line for the  second data set  (C) was where it would be outside
  this interval, the two data  sets could be  called different.

      Because of the marked disparity in the number of data points for the 1:0
 and 0:1 blend  tests (25 versus 2 data points), the slope of the 1:0 blend
 data as a function of load was used as the slope for all other emissions data
 graphically presented in this report.   A t-test procedure described by
 Natrella5 was used to confirm the reasonableness of this approximation.
 Results of this analysis indicated that at the 90 percent confidence level,
 the slopes of the 1:0 and 1:1 data sets would not be considered different.
 In fact,  even though sloped lines are  shown for each blend (based on the 1:0
 data),  a horizontal line (slope = 0) might also fit the data.   The 1:1,  1:2,
 and 0:1 data (designated as m:n)  were  fit with the 1:0 slope by realizing
 that  for the least  squared error under the common slope constraint the
 following is true:
 and
                               Y    = A    + B1  _X
                                m:n    m:n    1:0
A    = Y    - B_ nX
 m:n    m:n    1:0
                                                                       (5.2)
                                                                       v    '
where Y  and  X are  the average values  of the emission and load measurements,
respectively.

Data Normalization

     Since the  excess air  levels  varied throughout  the  testing,  all  results
were adjusted  to a common  reference to  remove  the dilution  effect.   Accord-
ingly, Equation 5.3 was used  to adjust  all  emissions to  the 50 percent  excess
air level.
                                              CO,
     4Murphy, T. D., Jr.  Design and Analysis of  Industrial Experiments.
Chemical Engineering, June 1977.  pp. 169-182.
     5Natrella, M. G.  Experimental Statistics.   National Bureau  of Standard
Handbook 91.  U.S. Government Printing Office, Washington, D.C.,  1963.
                                      98

-------
                                     90% CONFIDENCE
                                          INTERVAL FOR PLACEMENT
                                               OF A REGRESSION LINE
Figure 51.   Graphic representation of  probably similar  (A  & B)  and
             potentially dissimilar (A  &  C)  regression lines through
             data set.
                                  99

-------
      where      X2  =    emission level corrected to 50 percent excess  air

                 Xx  =    measured emission level at test conditions

                C02  =    carbon dioxide level at 50 percent excess air  as
                          calculated from fuel properties and stoichiometry

                C02  =    measured carbon dioxide levels at test conditions

            (1 + EA) =    stoichmetric air plus excess air

      Since carbon dioxide levels are direct measurements that correlate
 fairly well with excess air levels,  the carbon dioxide levels were used for
 all adjustments.  The expected carbon dioxide level at 50 percent  excess air
 can be determined from stoichiometry by using the fuel property data.   The
 reference carbon dioxide levels for all coal:dRDF blend and dKDF-only tests
 in March were 12.2, 12.6, and 12.3 percent.  These percentages correlate with
 the 1:1, 1:2, and 0:1 blends, respectively.

      The experiment was further complicated by varying fuel properties  when
 different coals and dRDF supplies were used.   The ultimate and proximate
 analysis for coal and dRDF shown in Tables 3  and 4 reflects the varying fuel
 properties during the test period.

      To facilitate detection of the  emission  changes when the coal:dRDF ratio
 was varied,  the fuel properties of the different coal and dRDF supplies were
 normalized by using the properties for the coal delivered for the  March tests
 and the properties for the average composition of dRDF delivered for all
 tests.

      Emissions were normalized by dividing the fraction of each element in
 the fuel per  joule equivalent by the amount of that element in the reference
 fuel.   This  correction assumes that  a constant fraction of sulfur,  for
 example,  is emitted as SOX regardless of  the  actual percentage of  sulfur in
 the fuel.  While this  correction ignores  secondary effects such as the  sulfur
 chemically bound with  the ash,  it compensates for primary effects.  The
 correction factors for particulates, SOX,  NOX,* and Cl are tabulated in
 Table  12.  The primary reduced data  in Tables A-l through A-4 of Appendix  A
 were multiplied by the appropriate factors  in Table A-5  before the statistical
 analysis  or plotting.   The  fluorine,  opacity,  and hydrocarbon emissions were
 not  corrected  to the normalized fuel properties  because  either the requisite
 data was  not  available (fluorine is  not measured in an ultimate analysis)  or
 the controlling parameter was uncertain.
     *The validity of correcting NOX organically bound nitrogen was question-
able since the NO formation was governed by combustion and flame cooling
rates, excess air levels, point of air addition, and recirculation effects as
well as fuel nitrogen.  Because of the low volumetric heat release rates
encountered and high excess air levels, the NO formation for this test should
have been governed by the fuel nitrogen only.

                                     100

-------
         TABLE 12.  FUEL ELEMENTAL COMPOSITION NORMALIZATION FACTORS
                    FOR ADJUSTING EMISSIONS TO A STANDARD FUEL
                                               Element

           Blend                   S         Cl       Ash        N2
December /January
1:0
1:1
March
1:0
1:1
1:2
May
1:0
1:1
1:2
0:1

.73
.51

1.00
1.00
1.01

1.23
1.22
1.20
1.00

.40
.72

1.00
.90
.93

2.00
1.20
1.09
1.00

.72
1.14

1.00
.99
.99

.42
.56
.62
.77

1.03
1.35

1.00
1.03
1.06

1.04
.93
.87
.56
PARTICIPATE EMISSIONS TEST RESULTS

     In the framework of compliance with emission regulations, the blend
effects on opacity, particulate concentration, size distribution, fly ash
resistivity, and overall ESP performance were evaluated.  The following
sections discuss these effects in the order given.

Opacity

     Figure 52 shows the opacity readings averaged over 8-hour test intervals
as a function of boiler load and coal:dRDF blend.  The comparison of the
confidence interval about the coal-only regression line with the best fit
curves for the 1:1, 1:2, and 0:1 firings indicates that the overall opacity
was reduced as the dRDF substitution was increased.

     Because of the large diameter of the stack relative to the amount of gas
discharged, the opacity appeared lower to ground-level observers than the
meter reading indicated.  As the plume left the stack, it was lazy, and it
immediately fanned.  While the opacity meter spanned a 3-m (10-ft) path, a
ground-level observer could see only about a 0.3-m (1-ft) path.  To adjust
the measured data to indicate the opacity which would be seen in a more
closely sized stack (4 feet versus the actual 10 feet), a second scale was

                                      101

-------
 included on Figure 52.   Beer's Law was used to apply the path  length  correction
 to the measured light attenuation.  Equation 5.4 is the  form of  Beer's Law
 which governs the transmittance of light across an attenuated  gap.

                                 I = I0EXP (-KL)                         (5.4)
 and
 where  I = perceived source  intensity
      Io = source  intensity
       K = extinction coefficient
       L = path length
       0 = opacity

     Assuming  that  shortening  the  path  length affects  only  the  extinction
 coefficient, Equation 5.5 is the manipulation of  Equation 5.4 to  show the
 impact of an altered path length on the measured  opacity.
                          02  =  1  -  EXP
(5.5)
where 02 =  opacity  at  L2
      Q! =  opactiy  at  Li

     Since  the Hagerstown boiler plant has  a  significant amount of carbon
carry over  when coal is fired, most of the  reduction  in the plume opacity can
be attributed to improved combustion conditions within the boiler when the
blends were fired.

Particlate  Concentration

     Figure 53 shows the particulate emission rate as a function of load and
blend.   Except for  the 0:1  (100 percent dRDF) firing, the particulate mass
flux in the flue gas was reduced with increasing dRDF substitutions, i.e.,
from the 1:1 to the 1:2 blend firing.  However, the reductions are not sig-
nificant at the 90  percent confidence level.  While the 0:1 firing produced
results that differed  from the data presented within  the 90 percent confidence
level,  the  limitation  of only two data points for the 0:1 firing precludes
definitive  conclusions.

     The data show  that when 100 percent dRDF was fired, the increase in the
fuel ash more than offset the reduction in  the fly ash carbon.  This improved
burnout is confirmed by the filters shown in  Figure 54.  These filters, which
were removed from the  EPA Method 5 train, show a color shift from the 1:0 to
the 0:1 blend firing.  The analysis of carbon in the fly ash, as shown in
Figure  42,  also indicates a reduction in carbon content with increased boiler
load and dRDF substitution for coal.
                                     102

-------
100 r
 80
%  60
CL
O
Q
UJ
DC
 40
 20
    - 38   *
    -31
  24  V
      cc
      O
      LL
  18  £

      O

"13  I
       - 9
    - 4
                                                              BLEND

                                                                1:0

                                                                1:1

                                                                1:2

                                                                0:1
                                                                  BOILER

                                                                   1   2

                                                                   O   •

                                                                   &   A


                                                                   a   •

                                                                      *
               0.10       0.20       0.30       0.40       0.50


                            £ — FRACTION OF RATING
                                                                  0.60
       Figure 52.  Effects  of blend and load  on stack opacity.
                                  103

-------
 O> OC
   a
o
CM IO
CD
05
CO
   CM
CM  i-
                                                      BLEND
                                                       1:0
                                                       1:1
                                                       1:2
                                                       0:1
                                                                               BOILER
                                                                                1  2
                                                                                O  •
                                                                                A  A
                                                                                O  •
   0.00
0.10
0.20         0.30         0.40
    «£   — FRACTION OF RATING
0.50
0.60
          Figure  53.  Effects of  blend and  load on particulate  mass
                       emission rate.  ,
                                       104

-------
                  EPA  Method  5
                       Filters
                      Coal/d'RDF
        1:0
                                           0:1
Figure 54.  Effect of blend on color of stack aerosol.
                         105

-------
 Size Distribution

      Typical of all data results,  Figure 55,  a size distribution  plotted on
 Rosin-Rammler paper, shows that the cumulative mass vs  size  distribution
 plots as a straight line.   The two probability plots of the  MRI cascade
 impactor data in Figures 56 and 57 are the averages of  all related values  for
 the March and May tests, respectively.  Appendix D  lists the data for each of
 the MRI cascade impactor runs.

      These plots indicate that as  dRDF is substituted for coal, the  aerosol
 size distribution shifts to the fines.  This  was expected for two reasons:
 First,  improved burnout of the aerosol produces smaller particulates for the
 same amount of ash.   Second,  the large number of fine paper  platelets in the
 dRDF causes the number of particles formed from burning a unit of fuel to
 increase.

 Fly Ash Resistivity

      The resistivity of the fly ash leaving a source is of interest  because
 the performance of  a precipitator  is governed by the total particulate concen
 tration,  the aerodynamic characteristics of the particles, and the resistivit
 During  design and operation the resistivity determines  the power  which can be
 applied to collect  the aerosol as  well as the extent to which the aerosol  can
 be  re-entrained into the flue  gas  stream after "collection."

      The coal-only  firing  produced a resistivity generally less than 106 ohm-
 This  unusually low  resisitivity is a direct result  of the high carbon content
 in  the  fly ash.   At  this level of  resistivity,  particles collect  at  the wall
 of  the  precipitator,  rapidly lose  their  charge,  and re-enter  the  gas stream.
 Consequently,  an ESP does  not  perform well when the particle  cloud has a low
 resistivity.   Figure 58 presents the resistivity results for  coal-only and
 blend firing.   The  1:1 blend fly ash had resistivities  about  1010 ohm-cm.
 This  resistivity is  within the range of  108 to  1010  ohm-cm required  for
 efficient  precipitator performance.   The fly  ash from the 1:2 and 0:1 blends
 may have  resistivities which are too high (1012  ohm-cm)  for  good  ESP
 collection efficiencies.   Table 13 illustrates  the  resistivities  of  the
 various blends  for  the March and May tests.

     The resistivity was not plotted against  carbon content or gas temperatui
 since the  data were  too limited to make  the results  meaningful.

Overall ESP Performance

     To test  the  aerosol control capability at  the  Hagerstown boiler plant
while firing  coal and  coal:dRDF blends,  a  field-portable, 5-cell  ESP was
installed  and  tested between April 28  and  May 17.   This  was the only period
when the precipitator  could be  used  at the  Hagerstown plant because of other
test demands.  The precipitator is owned by the  EPA  Industrial Environmental
Research Laboratory in  Research Triangle Park, North Carolina, and operated
by Monsanto Research Corporation.  The precipitator  system was housed in two
trailers,  one for the  five precipitator  cells with  independent power supplies
and the other for the  precipitator electrical monitors  and an aerosol

                                      106

-------
                 :m
               s »
               §:
                             vROSLIN-HAMMLER EXPONENT
                               PARTtCLE SIZE 0, MICflOMS
Figure  55.  Typical  MRI cascade impactor  results for  blend firing.
                                  107

-------
                                                               BLEND
                                                      MARCH
                                VROSLIN-RAMMLER EXPONENT
               1:0 -
               1:1
               1:2 -HUH-
               0:1 ---
                                      SIZE D MtCJIOWS
Figure  56.   Average size  distribution for  coal,  blend, and  dRDF firings
             during March  tests.
                                                      MAY
                    M[-	t'Ttiu^-jjffrrt-'iM!
                    of::::::ti±i^iI^J~;^   ""'Hi:
                            *i-tiggx» r\  , ^mr
                             aiWjr:   f -    ,!
--r:::-;—H
                                                 ,:n
                                                       *
fl» ii-
Figure 57.   Average size  distribution for  coal,  blend, and  dRDF firings
             during May tests.
                                     108

-------
c;
 i
   io12-
   10"-
     10
   I0 -
c/2  10 -
CO
bJ
cc
   io8H
1ST TEST
-0*
                                    BLEND
                                    I '0
                                    I : I
                                    I = 2
                                    0: I
                                    BOILER
                                      I 2
                                   -  o •
                                   -  A A
                                      am
                                   —   *
               O.I       0.2      0.3       0.4       0.5      0.6
                        jf, - FRACTION OF RATING


   Figure 58.  Effects  of blend and  load on aerosol  resistivity.
                                 109

-------
                TABLE 13.   EFFECT OF BLEND ON AEROSOL  RESISTIVITY
                                       Resistivity  (

                   Blend           March Tests       May  Tests
1:0
1:1
1:2
0:1
<107
8 x io9
—
—
6 x IO7
2 x io10
1 x IO12
1 x IO12
 laboratory.  As  shown  in  Figure  59,  the  precipitator  cells are of  the plate
 and  frame  construction with  the  charging electrodes installed in a pipe frame
 support  centered  between  parallel,  smooth-surface  collecting electrodes.

     Since  the charging electrodes  are suspended from dielectric blocks
 resting  across the  tops of the grounded  electrodes, an aerosol accumulating
 on the dielectric block tends to  short circuit  the cells.  Because of such
 short circuits during  the testing,  the number of operating cells progressively
 decreased.  Consequently, the results from  this test  cannot be considered
 representative of the  data which  might be collected from a commercially
 available ESP.  Additional ESP testing is planned  for the demonstration test.
 Appendix E  includes  the data acquired by Monsanto  during the May testing and
 a discussion of the  results.

 GASEOUS  EMISSIONS TEST  RESULTS

     The following  sections detail  the test results for the following gaseous
 emissions:  S02, oxides of nitrogen, halogens, and hydrocarbons.

 S02
     The sulfur oxide level in the flue gas is of particular interest to
precipitator designers since it is related to aerosol resistivity.  Figure 60
shows the blend effect on the overall sulfur dioxide emission rate as a func-
tion of boiler load.  The reduction in overall S02 emissions with the replace-
ment of higher sulfur coal by 0.6 percent sulfur dRDF is significant for the
1:2 and 0:1 blend firings.  While the 1:1 blend firings showed a reduction in
S02 when compared to coal, the difference did not exceed the 90 percent con-
fidence limits.  Very good agreement was obtained between the sulfur oxide
emissions as determined by the continuous monitoring electro chemical sensor
and the wet chemistry determinations.  Reference is made to Tables A-2 and A-3
in Appendix A.  The SO  for test days May 10 through May 13 were:  blend 1:1
                                     110

-------
                                               CHARGING
                                               ELECTRODE
                                               FRAME
                                             INSULATING
                                             FRAME
                                             SUPPORT
                                           SHORT CIRCUITS DEVELOPED
                                           AT THESE LOCATIONS

                                         GROUNDED
                                         ELECTRODE
Figure  59.  Cell  configuration in the  portable  ESP.
                           Ill

-------
 4500 r







2
0.
0.
£
CM
O
CO


4000


3500

3000
2500
2000
1500
1000
500
0
0

-


-

-

rrr^A*
\ • \ \ I • \ \~ \ * \ A°l "
\ \ \ tit* * ** t

— i t i t ^t i i > > * *
&
a
IF*
i i i I
00 0.10 0.20 0.30 0.40
BOILER
BLEND 1 2
1:0 	 o •
1:1 	 A A
1:2 fill 0 •
0:1 	 x


O
^_^o— — to- r
^UJ

0.50 0.60
                                — FRACTION OF RATING
    * S02 WC is S02 as determined by wet chemistry.
Figure 60.   Effects of blend and load on sulfur dioxide emissions.
                                112

-------
1334 ppm (meter), 1309 ppm (wet chemistry),  and blend 1:2 825 ppm (meter),
825 ppm (wet chemistry).  As would be expected, the reduction in sulfur
emission was proportional to the reduction of the sulfur in the fuel combusted.
Reference is made to Table 5 and Tables A-l through A-4 in Appendix A for
verification of the reduction in sulfur emissions as a function of the sulfur
in the fuel combusted.

Oxides of Nitrogen

     The results from the on-line NOX analyzer determinations are shown in
Figure 61 and Table 14.  The plot of NOX versus load as a function of blend
has considerable scatter.  The random pattern of the line placement for blend
firing suggests that there is no apparent relationship between the blend
firings and the NOX emissions.

     Figure 62 shows the NOX emissions replotted as a function of excess air.
All the results are similar in that they fall within the 90 percent confidence
level.  Table 14 summarizes all the NOX concentrations as determined by both
wet chemistry and the Theta Sensor trigas analyzer.

Halogens

     Of concern to many people is the emission of chlorine from the combustion
of dRDF.  Although chlorine levels (weight basis) in dRDF are about the same
as those found in some coals burned throughout the United States and Europe,
the coal:dRDF blends emit more chlorine per megajoule than most coals because
of their lower heat content.  The chlorine level per magajoule equivalent for
the coal:dRDF firing (1:2 blend) was seven times greater than the level for
coal-only firing.

     Figures 63 and 64 show the emission rates for chlorine and fluorine,
respectively, as a function of coal:dRDF blend and boiler loads.  Since the
chlorine emission regression lines for blend firing are well outside those
for coal-only firing, the blend firing apparently had substantially greater
chlorine emission rates.  Since fluorine concentrations were omitted in the
laboratory analysis of the base fuel, the significance of the fluorine
emissions for blend versus coal-only firing cannot be established.

Hydrocarbons

     Hydrocarbon emissions are particularly important because of their smog
forming potential.  Figure 65 shows the hydrocarbon emissions as a function
of boiler load and coal:dRDF blend.  The hydrocarbon emissions from the blend
firing was not significantly different from the coal firing.  The single data
point at 100 percent dRDF firing (55 ppm) suggests that further testing may
be required at higher dRDF substitution ratios to verify this increase.
                                      113

-------
     400
     300
a.
Q.
O
     200
     100
       0.00
0.10
0.20        0.30        0.40



 o£  — FRACTION OF RATING
0.50
0.60
  Figure 61.  Effects of blend  and load on nitrogen oxide  emissions.
                                    114

-------
TABLE 14.   RELATIONSHIP OF NOX  CONCENTRATION AND EXCESS  AIR
             PERCENTAGE FOR COAL,  BLEND, AND dRDF FIRINGS
Date
1:0 Blend (coal only)
3/21
3/22
3/28
3/31
4/1
5/3
5/4
5/5
5/16
1/20
1/20
1/21
1/24
1/25
NOX M(ppm)

362
257
197
277
300
230
141
105
424
358
358
228


NOX WC(ppm)






318

238
449



424
594
Percent
Excess Air

104
72
92
71
82
72
116
84
147
108
88
36
87
110
 1:1 Blend Firing
    3/24                       255                               102
    5/12                       326                360             109
    5/13                       405                386             101
 1:2 Blend
    3/29                       232                                94
    5/10                       223                192             138
    5/11                       305                273             132
0:1 Blend
    5/14                       272                274             133
    5/14                       255                247             106
Notes:   1.  NOXM is - NOX as determined by the  Theta Sensors, Inc. Tri-gas

           Meter.

        2.  NOXWC is - NOX as determined by the EPA Method 7 Wet Chemistry.
                                 115

-------
       600
       500
       400
    CL
    Q.

   O*   300
   z
       200
       100
BLEND

 1:0  —


 1:1  --
                          BOILER

                           1  2
 1:2


 0:1 —
                         -  o •



                         -  A A



                           O •
          0      20      40     60      80     100     120     140     160


                                   EXCESS AIR — %





Figure  62.   Effects  of  blend and excess air on nitrogen oxide  emissions.
                                    116

-------







0.
a
o


BOILER
BLEND 1 2
700


600
500
400
300
200

100
0
(
v.o 	 ° •
	 * V.I 	 A A
y. v.2 tttt a •
o-.i 	 *

a
Q A

A * n"
A
A
T__1 	 	 • 8
\ !• \ V \« \ \ \ \° v A v \\b\\
\ ^ \ \ ^P ^ -^ 	 i 	 1 	 1 O- ^ 	 i 	 ^ A ^ \ "*\ k \
300 0.10 0.20 0.30 0.40 0.50 0.60
                        — FRACTION OF RATING
Figure 63.  Effects of blend and load on chlorine emissions.
                             117

-------
     20 r
o.
Q.
     10
                                                   BLEND
                                                    1:0 —
                                                    1:1 —
                                                    1:2
                                                    0:1 —
                                                  BOILER
                                                   1  2
                                                — O •
                                                - A A
                                                   0 •
                                                            NOTE:  DATA FROM
                                                                 MARCH & MAY
                                                                 ONLY
                                               _l_
                                                          I
      .00
.10
.20        .30       .40       .50
      it — FRACTION OF RATING
.60
             Figure 64.   Effects of blend and  load on fluorine  emissions.
                                        118

-------
Q.
O.
6
    60
    50
    40
30
    20
     10
      0.00
              0.10
                                                              BLEND
                                                               1.0 	
                                                               1:1	
                                                               1:2 -t-+
                                                               0:1 	
                                             BOILER
                                             1 2
                                          — O •
                                          — A A
                                          •
-------
TRACE COMPOUND EMISSIONS TEST RESULTS

     While not yet regulated, the emissions of potentially carcinogenic
polycyclic hydrocarbons and hazardous heavy metals from stationary combustion
sources are coming under intense scrutiny.  Therefore, the flue gases were
evaluated for polycyclic organic compounds and heavy metal emissions.

Trace Organic Emissions

     Table 15 lists the results of the GC-Mass Spectrometer Analysis of the
Battelle Tenex samples.  The overall emissions of polycyclic organic materials
(POM's) for the coal-only and the blend firings were very low.  Morever, all
the monitored emissions were below the threshold limits proposed by the
National Academy of Science.  Since the increase in emissions for the 1:1
blend firing is based on the data of a sample taken during boiler switchover,
this increase is questionable.
         TABLE 15.  POM  CONCENTRATIONS  FOR  COAL  AND  BLEND  FIRINGS
COMPONENT
Anthracene/Phenanthrene
Methyl anthracenes
Fluoranthene
Pyrene
Methyl Pyrene/
Fluoranthene
Benzo (c)phenanthrene
Chrysene/Benz (a)
anthracene
Methyl chrysenes
7 , 12-Dimethylbenz (a)
anthracene
Benzo fluoranthenes
Benz (a) pyrene
Benz (e) pyrene
Perylene
Methyl Benzopyrene
3-Methyl Chloranthrene
Indeno(l,2,3,-cd)
pyrene
Benzo (ghi)perylene
Dibenzo(a,h) anthracene
Dibenzo (c, x) carbazole
Dibenz (ai and ah)
pyrenes
Coronene
1:1
ppb
0.0736
0.0126
0.0164
0.0039

0.00043
<0.0002

n/d
n/d

n/d
n/d
n/d
n/d
n/d
n/d
n/d

n/d
n/d
n/d
n/d

n/d
n/d
1:2
ppb
0.0516
0.0052
0.0064
0.0027

<0.0002
n/d

n/d
n/d

n/d
n/d
n/d
n/d
n/d
n/d
n/d

n/d
n/d
n/d
n/d

n/d
n/d
COAL
1:0
ppb
0.00086
0.00032
0.00030
0.0018

<0.0002
n/d

n/d
n/d

n/d
n/d
n/d
n/d
n/d
n/d
n/d

n/d
n/d
n/d
n/d

n/d
n/d
Ng/m
543
100
137
33

4
<1.0

n/d
n/d

n/d
n/d
n/d
n/d
n/d
n/d
n/d

n/d
n/d
n/d
n/d

n/d
n/d
1:2
Ng/m3
380
42
54
23

<1.0
n/d

n/d
n/d

n/d
n/d
n/d
n/d
n/d
n/d
n/d

n/d
n/d
n/d
n/d

n/d
n/d
COAL
1:0
Ng/m3
6
3
3
15

<1.0
n/d

n/d
n/d

n/d
n/d
n/d
n/d
n/d
n/d
n/d

n/d
n/d
n/d
n/d

n/d
n/d
                                     120

-------
Trace Inorganic Emissions

     Table 16 lists the concentrations of the fuel ash trace metals found in
two coal and two dRDF samples.  The concentrations were determined by first
preparing the specimens by oxygen plasma ashing and then analyzing them with
spark source mass spectrometry.  The enrichment ratios indicate the greater
amount of metals in dRDF than in coal.  The range of reported fuel metal
values clearly indicates the heterogeneous content of the dRDF and the
variability of the metalic material in coal.

     The detailed heavy metal ash analyses are presented in Appendix F and
summarized in Table 17.  This table lists the average emission rates for each
test battery.  The March and May data were separated to eliminate the effects
of normalizing the data to the reference coal.  Table 18 is a manipulation of
the data in Table 17 to present the emission rate data in terms of enrichment
functions.  For example, Table 18 shows that 43.3 times more lead was emitted
in the total particulates when firing dRDF only than when firing coal only.
In addition, while some metals were enriched, others had reduced emission
rates.

     Tables 17 and 18 also show the enrichment of certain metals in the
bottom ash, the reinjected fly ash, and the collected fly ash.  The presence
of these metals in the ash implies an increased possibility of heavy metal
leaching when boiler ash is landfilled or used for various applications.  The
significance of this leaching is unknown.

     The amounts of the various metals found for all blends were generally
normal except for the amount of arsenic which was fairly high.  Most notable
of the trends was the variation of several metals with the coal:dRDF ratios:
whereas the concentrations of Br, Mn, Pb, and Sb generally increased with
increasing dRDF substitution in the coal:dRDF blends, the concentrations of
As, Ni, and Vn decreased.  Although these trends are not definitive, they are
probably true since they were also observed in the data for the fly ash
leachates.

     While variations of the heavy metal concentrations with particle size
were poorly defined, the concentrations of As, Ga, Na, and probably Sb
generally increased with decreasing particle size.  In contrast, the concen-
trations of Br and Mn markedly increased with increasing particle size.
Appendix F provides a detailed summary of the heavy metal data.

     On the basis of previous work with coal aerosols and incinerator fly
ash,6 it would be expected that such metals as Br, Mn, Pb, and Sb would have
higher concentrations in RDF than in coal and that As, Ni, and V would have a
greater affinity to coal than to RDF.  Similarly, the affinity of As, Ga, Na,
and Sb to small particles would be expected since these metals can be vola-
tilized during combustion and then adsorbed onto the more developed surface
     6Kaakinen, J. W., et al.  Trace Element Behavior in Coal Fired Power
Plants.  Environmental Science and Technology, Volume 9.  pp. 862-869.


                                     121

-------
        TABLE 16.   TRACE METAL CONCENTRATIONS FOUND IN COAL AND dRDF FUEL
Element
                Coal Sample
                #1        #2
          d-RDF Sample
          #1         #2
                ppm
ppm
ppm
ppm
      dRDF:COAL
   Enrichment Ratio
Per kg Ash      Per MJ
Li
Be
B
F
Na
Mg
Al
Si
P
S
Cl
K
Ca
Sc
Ti
V
Cr
Mn
Fe
Co
Ni
Cu
Zn
Ga
Ge
As
Se
Br
Rb
Sr
Y
Zr
Nb
Mo
Ru
Rh
Pd
Ag
50
1
30
<5
= 1000
1000
High
High
300
= 1%
100
High
High
10
2000
40
100
100
High
20
100
10
5
20
<1
10
<1
3
100
300
10
100
5
10
<1
<0.3
<1
<0.6
54
4
23
=150
MC
MC
MC
MC
MC
MC
MC
MC
MC
14
MC
97
88
22
MC
7
19
13
16
32
<5
13
12
19
33
690
70
280
11
11
0
0
0
1
10
0.05
20
10
High
High
High
High
High
2000
2000
High
High
<1
2000
10
40
500
High
1
20
30
500
2
<0.3
4
<0.5
10
20
150
2
50
1
10
<0.2
<0.1
<0.2
1
<0.1
<0.1
<0.1
=36
MC
MC
MC
MC
MC
MC
MC
MC
MC
0.7
MC
8
260
>470
MC
2
15
41
300
4
0.1
4
0.4
4
3
74
1
13
2
4
0
0
0
0.4
0.097
0.03
0.38
0.30









0.12

0.13
1.60
7.95

0.11
0.29
3.09
38.10
0.12
0.07
0.35
0.07
0.64
0.17
0.23
0.04
0.17
0.19
0.67
0.20
3.00
0.20
0.88
.3
.1
1.1
.9









.4

.4
4.6
22.9

.3
.8
8.9
109.9
.3
.2
1
.2
1.8
.5
.7
.1
.5
.5
1.9
.6
8.7
.6
2.5
                                 (  continued )
                                     122

-------
TABLE 16.  (continued)
Element
Cd
In
Sn
Sb
Te
I
Cs
Ba
La
Ce
Pr
Nd
Sm
Eu
Gd
Tb
Dy
Ho
Er
Tm
Yb
Lu
Hf
Ta
W
Re
Os
Ir
Ft
Au
Hg
Tl
Pb
Bi
Th
U
Coal Sample
#1 #2
ppm ppm
<1
<1
<1
<2
<1
1
<3
100
10
20
1
4
<1
<0.6
<1
<0.4
<1
<0.4
<1
<4
<1
<0.4
<2
<2
<2
<1
<1
<1
<2
<2
<5
<1
<2
<1
<1
<1
2
STD
2
<0.8
<0.8
5
3
410
44
110
5
11
7
1
2
0.9
0
0
0
0
0
0
9
0
5
0
0
0
0
0
<0.8
0
8
0
17
8
d-RFD Sample
#1 #2
ppm ppm
<0.6
<0.06
20
1
<0.2
0.6
0.2
200
2
10
0.5
15
<0.3
<0.2
<0.3
<0.1
<0.3
<0.1
<0.3
<0.1
<0.3
<0.1
<0.3
<0.1
<0.3
<0.3
<0.3
<0.3
<0.3
<0.3
<0.3
<0.1
500
0.2
<0.1
<0.1
0.3
STD
8
9
0
0.1
0.2
330
3
3
0.5
0.5
0.4
0.1
0.1
0.1
0.2
0.1
0.1
0
0
0
0
0
0
0
0
0
0
0
<0.1
0
170
0.1
0.7
0.5
dRDF : COAL
Enrichment Ratio
Per kg Ash Per MJ
0.3

9.33
3.57
0.11
0.12
0.07
1.04
0.09
0.1
0.17
1.03
0.09
0.19
0.13
0.15
0.5
0.5
0.4
0.03
0.3
0.25
0.03
0.05
0.04
0.3
0.3
0.3
0.15
0.15
0.07
0.1
67
0.3
0.04
0.07
.9

26.9
10.1
.3
.3
.2
3
.3
.3
.5
3
.3
.5
.4
.4
1.4
1.4
1.2
.07
.9
.7
.08
.1
.1
.9
.9
.9
.4
.4
.2
.3
193
.9
.1
.2
            123

-------
                  TABLE 17.  AVERAGE  HEAVY METAL EMISSIONS IN ASH FROM BLEND FIRING TESTS
N3



Threshold Limit Level
ug/m1
Total 1'articulate
stark - lig/fli

Bottom Ash - ti£/k£


Multiclone Reinjectlon
Fly Ash - ug/kg



Thresliold Limit Level
uR/m1
Total Partlculate
stark - ug/nr


Bottom Ash - ug/kg



Multiclone Relnjectlon
Fly Ash - us/kg

Collector Ash
ug/kg



Hl.KNI)
„_

1:0
1 :1
1:2
1:0
1:1
1:2
1:0
1:1
1:2

BLEND
	

1:0
1:1
1:2
0:1
1:0
1:1
1:2
0:1
1:0
1:1
1:2
0:1
1:0
1:1
1:2
0:1
mi . it i
Samples
Analyzed


3


1
1


1
2
1
no. of
Samples
Analyzed


7
3
3
2
2
2
2
1
1
1
2
2
3
3
2


I'h
200

228
3975
7660
<12.5
26.3
128
16.3
97.5
109

Pb
200

230
4237
8217
9953
<12.5
46.3
65.0
169
20.0
92.5
165
363
20.8
217
274
1012


Cd
20

< 4 . 4 3
79.4
233
• .75
< . 8
0.75
'.75
1.12
1.5

Cd
20

4.33
72.4
220
267
<.75
£-75
.75
2.13
1.00
1.9
7.88
<0.75
3.58
6.00
25.4


As
500

173
45.9
44.9
11.0
39.3
36.7
34.4
39.3
66

As
500

184
153
126
49.4
16.1
35.0
28.8
56.3
48.1
49.5
34.9
46.3
76.3
80.4
43.5
103.5


HH
100

17 . 85
19.6
12.3
<0.4
<0.4
<0.4
<0.4
<.68
<0.4

Hg
100

<5.57
15.7
11.4
94.7
<0.4
'0.4
<0.4
<0.4
<0.4
0,58
S.415
i.440
<0.4
S.545
SJ..O
<.0.65
MARCH

Or
100

35.1
33.5
47.6
22.5
25.6
42.5
8.75
12.6
15.0
MAY
Cr
100

50.7
35.4
55.4
79.7
8.75
23.7
20.7
50.0
21.3
21.3
23.6
69.4
17.7
19.7
34.9
185


Nl
1000

32.6
32.1
41.0
20.0
27.4
139
15.0
16.9
18.7

Nl
1000

49.5
35.9
50.9
29.4
20.0
33.2
22.1
35.0
26.2
23.7
23.2
40.9
87.5
25.2
31.9
75.4


Mn


47. ;
64.6
101
51.0
138
250
105
193
300

Mn


30.4
62.6
115
275
40.5
170
135
43.5
45.0
120
132
81.3
49.7
145
328
123


Zn
5000

592
6012
8569
31.2
73.3
188
31.2
134
194

Zn
5QOO

596
5Cu3
8317
8033
68.8
112.5
73.4
539
50.0
356
292
1118
60.4
343
608
	


Cu


S.51.7
96.1
82. 5
15.0
152
200
<12.5
15
17.5

Cu
	

50.1
82.4
134
203
<18.8
221
>136
~205
15.0
22.5
42.2
88.4
15.0
27.8
39.9
149


Sn


•-1.46
3.36
4.99
0.50
3.58
5.00
1.88
2.50
2.25

Sn


51,45
2,70
3.47
6,07
0,50
3.31
3.62
3.6
1.0
3.0
3.0
4.84
2.09
3.58
1.63
8.08


Sb
500

<87.2
<52.2
<87.3
<25
<25
'25
<25
<25
'25

Sb
500

<65.6
<48.5
59.1
<107
<25
<25
<25
<25
<25
<25
<26
<25
<25
<25
<25
<26.3


Ag
10

<8.72
12.0
17.1
'2.5
<2.5
'2.5
<2.5
'-2.5
<2.5

Ag
10
5.75+120.
'6.56
<6.51
19.4
29.7
62.9
<2.5
<2.5
88.8
<2.5
<2.5
6.0
<2.5
<2.5
<2.5
13.7


Vn


<:87.2
<52.2
59.6
<25
<25
'25
'25
i.25
25

Vn


'65.6
<48.5
'59.1

-------
                     TABLE 18.   BLEND HEAVY METAL TO COAL-ONLY HEAVY METAL  RATIOS IN ASH  SAMPLES
to
U1
MARCH

Total Particulates
stack - ug/m3

Bottom Ash - pg/kg


Multiclone
Reinjection Fly Ash
pg/kg


Total Particulate
stack - pg/m3


Bottom Ash - pg/kg



Collector Ash - pg/kg



Multiclone Reinjection
Fly ash - pg/kg


BLEND
1:0
1:1
1:2
1:0
1:1
1:2
1:0
1:1
1:2

BLEND
1:0
1:1
1:2
0:1
1:0
1:1
1:2
0:1
1:0
1:1
1:2
0:1
1:0
1:1
1:2
0:1
Pb
1.0
17.4
33.6
1.0*
2.10
10.2
1.0
5.98
6.69

Pb
1.0
18.4
35.7
43.3
1.0*
3.70
5.20
13.5
1.0
10.4
13.2
48.6
1.0
4.63
8.25
18.2
Cd
1.0*
17.9
52.6
<


1.0*
1.49
2.00

Cd
1.0
16.7
50.8
61.7
1.0*
1.0*
1.0
2.84
1.0*
4.77
8.0
33.9
1.0*
1.33
2.53
10.5
As Hg
1.0 1.0*
.265 2.50
.259 1.57
1.0 <
3.57
3.34
1.0 <
1.14
1.92

As Hg
1.0 1.0*
.832 2.82
.685 2.05
.269 17.0
1.0 <
2.17
1.79
3.50
1.0 <
1.05*
.570
1.36
1.0 <
1.03
.726
.963
Cr
1.0
4.27
6.06
1.0
1.14
1.89
1.0
1.44
1.71
MAY
Cr
1.0
.698
1.09
1.57
1.0
2.71
2.37
5.71
1.0
1.11
1.97
10.5
1.0
1.0
1.11
3.26
Ni
1.0
.985
1.26
1.0
1.37
6.95
1.0
1.13
1.25

Ni
1.0
.725
1.03
.594
1.0
1.66
1.11
1.75
1.0
.288
.365
.862
1.0
.905
.885
1.56
Mn
1.0
1.35
2.12
1.0
2.71
4.90
1.0
1.84
2.86

Mn
1.0
2.06
3.78
9.05
1.0
4.20
3.33
1.07
1.0
2.92
6.60
2.47
1.0
2.67
2.93
1.31
Zn
1.0
10.2
14.5
1.0
2.35
6.03
1.0
4.29
6.22

Zn
1.0
9.51
14.0
13.5
1.0
1.64
1.07
7.83
1.0
5.68
10.1
	
1.0
3.12
5.84
22.4
Cu
1.0*
1.86
1.60
1.0
10.1
13.3
1.0*
1.2
1.4

Cu
1.0
1.64
2.67
4.05
1.0*
11.8
9.89*
10.9
1.0
1.85
2.66
9.93
1.0
1.5
2.81
5.89
Sn
1.0*
2.30
3.42
1.0
7.16
10.0
1.0
1.33
1.20

Sn
1.0*
1.86
2.39
4.19
1.0
6.62
7.24
7.20
1.0
1.71
0.780
3.87
1.0
3.0
3.0
4.84
Sb Ag
< 1.0*
1.38
1.96
< <


< <



Sb Ag
< 1.0*
1.0*
2.96
4.53
< 1.0**
<.435*
<.435*
15.4
< 1.0*
1.0*
1.0*
5.48
< 1.0*
1.0*
1.0*
2.4
Vn
<


<


1.0*
1.0*
1.0

Vn
<



<



1.0
1.0*
1.09
1.66
1.0
1.0
<1.0*
<1.0*
        < Below the detection limit
        * Extreme value deleted

-------
 area of the small particles.   The increasing Br and Mn concentrations  with
 increasing particle size cannot be explained.

      The MRI collector stages for coal-only, blend, and dRDF-only  firing  in
 May were accumulated and sent to Colorado State University for  analysis.
 Contained in Appendix G, the  University's complete report  presents considerable
 information about both the chemical characteristics and the potential  environ-
 mental impact of dRDF-coal fly ash.  The following trends  are based on the
 data of Table G-3 in Appendix G.

      First,  the amounts of most metals that  are soluble increase with
 increasing dRDF fraction of the original fuel.   While  this trend is apparent
 for Ca, Cu,  K^ Mg,  Mo,  Na, Si,  Cl~, N0~3,  and  SO,,2", it may also exist for B,
 Ba,  Cd, and  F .   Ni and P, and  possibly Cr and  Sr,  have a  reverse  trend.
 Since many of the major matrix  metals have increasing  solubility with
 increasing dRDF percentage, the addition of  dRDF to coal might  result  in
 greater bulk solubility (as well  as greater  trace metal mobilization)  than
 that evidenced in the fly ash of  pure coal.  This increased solubility may
 require special procedures for  landfill disposal.

      Second,  the metal  mobilization increased with decreasing particle size.
 While this trend is apparent  for  Cd,  Cr,  Cu, K,  Mn, Mo,  Na,  Ni, Pb,  and Cl ,
 it may also  exist for Ba,  Be, P,  and  F .   This  trend may be due to the con-
 densation  of  these  metals  from  vapor  onto  the particulate  surfaces or  to  the
 more efficient  formation of soluble oxides (i.e.,  calcining)  in small  particles.
 With the first  supposition, similar size  dependencies would be  expected for
 both the bulk and the soluble metal concentrations.  However, such dependencies
 would not  be  expected if solubility is the direct  result of chemical reaction
 at a particle surface.   In any  event,  the  available data are not sufficient
 to rule out  either  supposition.

      In the analysis  of  the fractional solubility  of coal-dRDF  fly ash, Al,
 Ba,  Mg, P, Si,  and  Sr have a very  low solubility  (<10 percent); Be,  Cd, K,
Mn,  and Na have  a moderate solubility (=20-80 percent);  and Ca  has a very
high solubility.

      Several  metals,  particularly Mn,  have an increasing fractional  solubility
with decreasing  particle size.  Since  Mn exhibits no dependence of concen-
tration on particle size,  its solubility increase with  decreasing  particle
size  is due to its more  efficient calcining.

Summary

     The following summarizes the major findings:

     1.   The specific concentrations  of trace metals in dRDF-coal  fly  ash
          are similar to those found  in pure coal fly ash.   The dRDF is the
          primary source of Br, Mn, Pb, and  Sb while the coal is the primary
          source  of As, Ni, and V.

     2.   Several metals,  particulary As, Ga, Na, and Sb,  tend  to  concentrate
          in small particles.

                                     126

-------
     3.    The volatization-condensation process which deposits volatile
          metals onto small fly ash particles is more effective in a plant
          firing a dRDF-coal mixture than in a plant firing coal only.   The
          greater volatile metal deposits in the coal-dRDF firing were
          probably due to the low combustion temperatures.

     4.    Except for Ni and P, the metals in the coal-dRDF fly ash increase
          in solubility with increasing dRDF content.

     5.    Both trace and matrix metals have a significantly greater solubility
          in small particles than in large particles.

     In summary, the results from the trace compound emissions test results
indicate that the quantities of the trace organic emissions for the blends
studied were so small that the ground-level concentrations would probably not
exceed 1 percent of the threshold level limits.  Hence, unless the data from
future tests indicate higher levels of trace organics, the emission levels
from coal-dRDF firings would be within acceptable limits.  The quantities of
metals present in the fly ash and bottom ash suggest that further studies
need to be carried out to establish if there is a health hazard due to
(1) increased bulk solubility (as well as greater trace metal mobilization)
of bottom ash with increased dRDF substitution and  (2) the adsorption of
volatilized metals during combustion onto the surface area of small (aerosol
size) particulates.
                                      127

-------
                                   REFERENCES
1.   Winegartner,  E. C.   Coal  Fouling  and  Slagging  Parameters.  American
     Society of Mechanical  Engineers,  1974.

2.   Fontana, M. G., and  M. D. Green.  Corrosion Engineering.
     McGraw-Hill,  N.Y.C., 1967.

3.   Watson, H. H.  American Hygiene Association Quarterly.  Volume 15,
     1954.  p. 21.

4.   Murphy, T. D., Jr.   Design and Analysis of Industrial Experiments.
     Chemical Engineering,  June 1977.  pp. 169-182.

5.   Natrella, M.  G.  Experimental Statistics.  National Bureau of
     Standard Handbook 91.  U.S. Government Printing Office,
     Washington D.C., 1963.

6.   Kaakinen, J. W., et  al.  Trace Element Behavior in Coal Fired Power
     Plants. Environmental  Science and Technology, Volume 9.  pp. 862-869.
                                      128

-------
                                                APPENDIX A

                                   EMISSIONS, FUEL, AND ASH DATA SUMMARIES
                            TABLE A-l.   FIELD TEST RESULTS FOR COAL ONLY 1:0 FIRING
NJ
VO


JATE X-
3/19
3/19
3/21
3/21
3/22
3/28
3/31
3/31
4/1
5/3
5/4
5/4
5/5
5/5
5/16
5/16
5/17
5/17 .
1/20
1/20
1/21
1/24
1/25
12/10
12/10
NOTES :






51
51
40
40
40
38
27
27
35
25
20
22
31
51
21
36
17
17
53
53
47
53
50


1.
2.
3.
4.
5.
6.


GR/FCF BOILER ' Cl F SO,.M
.311 1 72 8
.458 1 84 15
.267 2 1397
.325 2 75 7.2
.226 1 55 7.3
.197 1 38 9.3
.286 1 35 8.1 1000
.234 1 36 6.5
.281 1 44 7.5 1148
.295 2 25 5.4 1118
.339 2 19 7.9 1495
2 1080
.206 2 283
.199 2 19 6.8 367
.588 2 25 11.9 1302
.298 2 26 7. 3 1083
.448 2 1254
2 1118
.330 1 27 25 1734
.436 1 32 46 1734
.214 1 14 16 1155
.327 1 31 16 1265
1 1291
.229 1 98 4.6 1916
.240 1 115 23.6 1381
All values adjusted to 50% EA or 12* CO
GR/SCF is grains/standard ft3.
Lined out data are considered outliers.
PPM
soxwc
2312
1691

1397
1153
1040
1117
1324
1125



541
1341
1121
1149
4888
5946
2329
2217



2 .





NOXM NOXWC THC '

362

257
197
277

300
221
136

101

408

348
348
221


251




332
21

12
13

29
34
306


229

432 19

12
8
12
412 1
577 1






.9

.7
.7

.6
.1





.2

.0
.8
.6
.8
.6





%
EA

104
104
72
92
71
71
82
72
116
106
84
40
147
104
96
108
88
86
87
110





f
PPM g/hr g/hr fl-CM
OPACITY SO3 K4SJ.NJ1SUT COLLECT KBSIS.

38
38
32
44
68
68
46
50
70
70
54
54
70
70
91
91
39
42
37
41
39
59
59




<107
<107
110 <10'
110
76
76
75
82 103
102 140
20.7 152 90
70 106
99 123
252 (0.10) 592 (0.14)
235 (0.12) 579 (0.14)
491 (0.10) 666 (0.17)










oF
Tflame



2201
2236
2210
2197
2259
2209
2121
2020
2244
2440
2125
2302
2024










OF
Tflue
411
410C '/502C2
5100 2
509* 2
395*'
403
399
404
418
484
459
454
491
490
474
517
461
43Q
424
427







SO M is sulfur oxides measured with a meter (electrochemical transducer) .
SO*WC is sulfur oxides measured by wet chemistry.
In volumes headed by "g/hr," any second
fly ash density in g/cc.
value

(shown in

paranthei

36S)

represents






-------
                                TABLE A-2.   FIELD TEST RESULTS  FOR 1:1 BLEND FIRING
Uo
DATE
3/23
3/23
3/24
5/12
5/12
5/13
5/13
12/8
12/13
12/13
12/14
12/14
JL
.43
.43
.44
.34
.28
.39
. 30
.45
.45
.45
.43
.43
GR/SCF
. 138
.191
.196
.287
.303
.191

.172
.441
.224
.196
.177
BOILER
1
1
1
2
2
2
2
1
1
1
1
1

'Cl
311
321
265
195
172
146

281
437
182
58
309

F
9.1
7.0
13.7
8.4
7.4
11.8

27
22.3
33



SOXM
652

1003
1388
1503
1391
1054
1402
1809
1235
808
1234
PPM
SOXWC
794
755
1118
1502
1476
1268
992






NOXM


248
351

436

224
282

278


% PPM g/hr
NOXWC THC ^ EA OPACITY SO 3 REINJECT
8.3 74
18.4 74
13.4 102
387 109
109
415 20.0 101
101
9.0
337

225

27
27
22
59
59
42
42
70
73
73
74
74



141 (0
117 (0
131 (0
18.3 122 (0





87
87
82
.15)
.16
.14)
.14)





g/hr SJ-CM
COLLECT RESIS.
IxlO10
1x10 ' °
4xl09
384 (0.20) 7xl09
344 (0.18) 5xlOs
669 (0.16) 3x10'°
260 (0.17) 5x10'°





"F
T
Aflame
2226
2213
2275
2305
2264
2253
2210





OF
T
flue
423
429
430
506
493
505
497





    NOTE:
          See notes for Table A-l.

-------
                                      TABLE  A-3.   FIELD TEST RESULTS  FOR 1:2  BLEND  FIRING
U)
3/29
3/29
3/30
5/10
5/10
5/11
5/11
. 36
. 36
.26
. 36
. 30
.36
. 34
. 20 i
. 176
. 180
. 282
.320
.248

1 4 i8
1 32 1
1 301
2 198
2 238
2 243
2
14.0
12.4
8.8
11.7
11 . 1
11.6

800


828
980
777
715
994
9 8 ft
46]

928
810
842
722
                                                          301:
                                                                                                                          " F
19.7
15.8
34.2


19.1

'14
'14
')]
I 38
116
1 32
114
31
31
42
45 123
45 133
45 240
45 10.3 138
82
82
98
(0.
(0.
(0.
(0.


.17)
10)
.20)
.17)


155
164
241
241


(0.
(0.
(0.
(0.


.21)
.23)
.27)
.26)


6x10' '
8x10'
2x10' ;
4x10' '
2287
2306
2313
2401
2322
2308
2299
415
409
391
515
512
518
506
    NOTE:  See notes for Table A-l.

-------
                           TABLE A-4.   FIELD  TEST RESULTS FOR dRDF  0:1 FIRING


1 — 1
CO
N3

DATE
5/14
5/14


JL
.27
.26


GR/SCF
.348
. 356
	

BOILER ' Cl F SO M
2 654 9.4 251
2 610 7.8 275
	
PPM
SO WC
X
303
268
	

NO M
X
486
456


NO WC
X
489
441

'1, %. PPM
THC ^ EA OPACITY SO,
56.3 133 48
106 48

q/hr
REINJECT
215
230

(0.
(0.

,85)
.84)

q/hr
COLLECT
273
271

(0.
(0.

.76)
.70)

ii-CM
RESIS.
1x10 ' J
IxlO1 '

"F °F
T T
flame flue
2326 470
2285 473

NOTE:  See notes for Table  A-l.

-------
TABLE A-5.  CORRECTED FIELD TEST RESULTS FOR FOUR COAL:dRDF BLENDS



     (Emission data were  normalized  to  the March  reference coal)

BLEND DATE
1:0 (coal only) 3/19
3/19
3/21
3/21
3/22
3/28
3/31
3/31
4/1
5/3
5/4
5/4
5/5
5/5
5/16
5/16
5/17
5/17
1/20
1/20
1/21
1/24
1/25
12/10
12/10
*0utliers were
1:1 3/23
3/23
3/24
5/12
5/12
5/13
5/13
12/8
12/13
12/13
12/14
12/14
1:2 3/29
3/29
3/30
5/10
5/10
5/11
5/11
0:1 (dRDF only) 5/14
5/14
BOILER
.51
.51
.40
.40
.40
.38
.27
.27
.35
.25
.20
.22
.31
.51
.21
.36
.17
.17
.53
.53
.47
.53
.50


omitted
.43
.43
.44
.34
.28
.39
.30
.45
.45
.45
.43
.43
.36
.36
.26
.36
.30
.36
.34
.27
.26
1
1
2
2
1
1
1
1
1
2
2
2
2
2
2
2
2
2
1
1
1
1
1
1
1
from
1
1
1
2
2
2
2
1
1
1
1
1
1
1
1
2
2
2
2
2
2
ASH-
GR/SCF
.311
.458
.267
.325
.226
.197
.286
.234
.281
.124
.142

.087
.084
.247
.125
.188

.238
.314
.154
.235

.165
.173
PPM
Cl
72
8-4
75
55
38
35
36
44
50
38


38
50
52


11
13
6
12

39
46
SOXM
1397



1000

1148
1375
1839
1328
348
451
1601
1332
1542
1375
1266
1266
843
923
942
1399
1008
soxwc
2312
1691

1397
1153
1040
1117
1324
1384



665
1649
1379

1413
-3^68*
4341
1700
1618



THC
21.9

12.7
13.7

29.6
34.1





17.3



13.1
9.6
13.7
2.0
1.7


NOXM
362

257
197
277

300
230
141

105

424



358
358
228


259

NOXWC
332






318


238

449






424
594


statistical analysis.
.137
.189
.194
.161
.170
.107

.196
.503
.255
.223
.202
.201
.174
.178
.175
.198
.154

.268
.274
280
289
239
234
206
175

202
315
131
42
222
407
300
280
216
259
265

654
610
652

1003
1693
1834
1697
1286
715
923
630
412
629
808


994
1176
932
858
251
275
794
755
1118
1832
1801
1547
1210





1004
996
466
1114
972
1010
866
303
268
8.1
18.0
13.1


18.8

8.1




19.3
15.5
33.5


18.3

55.7



255
326

405

302
381

375

232


223

265

272
255



360

386


455

304




192

238

274
247
                                133

-------
TABLE A-6.  AS-RECEIVED COAL PROPERTIES
DATE
As Received
* Moisture
% Ash
i Volatile
% Fixed C.
Btu/lb
Dry Basis
I C
% H
% N.
% Cl
% s
% Ash
% 0^
Fusion
Initial
1st Softening
2nd Softening
Fluid
Mineral Analysis
Phos . Pent Ox.
Silica
Ferric Ox.
Alumina
Titania
Sodium Ox.
Potasium Ox.
Lime
Magnesia
Sulfur TriOx.
Undet.
DEC

2. 12
10.78
29.42
57.68
13,471

77. 15
4.77
1.26
.26
3.57
11.01
1.98

2060°F
2180°F
2280°F
2500°F

.48
35.43
34.94
22.39
. 56
.25
.99
1.63
.28
1.23
1.82
JAN

8. 00
11.78
18.33
61.89
12, 100

72.7
4.1
1.6
__
1.70
12.81
7.1

2460°F
2540°F
2500°F
2570°F

.92
43. 50
21.00
20.70
1.42
2.70
2. 37
.38
.58
.83

JAN

10.80
15.87
15.73
57.60
10, 910

69. 2
3.8
1.5
__
1.1
17.79
6.6

2540°
2610°
2660°
2720°












JAN

2.30
15.24
14.43
68.03
12,380

69.6
3. 5
1.6
__
2.6
15.60
7. 1

F 2280°F
F 2340°F
F 2390°F
F 2440°F












JAN
AVERAGE

7.03
14. 30
16.16
62.50
11, 797

70.5
3.8
1.57

1.80
15.40
6.93

2427°F
2497°F
2517°F
2577°F

.92
43.50
21.00
20.70
1.42
2.70
2.37
. 38
. 58
.83

MARCH

4.92
10.50
30.38
54. 20
12,675

75.21
4.90
1.34
.06
1.58
11.04
5.87

2330°F
2385°F
2430°F
2525°F

.41
52.05
12.74
25.64
.70
.47
1.87
2.18
. 36
1.66
1.95
MARCH

4.00
10.70
19. 87
65.43
12, 780

73. 1
4.2
1.6
.14
1.90
11.14
8.0

2340°F
2400°F
2450°F
2500°F












MARCH

2.20
10.50
18. 60
68. 60
13,210

73.9
4.1
1.7
.14
1.60
10.74
7.9

2330°F
2400°F
2460°F
2520°F












MARCH

4.00
9. 20
20.85
65.95
13,170

74.4
4. 3
1.7
.11
1.80
9. 58
8.2

2300°F
2360°F
2400°F
2450°F












MARCH*
AVERAGE

3.78
10.23
22.43
63. 55
12,959

74.15
4.38
1.59
.11
1.72
10.63
7.42

2325V
2386V
2435V
2499":"1

.41
52.02
12.74
25.64
.70
.47
1.87
2.18
.36
1.66
1.95
MAY

1.46
22.15
23.45
52.94
11,603

66.82
4.48
1.15
.08
1.01
22.48
3.98

2700+°F
2700+OF
2700+°F
2700+°F












MAY

1.00
28.27
16.91
53.82
10, 800

63.10
3.77
1.21
.03
1.09
28.56
2. 24

2700+°F
2700+°F
2700+°F
2700+°F

.30
62.23
3.83
26.83
.89
.27
2.52
.43
1.19
.02
1.49
MAY MAY
AVERAGE

1.35
15.43
27.29
55.93
12,715

72. 28
4.74
1.69
.05
1.56
15.64
4.04

2700+°F
2700+°F
2700+°F
2700+°F

.35
57.00
7.76
28.02
.91
.34
2.12
.64
1.10
. 62
1.14

1.27
21.95
22.55
54.23
11,706

67.4
4.33
1.35
.05
1.22
22.23
3.42

2700+°F
2700+°F
2700+°F
2700+°F

.33
59.62
5.80
27.43
.90
.31
2.32
.54
1.15
. 32
1.32

-------
                                 TABLE A-7.   MOISTURE AND ASH FREE COAL PROPERTIES
co
DEC
% Moisture
% Ash
Btu/lb
MAF Basis
% Vol
% Fixed C.
Btu/lb
% C
% H
% N
% Cl
% S
* 02
0
0
13,

33
66
15,
86
5
1

4
2


471

.78
.22
466
.70
.36
.42
.29
.01
.23
JAN
0
0
12,100

22.85
77.15
15,084
83. 38
4.70
1.83
-_
1.95
8.14
JAN
0
0
10,910

21.45
78.55
14, 878
84.18
4.62
1.82
--
1.34
8.03
JAN
0
0
12,

17
82
15,
82
4
1

3
8


380

. 50
.50
013
.46
.15
.90

.08
.41
JAN
AVERAGE
0
0
11.

20
79
14,
83
4
1

2
8


797

. f>
.4
996
.34
.49
.85

.12
.19
MARCH
0
0
12,

35
64
14,
84
5
1

1
6


675

.92
.08
986
. 55
. 51
.51
.07
.78
.60
MARCH
0
0
12,780

23.29
76.71
14, 982
82.27
4.73
1.80
.16
2.14
9. 00
MARCH
0
0
13,210

21.31
78. 58
15, 132
82.79
4. 59
1.90
.16
1.79
8.85
MARCH
0
0
13,170

24.02
75.98
15, 173
82.29
4. 76
1.88
.12
1.99
9.07
MARCH "
AVERAGE
0
0
12,

26
73
15,
82
4
1

1
8


959

.14
.84
069
.98
.90
.77
.13
.93
. 38
MAY


11,

30.
69.
15,
86.
5.
1.

1.
5.
0
0
603

70
30
189
20
78
48
10
30
13
MAY
0
0
10,

23
76
15,
88
5
1

1
3


800

.90
.10
270
. 32
.28
.69
.04
.53
.14
MAY
0
0
12,

32
67
15,
85
5
2

1
4


715

.80
.20
279
.68
.62
.00
.06
.45
.79
MAY
AVERAGE
0
0
11,706

29.13
70.87
13,246
86.73
5.56
1.72
.07
1.43
4.35
      Note:  Used as the reference fuel.

-------
                                  TABLE  A-8.   AS-RECEIVED dRDF PROPERTIES
DEC DEC DEC
AVERAGE
As Received
'(, Moi sture
•i Ash
* Volatile
* Fixed C.
Btu/lb
Dry Basis
't C
'i II
* ^2
y, ci
% s
* Ash
I 02
Fusion
Initial
1st Softening
2nd Softeninq
Fluid
Mineral Analysis
Phos.Perrt.Ox
Silica
Ferric Ox.
Alumina
Ti tania
Sodium Ox.
Potasium Ox.
Lime
Magnesium
Sulfur TriOx.
Undet .
10.72 16.08
16.95 22.98
59.89 53.18
12.44 7.76
6667 6309

43.98
5.29
. 35
.40
.19 .60
18.99 27.38
30. 80

1955°F 2080T
2055°F 2120°F
2175°F
2290°F 2260°F

. 87
55. 52
2. 27
13.45
.66
6. 82
1. 30
10. 75
1 .14
6. 03
1.19
13.4
19.97
56. 54
10.1
6488

43.98
5.29
.35
.40
. 40
23. 19
30.80

2 0 1 8 ° F
2088°F
2175°F
2275°F

, -87
55. 52
2. 27
13.45
. 66
6.82
1.30
10. 75
1. 14
6.03
1.19
MARCH
9. 55
24. 55
57.84
8.06
6008

37. 04
5. 04
.47
.49
. 22
27. 14
29.60

1940°F
2 0 0 0 ° F
2060°F
2180°F

.67
64.29
2. 34
6.74
. 55
9.60
. 60
10. 08
1.77
1.49
1.87
MARCH
15. 69
24.28
50. 32
9.71
5059

41.3
3.9
. 30
.40
. 30
28. 80
25. 00

2 0 8 0 ° F
2160°F
2210°F
2 2 6 0 ° F












MARCH













2080°F
2130°F
2 1 7 0 ° F
2200°F

.71
77.45
1.91
2. 77
1.10
1.46
. 34
6.90
.81
1. 13

MARCH MARCH
AVERAGE













2060°F
2120°F
2180°F
2220°F

. 82
73.00
4.41
3.78
1.33
5.93
. 65
5. 54
. 77
1. 03

12.62
24.41
54.08
8.89
5534

39.17
4.47
. 39
.45
.26
27.97
27. 30

2040°F
2103°F
2 1 5 5 ° F
2215°F

.73
71. 58
2. 89
4.43
.99
5.66
. 53
7. 50
1.12
1.22
1.87
MAY MAY
7.00 13.15
23.51 30.04
58.42 47. 57
11.07 9.24
6015 5068

38.48 34.04
5.38 4.27
1.23 .58
.31 .32
.19 .27
25i28 34.59
29.13 25.93

1960°F 2050°
2060°F 2150°
2080°F 2170°
2190°F 2260°

.73 .58
63.06 63. 58
1.57 4.27
6.43 12.23
.56 .73
9.69 5.81
.65 1.22
10.54 7.82
1.71 1.54
3.60 .93
1.46 1.29
MAY
16
32
41
8
.51
.72
.'81
.96
4716

34
3



39
20

F
F
F
F


64
2
6

7

10
1
b


.38
.96
.73
.44
.37
.19
.93






. 64
. 31
.09
. 52
.79
.08
.86
.87
. 51
. 08
.25
MAY
AVERAGE
12.22
28. 75
49.27
9.76
5266

35. 63
4. 54
.85
.36
.28
33.02
25. 33

2005°F
2105°F
2125°F
2225°F

.65
63. 65
2.64
8.39
.69
7.53
.91
9.74
1. 59
3.20
1. 00
*GRAND
AVERAGE
12.75
24'. 38
53.30
9.58
5763

39.59
4.77
.53
.40
.31
28.06
27.81


















Note :
       The  d-RDF"properties (grand average) were used  as a basis for normalizing the emissions results for  all test data.

-------
TABLE A-9.  MOISTURE AND ASH FREE dRDF PROPERTIES

% Moisture
% Ash
Btu/lb
MAP Basis
% Vol
% Fixed C.
Btu/lb
% C
% H
% N2
% Cl
% S
% 02
DEC
0
0
6667

82.80
17.20
9217
54.29
6.53
.43
.49
.23
38.02
DEC
0
0
6309

87.27
12.73
10353




.83

MARCH
0
0
6008

87.77
12.23
9117
50.84
6.92
.65
.67
.30
40.63
MARCH
0
0
5059

83.82
16.18
8427
58.00
5.48
.42
.56
.42
35.11
MAY
0
0
6015

84.07
15.93
8656
51.50
7.20
1.65
.41
.25
38.99
MAY
0
0
5068

83.73
16.27
8921
52.04
6.53
.89
.49
.41
39.64
MAY
0
0
4716

82.35
17.65
9290
56.54
6.51
1.20
.72
.61
34.42
AVERAGE
0
0
5692

84.54
15.46
9140
53.87
6.53
.87
.56
.44
37.80
S.D.

—
739

2.13
2.13
618
2.92
.58
.48
.12
.22
2.51
                          137

-------
                                  APPENDIX  B
            SUMMARY SHEETS  FOR ASME ABBREVIATED EFFICIENCY TESTS
                     AND  BOILERS 1 AND 2 SPECIFICATIONS
SUMMARY SHEET
       A.SME  TEST FORM
FOR ABBREVIATED  EFFICIENCY  TEST
                                                                    1:0
                                                                PTC 4.1-a(1964)
TEST NO BOILER NO.
DATE 5/4
O»NER OF PLANT LOCATION
TeST CONDUCTED BY
POLES MA* E & TYPE

DBJECTWE OF TEST
RATED CAPAC
DURA T (ON
TY
STOKER TYPE & SIZE
PULVERIZER. TYPE & SIZE BURNER, TYPE & SIZE
FUEL USED MINE COUNTY STATE
SIZE AS FIRED
PRESSURES & TEMPERATURES FUEL DATA
1
2
3
4
5
6
7
a
9
10
"
12
13
U
STEAM PRESSURE IN BOILER DRUM
STEAM PRESSURE AT S H. OUTLET
STEAM PRESSURE AT R. H. INLET
STEAM PRESSURE AT R. H OUTLET
STEAM TEMPERATURE AT S- H. OUTLET
STEAM TEMPERATURE AT R H INLET
STEAM TEMPERATURE AT R.H. OUTLET
WATER TEMP. ENTERING (ECON HBOILER)
STEAM QUALITY!". MOISTURE OR P. P.M.
AIR TEMP. AROUND BOILER (AMBIENT)
TEMP AIR FOR COMBUSTION
(This is Reference Temperature) T
TEMPERATURE OF FUEL
GAS TEMP. LEAVING (Boiler) (Eton.) (Air Hlr.)
GAS TEMP. ENTERING AH (II conditions to be
psio
psio
psi a
psio
F
F
F
F

F
F
F
F
F
153






230
.95

68

457

UNIT 0 UANTITIES
15
16
17
ia
19
20
21
22
23
24
25
ENTHALPY OF SAT. LIQUID (TOTAL HEAT)
ENTHALPY OF (SATURATED) (SUPERHEAT ED)
STM
ENTHALPY OF SAT. FEED TO (BOILER)
(ECON.)
ENTHALPY OF REHEATED STEAM R.H. INLET
ENTHALPY OF REHEATED STEAM R. H.
OUTLET
HEAT ABS/LB OF STEAM (ITEM 16-ITEM 17)
HEAT ABS, LB R.H. STEAMOTEM 19-ITEM 18)
DRY REFUSE (ASH PIT » FLY ASH) PER LB
AS FIRED FUEL
Btu PER LB IN REFUSE (WEIGHTED AVERAGE)
CARBON BURNED PER LB AS FIRED FUEL
DRY GAS PER LB AS FIRED FUEL BURNED
Biu/lb
Btu/lb
Biu/lb
Btu/lb
Btu/lb
Btu'lb
Btu/lb
Ib/lb
Btu/lb
Ib/lb
Ib/lb
HOURLY QUANTITIES
26
27
28
29
30
31
ACTUAL WATER EVAPORATED
REHEAT STEAM FLOW
RATE OF FUEL FIRING (AS FIRED -t)
TOTAL HEAT INPUT (Item 28 X Item 41)
1000
HEAT OU'PUT IN BLOW-DOWN WATER
HEA*L "'"" J''llem 20) •(!(•", J7. Item 21) • Item 30
OUTPUT '000
b'hr
b,-hr
Ib/nr
kBAr
UB/hr
kB/n,
198.2
L195.8
198.2


997. ft

.37
5800
.517
21. b

10,200

1952
22,850
—
10,176
FLUE CAS ANAL. (BOILERMECON) (AIR MTR) OUTLET
32
33
34
35
36
CO,
°J
CO
Nj (BY DIFFERENCE)
EXCESS AIR
-. VOL
-, VOL
% VOL
r. VOL
%
5.9
11.3
—
82.7
104
COAL AS FIRED
PROX. ANALYSIS
37
38
39
40
MOISTURE
VOL MATTER
FIXED CARBON
ASH
TOTAL
41
42
Btu per Ib AS FIRED
ASH SOFT TEMP.'
ASTM METHOD
% wt
1.3
22.55
54.23
21.95

11,706

COAL OR OIL AS FIRED
ULTIMATE ANALYSIS
43
44
45
46
47
40
37
CARBON
HYDROGEN
OXYGEN
NITROGEN
SULPHUR
ASH
MOISTURE
TOTAL
66.52
4.27
3.38
1.33
1.20
21.95
1.3

COAL PULVERIZATION
48
49
50
64
GRINDABILITY
INDEX-
FINENESS %THRU
50 M-
FINENESS 1. THRU
200 M-



INPUT. OUTPUT
EFFICIENCY OF UNIT %

51
52
53
44
41
OIL
FLASH
Sp. Gro
POINT f
• ity Deg. API-


VISCOSITY AT SSU-
BURNER SSF
TOTAL
% wt
Btupe

HYDROGEN
Ib

CAS
54
55
56
57
58
59
60
t-61
CO
CH4 METHANE
CjH, ACETYLENE
C,H« ETHYLENE
C,H.
ETHANE
H,S
COj
H,
HYDROGEN
TOTAL

62
63
41
TOTAL
*B wt
HYDROGEN



% VOL










DENSITY 68 F
ATM. PRESS.
Btu PER CU FT
Btu PER LB
ITEM 31
• 100



ITEM 29
HEAT LOSS EFFICIENCY
65
66
67
68
69
70
71
72
HEAT LOSS DUE TO DRY GAS
HEAT LOSS DUE TO MOISTURE IN FUEL
HEAT LOSS DUE TO HjO FROM COMB.OFH,
HEAT LOSS DUE TO COMBUST- IN REFUSE
HEAT LOSS DUE TO RADIATION
UNMEASURED LOSSES
Btu'lb
A. F. FUEL






TOTAL
EFFICIENCY = (100 - t.m 71)
% ol A F
FUEL
17.9
.1
4.0
18.3
3.7
1.5
45.5
54.5
'Not Reouireo1 for Elfic «ic* Tesfng
> For Po.nt of Meosurem.nl See Por. 7.2.8 1. PTC 4.1-1964
                                   138

-------
CALCULATION SHEET
             ASME  TEST   FORM
FOR   ABBREVIATED  EFFICIENCY
                            1:0
                PTC 4.1-b  (1964)

TEST      Revised September, 1965
CWNER OF PLANT TEST NO. BOILER NC. DATE
3C'
24
25
36

65
66
67
68
„
70
71
72
r ITEM 15 ITEM 171 kBlr
HE«T (TUTPUT N BOILEB "LC'.'iO'M »ATE p =' £ "i7 « • - = !; E> "• 'f"- r •= r f - , - 1
L 1300
/f impractical to we igb refuse, this
item can be estimated as follows 21.95
npv OFFIKF PFP inner .< r,PFCFUE' - ' ASH IN AS FIRED COAL f(OT[. |fr
,00 -'.COMB. IN REFUSE SAMPUE PITREFUSE

-UE DUST 1 ASH
DIFFER MATERIALLY
,__ r— —1 N COMBUSTIBLE CONTENT, THEY
££ PT«M ITEM«1 SHOULD BE ESTIMATED
CARBON BURNED 66.52 .37 x 5800 .517 SEPARATELY. SEE SECTION 7
FUEL 'DO [_ U.500 J I.UMKUIATIUNS.
DRY GAS PER LB llCOj « 80, • 7(Na « CO)
BURNED 3(CO> * C0) / v. 8
ITEM 32 ITEM 33 | ITEM 35 ITEM 34 ) ITEM 24 ITEM
11x5. 9 * « x jj.3 « 7^ 82.7' — / x .517 * 1.2
(ITEM 32 ITEM 34 \
.5.9 . « ) L
	 /
E,CE,t 0, - C° ,TVu,J, - 'TEM-3*
AIR* = 1 00 X 	 	 	 	 1 no x ^ =
.2682N, - (0 C0_ ) BZ./ 11.3 ITEM 34
* 3 3n83(ITFM3S) (ITFM13- "=•»•»)
1
HEAT LOSS EFFICIENCY
HEAT LOSS DUE LB DRY GAS .25 ITEM 25 ° . 25(| TEM ,31 _,|TEM , ,,
TO DRY GAS = PERLBAS xC x ('!., -•„;,)= _n"x«»4X ' Uit*ll) =
FIRED FUEL ' Uni, 21.6 457 68
MO*STURES|N1FUELI = AS F'RE"^! X 1 (ENTHALPY OF VAPOR AT 1 PSIA & T GAS LVG)
(ENTHA1 PY OF 1 IQUIDAT T AIR)] ~ ''^M 37 X[(FNTHAI PY OF VAPOR
1269 36 10° ,
AT 1 PSIA 1 T ITEM 13) -(ENTHALPY OF LIQUID AT T ITEM 11)1 = 	
HEAT LOSS DUE TO H,0 FROM COMB. OF H, = 9H, x [(ENTHALPY OF VAPOR AT 1 PSIA & T GAS
4. 27 LVG) - (ENTHALPY OF LIQUID AT T AIR)]
- „ , !TEM 44 x [,ENTHALPr nf VAPOR AT 1 P^IA «. T ITEM 13) - (ENTHALPY OF LIQUID AT


HEAT LOSS DUE TO ITEM 22 ITEM 23
COMBUSTIBLE IN REFUSE = 37 x 5800 =
HCAT LOSS DUE TO TOTAL BTU RADIATION LOSS PER HR
KADiAtlGN' LB AS FISt D FUE L ITI^ZB
UNMEASURED LOSSES ••
TOTAL
EFFICIENCY = (100 - ITEM 71)
47~l
21.6
,67 J 	
104%

Blu/lb
AS FIRED
FUEL
2.1.Q1.
.16:0.
474

ZU6 .







LOSS „
HHV
100 =
41
41
67
__X100 =
41
— xlOO =
41
69
	 x 1 00 -
41
70 x 100 =
41








LOSS
X
17.9
.1
4.P
. ia,3
. . .3. -.7
1.5

45.5

54.5

 II lo»ei ore noi meo.u.rd. u»e ABMA Slondo'd Rodrolion Lou Chorl. F,g. 8. PTC 4.1-1964

 • Unm.oiuod loins li»>«d in PTC 4.1 but not lobulot.d obov.mo, by pro'idcd (or by o»>ijn,ng o •

  og...d upon -o'u« fc>' I'"" 70.
                                                         139

-------
SUMMARY SHEET
         A.SME  TEST  FORM
FOR  ABBREVIATED EFFICIENCY TEST
1:1

PTC 4.1.a(1964)
TEST NO BOILER NO.
DATE 5/13
0»NER OF PLANT LOCATION
TEST CONDUCTED BY OBJECTIVE OF TEST
DURATION
BO LEO MAKE 8. TYPE RATE D CA PAD TY
STOKER TYPE & SIZE
PULVERIZER, TYPE & SIZE BURNER, TYPE & SIZE
FUEL USED MINE COUNTY STATE
SIZE AS FIRED
PRESSURES & TEMPERATURES FUEL DATA
1
2
3
4
5
6
7
8
9
10
11
12
13
14
STEAM PRESSURE IN BOILER DRUM
STEAM PRESSURE AT S. H. OUTLET
STEAM PRESSURE AT R. H. INLET
STEAM PRESSURE AT R. H. OUTLET
STEAM TEMPERATURE AT S. H. OUTLET
STEAM TEMPERATURE AT R H INLET
STEAM TEMPERATURE AT R.H. OUTLET
WATER TEMP. ENTERING (ECON MBOILER)
STEAM QUALITY?". MOISTURE OR P. P. M.
AIR TEMP. AROUND BOILER (AMBIENT)
TEMP AIR FOR COMBUSTION
TEMPERATURE OF FUEL
GAS TEMP. LEAVING (Boiler) (Econ.) (Air Htr.)
corrected to Quarontee)
piia
piia
pita
psio
F
f
F
f

f
F
F
F
F
155






221
.95

78
—
501

UNIT QUANTITIES
15
16
17
18
19
20
21
22
23
24
25
ENTHALPY OF SAT. LIQUID (TOTAL HEAT)
ENTHALPY OF (SATURATED) (SUPERHEATED)
STM.
ENTHALPY OF SAT. FEED TO (BOILER)
(ECON.)
ENTHALPY OF REHEATED STEAM R.H. INLET
ENTHALPY OF REHEATED STEAM R. H.
OUTLET
HEAT ABS/LB OF STEAM (ITEM 16-ITEM 17)
HEAT ABS/LB R.H. STEAM(ITEM 19-ITEM IB)
DRY REFUSE (ASH PIT « FLY ASH) PER LB
AS FIRED FUEL
Btu PER LB IN REFUSE (WEIGHTED AVERAGE)
CARBON BURNED PER LB AS FIRED FUEL
DRY GAS PER LB AS FIRED FUEL BURNED
3tu/lb
Btu/lb
Btu/lb
Btu/lb
Btu/lb
Blu'lb
Btu/lb
Ib/lb
Btu/lb
Ib/lb
Ib/lb
HOURLY QUANTITIES
26
27
28
29
30
31
ACTUAL WATER EVAPORATED
REHEAT STEAM FLOW
RATE OF FUEL FIRING (AS FIRED wt)
TOTAL HEAT INPUT (Item 28 x Item 41)
1000
HEAT OUTPUT IN SLOW-DOWN WATER
T|0!J*.L(ltem26.lt.m20).(ltem27»ltem2l).ltem30
OUTPUT 1000
b'hr
Ib/hr
Ib/hr
kB/h,
liB/hr
kB/n,

1196
189


1007

.36
5075
.41
12.62

20.018
—
3309
29.741
—
20,158
FLUE GAS ANAL. (BOILERHECON) (AIR HTR) OUTLET
32
33
34
35
36
CO,
o,
CO
N, (BY DIFFERENCE)
EXCESS AIR
r, VOL
••. VOL
% VOL
% VOL
*
8.1
9,9

82.0
82
COAL AS FIRED
PROX. ANALYSIS
37
38
39
40
MOISTURE
VOL MATTER
FIXED CARBON
ASH
TOTAL
41
42
Btu per Ib AS FIRED
ASH SOFT TEMP.-
ASTM METHOD
% wt
6.62
31.68
38.39
23.33

89.88

COAL OR OIL AS FIRED
ULTIMATE ANALYSIS
43
44
45
46
47
40
37
CARBON
HYDROGEN
OXYGEN
NITROGEN
SULPHUR
ASH
MOISTURE
TOTAL
54.06
4.10
9.83
1.06
L .86
23.33
6.62

COAL PULVERIZATION
48
49
50
64
GR1NDABIL1TY
INDEX-
FINENESS XTHRU
50 M*
FINENESS X THRU
200 M*




51
52
53
44
41
OIL
FLASH
Sp. Gro
POINT f
• ity Dej. API-


VISCOSITY AT SSU-
BURNER SSF
TOTAL HYDROGEN
Btu pe

Ib

GAS
54
55
56
57
58
j,
60
61
CO
CH4 METHANE
C,H, ACETYLENE
C,H« ETHYLENE
C.H.
ETHANE
H,S
CO,
H,
HYDROGEN
TOTAL

62
63
41
TOTAL HYDROGEN
Xwt



5 VOL










DENSITY 68 F
ATM. PRESS.
Btu PER CU FT
Btu PER LB


INPUT-OUTPUT ITEM 31 - 100
EFFICIENCY OF UNIT X ITEM 29
HEAT LOSS EFFICIENCY
65
66
67
68
69
70
71
72
HEAT LOSS DUE TO DRY GAS
HEAT LOSS DUE TO MOISTURE IN FUEL
HEAT LOSS DUE TO H,O FROM COMB. OF H
HEAT LOSS DUE TO COMBUST. IN REFUSE
HEAT LOSS DUE TO RADIATION
UNMEASURED LOSSES
Btu/lb
A. F. FUEL






TOTAL
EFFICIENCY = (100 - Item 71)
% of A F
FUEL
14.8
.9
5.1
20.3
1.8
1.5
44.4
55.6
'Not Required lor Efficiency Telling
t For Point =1 Me.iuremerit See Per. 7.2.8.1-PTC 4.1-1964
                                          140

-------
CALCULATION SHEET
             A.SME   TEST   FORM
FOR  ABBREVIATED   EFFICIENCY   TEST
    PTC4.1-b (1964)

Revised September, 1965
                                                                                                                       1:1

30
54
25
36

65
66
67
68
70
71
72
O. NEK OF PLANT T£ST NO. BOILER NO. PATE
I" ITEW 15 ITEV" '7~
HEAT CUT P. 7 IN E.OlLER Ell.0«-00' N »ATFR -L9 Oe »ATER &L.O»-"O»N PER H» x ' .... . . . - . _^
\ 1000
If impractical fa weigh refuse, ffiis
ifem con Se estimated as follows 23 33
100 - ?i COMB. IN REFUSE SAMPLE p|T REflJS€
WB n.
-UE DUST 4 ASH
DIFFER MATERIALLY
p- — , IN COMBUSTIBLE CONTENT, THEY
ITEM 43 F7TEM22 ITEM 23 1 SHOULD BE ESTIMATED
CARBON BURNED 5406 .36 „ 4205 .41 SEPARATELY. SEE SECTION 7.
PWVBMF.RED . 10() ^ U5M J , 	 COMPUTATIONS.
DRY GAS PER LB HCOj + 80, + 7(N, + CO)
BURNED 3(C0' * C0) / \ _ .
ITEM 32 ITEM 33 / ITEM 35 ITEM 34) I ITEM J4
I1x .8.1 *8x..9.9. *^__82. » .—-./ x ,91.
+ 3 s)
ITEM 47 I
* ..:s.6.. 12.62
/ITEM 32 ITEM34\ L 267 J
3 x 1. .9:f. . * .."..1
EltrEI- °i C ITEM'S ITEM 34
EXCESS i • Q Q -
A l R t = i oo x — 	 	 — 	 -innv "••' * =
.2682N, - <0 . i5L | |TEM ,.
2 2<»7 (ITPM Tr) (ITFU n )

HEAT LOSS EFFICIENCY
HEAT LOSS DUE LB DRY GAS ITEM 25 (ITEMJ31 (ITFM11I
TODRYGAS = PERLBAS xC X Civ, -•«,)=," xMZ ' ~' " =
FIRED FUEL " Unit 12.62
MOmu^N^U^ '%*££&&* l™™^ Of VA^AT 1 PSI^^G^S LVG)
(EMTH4' PY "F I iiJU'PAT T AIR)] - " fc 7 » [(pMTHAl Pv OF VAPOB
100 46
AT 1 PSIA & T ITEM 13) -(ENTHALPY OF LIQUID AT T ITEM 11)] = 	
HEAT LOSS DUE TO H,0 FROM COMB. OF H, = 9H, x [(ENTHALPY OF VAPOR AT I PSIA & T GAS
4_1 LVG) - (ENTHALPY OF LIQUID AT T AIR)]
. , , ITEM 44 x [(ENTHALPY OF VAPOR AT 1 PSIA 1 T ITEM 11) - (ENTHALPY OF LIQUID AT
100 T ITEM 11)] = 	
HEAT LOSS DUE TO ITEM 22 ITEM 23
COMBUSTIBLE IN REFUSE = . 36 x 5075 ~
HEAT LOSS DUE TO TOTAL BTU RADIATION LOSS PER MR
RADIATION' LB AS FIRED FUEL ITtxze
UNMEASURED LOSSES ••
TOTAL
EFFICIENCY = (100 -ITEM 71)
82%

B.u/l.
AS FIRED
FUEL
1335
82
458
18Z7. .







LOSS x
HHV
100 =
65
	 X 100 =
41
** x 100 =
41
asur«d lo.^i li«»i( in PTC 4.1 but »oi iabulot«d above mo» b> provided lor by oinjn.ng o ^
   O0,..d upon .olu. to, II.". 70.
                                                         141

-------
SUMMARY SHEET
         A.SME  TEST  FORM
FOR ABBREVIATED  EFFICIENCY TEST
  1:2
PTC 4.1-o{1964)
TE;T NC BOILER NO. DATE 5/n
OiNER OF PLANT LOCATION
TS ST CONDUCTED BY OBJECTIVE OF TEST DURATION
B2:L;R MA-.E & TYPF RATE D CA PACi Tr
STOKER TYPE & SIZE
PULVERIZER, TYPE s. SIZE BURNER, TYPE & SIZE
FUEL USED MINE COUNTY STATE SIZE AS FIRED
PRESSURES & TEMPERATURES FUEL DATA
1
2
3
4
5
6
-i
e
9
10
1 1
12
13
14

15
16
17
18
19
20
21
22
23
24
25

26
27
28
29
30
31
STEAM PRESSURE IN BOILER DRUM
STEAM PRESSURE AT S. H. OUTLET
STEAM PRESSURE AT R. H. INLET
STEAM PRESSURE AT R. H. OUTLET
STEAM TEMPERATURE AT S H. OUTLET
STEAM TEMPERATURE AT R H INLET
STEAM TEMPERATURE AT R.H. OUTLET
WATER TEMP. ENTERING (ECON )(BOILER)
STEAM QUALITY r. MOISTURE OR P. P.M.
AIR TEMP. AROUND BOILER (AMBIENT)
TEMP AIR FOR COMBUSTION
(Thij ii Reference Temp.rotw'e) T
TEMPERATURE OF FUEL
GAS TEMP. LEAVING (Boil.r) (Eton.) (Air Hlr.)
GAS TEMP. ENTERING AH (II conditions to b.
UNIT QUANTITIES
ENTHALPY OF SAT. LIQUID (TOTAL HEAT)
ENTHALPY OF (SATURATED) (SUPERHEATED)
STM
ENTHALPY OF SAT. FEED TO (BOILER)
(ECON.)
ENTHALPY Of REHEATED STEAM R.H. INLET
ENTHALPY OF REHEATED STEAM R. H.
OUTLET
HEAT ABS/LB OF STEAM (ITEM 16-ITEM 17)
HEAT ABS/LB R.H. STEAM(ITEM 19-ITEM 18)
DRY REFUSE (ASH PIT » FLY ASH) PER LB
AS FIRED FUEL
Btu PER LB IN REFUSE (WEIGHTED AVERAGE)
CARBON BURNED PER LB AS FIRED FUEL
DRY GAS PER LB AS FIRED FUEL BURNED
psio
?SIO
psro
ps.a
F
F
F
F

F
F
F
F
F

Btu/lb
Btu/lb
Btu/lb
Blu/lb
Btu/lb
3tu'lb
Btu/lb
Ib/lb
Btu/lb
Ib/lb
Ib/lb
HOURLY QUANTITIES
ACTUAL »ATER EVAPORATED
REHEAT STEAM FLO*
RATE OF FUEL FIRING (AS FIRED -ly
TOTAL MEAT INPtJT (It*™ 28 X Item 41)
1000
HEAT OUTPUT IN BLOW-DOWN WATER
HEAT"" (l"m J«'ll«'" 20)>(it«™ 37.11..- 31) •it.m. 30
OUTPUT 1000
Ib K.
b/h.
1 b/h r
kB/hr
it B hr
kB/hr
154






230
.95

70

512



1195.8
198.2


997.6

.35
4205
.371
14.3

i7O3.a

3776
30,684
—
17,696
FLUE CAS ANAL. (BOILERMECON) (AIR HTR) OUTLET
32
33
34
35
36
CO,
Q
CO
N, (BY DIFFERENCE)
EXCESS AIR
". VOL
', VCL
% VOL
% VOL
%
6.4
11.0
—
3?. 6
qa.6
COAL AS FIRED
PROX. ANALYSIS
37
38
39
40
MOISTURE
VOL MATTER
FIXED CARBON
ASH
TOTAL
41
42
Btu per Ib AS FIRED
ASH SOFT TEMP.
ASTM METHOD
% .t
7.94
38.26
29.09
24.73

83.82

COAL OR OIL AS FIRED
ULTIMATE ANALYSIS
43
44
45
46
47
40
37
CARBON
HYDROGEN
OXYGEN
NITROGEN
SULPHUR
ASH
MOISTURE
TOTAL
47.29
4.07
14.19
.90
.66
24.73
7.94

COAL PULVERIZATION
48
49
50
64
GRINDABILITY
INDEX-
FINENESS %THRU
50 M*
FINENESS % THRU
200 M*



INPUT-OUTPUT
EFFICIENCY OF UNIT %

51
52
53
44
41
OIL
FLASH POINT f
Sp. Grovity Deg. AP •


VISCOSITY AT SSU'
BURNER SSF
TOTAL HYDROGEN
% wt
Btu per Ib

CAS
54
55
54
57
58
V»
60
61
CO
CH« METHANE
CjH, ACETYLENE
C,K, ETHYLENE
CjH, ETHANE
H,S
CO,
H, HYDROGEN
TOTAL

63
63
41
TOTAL HYDROGEN
X .t



%VOL










DENSITY M F
ATM. PRESS.
Btu PER CU FT
Btu PER LB
ITEM 31 . 100
ITEM 39
Blu/lb
HEAT LOSS EFFICIENCY A. F. FUEL
65
66
67
68
69
70
71
72
HEAT LOSS DUE TO DRY GAS
HEAT LOSS DUE TO MOISTURE IN FUEL
HEAT LOSS DUE TO H,0 FROM COMB OF H,
HEAT LOSS DUE TO COMBUST. IN REFUSE
HEAT LOSS DUE TO RADIATION
UNMEASURED LOSSES
TOTAL
EFFICIENCY = (100 - Item 71)



%•! A. F
FUEL
18.8
1.2
5.4
17.6
1 .8
1.5
46.3
53.7
'Not R.quir.d lo' Efflcl.nc, T.ltlng
t For Po.nt ol M.otur.m.nt S.. Pot. 7.2.8.1-PTC 4.1-1964
                                            142

-------
CALCULATION SHEET
             ASME   TEST  FORM
FOR   ABBREVIATED   EFFICIENCY   TEST
    PTC4.1-B (1964)1:2

Revised  September, 1965

30
24
25
36

65
66
67
66
f-
70
71
72
OWNEK OF PLAHT TEST NO. BOILER NO. DATE
T ITEM 15 ITEM 1?"
^T •' ~ ;••_ - : ,'T II, B3ILES BLDW-DOWN ».'. TE " = _E OF » o -| p c . 3 , .- - . •. ;•£ R t-i r > < 	 - 	
L
item con fae estimated as follows 24.73 .35
- = v prr.. x ITEM 44 x ((ENTHALPY OF VAPOR AT 1 PSIA 8. T ITEM 13) - (ENTHALPY OF LIQUIO AT
100 T ITEM ID] = 	
HE AT LOSS DUE TO ITEM 22 ITEM23
COMBUSTIBLE IN REFUSE ^5 * 42Q5
NT A T L OS^ DUE TO TOTAL RTURAPtATIONlOSSPr^HP
k /. ! 1 1 A 1 ! ON • I b l 1 ' . R I. : ^ J f > ' ' '-• . *
UNMEASURED LOSSES ••
TOTAL
EFFICIENCY = (100 -ITEM 71)

Btu/lb
AS FIRED
FUEL
15KQ
100
455
1472





LOSS
TiffT
100 :
41
^- x 100 s
41
67
	 XIOO =
41
68
— xlOO =
41
69
jl
70 x 100 -
41






LOSS
X
.18,8
. . 1,2.
5.4
17.6

1.8
1.5
46.3

53.7

  If  I0,»** or. nor mroiured, u.e ABMA Sfondo'd Rod.of.oo Lou Chort. F.g. 8, PTC 41-1964

 " Unm«8«uf«d lo»»«I l>»fed rn PTC 4.1 but not lobut«l*d obov* moy by provided for by aitigmng o '
  ««r**d upon *otu« for Item 70.
                                                      143

-------
SUMMARY SHEET
         A.SME  TEST FORM
FOR  ABBREVIATED  EFFICIENCY TEST
   0:1

PTC 4.l-a{1964)
TEST NO BOILER NO.
0*NER 0= PLANT LOCATION
*EST- CONDUCTED BY OBJECTIVE OF TEST
DATE 5/14/77

DURATION
BOLfR MAKE 4 TYPE RA T E D CAPACI TY
STOKER TYPE 8. SIZE
PULVERIZER, TYPE & SIZE BURNER, TYPE & SIZE
FUEL USED MINE COUNTY STATE
SIZE AS FIRED
PRESSURES & TEMPERATURES FUEL DATA
1
2
3
4
5
6
7
8
9
10
I 1
12
13
14
STEAM PRESSURE IN BOILER DRUM
STEAM PRESSURE AT S. H. OUTLET
STEAM PRESSURE AT R. H. INLET
STEAM PRESSURE AT R. H. OUTLET
STEAM TEMPERATURE AT S H. OUTLET
STEAM TEMPERATURE AT R H INLET
STEAM TEMPERATURE AT R.H. OUTLET
WATER TEMP. ENTERING (ECON ((BOILER)
ST E AM QUA LI TYT. MOISTURE OR P. P.M.
AIR TEMP. AROUND BOILER (AMBIENT)
TEMP AIR FOR COMBUSTION
TEMPERATURE OF FUEL
GAS TEMP. LEAVING (Boil«r) (Econ.) (Air Htr.)
GAS TEMP. ENTERING AH (If condition, to be
psio
psio
psia
P!,0
F
F
F
F

F
F
F
F
F
152






226
.95

78

472

UNIT 0 UANTITIES
15
16
17
18
19
20
21
22
23
24
25
ENTHALPY OF SAT. LIQUID (TOTAL HEAT)
ENTHALPY OF (SATURATED) (SUPERHEATED)
STM.
ENTHALPY OF SAT. FEED TO (BOILER)
(ECON.)
ENTHALPY OF REHEATED STEAM R.H. INLET
ENTHALPY OF REHEATED STEAM R. H.
OUTLET
HEAT A8S/LB OF STEAM (ITEM I6-ITEM 17)
HEAT ABS/LB R.H. STEAM (ITEM 19-ITEM 18)
DRY REFUSE (ASH PIT * FLY ASH) PER LB
AS FIRED FUEL
Btu PER LB IN REFUSE (WEIGHTED AVERAGE)
CARBON BURNED PER LB AS FIRED FUEL
DRY GAS PER LB AS FIRED FUEL BURNED
Btu/lb
Btu/lt>
Btu/lb
Btu/lk
Btu/lb
Biu'lb
Btu/lb
Ib/lb
Btu/lb
Ib/lb
Ib/lb
HOURLY QUANTITIES
26
27
28
29
30
31
ACTUAL WATER EVAPORATED
REHEAT STEAM FLOW
RATE OF FUEL FIRING (AS FIRED wt)
TOTAL HFAT INPUT (Item 28 X liem 41)
1000
HEAT OUTPUT IN BLOW. DOWN WATER
H°ATL(l"m26""""20|'(l"m37'''""2"'''""30
OUTPUT 1000
Ib h,
Ib/hr
Ib/hr
fcBAr
VB/hr
kB/nr

1195.7
194.2


1001.5

.269
580
.298
6.404

tl,459

4991
5103
25^5J}Q
11.476
FLUE GAS ANAL. (BOILERMECON) (AIR HTR) OUTLET
32
33
34
35
36
co;
Gj
CO
N, (BY DIFFERENCE)
EXCESS AIR
% VOL
', VOL
% VOL
% VOL
%
7.3
11.5

83.2
112
COAL AS FIRED
PROX. ANALYSIS
37
38
39
40
MOISTURE
VOL MATTER
FIXED CARBON
ASH
TOTAL
41
42
Btu per Ib AS FIRED
ASH SOFT TEMP.'
ASTM METHOD
% wt
16.60
48.59
8.96
25.85_

5103

COAL OR OIL AS FIRED
ULTIMATE ANALYSIS
43
44
45
46
47
40
37
CARBON
HYDROGEN
OXYGEN
NITROGEN
SULPHUR
ASH
MOISTURE
TOTAL
30.90
1.7fi
?1 .78
.SS
.23
25.85
16.60

COAL PULVERIZATION
48
49
50
(4
GRINDABILITY
INDEX*
FINENESS %THRU
SO M-*
FINENESS % THRU
200 M-



INPUT-OUTPUT
EFFICIENCY OF UNIT *

51
52
53
44
41
OIL
FLASH
Sp. Gro
POINT F'
vity De9. API'


VISCOSITY AT SSU-
BURNER SSF
TOTAL HYDROGEN
X wt
Btu pe

Ib

GAS
54
55
VI
57
58
59
60
61
CO
CH. METHANE
C,H, ACETYLENE
C,H, ETHYLENE
C,H.
ETHANE
H,S
CO,
H,
HYDROGEN
TOTAL

62
63
41
TOTAL
X wt
HYDROGEN



SVOL










DENSITY 68 F
ATM. PRESS.
Btu PERCU FT
Btu PER LB


ITEM 31 > 100
ITEM 29
HEAT LOSS EFFICIENCY
65
66
67
68
69
70
71
72
HEAT LOSS DUE TO DRY GAS
HEAT LOSS DUE TO MOISTURE IN FUEL
HEAT LOSS DUE TO H,0 FROM COMB.OFH,
HEAT LOSS DUE TO COMBUST. IN REFUSE
HEAT LOSS DUE TO RADIATION
UNMEASURED LOSSES
B.u/lb
A. F. FUEL






TOTAL
EFFICIENCY = (100 - Item 71)
%of A. F
FUEL
12.4
T.7
7.5
3.1
3 7
1.5
31.9
68.1
'Not Required for Efficiency Totting
t For Point of Meolurement See POP. 7.2.I.I-PTC 4.1-1964
                                          144

-------
CALCULATION SHEET
             ASME  TEST  FORM
FOR  ABBREVIATED   EFFICIEKCY   TEST
    PTC4.1-b (1964)     0:1

Revised  September, !96S

3C
24
25
36

65
66
67
68
>,
70
71
72
OWNER OF PLANT TEST NO. BOILER NO. DATE
T ITEM 15 ITEM 17"
HE,»- C'J-°J~ it, BOILER BLC»-DO'.'- » A - E <; =.5 OF WATER faLC».:iOWN PEB *=! » ', 	 - 	
L 1C::
If impractical to weigh refuse, this
item con be estimated as follows 95 85 7fiQ
PPV pFr,,7 CTF" IN (1TFU T iicmj«j
81 . 2 11 . 5 2
HEAT LOSS EFFICIENCY
HEAT LOSS DUE LB DRY GAS ITEM 25 ^ITEMISI IITFMMI
TO DRY GAS = PERLBAS x C x ('!.,- 'oir) = ...XXB2J . , ' ,„ =
FIRED FUEL " Uni, 6.404 472 78
MO*STURES|N FUEL3 = AS nRE^FUEL X l (ENTHA1-PY OF VA116R. 6*T ' PS'*ft^*§ LVG)
(EN THAI PY OF I 'QUIP AT T AIR)] - ' ^M K [(ENTHAI PY OF VAPOR
100 46
AT 1 PSIA & T ITEM 13) -(ENTHALPY OF LIQUID AT T ITEM 11)] = 	
HEAT LOSS DUE TO H,O FROM COMB. OF H, = 9H, x [(ENTHALPY OF VAPOR AT 1 PSIA & T GAS
3.76 1178.9 LVG) - (ENTHALPY^F LIQUID AT T AIR)]
- 0 * ITEM 44 » [(ENTHALPY OF VAPOR AT 1 PSIA & T ITEM 131 _ (FNTHALPY OF 1 IQUIt) AT


HEAT LOSS DUE TO ITEM 22 ITEM 23
COMBUSTIBLE IN REFUSE = .269 x 580 =

HfAT LOSS DUF TO TOTAL BTU RAOIA1ION LOSS PF R HR
KA.TItT.ON- LB AS FIRED FUFL i - 1 M . I,
UNMEASURED LOSSES "
TOTAL
EFFICIENCY = (100 -ITEM 71)
* !7234! 6.404
Jw J= 	
..112%
B.u/lb
AS FIRED
FUEL
. "I .
188
383

156








LOU
TiHV
100 =
41
^1 X 100 =
41
67
	 X100 =
41
68
— X100 =
41
69
x 100 -
41
70 x 100 =
41






LOSS
.W,4.
3.7
7.5

..3,1.
.3.7
1.5

31.9

68.1

                                           . 9.2 - PTC 4.1-1964
• IMo'i!.',".'. 'o"'meoLr«
-------
                                     BOILER NO. 1
                       ERIE   CITY   IRON   WORKS
                                  Erie, Pennsylvania

                           FOR MARYLAND INSTITUTION FOR MEN
                               BREATHEDSVILLE, MARYLAND

                              Predicted Performance Only
               Guaranteed efficiency at 60,000 Ib/steam/hour is 75.0%
                October  1,  1963
                G.O.  96/63
                                             17
Percent of load
33          67
                                                                              1.0
                                                                                      Peak
1. Evaporation, actual, pounds per hour
2. Pressure in the drum, psig
3. Feedwater temperature °F
4. Temperature of flue gases leaving the
furnace °F
5. Temperature of flue gases leaving the
boiler °F
6. Excess air at boiler outlet %
(See note below)
7. Temperature of air at windbox °F
8. Dust collector draft loss, in H20
(See below)
9. Gas dust and damper draft loss (See
note below)
10. Furnace and boiler draft loss
11. Total draft required inches H20
12. Induced draft design static
13. Stoker and windbox air resistance
(See note below)
14. Duct and Damper air resistance
(No orifice in duct)
15. Total air resistance inches H20
16. Forced draft fan static design
17. Flue gas leaving the boiler, Ib/hr
18. Air required for combustion, Ib/hr
19. Fuel burned, Ib/hr
20. Furnace heat liberation Btu/cu ft
21. Heat release Btu/sq ft active grate
(at 129 sq ft grate)
22. Moisture in steam leaving the boiler %
23. Unit efficiency %
24. Losses:
A. Dry gas %
B. Moisture in fuel and hydrogen %
C. Moisture in air %
D. Unburned combustible % (See note
below*)
E. Radiation %
F. Unaccounted for %
G. Total %
13,083
175
220



422

96.45
80

0.20

0.10
0.20
0.50


0.11

0.10
0.21

25,837
24,080
1,274
4,320

93,600
1/2 of
77.71

12.34
4.02
0.33
0.80
0.80
3.30
1.50
22.29
NOTE: Items 8 and 9 are per Aerotec Industries,
Items 6 13 and 24-D are per Hoffman Comb.
June 12, 1963.
*See Item 24-D above


26,167
175
220



450

57.11
80

0.40

0.10
0.33
0.83


0.32

0.10
0.42

40,121
36,748
2,447
8,300

181,000
1 1/2 of 1
80.76

10.73
4.07
0.29
1.00
1.00
1.65
1.50
19.24
52,333 78,500
175 175
220 220

1,885

513 580

40.5 34.0
80 80

1.03 2.55

0.20 0.45
0.53 0.95
1.76 3.95
See Fan Design Below

0.70 1.40

0.20 0.45
0.90 1.85
See Fan Design Below
72,084 104,693
65,349 94,464
4,885 7,420
16,550 25,800

358,000 543,000
1/2 of 1 1/2 of 1
80.43 78.91

11.27 12.43
4.17 4.28
0.30 0.33
1.50 2.00
1.50 2.00
0.83 0.55
1.50 1.50
19.57 21.09
85,000
175
220



597

31.3
80

3.00

0.55
1.10
4.65


1.60

0.55
2.15

111,535
100,435
8,052
28,000

589,000
1/2 of 1
78.53

12.61
4.31
0.34
2.20
2.20
0.51
1.50
21.47
Inc. performance dated July 22, 1963.
Engineering


Company performance dated




The unburned combustible loss is as given by Hoffman Comb.-Engr. Company based on
reinjection from boiler hoppers and decantation collector.  Guaranteed overall boiler
and stoker efficiency is 75.0% at 60,000 pounds of steam per hour.
                                        146

-------
                                         BOILER NO. 2
                           ERIE   CITY   IRON   WORKS
                                      Er ie,  Pennsylvania

                               FOR MARYLAND INSTITUTION FOR MEN
                                   BREATHEDSVILLE, MARYLAND

                                  Predicted Performance Only
                   Guaranteed efficiency at 60,000 lb/steam/hour is 75.0%
                 October 1, 1963
                 G.O. 97/63
                                                 17
Percent of load
33          67
                                                                                  1.0
Peak
10,000
175
220

432
96.45
80
0.20
0.10
0.20
0.50

0.11
0.10
0.21

19,854
18,504
979
4,620
100,000
1/2 of 1
77.07
12.41
4.02
0.34
0.80
3.66
1.50
22.93
20,000
175
220

454
57.11
80
0.40
0.10
0.35
0.85

0.32
0.10
0.42

30,792
28,203
1,878
8,850
192,000
1/2 of
80.46
10.85
4.07
0.29
1.00
1.83
1.50
19.54
40,000 60,000
175
220

519
40.5
80
1.00
0.20
0.56
1.76
See fan design
0.70
0.20
0.90
See fan design
55,277
50,112
3,746
17,700
175
220
1,870
590
34.0
80
2.40
0.45
1.05
3.90
below
1.40
0.45
1.85
below
80,340
72,490
5,694
26,900
383,000 583,000
1 1/2 of 1 1/2 of 1
80.15
11.45
4.18
0.31
1.50
0.91
1.50
19.85
78.58
12.67
4.30
0.34
2.00
0.61
1.50
21.42
66,000
175
220

611
31.3
80
3.00
0.55
1.25
4.80

1.60
0.55
2.15

87,073
78,407
6,286
29,600
642,000
1/2 of 1
78.09
12.97
4.34
0.35
2.20
0.55
1.50
21.91
 1. Evaporation, actual  pounds per hour
 2. Pressure in the drum, psig
 3. Feedwater temperature °F
 4. Temperature of flue gases leaving the
     furnace °F
 5. Temperature of flue gases leaving the
     boiler °F
 6. Excess air at boiler outlet %
     (See note below)
 7. Temperature of air at windbox °F
 8. Dust collector draft loss, in H20
     (See below)
 9. Gas dust and damper draft loss (See
     note below)
10. Furnace and boiler draft loss
11. Total draft required inches H20
12. Induced draft design static
13. Stoker and windbox air resistance
     (See note below)
14. Duct and Damper air resistance
     (No orifice in duct)
15. Total air resistance inches H20
16. Forced draft fan static design
17. Flue gas leaving the boiler, Ib/hr
18. Air required for combustion, Ib/hr
19. Fuel burned, Ib/hr
20. Furnace heat liberation Btu/cu ft
21. Heat release Btu/sq ft active grate
     (at 129 sq ft grate)
22. Moisture in steam leaving the boiler%
23. Unit efficiency %
24. Losses:
     A.  Dry gas %
     B.  Moisture in fuel and hydrogen%
     C.  Moisture in air %
     D.  Unburned combustible% (See note
          below*)
     E.  Radiation %
     F.  Unaccounted for %
     G.  Total %

NOTE:     Items 8 and 9 are per Aerotec Industries, Inc. performance  dated July 22, 1963.
          Items 6, 13 and 24-D are per Hoffman Comb. Engineering Company performance dated
               June 12, 1963.

*See Item 24-D above
     The unburned combustible loss is as given by Hoffman Comb. Engr. Company based on
     reinjection from boiler hoppers and decantation collector.  Guaranteed overall boiler
     and stoker efficiency is 75.0% at 60,000 pounds of steam per hour.
                                              147

-------
                                   APPENDIX C

                     PROCEDURE FOR ESTIMATING STACK VELOCITY


      The  emissions were monitored at  the  stack using  isokinetic  sampling
 techniques.  Achieving isokinetic sampling at  MCI  was greatly  complicated by
 the  fact  that  the stack is  sized  for  all  three boilers firing  simultaneously
 at full load.  Only  one boiler is operated at  a time  at approximately one-
 half load.  As a result,  the  average  stack gas velocity is  too low  to be read
 on an S-type pitot tube.  Because of  the  dirty nature of  the stack  gas, use
 of a hot wire anemometer  or similar device capable of reading  the low flow is
 not  practical.  As an alternative approach,  the stack gas velocity  was com-
 puted on the basis of the boiler  load, the  experimentally determined boiler
 efficiency, the fuel characteristics, and  the  Orsat analysis at  the stack.
 This  procedure is an extension of  standard  boiler  monitoring techniques.  The
 calculation used to compute stack gas velocity is  as  follows:

 (1)   The fraction of nitrogen  in  the flue  gas  is determined from the Orsat
     analysis (which condenses all the water prior to  analysis)  by  recognizing
      that:

                            N2 =  100 - C02  - 02 -  CO                  (C-l)

 (2)  The weight of nitrogen and carbon per mole of dry flue gas  is  then
     computed by Equations  (C-2) and  (C-3).


                            lb/mole DFG = T^T x 28                   (C-2)
                             N2
                         lbc/mole DFG =        2 x 12                  (C-3)
(3)   The nitrogen-to-carbon ratio in the flue gas is computed as the
     ratio of the nitrogen and carbon levels in the dry flue gas.


                                   N2  _ Equation (C-2)               (C-4)
                                 lb,.     Equation (C-3)
                                   L.
                                    148

-------
(4)   The nitrogen-to-carbon ratio in the flue gas is  then multiplied by
     the carbon-to-fuel ratio computed by Equation (C-5)  and the air-to-
     nitrogen ratio to determine the air-to-fuel ratio.
                 Ib         mp                 np
                 TT^	 = 	T~  (%C )   +  	TZ—   (%CJ           (C-5)
                 Ib,.  ,    mp +np,     c      mp +np,      d
                   fuel   ^c   d              c   d
     where m is the volume fraction coal and n is the volume fraction
     dRDF; %C and p are the as-received carbon content and the bulk density
     of the respective fuels c(coal) and d(dRDF).

                    Ib .
                      air
                    -LU .,  ..
                      fuel


                      Note:       —  is the constant 1.30


(5)   The wet flue gas is then determined by recognizing that the total
     amount of flue gas must be equal to the fraction of the input fuel
     burned plus the pounds  of air added to the fuel.
               Ib of WFG    _ / lbfuel  _  lbash \    lbair            (C-7)
               Ib of fuel     \ Ib.  ,     Ib,  -,  J    Ib.  ,
                                  fuel       fuel/      fuel

(6)   The wet flue gas is then converted into a volumetric flux by multiplying
     by the standard flue gas density corrected to stack conditions.  Note
     that the actual density could be computed, but the error introduced by
     assuming standard combustion products is of a lower order than the
     sampling error of the composition of the fuel submitted to the labora-
     tory for ultimate analyses.


                    cu ft of WFG @ T        .„„   460+T
                    	s  =    WFG   	s              (  _.
                       Ib of fuel          0.071   560                v   '
        Note:   .071 is assumed to be the density of flue gas at 560°R,
               and T  is the stack temperature in °F.
                                     149

-------
          This result is then converted into an average gas velocity by
     recognizing that the gas flux will be the cubic feet of gas per pound of
     fuel burned multiplied by the rate of fuel consumption in the boiler
     with this product then divided by the cross-sectional area of the stack.
     These relationships [expressed in Equation (C9)] assume that the velocity
     flux is to be computed over the same heat balance period as the boiler
     efficiency test.  The ratio of pounds of blend divided by run time can
     be replaced by the amount of fuel needed to produce the steam generation
     rate (as read off the strip chart recorder) once the boiler efficiency
     is known for a given excess air level.


              U =  (cu ft of WFG/lb fuel) (Ib fuel/sec)  =  ft/sec    (c_9)

                                  f  °2


     where D is the stack diameter, and U is the gas velocity.

(7)   The calculated gas velocity can be readily converted into a velocity
     head (feet of air)  by employing a rearrangement of Bernoulli's equation:
                                Ah = .       ,                        (C-10)
                                     2g (Pm-p)


     where  p  is the gas density under stack conditions (Ib/cu ft),  p  is the
     density  of the manometer fluid (Ib/cu ft) ,  and g is the gravitational
     constant (ft/sec2).

         Equation (C-10)  is the velocity head as would be measured by a
     standard pitot tube.   As a result, this Ah  needs to be multiplied by a
     correction factor  («0.91) which relates standard pitot tube results to
     the measurements of an S-type pitot probe and by a constant  to convert
     from feet of  air head to inches of water.  This calculated apparent
     stack  velocity is  then used to calculate an isokenetic flue  gas
     sampling rate.
                                    150

-------
                APPENDIX D
           CASCADE IMPACTOR DATA
TABLE D-l.   DECEMBER CASCADE IMPACTOR RAW DATA
DKCF.MBF.K
Total Klapsed Impactor Assumed Part Impactor Stack
Flow Time Flow Rate Density Temp. Temp.
(CF) (Mill) (CFM) (t;/cc)' ('F) ("F)
1:0-1 14.47 30 2.6 1.0
Date: 12/6/77

1:0-2 A 1.02 60 2.6 1,0
Date: 12/6/77

1:0-3 15.91 30 .57 1.0
Date: 12/10/77

1:1-1 34.36 30 .58 1.0
Date: 12/8/76

1:1-2 28.26 30 .59 1.0
Date: 12/13/76

1:1-3 30.77 30 .59 1.0
Date: 12/13/76

345 345 D,0
mg
CUM 7,
345 345 D.,o
mg
CUM X
370 370 D,0
mg
CUM ?„
390 390 D50
mg
CUM 7,
390 390 D,0
mg
CUM 7.
380 380 D. o
mg
CUM ',',
1
J6
0
0
16
0
0
34
11.9
14.44
31.0
0
0
31.0
7.0
4.14
31
0
0
2
7.
0
0
7.
4.
7.
17
7.
24.
16.
A.
3.
16.
A.
6.
16
12.
9.
7


7
2
18

9
03
0
6
33
0
7
91

6
87
3
2.8
9.6
12.53
2.8
9.3
23.08
5.7
8.1
33.86
5.7
40.0
32.32
5.7
31.-.
25.47
5.7
34.8
37.15
4
1.
7,
21,
1.
7,
36,
2
12.
48.
2,
30.
54.
2.
32,
44.
2.
22.
54.
.3
.1
.80
3
.9
.58
7
.2
,67
,7
.1
13
7
3
56
7
5
78
5

48.
84.

12.
58.
1.
2.
52.
1.
0.
54.
1.
15.
53.
1.
16.
67.

7
0
46
7
8
46
5
9
18
9
3
35
4
7
84
4
2
A8

0
84
-

8
72

8
62

22
70

40
77

3
70
6

.46
-
.3
.1
.31
.6
.5
.50
.7
.2
.43
.7
.6
.84
.7
.3
.06
7

0
8A.46
...
O.A
72.99
j,
11.9
76.94
.4
13.5
80.22
.A
1.7
78. 8A
.A
22/6
87.77
Filter

11.9
100

15.8
100

19.0
100

27.3
100

35.8


15.6
100

-------
                           TABLE  D-2.   JANUARY CASCADE IMPACTOR RAW DATA
  JANUARY
  1:U-1
Dace: 1/20/77
   1:0-2
Date: 1/21/77
   1:0-3
Date: 1/21/77
   1:0-5
Date: 1/24/77
Total Elapsed Tmpactor
Flow Time I1' low Rate
(CF) (Min) (CFM)
19.92 30 .763


19.1 30 .feW


22.53 30 .54

'
13.95 30 .441


Assumed Part Impactor Stack
Density Temp. Temp.
(K/ec) no CF)
1.0 '390 390 D.,tJ
me
CUM 7,
l.Q 375 375 D.,,,
mg
CUM 7,
1.0 375 375 D™
mg
CUM 7
1.0 400 400 0,j0
mg
CUM '/,

1
30
2.0
2.54
31
0
0
33
0
0
37
0
0

2
13.
6.
10.
15
4.
12.
16
3.
6.
18
0
0



3
5
4
69

9
37

7
48



4.
28.
4h.
5.
11.
40,
5,
15,
32,
6,
6
16
9
3
69
6
1
,40
.9
.1
.92
.5
.1
.53


4
2
33
89
2
19
90
2
9
48
3
15
58
.25
.7
.57
.55
.9
.66
.7
.0
.69
.0
.6
.81


5
1.
3,
93.
1.
3,
98,
1
13,
71,
1
5
72
25
.4
89
45
,2
,74
.53
.0
.45
.7
.2
.90


6

0
93

0
98

4
79

2
79
.56

.89
.65

.74
.69
* 5
.33
.77
.4
.40



0
93.

0
98,

0
79,

0
79


35

89
.42

,74
.44

.33
.51

.40


Filter

4.
100

0,
100

11
100

7
100

,8


.5


.8


.6


-------
TABLE D-3.  MARCH CASCADE IMPACTOR RAW DATA
MARCH
Total Elapsed Impactor Assumed Part Impactor Stack
Flow Time Flow Rate Density Temp. Temp.
(CF) (Mln) (CFM) (g/cc) (°F) (°F)
1:0-1 11.33 30 .61 1.0
Date: 3/19/77

1:0-2 7.16 20 .58 1.0
Date: 3/19/77

1:0-3 10.44 30 .58 1.0
Date: 3/21/77
1:0-4 8.13 20 .67 1.0
Date: 3/21/77
1:0-5 7.62 20 .62 1.0
Date: 3/22/77
1:0-6 10.29 30 .54 1.0
Date: 3/28/77
1:0-7 7.01 20 .54 1.0
Date: 3/31/77
1:0-8 5.54 15 .60 1.0
Date: 4/1/77
1:0-9 5.22 15 .55 1.0
Date: 4/1/77
390 390 D 50
mg
CUM %
390 390 D,o
rag
CUM %
415 415 D,o
mg
CUM 7.
415 415 D50
mg
CUM 7.
370 370 D,o
mg
CUM %
370 370 D5o
mg
CUM %
360 360 D5o
mg
CUM %
390 390 D,o
mg
CUM X
390 390 D,o
mg
CUM
1
31
0.9
1.43
32
13.1
25.10
32
0.9
1.45
29.5
1.0
1.94
31
1.6
4.18
33
3.6
6.01
33
2.3
5.18
31
1.2
3.00
33
0.8
2.33
2
15
3.9
7.61
15.5
3.7
32.18
15.5
1.8
4.36
14.5
1.3
4.46
15
2.9
11.75
16.5
2.9
10.85
16.5
4.3
14.86
15
2.5
9.25
16
1.6
6.98
3
5.5
20.7
40.41
5.7
11.9
54.98
5.7
19.5
35.86
5,3
17.3
37.98
5.4
12.7
44.91
5.9
18.4
41.57
5.9
5.4
27,03
5.6
12.6
40.75
5.8
9.8
35.47

2
15
64
2
9
73
2
15
61
2
9
57
2
9
68
2
11
60
2
6
41
2
8
61
2
7
57
4
.5
.4
.82
.7
.9
.95
.7
.7
.23
.5
.9
.17
.4
.2
.93
.8
.2
.27
.8
.5
.67
.55
.1
.00
.75
.5
.27
5
1.
9.
79.
1.
5.
84.
1.
8.
74.
1.
7.
70.
1.
4.
81.
1.
8.
75.
1.
14.
73.
1.
6.
76.
1.
6.
74.
4
5
87
45
4
29
45
4
80
4
1
93
3
7
20
55
9
13
55
3
87
40
3
75
50
0
71
6
.64
3.8
85.90
.67
2.6
89.27
.67
2.2
78.35
.61
3.1
76.94
.64
1.9
86.16
.70
4.2
82.14
.70
2.2
78.83
.66
2.1
82.00
.69
2.6
82.27
7
.41
1.5
88.27
.34
0.7
90.61
.43
1.6
80.94
.40
5.8
88.18
.42
0.8
88.25
.45
2.6
86.48
.45
1.9
83.11
.42
1.1
84.75
.44
0.6
84.01
Filter

7.4
100

4.9
100

11.8
100

6.1
100

4.5
100

8.1
100

7.5
100

6.1
100
5.5
100
                                          (continued)

-------
TABLE D-3.  (continued)
MARCH
Total Elapsed Impactor Assumed Part. Impactor Stack
Flow Time Flow Rate Density
(CF) (Min) (CFM) (g/cc)
1:1-1 11.18 30 .62 1.0
Date: 3/23/77

1:1-2 8.76 30 .45 1.0
Date: 3/23/77

1:1-3 12.43 30 .68 1.0
Date: 3/24/77

1:1-1 8.54 25 .53 1.0
Date: 3/29/77

1:1-2 6.77 20 .51 1.0
Date: 3/29/77

1:2-3 6.3 20 .48 1.0

BLANK 10.71 30 .60 1.0
Date: 3/16/77

Temp . Temp
(°F) <°F)
385 385 D,0
mg
CUM %
400 400 D3o
mg
CUM %
390 390 D50
mg
CUM %
380 380 D50
mg
CUM %
390 390 D,o
mg
CUM Z
380 380 D,o
mg
CUM X
400 400 D50
mg
CUM X

1
31
0.
0.
37
0.
0.
31
4.
6.
33
0.
1.
34
0.
0.
35
3.

0.




3
68

2
44

7
89

9
87

2
48

39

6


2
15
1.5
4.09
18
1.1
2.86
16
2.2
10.12
16.5
0.9
3.74
17
0.7
2.18
17.5
7.5

0.1


3
5.4
15.4
39.09
6.7
8.9
22.47
5.6
13.9
30.50
5.9
8.9
22.25
6.0
10.1
26.48
6.4
28.72

0.5


4
2.
10.
62.
3.
9.
42.
2.
9.
44.
2.
7.
38.
2.
7.
44.
2.
48.

0.



5
4
73
0
1
51
5
4
28
75
7
25
75
5
58
7
30

3


5
1.30
4.4
72.73
1.70
4.6
52.64
1.30
3.5
49.41
1.55
2.6
43.66
1.60
3.6
53.28
1.60
62.92

0.7


6
.65
2.8
79.09
.77
2.4
57.93
.65
1.1
51.03
.70
3.9
51.77
.70
2.9
60.28
.74
66.32

0.7


7
.43
4.6
89.55
.50
4.4
67.62
.42
11.3
67.60
.45
7.2
66.74
.47
3.8
69.48
.49
70.23

0.3


Filter

4.6
100

14.7
100

22.1
100

16.0
100

12.7
100
11.4
100


3.1


-------
                                TABLE D-4.  MAY CASCADE IMPACTOR RAW DATA
Ul
MAY
1:0-1
Date: 5/3/77

1:0-2
Date: 5/4/77

1:0-3
Date: 5/5/77

1:0-4
Date: 5/5/77

1:0-5
Date: 5/16/77

1:0-6
Date: 5/16/77

BLANK 1-0
Date: 5/4/77

1:1-1
Date: 5/12/77

1:1-2
Date: 5/12/77

1:1-3
Date: 5/13/77

Total
Flow
(CF)
10.84


7.18


5.3


6.84


3.6


7.03


4.46


8.96


7.95


6.36


Elapsed
Time
(Min)
30


20


15


20


10


15


15


20


20


15


Impact or
Flow Rate
(CFM)
.56


.57


.62


.56


.60


.80


.45


.77


.66


.70


Assumed Part. Impact or Stack
Density Temp. Temp.
(g/cc) (°F) (°F)
1.0 360 360


1.0 360 360


1.0 420 420


1.0 400 400


1.0 380 380


1.0 435 435


1.0 360 360


1.0 415 415


1.0 400 400


1.0 400 400



DBO
mg
CUM X
Dso
mg
CUM X
D,0
mg
CUM X
D,o
mg
CUM X
D,o
mg
CUM X
D50
mg
CUM X
D,o
mg
CUM X
D,0
mg
CUM X
D,o
mg
CUM X
D,o
mg
CUM X
1
32.0
0.7
1.17
32.0
0.7
1.78
31
0.8
1.47
32
0.9
2.07
31
0.2
1.27
27
0.0
0

0.3

28
0.6
1.19
30
0.2
0.47
29
0.0
0
2
15.5
1.8
1.17
15.5
1.1
4.58
15
1.2
3.68
15.5
0.8
3.91
15.5
0.6
5.10
13
1.7
2.94

0.4

14
1.8
4.77
10.4
1.3
3.55
14
2.0
6.13
3
5. BO
17.4
33.22
5.80
9.6
29.01
5.5
14.3
29.96
5.8
11.9
31.26
5.7
1.4
14.01
4.8
25.7
47.40

1.0

4.9
16.8
38.17
5.4
12.3
32.70
5.2
9.9
36.50
4
2.60
14.7
57.76
2.60
7.5
48.09
2.5
14.0
55.70
2.6
12.9
60.92
2.6
4.4
42.04
2.2
12.9
69.72

0.5

2.2
10.5
59.05
2.4
9.9
56.16
2.3
7.4
59.20
5
1.50
9.6
73.79
1.50
4.1
58.53
1.4
10.0
74.08
1.5
7.4
77.93
1.45
3.2
62.42
1.2
4.1
76.82

0.5

1.2
4.1
67.20
1.35
4.7
67.30
1.3
1.8
64.72
6
0.69
3.9
80.30
0.69
3.5
67.43
.62
6.4
85.84
.69
1.9
82.30
.65
3.5
84.71
.54
3.0
82.01

0

.55
2.7
72.56
.60
2.7
73.70
.50
1.3
68.71
7 Filter
0.47 0.47
2.9 8.9
85.14 100
0.47
3.1 9.7
75.32 100
.40
1.7 6.0
88.97 100
.47
1.5 6.2
85.75 100
.42
1.6 0.8
94.90 100
.35
3.3 7.1
87.72 100

0.2 0.3

.35
3.6 10.2
79.72 100
.39
2.0 9.1
78.44 100
.37
2.5 7.7
76.38 100
                                                                           (continued)

-------
                                  TABLE D-4.   (continued)
Wl
MAY


BLANK
Date:

1:2-1
Date:

1:2-2
Total Elapsed Impactor Assumed Part. Impactor Stack
Flow Time Flow Rate Density
(CF) (Min) (CFM) (g/cc)
1-1 6.29 15 .69 1.0
5/13/77

7.33 20 .63 1.0
5/10/77

7.34 30 .63 1.0
Date 5/10/77

1:2-3
Date:

BLANK
Date:

0:1-1
Date:

0:1-2
Date:


8.68 20 .73 1.0
5/11/77

1-2 8.53 20 .72 1.0
5/11/77

8.0 20 .66 1.0
5/14/77

11.9 30 .65 1.0
5/14/77

Terap . Temp .
(°F) (°F>
400 400 D
mg
CUM %
420 420 D
mg
CUM %
425 425 D
mg
CUM %
410 410 D
mg
CUM %
420 420 D
mg
CUM %
380 380 D
mg
CUM %
390 390 D
mg
CUM %

1

0.4

30
0.8
1.63
30
1.2
2.51
28
1.2
2.01

0.5

29
0.0
0
29
0.4
0.45

2

0.1

15
1.6
4.89
15
1.2
5.02
14
1.7
4.87

0.3

14.5
0.5
1.17
14.5
1.8
2.49

3

0.6

5.4
11.6
28.51
5.4
2.4
10.04
5
20.1
38.59

1.1

5.2
0.0
1.17
5.2
18.9
23.87

4

0.

2.
8.
46.
2.
12.
35.
2.
13.
60.

1.

2.
7.
18.
2.
13.
39.



6

5
9
64
5
2
56
3
3
91

1

4
5
69
4
8
48

5

1.3

1.4
3.1
52.95
1.4
16.5
70.08
1.3
5.0
69.30

0.8

1.3
2.4
24.30
1.3
3,9
43.89

6

0.7

.62
2.8
58.66
.62
1.2
72.59
.57
2.4
73.32

0.7

.62
1.7
28.27
.62
4.2
48.64

7

0.6

.4
6.1
71.08
.4
2.1
76.99
.37
3.2
78.69

0.5

.39
7.1
44.86
.39
13.0
63.35

Filter

0.7


14.2
100

11.0
100

12.7
100

0


23.6
100

32.4
100

-------
                                  APPENDIX E

                    DISCUSSION OF MONSANTO'S ESP TEST DATA
     SYSTECH analyzed Monsanto's ESP test data to determine how the fly ash
from coal-only and 1:1 and 1:2 blend tests affected the precipitator
performance.  Figure E-l plots the results of this analysis with inverse
penetration as a function of the specific collector area, the applied voltage,
and the square root of the current.  These parameters were chosen as axes
because the Deutch Equation (E-l) indicates that these axes should yield a
straight line on semilog paper as demonstrated in the following.
                                         - e -                        (E-l)
where ri = probability of particle capture
      P = penetration
      A = collector electrode area ft2
      Q = gas glow CFM
      W = migration velocity
     Further reduction of Equation (E-l) yields the following:


                                 In P = - ^                          (E-2)

               and

                                 In ^ = ^                            (E-3)
     The migration velocity is a function of electric field strength (E ) and
particle charge (q).


                                   W a q E                             (E-4)


     For a uniform field, the field strength (E ) is the applied voltage  (V)
to the collector electrodes divided by the gap between the electrodes.  Also,
the charge per particle  (q) is a function of the electron cloud density which
is proportional to the square root of the current flow (i) for collision
                                     157

-------
100
 90
 80
 70
 60
 50

 40


 30
20
10
 9
 8
 7
 6

 5
                                          BLEND  SYMBOL

                                            1:0     •

                                            1:1     A

                                            1:2     •
            I
                               I
                                                  I
          10,000    20,000    30,000    40,000
                          (SCA) kV VT
                           50,000
      Figure E-l.
Precipitability of blend fly ash component to
coal-only conditions.
                               158

-------
charging.  Therefore, substituting these relationships into Equation (E-4)
yields the following equation:
                                  W a V  y/i
(E-5)
     Since A/Q is commonly referred to as SCA, Equation (E-3) can now be
expressed as:


                                  1         /	                         /
                               In ^ a SCA V,/i                        (E-6)
     Table E-l summarizes all of the Monsanto data, and Figure E-l presents
a logarithmic plot of 1/P versus SCA for the various blends.  If the precipi-
tator performs normally, regardless of the variation in these parameters, the
results should plot as a straight line.

     However, the coal-only ash had resistivities too low for proper
precipitation.  The 1:1 and the 1:2 ash were slightly less and slightly more
precipitable, respectively, than the coal-only ash.

     The difference between the resistivity for the first 1:2 test data and
the resistivity for the rest of the data is probably real.  After the
coal-only runs were completed, the precipitator was thoroughly cleaned.
Consequently, most of the power applied to the cells for this first 1:2 test
probably passed through the air gap.  During the 1:1 test, cell after cell
was taken off line because of the short-circuiting over the dielectric
bridge.  Consequently, while the applied voltage or field was correct, a
significant amount of current probably leaked through the insulating hangers
to the grounded electrodes.  If it is assumed that half of the current
bypassed the air gap, then the aerosol data for the 1:1 test aligns with the
aerosol data for the 1:2 test.  The collection data for the 1:1 test, there-
fore, has questionable validity.

     Because the first 1:2 run had more fly ash precipitation than the coal-
only run, it is likely that substituting dRDF for coal will not seriously
degrade the precipitator performance.  This conclusion, however, must be
verified by further testing.
                                     159

-------
TABLE E-l.  FIELD PORTABLE ELECTROSTATIC PRECIPITATOR DATA
DATE



5/5/77

1:0
coal
only





5/10/77

1:2
Blend






FLOW
(ACFM)


1500









1500









SCA
(Ft.2/KCFM)


320









320









Temp.
(°F)


480









470









AVERAGE
ELEC. COND.
CELL VOLT C.D.
(KV) (UA/Ft?)
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
27
34
32
32
32





32
36
34
32
36





15
30
22
16
30





4
12
15
10
42





INLET COND.
(gr/SCF) MEAS.






Avg.









Avg.





0.16

0.19

0.175





0.251
0.
0.201

0.226





OUTLET COND,
MEASURED
(gr/SCF)

0.0012

0.0082

0.0047





0.0040

0.00047

0.00226





EFFICIENCY
(%)


99.25

95.7

97.3





98.4

99.8

99.0





                             (continued)

-------
TABLE E-l.  (continued)
DATE



5/11

1« o
• f.
Blend






5/12

1:1
Blend






FLOW
(ACFM)


3000









1500









SCA
(Ft.2/KCFM)


160









320









Temp.
(°F)


530









470









AVERAGE
ELEC. COND.
CELL VOLT C.D.
(KV) (UA/Ft.2)
1
2

3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
32
37

30
38
37





23
25
24
30
32





5
7

0.8
12
17





10
15
'9
8
15





INLET COND.
(gr/SCF) MEAS.






AVG.









AVG.





0.15
0.098


0.124





0.20

0.18

0.19





OUTLET COND.
MEASURED
(gr/SCF)

.0178
.0114


0.0146





0.059

0.030

0.0445





EFFICIENCY
(%)


88.1
88.3


88.2





70.6

83.3

76.6





         (continued)

-------
TABLE E-l.  (continued)

DATE


5/13

1:1
Blend






5/16

(Note:
1:0 coal
only
5/16

(Note:


FLOW
(ACFM)

3000









1200

Operating


1200 ^

Operating


SCA
(Ft.2/KCFM)

160









320

4 cells)


240

3 cells)


Temp.
(°F)

500









450




450



AVERAGE
ELEC. COND.
CELL VOLT C.D.
(KV) (UA/Ft.2)
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
24
25
22
23
42





27
33
33
28
.»
26
32
32
—
12
10
8
8
20





10
20
28
18
— —
11
14
14
__
INLET COND.
(gr/SCF) MEAS.






AVG.














0.12
0.23


0.175





0.164




0.127



OUTLET COND.
MEASURED
(gr/SCF)

0.0425
0.0331


0.0378





0.00605




0.0103



EFFICIENCY
(%)


64.6
85.6


78.4





96.3




91.9




-------
                                                 APPENDIX F
                                    HEAVY METALS EMISSIONS DATA SUMMARIES
                     TABLE F-l.  AA ANALYSIS  OF AEROSOL  CAPTURE  FOR COAL-ONLY  (1:0)  FIRING
OJ

Part
4-34
4-35
4-37
4-3H
4-39
4-40
4-41
4-42
3-42
3-43
3-44
3-45
3-46
3-47
3-48
Hg/m3
Filter
3/19
3/19
3/21
3/22
3/28
3/31
3/31
4/1
5/3
5/4
5/5
5/5
5/16
5/16
5/17
Probe Wash
4-51 3/19
4-53
4-54
4-55
4-56
4-57
4-58
3-38
3/22
3/22
3/28
3/31
3/31
4/1
Composite
May
Pb

156
254
147
758
169
190
190
200.
184
165
222
369
144
517
116
28.1

13.8
18.7
32.3
27.4
17.8
16.1
Cd

2.60
4.54
2.62
2.43
<3.38
7.15
2.38
3.73
2.30
2.35
3.13
5.76
2.41
8.01
2.00
4.02
1.32
2.07
1.40
1.59
.697
1.02
2.15
As

168
439
237.
115.
143
65.6
87.3
187.
196.
162
195.
199
76.8
151
71.2
184.5
44.2
45.3
38.5
23.7
13.1
11.2
16.1
Hg Cr

<0.69 29.8
27.3
36.7
36.4
<0.56 22.5
<1.03 7.95
<0.99 11.9
58.6
69.0
20.6
287.
60.2
12.0
32.8
46.0
44.7
25.7
82.5
409.
74.7
38.3
44.1
134.2
Ni

24.2
18.4
34.0
30.3
28.2
15.9
23.8
40.0
65.2
35.3
36.5
60.2
11.4
34.3
35.6
5.43
31.2
88.9
5253.
67.9
44.8
37.1
183.
Mn

26.0
46.3
59.7
41.9
37.2
21.5
22.2
59.2
29.9
24.7
20.4
31.9
27.5
28.8
10.4
17.5
9.99

62.1
13.6
8.07
10.2
—
Zn

474.
613.
445.
531
507
397
337
546
115
<58.9
679.
1047.
337
713 .
286.
101.
53.8
346
105
70.8
47.3
42.4
56.4
Cu

33.5
45.4
41.9
42.5
<56.3
<39.7
<39.7
<37.3
38.3
35.3
52.3
68.0
24.1
51.2
24.9
12.1
7.05
7.90
21.0
5.66
<4.98
3,96
6.4
Sn

1.86
1.82
.524
1.21
1.13
<.79
<.79
1.33
.767
.883
1.57
3.14
.722
1.19
<.498

.277
.296
.467
.509
.300

.107
Sb

<37.2
<90.9
<32.4
<60.7
<113.
< 79.5
< 79.3
< 53.3
< 76.7
< 58.9
< 52.2
< 52.3
< 48.2
< 32.0
< 49.5
< 10.1
< 6.94
< 9.86
< 23.3
< 11.3
< 9.96
< 5.66
< 1.08
Ag

<3.72
<9.09
<5.24
<6.07
<11.3
<7.9
<7.93
<5.33
<7.67
<5.89
<5.22
<5.23
<4.82
<3.20
<5.0
<13.7
<9.71
< .988
<2.33
<1.13
< .996
<1.13
< .537
Vn

37.2
90.9
<52.4
<60.6
<113
<79.5
<79.3
<53.3
<76.7
<58.9
52.3
52.3
<48.2
37.0
<50.0
10.1
<7.05
<9.9
<23.3
<11.3
<9.96
<5.66
<.5.37

-------
               TABLE F-2.   AA ANALYSIS OF AEROSOL CAPTURE  OF 1:1 BLEND FIRING
ug/m
       1'b
               Cd
                       As
Part.
4-43
4-44
4-45
3-49
3-50
3-51
Probe
4-59
4-60
4-61
j-39
Filter
3/23
3/23
3/24
5/12
5/12
5/13
Wash
3/23
3/23
3/24
Composite
May

2705.
4638
3492
5609
2847
2438

48.1
112.
1Z2.
108

69.
84.
62.
74.
64.
47.

2.
1.
2
2

1
3
8
5
0
6

,09
.68
.34
.16

55.1
46.7
23.4
159.
165.
64.5





                               Hg
                              <0.65
                              <0.54

                              <0.68
                              <0.50
Cr
25.0
37.1
29.1
33.5
28.9
29.5
34.5
11.2
13.2
13.5
Ni
27.6
34.5
25.2
37.7
31.1
23.6
35.6
11.2
22.0
116.
Mi
43
78
54
62
64
33
18
6
11
10
i
.6
.0
.6
.8
.0
.0
.4
.14
.0
.2
Zn
6011
5994
4341
5441
4892
4325
207
196
173
132.
Cu
75.1
101.
85.3
83.7
66.7
62.9
5.23
<27.9
<14.6
7.73
Sa Sb
2.50 <30.1
4.77 <53.0
1.94 <38.8
3.35 <41.9
2.67 <44.5
.786 <39.3
.418 <10.5
<.559 <55.9
<29.3
.232 <7.73
Ag
10.
12.
9.
8.
<4.
<3.
2.
<5.
<2.
<.

5
7
30
37
45
93
09
59
93
773

<50
<53
<38
<41
<44
<39
<10
<55
<29
<7
Vn
.1
.0
.8
.8
.5
.3
.5
.9
.3
.73

-------
                         TABLE  F-3.   AA ANALYSIS OF AEROSOL CAPTURE FOR 1:2 BLEND FIRING
(J\
>J
Part.
4-46
4-47
4-48
3-52
3-53
3.54
Probe
4-62
4-63
4-64
3-40
.«/.'
Filter
3/29
3/29
3/30
5/10
5/10
5/11
Wash
3/29
3/29.
3/30
Comp
May
Pb

10,140.
5,151.
3,621.
6,345.
5,595.
7,893.

337
218
249
257
Cd

269.
216.
94.0
162
275
90.5

6.59
5.41
4.04
5.82
As

60.6
23.3
36.5
103
105
95.5

12.1
12.2
4.7
20.4
Hg Cr

38.6
<0.65 43.7
<0.90 34.8
61.3
<0.80 39.6
35.2

161
79.6
49.4
68.0
Ni

35.8
14.6
48.7
78.3
29.7
20.1

130.
78.1
67.4
113.
Mn

72.7
76.3
98.2
96.4
107.
75.4

31.2
60.4
29.7
20.6
Zn

11020
3686
6264
7191
6435
6539

295.
263.
371.
318.
Cu

110
36.9
55.7
140.
94.1
95.5

<17.4
<15.0
<22.5
12.1
Sn

2.2
2.13
7.66
2.33
4.46
1.51

.347
2.11
1.35
.605
Sb

<55.1
<89.9
<69.6
<42.3
<49.5
<50.3

<34.7
<30.1
<44.9
<7.56
Ag

23.7
11.7
6.96
19.5
13.9
14.1

< 3.47
<3.01
<4.49
<.756
Vn

<55.1
<19.4
<69.6
<42.3
<49.5
<50.3

<34.7
<30.1
<44.9
<7.56

-------
TABLE F-A.  AA ANALYSIS OF AEROSOL CAPTURE FOR 0:1 BLEND FIRING
Pb Cd As Hg Cr Ni Mn Zn Cu Sn Sb Ag Vn
Part. Filter
3-55 5/14
3-56 5/14
Probe Wash
Composite
May

9557 240 39.5 <1.06 60.8 26.1 200. >821 165. 3.48 <86.9 24.3 =86.9
6351 187 39.6 66.9 20.9 241 5016 159. 6.27 <83.6 23.4 <83.6

326 15.8 11.8 34.0 31.3 — 605 21.8 .816 <27.2 <2.7 ( <27.2


-------
              TABLE F-5.   AA ANALYSIS OF ASH SAMPLES FROM 1:0 FIRING

B.A.
4-66
3-111
3-105
F.A.
3-107
3-84
3-99
F.A.
4-70
4-68
3-83


3/28
5/3
5/17
Coll #1
5/4
5/5
5/17
Relnj «
IfD 3/28
3/31
5/3
Pb
mg/kg
<12.5
<12.5
<12.5
30.0
17.5
15.0

134.2 	 -
16.3
20.0
Cd
mg/kg
<.75
<.75
<.75
<.75
<.75
<.75

2.87
<.75
<.75
As
mg/kg
11.0
17.9
14.3
102.4
77.0
49.5

- 625.
34.4
48.1
Hg
mg/kg
<0.4
<0.4
<0.4
<0.4
<0.4
<0.4

'0 '<
<0.4
<0.4
Cr
mg/kg
22.5
10.0
7.5
20.0
15.0
18.1

25 0
8.75
21.3
Hi
mg/kg
20.0
23.7
16.2
37.5
25.0
25.0

25 6
15.0
26.2
Mn
mg/kg
51.0
40.5
40.5
69.7
42.0
37.5

300
105
45.0
Zn
mg/kg
31.2
37.5
100.
50.0
81.2
50.0

inc.
31.2
50.0
Cu Sn
mg/kg mg/kg
15.0 .50
25.0
<12.5 .50
20.0 2.38
12.5 2.13
12.5 1.75

27 5 2 37
<12.5 1.88
15.0 1.0
Sb
mg/kg
<25
<25
<25
<25
<25
<25

^25
<25
<25
Ag
mg/kg
<2.5
5.75
120.
<2.5
<2.5
<2.5


<2.5
<2.5
Vn
mg/kg
<25
<25
<25
25.0
25.0
25.0


<25
25.0
Notes:  B.A. denotes bottom ash and F.A. denotes fly ash.

-------
                               TABLE F-6.   AA  ANALYSIS OF ASH SAMPLES FROM  1:1  FIRING
oo
B.A.
3-102
3-103
4-67
F.A.
3-85
3-91
3-92
3-93
F.A.
4-72
4-71
3-94

5/12
5/13
3/23
Coll #1
5/9
5/12
5/12
5/13
Reinj. #2
3/23
3/24
5/13
Pb
mg/kg
36.3
56.3
26.3
148.
250.
300.
168.
97.5
97.5
92.5
Cd
rag /kg
.75
<.75
<.80
1.75
4.7
5.37
2.5
1.25
1.0
1.0
As Hg
mg/kg mg/kg
40.0 <0.4
30.0 <0.4
39.3 :0.4
50.0 <0.4
100. <0.75
124. <0.5
47.5 0.53
38.5 0.68
39.0 *r6*
49.5 0.58
Cr
mg/kg
25.0
22.5
25.6
20.6
22.5
23.1
12.5
13.8
11.3
21.3
Ni
mg/kg
40.0
26.2
27.4
25.0
35.0
35.0
21.2
17.5
16.2
23.7
Mn
mg/kg
185.
155.
138.
101
197.
196.
86.3
161
225
120
Zn
mg/kg
150.
75.0
73.3
197
431
512
231
125.
144.
156.
Cu
mg/kg
218.
225.
152.
-15.0
40.0
36.3
20.0
15.0
15.0
22.5
Sn
mg/kg
2.63
4.0
3.58
2.13
4.38
5.3
2.5
2.0
3.0
3.0
Sb
mg/kg
<25
<25
<25
<25
<25
<25
<25
<25
<25
<25
Ag
mg/kg
<2.5
<2.5
<2.5
<2.5
<2.5
<2.5
<2.5
<2.5
<2.5
<2.5
Vn
mg/kg
<25
<25
<25.5
-25
35
30
<25
25.0
<25
25.0
                    Notes:  B.A.  denotes bottom ash,  F.A. denotes fly ash, and £ represents that  the number reported is an
                           average of two in which one was below the detection  limit and the other was just above.

-------
               TABLE F-7.   AA ANALYSIS OF  ASH SAMPLES FROM 1:2 FIRING
BA
4-65
3-100
3-101
F.A.
3-86
3-89
3-90
F.A.
4-69
3-87
3-88

3/30
5/10
5/11
Coll #1
5/10
5/11
5/11
Relnj «2
3/29
5/10
5/11
Pb
mg/kg
128
47.5
82.5
275
246
300
109
132
198
Cd
mg/kg
.75
.75
.75
5.75
6.25
6.0
1.5
1.3
2.5
As
rag /kg
36.7
30.2
27.5
49.5
37.5
66
32.4
37.5
Hg
mg/kg
<0.4
<0.4
<0.4
<0.4
<0.4
2.3
<0.4
<_0.43
<0.4
Cr
mg/kg
42.5
18.8
22.5
26.1
38.7
40.
15.0
28.4
18.7
Ni
mg/kg
139
16.3
27.8
22.0
37.5
36.2
18.7
20.2
26.2
Mn
mg/kg
250
135
135
317.
379.
288.
300
146.
250
Zn
mg/kg
188.
81.2
65.6
598
619.
606
194
203
381.
Cu
mg/kg
200
186
29.7
50.0
40.
17.5
24.3
60
Sn
mg/kg
5.0
3.0
4.25
1.08
2.69
1.13
2.25
3.0
Sb
mg/kg
<25
<25
<25
<27.5
<25
<25
<25
<27
<25
Ag
mg/kg
<2.5
<2.5
<2.5
<2.5
<2.5
<2.5
<2.5
<2.7
<2.5
Vn
mg/kg
<25
<25
<25
27
30
25
25.
<27
25.
Noces:  B.A.  denotes bottom ash, F.A. denotes fly ash, and £ represents that the number reported is  an
       average of two  in which one was  below the detection limit  and the other was just above.

-------
                     TABLE  F-8.   AA  ANALYSIS OF ASH  SAMPLES FROM 0:1  FIRING
B.A.
3-104 5/14
Pb
mg/kg
169.
Cd
mg/kg
2.125
As
rag /kg
56.3
Hg
mg/kg
£0.4
Cr
mg/kg
50.0
Ni
mg/kg
35.0
Mn
mg/kg
43.5
Zn
mg/kg
539.
Cu
mg/kg
205
Sn
mg/kg
3.6
Sb
rag /kg
<25
Ag
mg/kg
88.8
Vn
mg/kg
<25
F.A.  Coll  #1
3-96    5/14
3-97    5/14
 925.
1100.
26.0
24.8
122.
 85.3
£0.5
 0.61
176.
193.
81.9
68.8
130.
117.
                                                                                                176.
                                                                                                127.
                                                                                     11.3
                                                                                      4.86
                                                                          <25.
                                                                          <27.5
                                                                           12.7
                                                                           14.8
                                                                            40.0
                                                                            43.2
F.A.  Reinj #2
3-95    5/14
3-98    5/14
350      8,50
375      7.25
52.5
40.0
£0.48
<0.4
65.0
73.7
                                                                    45.0
                                                                    36.9
                                                          82.3
                                                          80.3
                                                         1138.
                                                         1097.
                                                           95.0
                                                           81.7
                                                          4.38
                                                          5.3
                                                          <25
                                                          <25
                                                         6.5
                                                         5.5
                                                         25.0
                                                         <25.
    Notes:   B.A. denotes bottom ash,  F.A. denotes fly ash, and < represents  that the number reported is an
            average of two In which one was below the detection limit and  the other was just above.

-------
TABLE F-9.   TOTAL METAL  EMISSION RATE IN THE  FLUE GAS

March
4-34
4-35
4-37
4-38
4-39
4-40
4-41
4-42

May I
3-42
3-43
3-44
3-45
3-46
3-47
3-48


1:0
3/19
3/19
3/21
3/22
3/28
3/31
3/31
4/1
X
:0
5/3
5/4
5/5
5/5
5/10
5/16
5/17
X
Pb

202
327
171
•96i
220
233
218
222
228

295
210
242.
333
171
-5W
129
230
Cd

3.37
5.85
3.05
3.08
<4.40
8.76
2.73
4.14
<4.43

3.69
2.99
3.42
6.21
2.86
8.98
2.23
4.33
As

218
***,
276.
146.
186.
80.4
100.
208.
173

315
206.
213.
215.
91.0
169.
79.4

Hg

11.1
9.4
10.4
8.0
<5.3
3.3
<8.1
7.2
<7.85

<8.1
6.0
2.8
4.8
<5.1
7.3
<4.9
£5.57
Cr

38.7
35.2
42.7
46.2
29.3
9.74
13.6
65.1
35.1

111.
26.2
•3*3-.
64.9
14.2
36.8
51.3
50.7
TOTAL PARTICULATE
/ig/m3
Nl Mn Zn

31.4
23.7
39.6
38.4
36.7
19.5
27.3
44.4
32.6

105.
45.0
39.9
64.9
13.5
38.4
39.7
49.5

33.7
59.6
69.5
53.1
48.4
26.3
25.4
65.7
47.7

48.0
31.5
22.3
34.4
32.6
32.3
11.6
30.4

615
790.
518
673
660
486
386
606
592

185.
*V-r
742
1129
399.
799.
319.
596
P.P.
Cu

43.5
58.5
48.7
53.9
<73.3
<48.6
<45.5
41.4
-51.7*

61.5
45.0
57.1
73.3
28.6
57.4
27.8
50.1
/ig/m3
Sn

2.41
2.34
0.610
1.53
1.47
<.968
<.905
1.48
<1.46

1.23
1.10
1.71
3.39
.855
1.33
<.555
£1.45
Tot Part g
P.F.g
Sb

<48.3
<117.
<61.0
<77.0
<147
<97.4
<90.9
<59.2
<87.2

123.
<75.
<57.0
<56.3
<57.1
<35.9
<55.2
<65.6
Ag

<4.83
<11. 7
<6.10
<7.70
<14.7
<9.74
<9.09
<5.92
<8.72

<12.3
<7.5
<5.70
<5.63
<5.71
<3.59
<5,57
<6.56
Vn

48.
117
<61
<77.
<147
<97.
<90.
<59.
<87.

<12.
< 7.
<57.
<56.
<57.
<35.
<55.
<65.


3


0

4
9
i
2

3
5
0
3
1
9
7
6
Total
Part.

.8215
.5618
.6814
.4946
.2203
.3747
.3851
.5850


.4090
.5716
.6955
.6578
.9075
1.0586
.6837

Part.
Ftlter

.6329
.4361
.5857
.3901
.1692
.3059
.3361
.5268


.2548
.4488
.6368
.6101
.7660
.9447
.6131

                                                                      (continued)

-------
                                                  TABLE  F-9.   (continued)
TOTAL PARTICULATE
Ug/m3

March
4-43
4-44
4-45

May
3-49
3-50
3-51
I-1
— J
NJ
March
4-46
4-47
4-48

May
3-52
3-53
3-54

May
3-55
3-56


1:1
3/23
3/23
3/24
X
1:1
5/12
5/12
5/13
X
1:2
3/29
3/29
3/30
X
1:2
5/10
5/10
5/11
X
0:1
5/14
5/14
X
Pb

3035
4917
3973
3975

6656
3512
2543
4237
12289
6054
4636
7660.

7116.
7338
10196
8217

12046
7859
9953.
Cd

77.5
89.4
71.4
79.4

88.4
79.0
49.7
72.4
326.
254
120
233

182
361
117
220

303
231
267
As

61.8
49.5
26.6
45.9

189.
204
673
153
73.4
27.4
33.9
44.9

116.
138
123
126

49.8
49.0
49.4
Hg

21.3
11.1
26.5
19.6

20.0
15.0
12.2
15.7
15.4
14.7
6.9
12.3

13.7
9.1
11.5
11.4

90.8
98.6
94.7
Cr

28.0
39.3
33.1
33.5

39.8
35.7
30.8
35.4
46.8
51.4
44.6
47.6

68.8
51.9
45.5
55.4

76.6
82.8
79.7
Ni

31.0
36.6
28.7
32.1

44.7
38.4
24.6
35.9
43.4
17.2
62.3
41.0

87.8
38.9
26.0
50.9

32.9
25.9
29.4
Mn

48.9
82.7
62.1
64.6

74.5
79.0
34.4
62.6
88.1
89.7
126.
101

108.
140
97.4
115.

252.
298
275
Zn

6744
6354
4939
6012

6457
6035
4512
5668
13355
4332
8020
8569

8065
8440
8447
8317

9858
6207
8033
Cu

84.3
107
97
96.1

99.3
82.3
65.6
82.4
133
43.4
71.3
82.5

157.
123
123
134

208
197
203
PF ug/tn1
Sn

2.80
5.06
2.21
3.36

3.98
3.29
0.82
2.70
2.67
2.50
9.81
4.99

2.61
5.85
1.95
3.47

4.38
7.76
6.07
3 Tot

Sb

<56.2
<56.2
<44.1
<52.2

<49.7
<54.9
<47.0
<48.5
<66.8
106.
<89.1
<87.3

<47.4
<64.9
<65.0
<59.1

<109.
<103

Part &
PFg
Ag

11.8
13.5
10.6
12.0

9.93
<5.49
<4.10
£6.51
28.7
13.8
8.91
17.1

21.9
18.2
18.2


30.6
28.9
29.7

Vn

<56.2
<56.2
<44.1
<52.2

<49.7
<54.9
<41.0
<48.5
<66.8
<22.8
<89.1
<59.6

<47.4
<64.9
<65.0
<59.1

<109
<103


Total
Part.

.4104
.4923
.5222


.7092
.7000
.4035

.4518
.4694
.2845


.8542
.7227
.5889


.3518
.3246


Part.
Filter

.3658
.4644
.4590


.5976
.5674
.3868

.3728
.3994
.2222


.7616
.5510
.4559


.2791
.2623

Notes:   B.A. denotes bottom ash, F.A.  denotes fly ash,  and < represents  that the number reported is  an
        average of two in which one was  below the detection limit and  the other was  just above.

-------
                                  APPENDIX G

          PHYSICAL AND CHEMICAL CHARACTERIZATION OF RDF/COAL FLY ASH


 INTRODUCTION

     As part of the effort to identify the potential for environmental
 impact by fly ash generated by combustion of a coalrdRDF blend, fly ash
 samples were sent to Colorado State University for analysis.  The aims of
 this testing were to identify any differences in the composition of fly ash
 generated by different coal:dRDF fuels, to ascertain whether concentrations
 of certain materials existed at the surface of the particles, and to deter-
 mine if fly ash composition was a function of particle size.  D. F. S. Natusch
 of Colorado State University was selected to perform the analysis because of
 his extensive background in this area.  The tentative conclusions put forth
 in this section are the result of Dr. Natusch's analysis.

     Specifically, the investigation was designed to answer the following
 questions:

     1.   What are the morphological and compositional characteristics of fly
          ash generated by burning dRDFrcoal fuel blends?

     2.   What is the elemental composition of fly ash generated by burning
          dRDFrcoal fuel blends with respect to (a) the size of the particles
          and (b) the ratio of dRDF to coal?

     3.   What factors appear to be responsible for the partitioning of
          elements present in the fly ash as a function of size and RDF
          content?

     4.   To what extent can individual elements present in the fly ash be
          mobilized into solution as a result of an aqueous leaching process?

     In the following sections we present a brief description of the analyti-
cal methodology employed, a list of the results obtained, and a short dis-
cussion of the meaning of these results.

PHYSICAL CHARACTERISTICS OF RDF/COAL FLY ASH

     The samples of dRDFrcoal fly ash received consisted of four sets of
seven samples.  Each sample set was obtained from burning a different ratio
of dRDF to coal ranging from pure coal to pure dRDF, and the seven samples
corresponding to each set were obtained from the seven impaction stages of an


                                     173

-------
 MRI stack sampler.  Each of the seven impactor stages for each blend ratio
 consisted of composited scrapings from two to six stack samples.  The dis-
 tribution of particle mass within this 28-sample set is presented elsewhere
 in this report.

      Each of the 28 subsamples was investigated using a scanning electron
 microscope (Hitachi, Model HHS-2R).   Each sample was mounted on double-sided
 Scotch tape and coated with carbon and gold for the purpose of observation.

      The results showed the fly ash to consist of four different structural
 types.  The first type consisted of a "spongy-looking" material that may have
 been formed as a result of condensation or agglomeration of extremely small
 particles onto the surface of large particles (Figure G-l).  The second type
 was in the form of extremely thin sheets, some of which were found to roll or
 unroll under the influence of the SEM electron beam (Figure G-2).  The third
 type was spherical in form and similar to conventional fly ash particles
 obtained from the combustion of coal.   It is interesting to note, however,
 that the size distribution, as observed qualitatively under the microscope,
 indicated that these spherical particles are significantly smaller than those
 normally found in a conventional coal-fired power plant.  In addition, it was
 noted that even when pure coal was employed as fuel,  the geometry of the so-
 called spherical fly ash particles was significantly different from those
 obtained in a full-size power plant,  indicating that  combustion conditions
 employed in this particular experimental unit were,  in fact,  somewhat unique
 (Figure G-3).   The fourth type of particle was in the form of flaky material
 (Figure G-4)  with somewhat rounded edges which indicate the possibility of
 some melting  during the combustion process.

      It was observed that each of these morphological types was present in
 all  of the  subsamples  investigated.  There were,  however,  some variations in
 the  relative  amounts of each structure obtained in particular subsamples.
 For  example,  the spherical particles were found to occur in much greater
 profusion in  the small size fractions  of the  samples  from the 1:1 and 1:2
 RDF-to-coal blends.

      It is apparent  from the foregoing results that  fly ash generated from
 burning blends  of  dRDF:coal is significantly  different in form from that
 which  results  from the combustion of pure coal.   In  particular,  the absence
 of a large population  of  spherical particles  indicates that the combustion
 temperature employed in this particular  system was insufficient to generate
 molten fly ash  material as  occurs  in the combustion  zone of conventional
 coal-fired units.  This phenomenon provides  relatively high specific surface
 areas  due to  the predominance  of  the spongy-type  material  illustrated in
 Figure  G-l, and consequently the  operation of  strong  surface-associative
 effects  in the  distribution of potentially toxic  species might be expected.

     The second type of physical  characterization performed was a determi-
 nation  of the distribution  of  the  specific surface area of the fly ash
 particles as a  function of  both particle  size  and the  dRDF:coal ratio.   This
was achieved using a Quantachrome  Quantasorb Model QS-7  which enables deter-
mination of a specific  BET  surface area  based  on  the  adsorption of nitrogen.
Under normal operating  conditions, samples are  initially outgassed at 300°C

                                    174

-------
Figure G-l.  0% RDF sample, S4, magnification 450 X.
             (Note spherical particle at about 4 o'clock)
 Figure G-2.  67% RDF sample, S4, magnification 3500 X.
                           175

-------
Figure G-3.  100% RDF sample, S4, magnification 10,000 X.
  Figure  G-4.   50% RDF sample,  S4,  magnification 2000 X.
                          176

-------
to remove any absorbed material prior to the determination of the surface
area.  Due to lack of material, however, outgassing at 300°C was not employed
in the present instance for fear of losing potentially volatile elements such
as arsenic and cadmium.  For this reason, outgassing at room temperature was
employed with the result that the samples were not, in our opinion, com-
pletely outgassed.  Nevertheless, the specific surface area values obtained
are considered to be relatively consistent within the individual sample set
investigated.  They should not, however, be taken as absolute values.

     The results obtained are illustrated in Table G-l from which it will be
seen that insufficient material was available for determination of all size
fractions and dRDFrcoal ratios.  Nevertheless, the results do establish
several points of interest.  First, it is apparent that the stack sampler
employed for particle collection in this study is, in fact, producing a good
differentiation of the fly ash on the basis of particle size.  In other
words, there is a clear dependence of the specific surface area of these
particles on aerodynamic particle size.  Secondly, there appears to be a
general trend of increasing specific surface area with increasing RDF-to-coal
content.  In short, the higher the RDF content of the fuel the greater the
specific surface area of the fly ash particles which result.
  TABLE G-l.  SURFACE AREAS  (m2/g) OF dRDF:COAL FLY ASH FOR THE
              SEVEN SIZE FRACTIONS COLLECTED USING AN MRI STACK SAMPLER
dRDF by                                 Size fraction
volume    SI        S2        S3        S4        S5        S6        S7
0%
50%
67%
100% 2.7
4.96
7.26
	 	
5.5 15.3
5.73
9.02
9.61
17.9
8.38
	
11.4
20.8
10.2
9.90
	
23.0
10.1
	
	
30.0
                             -Increasing size-
                                     177

-------
      Table G-2 presents the size fraction cut-off for each of the impactor
 plates  (SI through S7).  For example, Stage S2 is 50 percent efficient in
 collecting particles in the size range of 15 to 30 microns.
              TABLE G-2.  NOMINAL CUT DIAMETERS FOR IMPACTOR TESTS
           Stage                    Partical Size at 50 Percent Efficiency
                                                 (microns)
SI
S2
S3
S4
S5
S6
S7
Filter
30
15
6
2.4
1.5
0.65
0.37
Less than







0.37
 NOTE:     For further information see Meteorology Research,  Inc.,  Instruction
           Manual, Inertial Cascade Impactor,  Model 1502 and  1503,  1976.
 ELEMENTAL COMPOSITION

      Elemental analyses of the various subsamples presented were carried out
 in  several ways.   Semiquantitative analysis of a number of the samples  was
 undertaken using  direct current arc emission spectrometry in which  the
 spectra were  recorded photographically on a Baird-Atomic 3-meter grating
 spectrometer,  Model 6X-1.   Samples were mixed with "Spex Mix" spectroscopic
 graphite using a  "wig-L-bug"  and were completely vaporized using a  direct
 current arc.   The integrated  spectra obtained were analyzed using a manual
 densitometer.   The results obtained for the S3 size fraction are presented in
 Table G-3.

     Samples were digested in an acid mixture consisting of 3.5  mis of  aqua
 regia,  2.5 mis of 48  percent  hydrofluoric acid,  and 0.5 mis of water.   The
 resulting digest  was  neutralized using approximately 2  grams of  boric acid to
 remove  excess  hydrofluoric in the form of boron trifluoride.   These samples
were then analyzed for  18  elements using an automated Spectrometrics plasma
 emission spectrometer utilizing an Echelle monochromator.   The results
 reported by element are  listed in Table G-3.

     Specific  analyses  for  arsenic were performed  by generating  the arsenic
hydride and identifying  this  concentration using  conventional flame atomic
absorption spectrometry.   The  results  obtained  for arsenic  are also included
in Table G-3.


                                       178

-------
TABLE G-3.   ELEMENTAL ANALYSIS OF INDIVIDUAL RDF  COAL  FLY ASH SAMPLES
              CONCENTRATIONS  IN yg/g
               dRDF    Analysis
 Element     by Volume  Method     SI
                    Size  Fraction From Sampler
                  S2      S3       S4      S5
                              S6
                      S7
Aluminum
Antimony
                 0%
                50%
                67%
               100Z
                 OZ
                50Z
                67Z
               100Z
PES
INAA
DCAES
PES
INAA
DCAES
PES
INAA
DCAES
PES
INAA
DCAES
PES
INAA
DCAES
PES
INAA
DCAES
PES
INAA
DCAES
PES
INAA
DCAES
             84,000
    57,900     9,970
            101,000  55,900
    44,300    46,400  42,700  41,200
                                    22
                                                  (1)
                                                  (5)
      (10)

24     31
      (20)
                                                             58
                                                            200
                                                            341
3.2*   5.5*    5.8*    111
                                                               (continued)
                                          179

-------
                              TABLE  G-3.   (continued)
                dRDF    Analysis
  Element     by Volume   Method     SI
                      Size Fraction from Sampler
                    S2      S3       S4       S5
                                    S6
                                    S7
Arsenic
Barium
                 0%
                50%
                67%
               100%
                 0%
                50%
                67%
               100%
 AAS
 INAA
 DCAES
 AAS
 INAA
 DCAES
 AAS
 INAA
 DCAES
 AAS
 INAA
 DCAES
PES
INAA
DCAES
PES
INAA
DCAES

PES
INAA
DCAES

PES
INAA
DCAES
                                    312
  548
                                                   (400)
                                                  (300)
 (350)
  500
  888
 (200)
        1630
                                 <105
<334
 992
1100
1150
1211
1220
 762


1230
 488


1530
 959

1050
1042
 461
1392
1700
1571
1000
 691
                           1220
1390
 881
 679
1050
 252
                                                                 (continued)
                                           180

-------
                              TABLE G-3.   (continued)
                dRDF   Analysis
 Element     by Volume   Method
            SI
             Size Fraction from Sampler
         S2      S3       S4       S5      S6      S7
Beryllium
Bromine
                 50Z
                 67Z
                100Z
                  OZ
                 50Z
                 67Z
                100Z
PES
1NAA
DCAES

PES
INAA
DCAES

PES
INAA
DCAES

PES
INAA
DCAES
PES
INAA
DCAES

PES
INAA
DCAES

PES
INAA
DCAES

PES
INAA
DCAES
                 (9)


                 (9)
               <53.4

                 (7)
                13.0

                (10)
                                                            21.7
  9.15
<29.0
  3.27
42.6
18.9    <36.2
973
                                                           85.
                                                          229
                                                          357
                                          256     107      146       95      140     110
                                                                 (continued)
                                          181

-------
                               TABLE G-3.   (continued)
                dRDF    Analysis
 Element     by Volume   Method
            SI
 Size Fraction  from Sampler
S2      S3       S4       S5
                  S6
S7
Cadmium
Calcium
                0%
               50%
               67%
              100%
                0%
               50%
               67%
              100%
 PES
 INAA
 DCAES
 PES
 INAA
 DCAES
 PES
 INAA
 DCAES
 PES
 INAA
 DCAES
PES
INAA
DCAES
PES
INAA
DCAES
PES
INAA
DCAES
PES
INAA
PCAES
        (3)
      <931

       (20)
       (20)
      <92.1

        (3)
              16,700


     41,000   14,100
               139
<108
               <505     394
        <167     <631
              32,400   27,000
     12,000   14,000    9,750  10,500
                                                                 (continued)
                                          182

-------
TABLE G-3.   (continued)
dRDF Analysis
Element by Volume Method SI
Chromium
0% PES
INAA
DCAES
50% PES
INAA
DCAES
67% PES
INAA
DCAES
100% PES
INAA
DCAES
Cobalt
0% PES
INAA
DCAES
50% PES
INAA
DCAES
67% PES
INAA
DCAES
100% PES
INAA
DCAES
Size Fraction from Sampler
S2 S3 S4 S5 S6 S7


(150)


(300)


(300)


(70)


(60)


(50)


(30)


(60)
                               (continued)
            183

-------
TABLE G-3.   (continued)
Element
Copper
Dysprosium

dRDF Analysis Size Fraction from Sampler
by Volume Method SI S2 S3 SA S3 S6 S7
OZ PES 91.3
INAA
DCAES (100)
50Z PES <3,630 <283
INAA
DCAES (60)
67Z PES <134 <4,320
INAA
DCAES (50)
100% PES <747 <1,780 <168
INAA
DCAES (100)
OZ PES
INAA 18.2
DCAES
50Z PES
INAA 10.9
DCAES
67Z PES
INAA 7.5
DCAES
100Z PES
INAA <0.039 5.0 9.7 11.1 10.5 10.0 3.9
DCAES
                           (continued)
            184

-------
TABLE G-3.   (continued)
dRDF
Element by Volume
Europium
0%
50%
67%
100%
Gallium
0%
50%
67%
100%
Analysis Size Fraction from Sampler
Method SI S2 S3 S4 S5 S6 S7
PES
INAA 3.13
DCAES
PES
INAA 2.2
DCAES
PES
INAA 1.55
DCAES
PES
INAA <0.37 <0.74 1.79 1.93 1.90 2.08 0.73
DCAES
PES
INAA 120
DCAES
PES
INAA 80
DCAES
PES
INAA 49
DCAES
PES
INAA 32 22 69 80 114 173 165
DCAES
                            (continued)
              185

-------
                              TABLE G-3.    (continued)
                dKDF    Analysis
 Element     by Volume   Method
            SI
 Size Fraction from Sampler
S2      S3       S4      S5
        S6
                        S7
Lanthanum
Lead
                 0%
                50%
                67%
               100%
                50%
                67%
               100%
 PES
 INAA
 DCAES
 PES
 INAA
 DCAES

 PES
 INAA
 DCAES
 PES
 INAA
 DCAES
PES
INAA
DCAES

PES
INAA
DCAES

PES
INAA
DCAES

PES
INAA
DCAES
                                   9.3
19      46
                                                   (800)
                                                 (1,000)
                                                            68
                                                            60
37
                                                <10,400    757

                                                   (300)
49      48
                                                           189
              <7,830   <7,745
                                                                                    25
      <1,030   <1,290   <1,860  <7,042

        (500)
                                                           (continued)
                                           186

-------
TABLE  G-3.   (continued)
dRDF Analysis
Element by Volume Method SI
Magnesium
0% .PES
INAA
DCAES
50% PES
INAA
DCAES
67% PES
INAA
DCAES
100% PES
INAA
DCAES
Maganese
0% PES
INAA
DCAES
50% PES
INAA
DCAES
67% PES
INAA
DCAES
100% PES
INAA 947
DCAES
Size Fraction from Sampler
S2 S3 S4 S5 S6 S7

4,000
7,260 4,550
6,810 6,630
3,680 4,060 2,970 2,910

<391
378
(300)
<4,530 <548
555
(300)
<257
741
(400)
<3,260 <58.5 <321
666 206 189 183 167 79
(200)
                             (continued)
              187

-------
TABLE G-3.   (continued)
Element
Molybdenum
Nickel

dRDF Analysis Size Fraction from Sampler
by Volume Method SI S2 S3 S4 S5 S6 S7
0% PES <13.5
INAA
DCAES
50% PES <18.9
INAA
DCAES
67% PES <88.6
INAA
DCAES
100% PES <20.3 <111
INAA
DCAES
0% PES
INAA
DCAES (1,000)
50% PES 7,260
INAA
DCAES (600)
67% PES
INAA
DCAES (600)
100% PES 4,610 <37,200 214
INAA
DCAES (300)
                           (continued)
            188

-------
TABLE G-3.   (continued)
dRDF
Element by Volume
Phosphorous
0%
50%
67%
100Z
Potassium
0%
50%
67%
100%
Analysis
Method SI

PES
INAA
DCAES
PES
INAA
DCAES
PES
INAA
DCAES
PES
INAA
DCAES
PES
INAA
DCAES
PES
INAA
DCAES
PES
INAA
DCAES
PES
INAA 2,500
DCAES
Size Fraction from Sampler
S2 S3 S4 S5 S6 S7

3,710
<11,900 2,340
3,260 3,450
1,360 1,770 887 1,360
18,400
12,900
20,800 15,900
18,000
17,000 26,200
15,900
13,300 15,600 11,400 10,000
3,000 1,010 11,600 11,500 11,800 5,400
                              (continued)
          189

-------
TABLE G-3.   (continued)
dRDF Analysis
Element by Volume Method SI
Silicon
0% PES
INAA
DCAES
50% PES
IXAA
DCAES
67% PES
INAA
DCAES
100% PES
ISAA
DCAES
Sodium
0% PES
INAA
DCAES
50% PES
ISAA
DCAES
67% PES
INAA
DCAES
100% PES
INAA 1,990
DCAES
Size Fraction from Sampler
S2 S3 S4 S5 S6 S7
74,600

33,000
19,000 88,900
24,300
3,300
1,960
28,200
45,600 4,000
28,200
2,920 23,500
1,470 2,680 2,930 3,000 3,350 1,720
                            (continued)
         190

-------
                            TABLE G-3.   (continued)
 Element
  dRDF     Analysis
by Volume   Method
                                   SI
  Size Fraction  from  Sampler
S2      S3        S4       S5      S6
                                                                                    S7
Strontium
                50%
                67%
               100%
           PES
           INAA
           DCAES

           PES
           INAA
           DCAES

           PES
           INAA
           DCAES
           PES
           INAA
           DCAES
                                         <193
               1,180
               1,270

        <981     881
               1,518

                 943
               1,237

         740     879
         989   1,107
479
410     661
638   1,010
420
Titanium
                 0%
                50%
                67%
               100%
           PES
           INAA
           DCAES

           PES
           INAA
           DCAES

           PES
           INAA
           DCAES

           PES
           INAA
           DCAES
                                                 (6,000)
                                                (12,000)
                                                (10,000)
                                                                 (continued)
                                          191

-------
                             TABLE G-3.   (continued)
Element
Thalium




Vanadium




dRDF Analysis
by Volume Method SI
0% PES
INAA
DCAES
50% PES
INAA
DCAES
67% PES
INAA
DCAES
100% PES
INAA
DCAES
0% PES
INAA
DCAES
50% PES
INAA
DCAES
67% PES
INAA
DCAES
100% PES
INAA
DCAES
Size Fraction from Sampler
S2 S3 S4 55 S6 S7
(0.5)
(2.0)
(0.6)
(2.0)
(1,500)
(500)
(300)
(800)
NOTE:  PES    - Echelle Plasma Emission Spectrometry
       INAA   - Instrumental Neutron Activation Analysis
       DCAES  - DC Arc Emission Spectroscopy
       AAS    - Atomic Absorption Spectrometry

* These values may be low due to arsenic Interference.
                                        192

-------
     Selected fly ash samples were subjected to instrumental neutron acti-
vation analysis utilizing a Triga II reactor having a neutron flux at the
sample of approximately 1012 neutrons/second/cm2.  Since the primary objective
of this analysis was to obtain some relative measure of the accuracy of the
other analytical techniques and, specifically, to determine the arsenic and
antimony present, only a short irradiation was employed.  Following removal
from the reactor, the samples were analyzed using a Ge/Li detector in
conjunction with a 4096 channel multichannel analyzer.  The results were
transferred to magnetic tape and analyzed using the PIDAQ program.  These
results are presented in Table G-3.

     Finally, several subsamples were subjected to individual particle
analysis using a Kevex Model 5000A X-ray energy dispersive spectrometer (XES)
associated with the scanning electron microscope.

     A total of 27 elements were determined using the techniques listed
above.  It will be noted that the analyses were not performed by X-ray
fluorescence spectrometry as originally envisaged due to the small amounts of
sample provided.  This decision was reached after initial results indicated
unacceptably low counting statistics for most elements.  As a result, a
semiquantitative screening analysis was performed using DC arc' emission
spectroscopy (DCAES).  This technique was applied only to the size fraction
S3 of each RDF:coal blend sample.  The results are considered to have only an
order of magnitude precision so are presented in brackets in Table G-3.

     The size fractions S3 to S6 were analyzed by plasma emission spectro-
metry.  The results were generally disappointing due to the high blank
levels encountered.  Consequently, only about half of the analyses performed
are considered to be meaningful, and only those are included in Table G-4.
(It will be noted that good analytical data were obtained for the aqueous
leachates presented later since extensive acid digestion was not required.)
The precision of the PES results is generally about 10 percent although in
this complex analytical matrix accuracy for the trace elements Be, Cd, Cu,
Pb, Mn, Mo, and Ni is likely to be considerably poorer.

     Analyses performed by Instrumental Neutron Activation Analysis all
have a precision (based on counting statistics) of less than 10 percent.  The
accuracy appears to be comparable.  Precisions of 1 to 2 percent are
associated with the results obtained for AS, Mn, and Na.

     It will be noted that quite good agreement is obtained between the
techniques employed for the elements As, Ba, K, and Sr.  Agreement is poor
for the other elements.  It must be strongly stressed, however, that in
situations such as this where the original samples are composites of grab
samples, sampling statistics are normally extremely poor.  Since different
analytical techniques require different amounts of sample, this means that
the precision associated with each procedure will vary greatly even though
the intrinsic analytical precision is good.  In brief, the spread of results
depicted in Table G—4 is fairly typical.  Nevertheless,, the relative pre-
cision and accuracies do enable trends to be observed.
                                     193

-------
       The  actual  amounts  of  the various  elements  listed in Table G-3 are
 hardly  remarkable  except in the  case of As, which is fairly high.  However,
 some  interesting trends  are observed.   Most notable is the apparent variation
 of  several  elements with variations in  the dRDF:coal ratio.  Thus, Br, Mn,
 Pb, and Sb  show  a  tendency  to increase  in concentration with increasing dRDF
 levels  whereas As, Ni, and  V show an opposite tendency.  The trends are
 hardly  definitive  but are probably real as indicated by similar trends
 observed  in these  materials in the fly  ash leachates.

       Variations  of concentration with particle size are also not well defined.
 There is, however, a general tendency for the elements As, Ga, Na, and Sb to
 increase  in specific concentration with decreasing particle size.  A pro-
 nounced increase in the  concentration of Br and Mn in large particles is
 also  observed.

       The  trends  indicated above  are not unexpected.  Thus, one would expect
 elements  such as Br, Mn,  Pb, and Sb to  be present at higher levels in dRDF
 than  in coal, and  the preferential association of As, Ni, and V with coal is
 acceptable.  Similarly,  the  preference  of As, Ga, Na, and Sb for small
 particles is to  be expected  since these elements are capable of being vola-
 tilized during combustion and then preferentially adsorbed onto small
 particles.  The behavior of  Br and Mn is not understood,  however.

      To investigate the association of different elements with individual
 particles, the size fraction S4 was subjected to energy dispersive X-ray
 emission analysis under a scanning electron microscope.  A number of indi-
 vidual particles were analyzed and found to contain Al, Ca, Fe, K, Si, Ti,
 and Zn as  consistent matrix elements (note:   these are the only analyses for
 Fe and Zn).   The presence of As,  Na,  P,  and S was indicated in some particles,
 but signal intensities were too weak for absolute identification.

 MECHANISM  OF TRACE  ELEMENT DISTRIBUTION IN FLY ASH FROM FUEL BLENDS

      It  has now been reasonably established that certain  of the more
 volatile trace elements  (or the compounds in which they are present) are
 volatilized at temperatures encountered  during many combustion processes.
 The  resulting  vapor-phase metallic  species then either condense or adsorb
 (probably  the  former)  onto the  surface  of co-entrained fly ash particles as
 both vapor and particles  move away  from  the  high temperature combustion
 zone.  This  process results  in  the  preferential redistribution of volati-
 lizable  species  into  small particles  due to  the fact  that small particles
 have a larger  specific surface  area  than large particles.

     This  volatilization-condensation phenomenon has  several undesirable
 environmental  consequences.   First,  it results in many of the more toxic
 elements becoming preferentially  associated with small particles which are
most readily emitted from most combustion  operations,  which can have long
atmospheric  lifetimes, and which  are preferentially deposited in the pulmonary
region of  the human lung  when inhaled.   Secondly,  the  condensation phenomenon
results  in the presentation  of toxic species  on  the surface  of  particulate
matter,  thereby making it most readily available  to the external environment


                                     194

-------
 (for example, extracting body fluids).  Because of these effects, it is of
 considerable interest to determine the extent to which such a volatilization
 condensation mechanism may operate.  Consequently, it is appropriate to
 determine trends in elemental concentrations with particle size.

     The dependence of elemental concentration on particle size is indicated
 in Table G-3.  While there is some evidence that certain elements (e.g., As,
 Ga, Na, and Sb) increase in specific concentration with decreasing particle
 size (i.e., increasing particle specific surface area), these trends are not
 sufficiently convincing to establish the existence of a volatilization-
 condensation process.

     To obtain direct information on surface concentrations, a number of
 particles were examined using auger electron spectroscopy.  Since the
 analytical volume for this technique extends only about 10 to 20 A below
 the particle surface, the spectra obtained are derived entirely from surface
 associated material.  In order to obtain depth profiles, inner layers of the
 particles are exposed by etching into the particles using a stream of
 positively charged argon ions.

     A representative scan from the auger spectrometer is presented in
 Figure G-5 for the S4 size fraction derived from 100 percent dRDF.  This
 shows that the elements Al, C, Ca, Fe, P, S, Si, and Ti are the predominant
 surface constituents of these particles.  Depth profiles for the most
 readily identifiable elements associated with samples derived from 0, 50,
 and 100 percent dRDF:coal blends are presented in Figures G-6 through G-8.
 Precise depth scales were not established; however, a sputtering rate of
 approximately 30 A/minute was employed.

     These depth profiles do not show any pronounced surface predominance
 for the elements observed.  There is an indication that S and Si may be
 surface enriched and C surface depleted in some samples; however, these
 results could well be artifactual in the case of S and Si.  Although not
 shown in Figures G-5 through G-8, there appears to be an increase in the
 weak Fe and Ti signals with depth.  Chlorine, though observed initially, is
 rapidly removed by the electron beam.

     It can be concluded that these surface studies provide no evidence for
 the occurrence of a volatilization-condensation mechanism.  This does not
mean that such a mechanism does not exist but simply that it does not apply
 for the elements observed by auger electron spectroscopy.

MOBILIZATION OF TRACE ELEMENTS IN SOLUTION

     While the elemental composition of fly ash from dRDFrcoal fuel blends
was the primary measurement required in the present study, it is important
 to establish the extent to which the species present can be mobilized in
solution.  This is because the toxic trace elements exert an adverse
environmental impact only if they can be transferred from the solid material
 to a liquid solution.
                                     195

-------
                  (original data  supplied by
                   the University of Colorado)
Figure G-5.  Auger  electron spectrum of 100 percent dRDF fly ash.


JR6
r-0-j-iH





Co
E. -fo«r" S



'SPECIMEN
; 1 ; j Element [ Etching
i ' v {*•/ : 1 Gun 2
?%> KDP Fly AS^ 3 c jy ' ' *V|P
- •— 	 j 	 •_.!-. A 	 i£_: ^i^i"
j — — f>— R»le:
; . . | • i • •
        (original data supplied  by the University of Colorado)
                         TIME
                                  MINUTES/DIVISION
      ip-   3S   v,
 	1	1	-•—— •—	•	                     / i
  Figure G-6.   Auger  depth profile for  0  percent dRDF fly ash.

                              196

-------
                          SO*
                                                             SPECIMEN
                                                             )M-i4
                                                               *L_
                                                                          Etching
                                                                           T
s !
                                              L-::1-::::!;:i:  ::.:.:t-^
     (original  data  supplied by the  University of  Colorado)
                                                                           -f	
                                TIME.    2__   MINUTES/DIVISION
  [Ep-  r
               ir  ;Vnxxi= z   RC=
                                                 IOX   NEUT.     DATE: 2-'
     Figure G-7.   Auger depth  profile for  50 percent dRDF fly ash.
                                                             I SPECIMEN
                                                              0='- S*
                                      8
                                               F«
                                                                           Etehmf
           (original data supplied by  the University of  Colorado)
                                 TIME.   2.   MINUTES/DIVISION
                                              SEMS- IPX
                                                               o«TE: 3.-l'i-
Figure G-8.
                     Auger depth profile for  100 percent  dRDF fly ash.
                                     197

-------
       In order  to  obtain information about the solubility characteristics of
 the  fly ash, the  size fractions S3 through S7 were ultrasonically agitated
 (Heat Systems  Model W 200R Sonicator Cell Disrupter) for 2 hours with
 15 mis of triply  distilled water.  Sample masses varied from 0.0019 to
 0.0230 g.   (Previous studies have established that water soluble material
 can  be quantitatively extracted under these conditions.)  Following sonica-
 tion, the samples were filtered through a 0.45-um millipore filter.  Elementa
 analyses of the filtrate were then performed using plasma emission spectro-
 metry, and anion  analyses were performed by ion chromatography using a
 Dionex Model 10 ion chromatograph.  A representative ion chromatogram showing
 the  presence of fluoride, chloride, nitrate, and sulfate in the leachate
 from the fly ash  is presented in Figure G-9.

      The results obtained from these analyses are presented in Table G-4.
 Correction has been made for all blank levels and precision is generally
 j<10  percent.

      These solubility studies provide considerable information about both
 the chemical characteristics and the potential environmental impact of the
 fly ash.   Thus, with reference to the data in Table G-4, a number of trends
 can be observed.

      First,  it is apparent that for most species the amount of material
 which is  soluble increases with increasing RDF levels in the original fuel
 blend.   This trend can be seen clearly for Ca, Cu, K, Mg, Mo,  Na, Pb, Si,
 Cl ,  N03  ,  and SO*,2  and may occur for B,  Ba,  Cd, and F~.  A reverse trend
 is observed for Ni and P and, possibly,  for Cr and Sr.   The fact that many
 of the major matrix elements exhibit increasing solubility with increasing
 dRDF  percentage indicates that the addition of dRDF to coal will result in
 greater bulk solubility (as well as greater trace element mobilization) as
 compared with  pure coal fly ash.

      The second obvious trend is towards increasing mobilization of species
 with  decreasing particle  size.   This is  apparent for Cd, Cr,  Cu, K, Mn, Mo,
 Na, Ni, Pb,  and Cl~ and may occur  for Ba,  Be,  P, and F~.  This trend may be
 due to condensation of  these species from  the  vapor phase as  discussed
 earlier, or  it  may result  from the more  efficient formation of soluble
 oxides (i.e., calcining)  in small  particles.   In the former case one would
 expect to see similar size  dependenices  for  both the bulk (Table G-3) and
 the separated soluble sample  (Table  G-4).   This  would not be  the case if
 solubility is the  direct  result  of chemical  reaction at a particle surface.
 Unfortunately,  the available  data  are  not  sufficient to rule  out either
 mechanism.

      Further consideration  of  the  data in  Tables G-3 and G-4  provides some
 interesting insights into the  fractional solubility of  fly ash from dRDF:coal
 blends.  As an  initial general statement,  it can be said that  matrix elements
 such  as Al, Ba, Mg, P, Si, and Sr  exhibit  quite  low solubility (<10 percent)
whereas minor and  trace elements such as Be, Cd,  K,  Mn,  and Na are fairly
 soluble (approximately 20 to 80 percent).  Calcium is a notable exception to
 this  rule insofar  as it exhibits high solubility.   In considering this


                                     198

-------
                           iijljii'ijiiiip!;1^:^;!1;-!!;^'!

                           i'lii!;!!!   ;i'i;^J: .•''•' iiii! i!:-:!
                                      .;  (original  data supplied by

                                      j  the University of  Colorado)

                                                    ; •!•!.•;.i
                 :i:il-::.   :..
                                      ;:.+!. ...I..!.; i -iJ *•»!,.! ••.;•
                                      .ii!::!i:&fl!j!i;Si;.!:!
                                               !',' i: i i:! • • i  i
                :•. 1...I. , 1 ;.M
               :.'' "! i •' • J!' • I ! •
                I;!::  i
':;:>;.; /;;i;i!:; •  liljliP 'Hi'!"•!'*
'  ll V1/1!^''!'!]!!:;;!;!'!!!!!^!''!!
       (2:10 dilution,  0-3 micromhos  full scale deflection)
Figure G-9.   A representative Ion Chromatograph of  Sample:   1-0-S7
                                   199

-------
TABLE G-4.  CONCENTRATIONS OF SEVERAL METALLIC ELEMENTS AND ANIONS
            IN AQUEOUS LEACHATES FROM dRDF:COAL FLY ASH
            (yg/g of fly ash leachate)
Element
Aluminum





Barium





Beryllium





Boron





Size

S3
S4
S5
S6
S7

S3
S4
S5
S6
S7

S3
S4
S5
S6
S7

S3
S4
S5
S6
S7
dRDF Fuel
0% dRDF
by volume

18.8
138
148
24.6
3.03

3.02
17.2
42.3
13.9
<12.3

0.123
<2.61
2.01
2.46
9.09

<1.62
20.4
22.1
12.3
<199
Blend Composition
50% dRDF 67% dRDF
by volume by volume

2.12
12
3.90
263
158

2.74
25.8
44.2
68.4
77.5

<0.342
<5.00
<1.31
<5.32
2.50

<1.79
25.5
77.9
<346
<165

6.15
8.40
—
—
—

3.93
11.0
—
—
—

<0.427
<4.70
—
—
—

<2.24
32.2
—
—

100% dRDF
by volume

31.3
112
194
181
<10.1

3.57
61.0
169
171
35.9

<0.714
<4.86
<2.81
4.76
1.71

<3.74
56.5
22.2
<313
<56.2
                                                 (continued)
                              200

-------
TABLE G-4  (continued)
Element
Cadmium





Calcium





Chloride





Chromium





Size

S3
S4
S5
S6
S7

S3
S4
S5
S6
S7

S3
S4
S5
S6
S7

S3
S4
S5
S6
S7
dRDF Fuel
0% dRDF
by Volume

1.17
<7.20
<12.8
<15.6
<167

343
750
1,140
862
830

<560
<137
<346
28.7
417

0.926
<1.16
6.04
5.74
33.3
Blend Composition
50% dRDF 67% dRDF
by Volume by Volume

<0.189
<13.9
<24.8
<100
120

620
2,300 2,
3,020
2,550
548

223
56.9
<669
<2,710
<1,290

0.822
<2.24
5.19
<21.3
10.0

<0.236
<13.0




769
370
—
—
—

72.6
792
—
—
—

0.684
<2.09
—
—

100% dRDF
by Volume

<0.395
<13.4
<53.0
252
527

1,680
7,750
16,700
8,490
899

2,310
6,310
13,000
33,700
54,900

1.71
<2.16
<11.2
28.6
6.84
                      (continued)
         201

-------
TABLE G- 4   (continued)
Element
Copper





Fluoride





Lead





Magnesium





Size

S3
S4
S5
S6
S7

S3
S4
S5
S6
S7

S3
S4
S5
S6
S7

S3
S4
S5
S6
S7
dRDF Fuel
0% dRDF
by Volume

<0.926
2.33
<3.39
<4.14
27.3

<443
12.6
<72.8
28.5
107

<5.04
<1.04
18.1
13.9
60.6

33.1
178
310
281
248
Blend Composition
50% dRDF 67% dRDF
by Volume by Volume

0.274 0.684
<5.50 <5.15
<6.56
5.26
228

<491 72.6
27.8 792
<141
<1,140
<271

3.70 4.79
15.5 6.10
22.1
1,740
7,890

18.1 25.1
308 350
474
353
70.0
100% dRDF
by Volume

<3.14
<5.35
<14.0
76.2
292

275
221
49.3
2,130
<92.7

7.71
12.0
283
2,420
4,720

94.6
328
1,240
1,300
100
                      (continued)
       202

-------
TABLE G-4   (continued)
Element
Manganese





Moybydenum





Nickel





Nitrate





Size

S3
S4
S5
S6
S7

S3
S4
S5
S6
S7

S3
S4
S5
S6
S7

S3
S4
S5
S6
S7
dRDF Fuel
0% dRDF
by Volume

2.22
10.6
26.8
25.4
57.6

0.864
2.15
14.8
24.6
60.6

0.556
5.40
17.4
17.2
24.2

60,000
1,550
1.580
6,090
5,380
Blend Composition
50% dRDF 67% dRDF
by Volume by Volume

4.79 5.47
29.0 " 25.2
36.4
26.3
30.0

1.03 1.11
<2.14 2.90
9.09
<37.5
2.50

0.822 3.08
5.15 2.26
6.49
10.5
70.0

142,000 199,000
2,160 1,980
7 , 690
10,400
<19,400
100% dRDF
by Volume

2.43
4.00
74.0
119
36.8

4.14
9.00
27.8
42.9
8.55

1.57
<0.665
<33.8
9.52
1.71

207,000
2,620
<21,500
29,800
8,890
                       (continued)
        203

-------
TABLE G-4   (continued)
Element Size
Phosphorous
S3
S4
S5
S6
S7
Potassium
S3
S4
S5
S6
S7
Silicon
S3
S4
S5
S6
S7
Sodium
S3
S4
S5
S6
S7
dRDF Fuel Blend Composition
0% dRDF 50% dRDF 67% dRDF
by Volume by Volume by Volume

<31.0
46.6
205
101
870

257
870
1,460
1,480
1,280

25.2
105
170
161
288

444
545
584
295
<3,280

<34.4 <11.7
40.7 34.2
18.2
<1,450
27.5

525 160
1,200 835
2,710
7,950
15,400

15.0 38.6
129 186
184
21.1
<200

514 1,050
3,340 3,120
7,900
18,100
32,600
100% dRDF
by Volume

18.9
4.66
142
890
200

630
2,260
11,100
24,000
38,000

163
325
1,120
948
212

1,920
6,700
29,200
63,600
82,400
                      (continued)
          204

-------
                          TABLE G-4 (concluded)
                             dRDF Fuel Blend Composition

                        0% dRDF     50% dRDF      67% dRDF     100% dRDF
Element          Size    by volume   by Volume     by Volume    by Volume
Strontium
Sulfate
                 S3          4.75          7.53        5.47         10.1
                 S4         34.8          49.6        25.8          40.0
                 S5         72.5          70.1          —          80.6
                 S6         32.8          42.1          —          19.0
                 S7        130           <25.3          —           3.42
                 S3     43,100        82,500      80,800       110,000
                 S4     10,800        17,900      17,700        25,400
                 S5     11,600        23,800            —      40,800
                 S6     11,900        56,800            —      66,200
                 S7      8,640       107,500            —      75,000
                                    205

-------
 fractional solubility, it is noteworthy that there is a trend of increasing
 fractional solubility with decreasing particle size for several elements,
 the most notable of which is Mn.  Since Mn exhibits no dependence of concen-
 tration on particle size in Table G-3, one can tentatively conclude that,
 for this element at least, its solubility increase with decreasing particle
 size is simply due to the more efficient calcining, and thus the greater
 bulk solubility, of small particles.

 CONCLUSIONS

      The following tentative conclusions can be drawn:

      1.    The specific concentrations of trace elements present in the fly
           ash from blend combustion are quite similar to those found in pure
           coal fly ash.   The dRDF seems to be the main contributor of Br,
           Mn, Pb,  and Sb in this particular case while the coal is the
           primary  source of As,  Ni, and V.

      2.    There is a tendency for several elements to be preferentially
           concentrated in small  particles, notably As, Ga, Na, and Sb.

      3.    It  is apparent that the volatilization-condensation mechanism
           which is responsible for partitioning volatile elements into small
           fly ash  particles in a coal-fired power plant is less effective  in
           the plant used to burn the  dRDF:coal blend.   This is probably due
           to  the lower temperatures achieved in the blend combustion since
           the most volatile elements  still exhibit volatilization-condensation
           partitioning.

      4.    The solubility of the  fly ash increases with increasing dRDF
           content.

      5.    The solubility of  both trace and matrix species present in small
           particles  is significantly  greater than in large particles.

     The overall conclusion  to be  drawn is that  utilization of dRDF supple-
ments to coal  for  energy  generation will  generally increase the amounts,
mobility, and  toxic  potential of  inorganic species associated  with  the
emitted fly ash as compared with  that  associated  with  pure coal fly ash.
                                     206

-------
                                  APPENDIX H

                      PRECEDING COAL:dRDF BLEND STUDIES
 INTRODUCTION

     In recent years European and American industries and government agencies
 have become increasingly interested in the commercial conversion of urban
 solid waste into a stoker coal substitute.  Since little of the European
 research data are available, this appendix is limited to the American
 developments.

     The following synopses of American tests conducted between 1972 and 1976
 represent all information that could be gathered on these tests.  While the
 data in these synopses are incomplete, they give a qualitative and quantita-
 tive insight into the recent developments.

 FORT WAYNE MUNICIPAL ELECTRIC STATION TESTS1

     The municipal electric station in Fort Wayne, Indiana, conducted these
 tests with RDF cubettes supplied by the National Recycling Corporation (NRC)
 of the same city.

 Fuel Preparation System

     As shown in the flow diagram of Figure H-l, NRC prepared the cubettes
 from paper and paper board scraps collected from local paper mills.  The NRC
 plant includes a modified John Deere stationary alfalfa cubetter which has a
 nominal capacity of 4.6 to 9.1 Mg/hr (5 to 10 TPH).  This capacity varies
 with the relative density of the scraps being densified.  Up to 20 TPH of
waste can be injected into the cubetter and processed into fuel.  White
metals, yellow metals, glass, and ceramics were excluded from the cubettes
 since their low softening temperatures would likely cause clinkering and make
 the furnace ash handling difficult.  The cubettes were approximately
 1-1/2 x 1-1/2 x 2 in. and were free of metals and glass.  Occasionally
moisture was added to improve the binding qualities of the scrap waste.
     Hollander, H. I. and N. F. Cunningham, "Beneficated Solid Waste
Cubettes as Salvage Fuel for Steam Generation," Proceedings 1968 National
Incinerator Conference, ASME, pp. 75-86.


                                     207

-------
                    NATIONAL  RECYCLING  CORPORATION

                            FORT WAYNE PLANT
STORAGE
FOR
RECYCLE
4
METALS
3 TON/HR
.^ STEEL
MILLS
•^REFINERS


LANDFILL
4 r
K


STORAGE
OR RECYCLE
                                                                              •REFINERS

                                                                              •GLASS PLANTS
                MUNICIPAL
                WASTE
                           30 TON/HR
                INDUSTRIAL
                WASTE
SIZE REDUCER
   1000 HP
                        BULKY
                        WASTE
N3
O
oo

CLASSIFIER
a
SEPARATOR

MAGNETIC
SEPARATOR
i
L^
i
4*
\^
                                                                                             NON FERROUS
                                                                                          METALS, GLASS, CERAMICS,
                                                                                             STONE, DIRT, ETC.
                                                                                                  t
                                                                                                7 TON/HR
                                                            6 TON/HR
                                                            PLASTICS a
                                                            LIGHT PAPER
SEPARATOR
a
SCREENER

»i


ROTARY
SCREEN
                                                                                 1
                                                                                                              4 TON/HR   10 TON/HR
                                                                       4 TON/HR
                                             REJECTS
                                             PAPER, WOOD, RUBBER, PLASTICS, ETC.
                                                                       10 TON/HR
                                                                                                RECOVERED LOW GRADE PAPER
                                                            CUBETTER
                                                  BALER
                                                                                          T
(PRINCIPALLY MUNICIPAL WASTE)


  RECOVERED FIBER STOCK
 ^>^^ «^        i     ••«•
  ( PRINCIPALLY INDUSTRIAL WASTE )
       10 TON/HR
                                                             CUBETTES
                                                       is TON/HR - MUNICIPAL
                                                        5 TON/HR - INDUSTRIAL
                                                 10 TON/HR
                                                     I
                                                 BALED FIBER
                                                   STOCK
                                                                           _STEAM

                                                                            PLANTS
                                                                         PAPERMILLS

                                                                         BOARDMILLS
                                     Figure H-l.
                Flow chart,  and  mass balance  for  cubette  production  at  the
                National Recycling Corporation.

-------
Furnace and Boiler Facility

     The tests were performed in one of the four furnace-boiler units.  Each
unit includes a multiple retort, an underfed retort stoker-fired furnace with
a Sterling boiler, an economizer, and an air preheater.  The boilers produce
2654 kPag (385 psig), 371°C (700°F) steam to drive turbine generators which
have a combined capacity of 40 megawatts.  None of the plant equipment had to
be modified to accommodate the tests.

Test Program

     Approximately 36 Mg of cubettes were burned during two tests.  A fuel
analysis revealed that the Btu content of the as-received cubettes ranged
from 15.9 to 19.8 MJ/kg (6850 to 8530 Btu/lb).  The average fuel properties
are shown in Table H-l.  The tests were run at a 3:1 (coal:cubette) ratio.
While the cubette firing improved the appearance of the fuel bed, the smoke
opacity remained the same as when firing coal only.  Although the tests were
successful, no further cubette firing has been reported.
         TABLE  H-l.   ANALYSIS OF FUEL BURNED IN FORT WAYNE TESTS
        Characteristic
     Content
        Moisture
        Volatile Matter
        Fixed  Carbon
        Ash
        HHV  as Fired
        Sulfur
        Chlorine
        Hemispheric Reducing
            Atmosphere Temperature
      15%
      65%
      14%
       6%
6800 Btu/lb (15.8 MJ/kg)
       0.25%
       0.20%

    1148°C
                                     209

-------
  SUNBURY STEAM ELECTRIC STATION TESTS2

       The Sunbury Steam Electric Station  of  the Pennsylvania Power and Light
  Company conducted these tests  with pellets  prepared by Elo & Rhodes in
  Easton,  Pennsylvania.

  Fuel  Preparation System

       Elo & Rhodes had  ground mixed municipal solid waste to less than 2.5 cm,
  pelletized the ground  particles, and then delivered the pellets to a storage
  bin.  According  to the fuel property summaries in Tables H-2 and H-3, the
  waste benefication reduced the  ash level to 20.6 percent and the moisture to
  10.3 percent.  The high heating value of the pellets is suspect because on a
  moisture  and ash-free  basis, the heating value for the Elo & Rhodes pellets
  is 22.6 MJ/kg (9700 Btu/lb) while the value for the NCRR pellets is 21.2 MJ/kg
  (9100 Btu/lb).

 Furnace and Boiler Facility

      The test boiler at the Sunbury plant was the No.  4 boiler which is
 equipped with three ball-in-tube mills that are normally used for pulverizing
 coal.   While a Raymond Bowlmill was acquired for the pellet pulverizing, the
 ball-in-tube mills had to be used since the bowl mill  did not perform satis-
 factorily.  The maximum generator capacity under normal conditions was 140 Mw,
 but only 10 of the 12 burners were operable, and the boiler ratings could not
 be attained.   The pellets were nominally 5/8 in.  in diameter and 1 in.  long.

 Test Program

     Forty tons  of pellets were burned  during the test.   It was difficult to
 unload the pellets from the hopper cars because they had  packed and bridged
 over the bottom  of the  sliding gates  on the cars.   The pellet  handling
 produced excessive dust in the plant.

     As  the  firing of  the coal-pellet blend stabilized,  the boiler output
 decreased from the 120  Mw with coal-only  firing to 104 Mw with the blend
 firing.   This  drop was  due to  the lesser  pulverizer capacity that  reduced the
 fuel input to  the boiler.   The  pellets  were fired  for  6 1/2 hours.   Although
 the exact  mixture of pellets and coal was not determined, it was estimated
 that the  pellets  accounted for  45 percent of the  total heat input.   Table H-4
 summarizes the monitored gaseous stack  emissions.   While  the  S03 concentration
 doubled,  the total S02  concentration remained the  same.   The  chlorides
 increased  as expected.   In general, the tests were  successful.   However,  the
 Sunbury staff has not indicated  any interest in continuing  the  pellet  firing.
     2Author-unknown, "Final Report on Burning of Processed Refuse Pellets  in
No. 4 Steam Generator on May 29 and 30, 1975" Sunbury Steam Electric Station
report, undated, 9 pages.
                                     210

-------
             TABLE H-2.  ANALYSIS OF FUEL BURNED IN SUNBURY TESTS
          As Received
            Processed Refuse*   Bituminous to Mills
Proximate Analysis

Moisture
Ash
Volatile Matter
Sulfur
Heating Value
Ultimate Analysis
Btu/lb
 Mj/kg
   10.3
   20.6
   51.8
    0.4
6,680
   15.5
     4.7
    16.0
    26.2
     3.1
12,011
    27.9
Nitrogen
Hydrogen
Carbon
Oxygen
Ash
Sulfur
% 0.74
% 5.16
% 39.42
% 31.18
% 23.00
% 0.50
0
4.16
65.81
11.10
14.90
4.03
 * Average  of  two samples

         TABLE H-3.   PARTIAL ANALYSIS OF ASH IN SUNBURY TESTS
                                              Refuse
                                      Bituminous
Silicon Dioxide %
Aluminum %
Iron and Titanium %
Calcium %
Magnesium %
Sodium %
Potassium %
Chlorides %
Remainder %
41.94
6.77
4.89
6.95
1.86
3.24
2.14
0.04
32.17
36.96
8.19
23.48
0.66
0.61
0.39
1.50
trace
28.21
                                             100.00
                                       100.00
                                   211

-------
     TABLE H-4.  FLUE GAS ANALYSIS AFTER THE INDUCED DRAFT FAN AT SUNBURY
Date
5/29/75
5/29/75
5/29/75
Time
8:00 AM
10:00 AM
11:20 AM
1:00 PM
2:00 PM
4:00 PM
SO 3
ppm
Baseline 9.3
1 18.7
2 18.5
S02
ppm
935
994
1008
CL
0
6 & 6
8 & 11
 PIQUA, OHIO, BLACK-CLAWSON TESTS3

      These tests with coal-pellet firings were conducted by Black-Clawson
 Fibreclaim,  Inc., at the Piqua Municipal Power Plant.

 Fuel Preparation System

      The Black-Clawson Fibreclaim Company in Franklin,  Ohio,  prepared  22  tons
 of nominal 3/8-in.-diameter pellets  for  the  Piqua plant tests.  They trucked
 wet fiber that  had  been produced  in  their solid waste processing  plant  to the
 Toledo Alfalfa  Company,  Middletown,  Ohio, where the  fiber was dried in  a  Heil
 Model 125 triple-pass rotary dryer.   Upon the return of the fiber to Franklin,
 Black-Clawson produced the pellets in a  California Century  pelletizer mill.
 A  composite  sample  of the  pellets had the fuel properties listed  in Table H-5.

 Boiler and Furnace  Facility

      Boiler  No.  4 in  the Piqua Municipal Power Plant was  used for the  test.
 Manufactured in 1947  by  the Combustion Engineering Company,  this  boiler has a
 rating of 18.9  kg/sec (150,000 Ib/hr).   Normally,  the maximum steam pressure
 is  3130 kPag (454 psig)  at  440°C  (750°F).  The stoker is  a  Lloyd/Combustion
 Engineeering chain  grate.   The economizer was designed  and  built  by Combustion
 Engineering.  The air preheater was  rated at  a 2.7 GJ/hr  (2.6 MMBtu/hr) input.
Emission  control  devices were not  installed in this  facility.
     3Marsh, Paul, Black-Clawson Fibreclaim, Inc., "Preliminary Test Report
on Handling and Combustion Characteristics of Franklin Pelletized Fuel and
Coal Mixes," November, 1975, 17 pages.
                                     212

-------
               TABLE H-5.  CHARACTERISTICS OF THE PIQUA dRDF
Moisture %
Ash %
Volatile %
Fixed Carbon %
Sulfur %
Heating Value Btu/lb
MJ/kg
16.5
9.02
63.7
10.38
0.22
6382
14.8
Test Program

     The 20 Mg (22 tons) of pellets were mixed with an equal volume of coal
by a bulldozer in the coal yard.  The mixture was then pushed with a front-
end loader into the bucket elevator and then transported by a drag chain
conveyor to the overhead bunkers from which it was metered to the grate
through a weigh lorry.  No modifications were made to the existing coal
handling system.   There were no mechanical problems during the test.  Except
for the normal airflow and bed-depth adjustments for a particular fuel, the
boiler operated as for coal-only firing.

     The blend was approximately 1:1 by volume, and the pellet substitution
provided 20 to 24 percent of the heat generated.  The steam pressure and
temperature were maintained during a 7-hr test.  The plant's normal coal
analysis is shown in Table H-6.

Conclusions

     The 1:1 tests demonstrated the feasibility of using dRDF as a coal
supplement.  Additional tests with a 2:1 mix, along with a detailed combus-
tion and emission assessment, were judged desirable.
                                     213

-------
             TABLE H-6.   ANALYSIS  OF  THE COAL  CO-FIRED IN PIQUA, OHIO
            Parameter                As  Received              Dry Basis
Moisture Percent
Volatile Matter Percent
Fixed Carbon Percent
Ash Percent
Sulfur Percent
Btu/lb 11,
MJ/kg
Ash from the coal
4
37.51
46.70
11.79
3.36
680
27.2

Initial Deformation Temperature
Second Softening Temperature
Fluid Temperature







12


1263°C, 2306°F
1318°C, 2405°F
1471°C, 2680°F

39.07
48.65
12.28
3.50
,170
28.3




 WRIGHT-PATTERSON AFB BLACK-CLAWSON TESTS4

      These  tests were conducted  by Black-Clawson  Fibreclaim, Inc., at the
 Wright-Patterson Air Force  Base  central heating plant, Building 770.

 Fuel  Preparation System

      As  for the  Piqua Tests,  Black-Clawson similarly prepared pellets for the
 Wright Patterson AFB tests.   Typical properties of  the latter pellets are
 listed in Table  H-7.

 Furnace  and Boiler Facility

      The central heating plant contains two Edgemoor Ironworks (36,300 kg/hr)
 (80,000  Ib/hr) boilers that produce 862 kPag  (125 psig) of saturated steam.
 Installed in 1956, these boilers are fired with Detroit Rotograte spreader
 stokers.  The emission control equipment includes a cyclone separator for
 reinjected  fly ash,  multiclones  for coarse particulate control, and an
 installed,  but inoperative, electrostatic precipitator.

Test Program

     The pellets  trucked to Wright-Patterson AFB were placed in hopper cars
for delivery to  the  rail car  dumper and then into the power plant hoppers.
     ^Jackson, J. W., "A Bioengineering Study of Emissions from RDF,'
UASFEHL, McClellan AFB, ADA024661, 1976.
                                      214

-------
     TABLE H-7.   PROPERTIES  OF  THE  dRDF AND  STOCK USED  IN THE WPAFB TEST
                                   Feed Stock                  dRDF
                                   As Received              As Received
     Energy - Btu/lb                  3300                     5800
               MJ/kg                     7.6                     13.5
     Moisture Percent                   55.5                     21.5
     Ash Percent                         6.2                     11.2
     Chloride Percent                    0.07
     Sodium                              0.03
     Softening Temperature       1266°C (2310°F)
     Density Kg/m3                      46                       58.2
     Sulfur Percent                                               0.12
     Volatile Percent                                            60.2
     Fixed Carbon                                                10.46
Two methods of mixing the pellets with coal were used to investigate the
mixing behavior.  In the first method, a pile of blended fuel was prepared in
the coal yard by having a car unloading crane first alternately scoop and
deposit dRDF and coal and then lift and drop the deposit for further mixing.
The mixture was then loaded into a hopper car and delivered to the coal
bunker.  In the second method, alternate scoops of the two fuels were simply
loaded into the hopper car with the fuels being mixed as they were unloaded
from the car and passed through the materials handling system to the coal
bunker.  Both methods produced a visibly well-mixed fuel.  After the blends
were loaded onto the coal cars, they were covered to prevent rain from
reducing them to their original pulped form.  Calculated blend properties are
presented in Table H-8.

     Two blend ratios were burned:  1:1 (coal:dRDF) by volume for 34 hours
and 1:2 by volume for 6 hours.  Although the emissions were monitored, they
were uncontrolled since the installed electrostatic precipitator had been
inoperative for several years.

     As indicated by the summary of the stack emissions in Table H-9, the
results were encouraging.

     In contrast to the Sunbury tests, S02 was reduced in the Wright-Patterson
AFB tests.  Moreover, the unburned hydrocarbons were drastically reduced in
the latter tests.  Halides and heavy metal emission test data, however, could
not be interpreted conclusively, although increases were detected.
                                     215

-------
           TABLE H-8.  AVERAGE PROPERTIES OF WPAFB dRDF
     Properties                 Coal         1:1 Mix        2:1 Mix
Heating Value MJ/kg
(As Fired) Btu/lb
Moisture Percent
Bulk Density Ib/CF
kg/m3
Ash Percent
Sulfur Percent
Chlorine Percent
Fixed Carbon Percent
Volatiles Percent
Hydrogens Percent
Carbon Percent
30.0
12,900
4.88
53.1
851
7.5
0.67
0.09
55.6
31.9
4.9
73.3
26.3
11,327
7.73
44.2
708
9.09
0.53
0.14
43.8
39.3
5.09
63.5
22.1
9,518
14.01
42.4
679
8.86
0.43
0.14
36.5
40.6
4.91
56.5
              TABLE H-9.  STACK EMISSIONS (COMPARED TO COAL)
  Emissions                1:1 Mix                    1:2 Mix
Particulates          Unchanged                Highly variable, on
                                               the average, unchanged

  S02                 Reduced by               Reduced by
                      approximately 50%        approximately 60%

  NO                  Drastically reduced      Drastically reduced
    X                 by approximately 80%     by 95+ %
                                 216

-------
      The boiler operation was  normal during the 1:1 mix test after a few
 minor control adjustments.   However, during the 1:2 mix test, a loss of
 control over the fuel distribution  in  the boiler caused the fire to concen-
 trate near the rear of the  firebox.  In  the automatic scale operation, some
 fuel segregation produced a higher  concentration of pellets in the middle
 spreaders.   Clinkers formed near  the heavy fire concentration at the rear of
 the boiler.   The bottom ashes  produced by the 1:1 blend firing differed
 little from the bottom ashes produced  by the coal-only firing.  With the 2:1
 mix,  however,  the ashes were frequently  fused into clinkers.

      A furnace inspection after the test revealed that some slag had
 deposited on the rear wall  refractory  and on the rear portions of the side
 walls.   These  deposits may  have been caused by the poor fuel distribution
 during the 1:2 blend tests.  Small  slag  cones which had formed at the base of
 the slag deposit indicated  that ash had  melted.  Another evidence of fouling
 was the formation of flake-like deposits on the fire side of the wall tubes.
 These deposits generally sloughed off  the tubes when the boiler cooled.
 After the grates were swept clean, black stains were found with those close
 to  the grate air holes being the  most  prominent.

 Conclusions

      While  the 1:1 mixture  firing was  generally satisfactory, the 2:1 mixture
 firing caused  poor  fuel distributions  which likely could have been solved by
 minor facility modifications.  A  larger  and more dense pellet should be
 tested  since its handling characteristics might solve some of the fuel
 distribution deficiencies at the  higher  coal:KDF ratio.  The potential for
 scaling and  waste should be evaluated  further.  The stack emission changes
 were  generally acceptable,  and the decrease in hydrocarbons was especially
 noteworthy.  Lead emissions, which increased significantly during the tests,
 were  generally submicron aerosols that indicated the deposition of lead
 vapors.   The NOX and  chloride-fluoride emissions in both the 1:1 and the 2:1
 blend tests  were significantly greater than those in the coal-only firings.

 SANDWELL  INTERNATIONAL TESTS5

      These tests  with fluff and densified RDF firings were conducted by
 Sandwell  International,  Inc., at  a facility owned by the Eugene Water &
 Electric  Board,  Eugene,  Oregon.

 Fuel  Preparation  System

      The Vista Fiber  and  Chemical Company in Los Gatos, California, produced
 21 tons of nominal 1-in.-diameter dRDF pellets for these tests.  The material
 density of the individual pellets was  881 kg/m3 (55 lb/ft3), and the bulk
 density was  approximately 593 kg/m3 (34  lb/ft3).  The average calorific
value of  the fuel was  12.0 MJ/kg  (5156 Btu/lb) as received and 15.0 MJ/kg
 (6436 Btu/lb)  on  a dry basis.
     5Sandwell International, Inc., "Eugene Water & Electric Board, Eugene,
Oregon, Solid Waste Fuel Modifications, Second Series Burn Tests—Final
Report," Report W3508/2, December 23, 1974.

                                      217

-------
  Boiler and Furnace Facility

      The facility consists of three wood waste and bark boilers of unreported
  design or manufacture.  Testing was done in the No. 3 boiler.  This boiler
  has a nominal capacity of 19.5 kg/sec  (155,000 Ib/hr) of steam and produces
  427°C  (800°F) superheated steam.  The boiler has an air preheater and is
  fired with a spreader stoker/traveling grate combination.

  Test Program

      The 21 tons of pellets were fed into the furnace during a 90-minute
  period or at a feed rate equivalent to 304 Mg/day (335 TPD).   As the furnace
  exit gas temperatures increased from 593°C to 760°C (1100°F to 1400°F), the
  speed of the feeders was reduced.  The steam output reached 18.9 kg/sec
  (150,000 Ib/hr) at 1169 kPag (300 psig) and could have attained a higher
  output if additional fuel had been available.  Although the emissions during
  the dRDF tests were not analyzed, the fluff RDF tests indicated that the
 particulate emissions, especially the fine particulate content, were generally
 greater than those from coal only.

 WISCONSIN SOLID WASTE RECYCLING AUTHORITY TESTS6

      Of the several dRDF tests sponsored by the Wisconsin Solid Waste
 Recycling Authority,  three are synopsized as follows:

 University of Wisconsin Tests

      These tests with dRDF pellets were conducted at the University of
 Wisconsin heating plant.   For these tests,  Gruman Eco  Systems in St.  Louis
 prepared  the  pellets  by densifying in a Sprout Waldren pelletizer the RDF
 produced  at the  municipal  RDF pilot plant.   The pellets were  3/4 in.  in
 diameter  and  up  to 3  in. long with a bulk density of 62.4 kg/m3 (39 lb/ft3)
 and a heating value of 14  MJ/kg  (6000 Btu/lb).   The test furnace-boiler was a
 Wickes  waterwall furnace equipped with a Detroit Vibragrate stoker and a
 water tube boiler rated at 5.7 kg/sec (45,000 Ib/hr) of 1862  kPag (125 psig)
 saturated  steam.   The underfire  air was delivered to the grate through a
 5-compartment wind box.  A baffle chamber provided the means  for some removal
 of  particulate emissions.

     Three volumetric coal:dRDF  blends  were  burned during the test:   1:1,
 1:3, and 0:1  (100 percent  pellets).   In each test,  the furnace-boiler per-
 formed satisfactorily under automatic control with the facility operating
 between one quarter and one half  of  the  design  load.   The burnout was
 excellent, no emissions were visible; and no clinkers  were formed.  Although
 the overfire air  was  not adjusted,  the  underfire  air was throttled so that it
was introduced primarily through  the  first zone  of the 5-compartment  wind
box.  No emissions data were collected.
     6Private Correspondence between Warren Porter of Wisconsin  Solid Waste
Recycling Authority and H. G. Rigo, 1976.


                                      218

-------
Appleton Division Tests

     These tests were conducted in the boiler house of the Appleton Division.
A blend of dRDF and bark was fed through a gravity feed chute into a Dutch
oven-fired, refractory-walled bark burner.  With the dRDF fed at a rate of
2.73 kg/hr (3 tons/hr), the furnace heat release rate rose above the design
limit after 2 hours of firing, and consequently the blend feeding had to be
stopped.  A pile of unburned material in the middle of the furnace was high
enough to block the overfire air ports and to cause poor combustion, smoking,
and clinkering.  However, when the test was repeated at a slightly lower feed
rate, the combustion was satisfactory.

     During the second test, a single stack test had particulate emission
rate of 94 mg/MJ (0.219 Ib/MBtu).  Most of the ash in the dRDF remained in
the boiler as bottom ash.  There was no visible plume.

Menasha Paperboard Mill Tests

     These tests were conducted in the boiler house of the Menasha Paperboard
Mill.  For these tests, 11.6 MJ/kg (5000 Btu/lb) dRDF prepared by the Vista
Fiber and Chemical Company was blended with 31.3 MJ/kg (13,444 Btu/lb) coal
in the existing fuel handling system.  The coalrdRDF blend, which had a
15 percent dRDF substitution rate, was fired in a spreader stoker boiler.
The plant evaporation rate decreased from 9.75 to 9.05 kg steam/kg fuel as
the blend entered the boiler.

     The plant power chief stated that although the blend firing appeared
feasible since it required no feed equipment changes, its particulate
emission was high.

CHANUTE AIR FORCE BASE TESTS

     The U.S. Army Construction Engineering Research Laboratory conducted
these tests with pellets at Chanute Air Force Base for the Naval Facilities
Engineering Command.  These tests provided a unique experience since they
were conducted in a chain grate stocker boiler and most of the pellets had
deteriorated into fines.

Fuel Preparation System

     For these tests, the Vista Fiber and Chemical Company of Los Gatos,
California, prepared 181 Mg (200 tons) of dRDF pellets from mixed municipal
solid waste.   In this preparation, the waste was course shredded, magneti-
cally separated, fine shredded, reshredded, air classified, screened, and
pelletized in a California Pellet Mill pelletizer.
     7Letter to Warren Porter from C. Eaton on December 27, 1976, concerning
the test.
                                      219

-------
      After the 181 Mg (200 tons)  of pellets had aggregated at the  California
 plant, they were loaded into box cars which were routed to Charleston,  South
 Carolina.   Then, after a delay with the pellets unprotected,  the box cars
 proceeded  to Rantoul, Illinois, where the pellets were unloaded and  delivered
 to Chanute Air Force Base.  Upon their arrival at the boiler  site, the  pellets
 had so deteriorated that the RDF was mostly fines,  and the remaining pellets
 had rough  sides rather than the smooth sides characteristic of newly formed
 pellets.

      When  the 181 Mg (200 tons) of  deteriorated dRDF  were  placed in  bunkers
 to a depth of 7.9 meter  (26 feet),  the fuel bridged and rat holed.   Since the
 sides of the rat hole were stable,  bins were unloaded by flushing  the bunkers
 with a fire hose.   After 3 weeks  of storage,  the dRDF ignited by spontaneous
 combustion.   While bunker flooding  with water extinguished the fire,  much of
 the remaining dRDF had deteriorated further.

 Boiler and Furnace Facility

      The Chanute Air  Force Base heating plant houses  several  low-pressure
 saturated  steam boilers  fired with  chain grate stokers.  The  wind box is
 unsegmented,  and the  front and  rear overfire  air jets  are  modulated  as  a
 battery.   The  overhead parabolic bunkers feed a weigh  lorry.

Test Program

     After the bunkers were unloaded,  some  pellets  were  salvaged for  short-
duration tests.  During  these tests  (fired  at  100 percent  pellets),  the chain
grate  could not  be  fed fast enough  to  maintain load.  Also, since the fire
filled the front of the  furnace, the pellets  burned too  rapidly.  No  further
information on these tests was  acquired.
                                     220

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
 1. REPORT NO.
  EPA-600/2-80-095
                                                           3. RECIPIENT'S ACCESSIO(*NO.
 4. TITLE ANDSUBTITLE

  A FIELD TEST USING  COAL:dRDF BLENDS IN SPREADER
  STOKER-FIRED BOILERS
             5. REPORT DATE
              August 1980 (Issuing Date)
             6. PERFORMING ORGANIZATION CODE
 7. AUTHOR(S)
  Gerald H. Degler, H.  Gregory Rigo, and
  Boyd T. Riley, Jr.
                                                           8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
  Systems Technology  Corporation
  245 North Valley Road
  Xenia, Ohio  45385
             10. PROGRAM ELEMENT NO.
               C73D1C
             11. CONTRACT/GRANT NO.

               68-03-2426
 12. SPONSORING AGENCY NAME AND ADDRESS
  Municipal Environmental Research Laboratory—Cin.,OH
  Office of Research  and  Development
  U.S.  Environmental  Protection Agency
  Cincinnati, Ohio  45268
             13. TYPE OF REPORT AND PERIOD COVERED
               Final (6/76 - 7/78)
             14. SPONSORING AGENCY CODE
               EPA/600/14
 15. SUPPLEMENTARY NOTES
  Project Officer:  Carlton C.  Wiles  513/684-7871
16. ABSTRACT
  This program was conducted to characterize  and demonstrate the technical,  economic,
  and environmental feasibility of combustion densified forms of refuse derived fuel
  (dRDF)  blended with  coal in spreader stoker-fired boilers.  A total  of 258.5 Mg
  (285 tons) of pelletized 1/2-inch-diameter  x 3/4-inch-long dRDF was  co-fired with
  coal in 2.7 x 7.5 kg/sec (60,000 Ib/hr) and 3.6 x 10 kg/sec (75,000  Ib/hr)  of
  1.03 MPa (150 psig)  saturated steam.  The results indicate that coal:dRDF  blends up
  to  1:2  can be handled and burned in conventional spreader stoker-fired boilers with-
  out major equipment  modification.  As more  dRDF was substituted for  coal,  the flame
  volume  increased, the opacity decreased, the fly ash carbon burnout  improved, and
  the turndown ratio of boiler operation increased.  The emissions from the  blend
  firing  decreased slightly in mass flux, dropped significantly in particulate size
  and stack opacity, and had satisfactory particulate resistivities.
 7.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
 Refuse
 Energy
 Combustion
 Air pollution
 Evaluation
                                              b.lDENTIFIERS/OPEN ENDED TERMS
 Municipal solid wastes
 Densified refuse derived
   fuels
 Resource recovery
                           c. COSATI Field/Group
       13B
 3. DISTRIBUTION STATEMENT


 RELEASE TO  PUBLIC
19. SECURITY CLASS (ThisReport)
     UNCLASSIFIED
21. NO. OF PAGES
       235
20. SECURITY CLASS (This page)
     UNCLASSIFIED
                           22. PRICE
EPA Form 2220-1 (9-73)
                                             221
           •it U.S. GOVERNMENT PRINTING OFFICE:  1980--657-165/0071

-------