EPA
ERDA
USBM
U.S. Environmental Protection Agency Industrial Environmental Research      EPA~600/7"7o™073
Office of Research and Development  Laboratory                _  *| H f\~rQ
                 Research Triangle Park, North Carolina 27711 ApTll 1 9/O
U.S. Energy Research
and Development Administration
Fossil Energy
Washington, D.C. 20545
U.S. Bureau of
Mines
Coal Preparation and Analysis Laboratory
Pittsburgh, Pennsylvania 15213
             PROCEEDINGS
             OF THE ENGINEERING

             FOUNDATION CONFERENCE
             ON CLEAN COMBUSTION OF COAL
             Interagency
             Energy-Environment
             Research and Development
             Program Report

-------
                 RESEARCH REPORTING SERIES


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

    1. Environmental Health Effects Research

    2. Environmental Protection Technology

    3. Ecological Research

    4. Environmental Monitoring

    5. Socioeconomic Environmental Studies

    6. Scientific and Technical Assessment Reports (STAR)

    7. Interagency Energy-Environment Research and Development

    8. "Special" Reports

    9. Miscellaneous Reports

This report has been assigned to the  INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND  DEVELOPMENT series. Reports in this series result from the
effort funded  under the 17-agency Federal Energy/Environment Research and
Development Program. These studies  relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to  assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport  of energy-related pollutants and their health and ecological
effects;  assessments of, and development of, control technologies  for energy
systems; and integrated assessments of a wide range of energy-related environ-
mental issues.
                        EPA REVIEW NOTICE


This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect
the views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.

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

-------
                                                   EPA-600/7-78-073
                                                           April 1978
    PROCEEDINGS OF THE  ENGINEERING
        FOUNDATION CONFERENCE ON
         CLEAN  COMBUSTION  OF COAL
                             Victor S. Engleman
                            Conference Chairman

                           Science Applications, Inc.
                            1200 Prospect Street
                           La Jolla, California 92038


                        EPA Purchase Order DA-7-03662B
                         Program Element No. EHE624A


                        ERDA Grant No. EF-77-G-01-6003
                       USBM Purchase Order No. PO172317


                        EPA Project Officer: G. Blair Martin
                         ERDA Project Officer: A. Macek
                      USBM Project Officer: A.W. Deurbrouck

                    Industrial Environmental Research Laboratory
                      Office of Energy, Minerals, and Industry
                        Research Triangle Park, N.C. 27711


                               Prepared for

Coal Preparation and Analysis Laboratory                 Office of Research and Development
       U.S. Bureau of Mines                        U.S. Environmental Protection Agency
       Pittsburgh, PA 15213                            Washington, D.C. 20460

                               Fossil Energy
                  U.S. Energy Research and Development Administration
                            Washington, D.C. 20545

-------
                                CONTENTS

SESSION

          INTRODUCTION	   1
               Conference Chairman:  V.S. Engleman,
                          Science Applications, Inc.

  IA      TECHNICAL, ECONOMIC AND ENVIRONMENTAL PROBLEMS
          IN CLEAN COMBUSTION OF COAL	   3
               Session Chairman:  Andrej Macek, U.S. ERDA

          TECHNICAL AND ECONOMIC PROBLEMS FOR CLEAN
          COMBUSTION OF COAL	   5
               S.I. Freedman, U.S. ERDA

          COAL COMBUSTION AND FUTURE EMISSION REGULATIONS 	  21
               Thomas Schrader, U.S. EPA

          RECENT DEVELOPMENTS IN COAL COMBUSTION TECHNOLOGY ....  27
               James I. Joiibert, Pittsburgh Energy Research
                          Center
  IB      STRATEGY AND APPROACH TO SPONSORED R&D	U3
               Session Chairman:  Andrej Macek, U.S. ERDA

          STRATEGY IN COAL PREPARATION RESEARCH PLANNING	1*5
               W.E. Warnke, U.S. Bureau of Mines

          STRATEGY AND APPROACH TO ERDA RESEARCH AND
          DEVELOPMENT ON CLEAN COMBUSTION OF COAL	^9
               S. William Gouse, U.S. ERDA

          EPA R&D PROGRAM RELATING TO CONVENTIONAL COAL
          COMBUSTION	63
               Frank T. Princiotta, U.S. EPA
  II      PRECOMBUSTION PROCESSES 	  77
               Session Chairman:  P. Stanley Jacobsen, Colorado
                          School of Mines Research Institute

          COAL PREPARATION HISTORY AND DIRECTION	79
               Robert L. Llewellyn, Roberts & Schaefer Company

          THE INFLUENCES OF MINING PRACTICES ON COAL
          PREPARATION	83
               C.A. Goode, Bureau of Mines

          CURRENT COAL PREPARATION RESEARCH AND DEVELOPMENT ....  87
               Richard P. Killmeyer, Jr., Bureau of Mines

          COAL TRANSPORTATION IN 1985	91
               David J. Hoexter, U.S. Department of Commerce

-------
ii                      CLEAN COMBUSTION OP COAL


                          CONTENTS  (Continued)

SESSION                                                            FAGE_

  II      PRECOMBUSTION PROCESSES (Continued)

          COAL DESULFURIZATION TEST PLANT STATUS - JULY  19TT-  ...   97
               L.J.  Van Nice, M.J.  Santy, E.P.  Koutsoukos,
               R.A.  Orsini and R.A.  Meyers, TRW Systems  and
                          Energy

          STATUS AND PROBLEMS IN THE DEVELOPMENT OF HIGH
          GRADIENT MAGNETIC SEPARATION  (HGMS) PROCESSES
          APPLIED TO COAL BENEFICIATION ...... ,  ........  109
               Y.A.  Liu and C.J.  Lin, Auburn University

          A THEORETICAL APPROACH TO WASHABILITY CURVES IS
          COMPARED TO THE OTISCA PROCESS SEPARATION OF
          FINE COAL  ........................  131
               D.V.  Keller, Jr.,  Otisca Industries, Ltd.
  Ill     COMBUSTION TECHNOLOGY ..................
               Session Chairman:   G.  Blair Martin, U.S. EPA

          SOME CHARACTERISTICS  OF  COAL  COMBUSTION SYSTEMS  .....
               Janos M.  Beer, Massachusetts  Institute of
                          Technology

          POLLUTANT FORMATION DURING  COAL COMBUSTION ........  171
               M.P.  Heap and R.  Gershman, Energy and
                          Environmental Research Corporation

          THE DUAL REGISTER  PULVERIZED  COAL  BURNER:  FIELD
          TEST RESULTS .......................  185
               E.J.  Campobenedetto , Babcock  & Wilcox Company

          COAL-OIL MIXTURE COMBUSTION IN BOILERS - AN UPDATE.  .  .  .  193
               Sushil K.  Batra,  New England  Power Service Company

          STOKERS  FOR INDUSTRIAL BOILERS:  ASSESSMENT OF
          TECHNICAL,  ECONOMIC, AND ENVIRONMENTAL FACTORS ......  205
               Robert D.  Giammar,  Battelle,  Columbus Laboratories

          INITIAL  OPERATION  OF  THE 30 MWe RIVESVILLE MULTICELL
          FLUIDIZED BED  STEAM GENERATION SYSTEM ..........  219
               Robert L.  Gamble and Newton G. Watt is,
                          Foster Wheeler Energy Corporation

          DEVELOPMENT OF AN  EFFICIENT SOLIDS FUEL BURNER ......  227
               Norman A.  Lyshkow,  Combustion Equipment Associates,
                          Inc.

-------
                                                                    iii


                          CONTENTS (Continued)

SESSION                                                            PAGE

  IV      POSTCOMBUSTION CLEANUP	235
               Session Chairman:  Sidney R. Orem,
                          Industrial Gas Cleaning Institute

          ELECTROSTATIC PRECIPITATION STATE OF THE ART	237
               R.S. Atkins and D.V. Bubenick, Research-Cottrell

          STATE OF THE ART — FABRIC FILTRATION	255
               Richard L. Adams, Wheelabrator-Frye, Inc.

          STATUS OF FLUE GAS DESULFURIZATION, THE FEDERAL
          RESEARCH, DEVELOPMENT AND DEMONSTRATION PROGRAM 	 273
               Julian W. Jones and Michael A. Maxwell, U.S. EPA

          STATUS OF FLUE GAS TREATMENT TECHNOLOGY FOR CONTROL
          OF NOX AND SIMULTANEOUS CONTROL OF SOX AND NOX	293
               J. David Mobley and Richard D. Stern, U.S.  EPA
  V       WHERE DO WE GO FROM HERE?	311
               Session Chairman:  Victor S. Engleman,
                          Science Applications, Inc.

          THE NATIONAL ENERGY PLAN	313
               C. William Fischer, Department of Energy

          THE OUTLOOK FOR COAL THROUGH 1985	323
               Robert L. Major, AMAX Coal Company

          WHERE DO WE GO FROM HERE IN R&D?	  335
               S. William Gouse, U.S.  ERDA

-------
CLEAN COMBUSTION OF COAL

-------
                             INTRODUCTION
     Coal is a major energy resource for the United States.   The use of
coal as an energy source will increase in importance over the intermediate
term.  While synthetic fuels from coal will be needed for specific appli-
cations, direct combustion of coal will probably continue to be the most
efficient and most economical use of coal.  However, the environmental
problems associated with coal combustion must be solved economically and
efficiently at the same time for the continued growth of utilization of
this vital energy resource.

     Clean combustion of coal must address issues of air quality, surface-
and groundwater quality and solid waste management.   Control technology
for clean combustion of coal must include precombustion, combustion, and
postcombustion processes.  Air quality impacts from coal combustion
include gaseous and particulate combustion product emissions as well as
potential fugitive emissions from handling and storage.  Potential
pollutant emissions include NOX, SOX, particulates and trace elements.
Generally hydrocarbon and carbon monoxide emissions  are low from large
coal combustors but new combustion- technologies should be checked for
emissions of these pollutants.  Water quality impacts would result
primarily from leaching of chemical compounds from the solid residues of
the combustion process;  this problem therefore goes hand in hand with
the solid waste management problem.  As a minimum, the ash in the coal
must be managed.  In technologies which produce additional solid waste,
(precombustion, combustion, or postcombustion) an additional solid waste
management problem is created.

     Energy efficiency, availability, and economics  must be considered
in establishing acceptable methods for clean combustion of coal.  One
could postulate the emissions limit going to zero while paying the price
of the energy efficiency going to zero as well.  Therefore,  it is not
only a matter of technical, economic, and environmental considerations,
but also policy considerations.  This is  recognized  in the program for
this conference, and while the engineering aspects of coal combustion
are emphasized, the vital importance of energy policy and its impact on
clean combustion of coal is highlighted.

     Special thanks to the Organizing Committee for their role in
establishing the program and helping to arrange for speakers.

             Andrej Macek - ERDA .
             Albert W. Deurbrouck - Bureau of Mines
             G. Blair Martin - EPA
             Sidney R. Orem - Industrial Gas Cleaning Institute
             Joseph W. Mullan - National Coal Association

-------
2                      CLEAN COMBUSTION OF COAL


     The suggestions provided "by the Advisory Committee were invaluable
in  selecting the highly qualified slate of speakers and the equally
qualified group of conference participants.  Their efforts are grate-
fully acknowledged.

             E.K. Bastress - ERDA
             R. Beck - ERDA
             J.O. Berga - National Research Council
             R.A. Carpenter - National Research Council
             D.E. Gushee - Congressional Research Service
             R. Hangebrauck - EPA
             G.R. Hill - EPRI
             G.A. Mills - ERDA
             E. Plyler - EPA
             W.E. Warnke - Bureau of Mines
     The experienced staff of the Engineering Foundation Conferences
with guidance and assistance both before and during the conference were
instrumental in its success.  Special notes of thanks to

              Sandford S. Cole, Conferences Director
              Harold Comerer
              Dean Benson.

In addition, the staff at the conference site at Franklin Pierce College
provided invaluable assistance during the course of the conference.

     The publication of the proceedings are due in large part to the
patience and perseverance of Ms.  Billie St. Pierre who saw that the manu-
scripts were put in their proper final form.

-------
          SESSION IA - TECHNICAL, ECONOMIC AND ENVIROMMEHTAL
                 PROBLEMS IN CLEM COMBUSTION OF COAL

              SESSION CHAIRMAN:   ANDREJ MACEK, U.S.  ERDA
     In broad terms, we will have to learn to burn coal under very
restrictive conditions.  The restraints will be imposed by national
policy considerations and environmental requirements.   This opening
session will include a general description of coal combustion technol-
ogies which are, or might become, candidates for clean combustion, and
of the problems associated with these technologies.  Policy considera-
tions, which are also a key element in clean combustion of coal,  will
be included in the closing session of the conference.   Two of the papers
in this session describe four candidate technologies:   fluidized  bed
combustion of coal, combustion of solvent-refined coal, combustion of
coal/oil mixtures, and combustion of coal for direct-powered magneto-
hydrodynamic generators.  In addition, there is a discussion of the
potential environmental constraints in the forseeable  future.

-------
CLEAN COMBUSTION OF COAL

-------
     TECHNICAL AND ECONOMIC PROBLEMS FOR CLEAN COMBUSTION OF COAL

                                  by

                            S. I. Freedman
     Assistant Director, Combustion and Advanced Power Development
             Division of Coal Conversion and Utilization
                             Fossil Energy
          U.S. Energy Research and Development Administration
                           Washington, D.C.
INTRODUCTION

     Ever escalating demand for limited and dwindling supplies of con-
ventional fuel resources has been recognized as a key factor in pre-
cipitating the worldwide financial and energy crises.  The gravity of
these crises has forced nations around the world to look for ways and
means to improve their energy supply base to meet the ever increasing
demands.  To meet the challenge, President Carter declared the "moral
equivalent of a war" for energy independence.

     Although nuclear technologies might satisfy a substantial portion
of world energy needs in a distant future, there exists an immediate
necessity for the development of energy technologies to conserve and
efficiently use a wide base of natural resources.

     Current estimates indicate the U.S. coal reserves, which are econ-
omically mineable with conventional technology, at about 600 billion
tons and the overall coal resource at 3.2 trillion tons.  Depending on
the efficiency of mining, population growth, and growth in energy use
per capita, and efficiency in end use, we have enough coal to supply
our energy needs for several hundred years; sufficient, it is hoped, to
carry us through a transition to a non-fossil fuel energy source.  Ad-
vances in coal mining technologies and increased coal price can increase
the estimate for coal which can be economically produced considerably
beyond the 600 billion ton figure.  Recognizing this abundance of coal,
the U.S. Energy Research and Development Administration (ERDA) is greatly
emphasizing the development of technologies to utilize efficiently a
wide variety of coals in an environmentally acceptable manner.

     The ongoing ERDA efforts in the development of fossil energy tech-
nologies are a continuation of the programs of its predecessor organiza-
tions, the U.S. Department of Interior's Office of Coal Research and
Bureau of Mines.  These technologies include:  (1) Coal gasification,
(2) Coal liquefaction, (3) Direct combustion of coal, and (4) Advanced
power systems using coal and coal-based fuels.

     ERDA is supporting the development of these technologies by funding
various projects ranging in size from exploratory research to pilot and
demonstration projects.

-------
  6                     CLEAN COMBUSTION OF COAL


      ERDA's main  objective is to bring up the technologies to technical
 maturity  for  commercial implementation.  Although coal gasification and
 liquefaction  are  of  considerable importance, this paper emphasizes the
 ERDA efforts  in direct combustion technologies.

      Direct combustion of coal in boilers and furnaces is a well estab-
 lished practice for  heat and power generation.  Only a certain fraction
 of the U.S. coal  reserves, with low-sulfur content, termed compliance
 coal,  can be  burned  directly in conventional furnaces to meet the envir-
 onmental  regulations if burned in conventional burners.  As Dr. James
 Schlesinger aptly points out, "Americans want a clean environment and do
 not want  to turn  the country into something equivalent to Pittsburgh in
 the 1930's."   Of  course, all modern coal fired equipment is built for
 high combustion efficiency and includes a bag house or precipitator to
 reduce the emission  of particulates, thereby avoiding the two most obvious
 sources of pollution from coal fired equipment, soot and fly ash.  To
 achieve increased coal utilization while meeting stringent environmental
 regulations,  the  Government is directing the development of promising
 coal utilization  technologies.

      These technologies can be categorized as follows:

      • Precombustion treatment of coal for ash and sulfur removal.
      • Post  combustion treatment for S02 removal.
      • In-situ S02  removal at the source during coal combustion.

     A brief  description of these technologies follows.

 Precombustion Treatment of Coal

     Precombustion treatment of coal includes a number of physical and
 chemical approaches  such as:

     •  Physical  cleaning.
     •  Chemical  cleaning.
     •  Solvent refining.

     Precombustion treatment,  essentially aimed at removing ash (mineral
matter) and other obnoxious constituents such as sulfur in its pyritic
form,  increases the  specific heating value (Btu/lb) and reduces the sulfur
content of some coals to an environmentally acceptable level.

     Physical and chemical approaches for cleaning coal are in different
stages of development:   A full-scale plant to demonstrate a multi-stream
physical cleaning concept  is being built at Homer City, Pennsylvania, and
is scheduled  to begin operation this year (1977).  A pilot plant to test
sulfur removal through chemical leaching is being built in California.
Commercial implementation of this technology, depending on favorable
pilot plant results, can be anticipated in the early 1980's.  The impact
of these technologies is to increase the availability of compliance coals
by a maximum of approximately 175 million tons per year.  This still
leaves the majority  of American coal being mined as non-compliance coal.
Physical cleaning is estimated to cost about $1.50 to $4.50 per ton while
chemical cleaning would cost about $10-$15 per ton.

-------
                     TECHNICAL AND ECONOMIC PROBLEMS
     Some variable amount of our existing coal reserves can be considered
as compliance coal, based on the S02 emissions criteria that apply locally,
and can be used in existing equipment.  The Federal EPA rules on SC>2 emis-
sions for new sources are quite definite, 1.2 Ibs of S02 per million Btu.

     Low-sulfur coal may be defined as coal that occurs naturally with
a sulfur content low enough that the coal can be burned without violating
applicable S02 emission standards.  As S02 emission standards vary, the
same coal may be considered compliance for one application but not for
another.  The extent to which low-sulfur coal is used to meet national
ambient air quality standards (NAAQS) and New Source Performance Standards
(NSPS) will depend primarily on the availability.  It should also be noted
that S0-j in combustion gases, which accompanies S02> enhances the col-
lection efficiency of electrostatic precipitators, so the use of low-
sulfur coal at existing facilities necessitates modifications of electro-
static precipitators to control particle emissions.  Low-sulfur coals
with low ash fusion temperatures can cause fouling in boilers, reducing
capacity and increasing maintenance costs.  This has happened in several
instances when western coals were used in boilers designed for eastern
coals.

     The availability of low-sulfur coal depends on the expeditious dev-
elopment of our domestic resources and of the transportation network re-
quired to get the fuel to the markets.  In 1974, almost 390 million tons
of coal were consumed by electric power plants; about half would comply
with S02 emission regulations in effect on July 1, 1975.  In 1980, utility
coal demand is projected to be about 620 million tons with an 8% annual
growth rate.  In an October 1974 EPA report, low sulfur coal production
in 1980 was projected to supply less than 44% of the 620 million ton
demand.

     The production capacity for low-sulfur coal and its use will depend
on many factors, including:

     (1)  The relative economics of low-sulfur coal in comparison with
          other control approaches.

     (2)  The resolution of disputes over compliance schedules and
          state sulfur regulations.

     (3)  Political/economic decisions concerning low-sulfur coal
          recovery and distribution.

     (4)  The compatability of low-sulfur coal usage with existing
          boilers and electrostatic precipitators.

     The costs of complying with SOn standards by using low-sulfur coal
include:  (1) The increased cost of the low-sulfur coal, (2) Transporta-
tion cost, and (3) The cost of power plant modifications which may be
necessary.  While low-sulfur coal is generally more expensive than high-
sulfur coal, the price differential currently is highly variable.
Transportation cost often causes significant increases in low-sulfur
coal costs.  Another factor to consider is that about one and one-half
tons of the western coal are needed to equal the heating value of one
ton of eastern coal.

-------
                        CLEAN COMBUSTION OF COAL


 Post  Combustion Treatment for SO? Removal

      Flue  gas  desulfurization (FGD) is a post combustion method of re-
 moving  S02 from combustion gases.  FGD processes contact the combustion
 gases with a sorbent to react with the S02.

      The numerous FGD processes can be categorized as nonregenerable or
 regenerable processes.  In nonregenerable processes, the sorbent reacts
 with  absorbed  S02 and is not regenerated or reused.  These processes
 produce a  sludge which consists of a mixture of fly ash and water plus
 calcium sulfite and calcium sulfate which result from the S02/sorbent
 reactions.  Several alternatives for disposition of the sludge are avail-
 able.   The alternatives include ponding of untreated sludge, landfilling
 of  untreated and treated sludge, and commercial utilization.

      Regenerable processes recover SOo from the combustion gases and
 convert it into marketable by-products such as elemental sulfur or
 sulfuric acid.

      The best  developed FGD processes are the nonregenerable lime and
 limestone  scrubbing processes.  In these processes, a lime or limestone
 slurry  absorbs S02 from the flue gases, and subsequent reactions produce
 calcium sulfite and calcium sulfate.  A variation of lime/limestone scrub-
 bing, known as the "double alkali" process, uses a clear solution of a
 soluble salt rather than a slurry to absorb S02.  Subsequent reactions
 to  produce calcium sulfite and calcium sulfate sludge occur outside the
 absorber.

      The most  significant of the regenerable processes are magnesium
 oxide scrubbing and sodium sulfite scrubbing (the Wellman-Lord process).
 The magnesium oxide process utilizes an MgO slurry to absorb S02 and
 produce magnesium sulfite and magnesium sulfate.  The Wellman-Lord pro-
 cess employs a clear solution of sodium sulfite to absorb S02 and react
with it to  form sodium bisulfite.

     The first full-scale U.S. installations of the lime and limestone
scrubbing processes were operated in the late 1960's, and the processes
that evolved have controversially been considered to be commercially
available since 1974.

     The economics of FGD are the most critical factor affecting its
widespread use and also the factor that arouses the most debate.  Problems
of inflation,  extent of redundancy, site-specific requirements, vendor
optimism/user skepticism, variations in process options, variations in
the kinds or quantities of by-products, differences in operating condi-
tions,  and other factors have contributed to the complexity of the
situation.   The cost estimates for scrubbing equipment have escalated
from about $18/kw in 1968 to about $89/kw in 1977.  On top of the equip-
ment costs have to be added the cost of heat and power to operate the
scrubbers,  the additional site labor costs, interest during construction
costs and escalation costs.  The difference in capital costs between
two new plants which are identical except that one is equipped with a
scrubber is presently $150/kw.  This 400% increase is not wholly at-
tributable to inflation.  The figure does underline the difficulty of
accurately projecting control costs over a period of a few years for
an emerging new technology.

-------
                    TECHNICAL AND ECONOMIC PROBLEMS                   9


     A member of my staff recently had the opportunity to visit a scrubber
installation and closely examine the operation.  The plant visited was
representative of the new large coal-fired plants.

     The plant consists of two 825 MW units.   Steam condition is 3500 psia
1000°F/1000° and steam is supplied by a supercritical boiler to
turbo-generators.  The units burn pulverized coal.  Operational control
is by digital computer.  Natural Draft cooling towers are used.

     Flue gas desulfurization is accomplished by Wet Lime Scrubbers.

     Capital costs for each of the two units were:

                Plant and Equipment          $510 x 106
                Scrubbers plus Cooling
                  Tower                      $240 x
                                             $750 x 106

     The scrubbers are of the two vessel venturi type.  The first vessel
is for particulate control and the second does the sulfur absorbing.  A
demister follows the second vessel, then a reheater and the stack.  There
are six equal size parallel trains for each unit while each train handles
approximately 150 MW.  All six units must be worked to handle the full
output.  The vessels and piping are all rubber lined.  The limestone is
calcined separately at an installation 300 yards away, slaked, doped
with MgO and piped to the units.  Wet sludge is disposed of in a man-
made lake with an estimated time to fill up of twenty years.

     Pittsburgh seam coal, approximately 3 percent sulfur is used, most
deep mined, some Ohio surface mined.

     The design operation calls for a sulfur removal guarantee of 92%.
43 MW of electrical power is 'used to drive the scrubber and 1200 gal/hr
of #2 distillate fuel oil is burned to reheat the 125°F air from the
demister to the 165°F stack.

     Nominal crew for the two units is 136 men.  This does not include
the extra men required to maintain and modify the scrubbers.

     Commercial operation started December 1975 and the scrubbers were
not yet  running satisfactorily in June 1977.

     Performance achieved to date has been very good considering the
circumstances.  The turbine-generator heat rate is 8881 Btu/kWh (which
is probably due to the impluse turbines) and 9981 Btu/kWh for the whole
plant.  The achieved capacity factor is 71%.

     The turbine, generator, and boilers work well.  The scrubbers do not
work well when they are operating.  With six trains in parallel on each
825 MW unit, they can get some power out through the use of at least one
train 91% of the time.

-------
  10                     CLEM COMBUSTION OF COAL


      Sulfur  removal runs ca 60% in lieu of the guarantee of

      Cost  of electricity at the busbar is:

                 Plant                       9 Mil/kWh
                 Scrubber                    3 Mil/kWh

                          Total Unit        12 Mil/kWh

      The lime requirement is 7.8 percent by weight of the coal used.  For
 a 3  percent  sulfur coal, this gives a Ca/S mole ratio of approximately
 1.5.

      They  have had endless operating troubles, and the only apparent
 way  to  minimize  the problem is very, very careful control of the pH of
 the  solution in  the particulate and S02 absorption vessels.  The pH can
 vary from  2  to 10 in lieu of the desired 7.

      The rubber  lining in the vessel and piping suffers mechanical damage
 from the displaced deposits.  Due to the acid condition, this has eaten
 through the  3/8  carbon steel plate in as little as three days.

      The demister chevrons and cascades plug in three days to a month.
 The  original demister structure failed as did the succeeding two genera-
 tions of replacements.

      The sludge  carry-over frequently plugs the lines, requiring shutdown
 to clear them.

      Additional  manpower over and above the 136 normal, runs twenty to
 fifty men.   These are required to keep the scrubbers maintained and are
 a mix of power plant, vendors, and contractor personnel since the scrub-
 bers  are not  accepted yet.

      In summary, the new coal plant is working well except for the wet
 lime  scrubbers.  They do not work well, and add greatly to the cost at
 this  time.

 In-Situ S02 Removal at the Source During Coal Combustion

      One ERDA approach is to develop new alternative technologies which
 control emissions during the combustion process or regulate input of
potential pollutants.  This is being pursued by fluidized-bed combustion
 (FBC) for heat,  steam and power generation in the presence of sulfur
dioxide sorbent.

     The increased use of coal by this technology for heat and power
will benefit  the public.  Coal is appropriate for use in moderate and
large-sized  fuel burning installations where it is economical to have
equipment necessary to handle large amounts of solids (i.e., coal, lime-
 stone, ash).   Increased coal use can reduce oil and natural gas use by
 the utility and  industrial sector and make domestic supplies of these
 fuels available  for the unique requirements of the residential, commer-
 cial  and transportation sectors of the economy.

-------
                     TECHNICAL AND ECONOMIC PROBLEMS                  11


     The user sectors which these programs will supply with technology
are:
     •  Industrial and institutional heat and steam.
     •  Utility steam-electricity generation.

Even with a low-energy growth future, the need for new and replacement
fossil-fueled equipment facilities for these user sectors represent a
major market.

     Emission control for fluidized-bed boilers centers in the combustion
zone.  Sized coal is burned at atmospheric pressure or at elevated pres-
sures in a fluidized bed of inert ash and limestone or dolomite.  The
limestone or dolomite is calcined and the calcium oxide reacts with S02
to  form a solid sulfate material which can be disposed of with the ash.
Sorbent regeneration could produce a highly concentrated SC>2 gas from
which elemental sulfur could be readily removed using existing processes.
Fluidized-bed boilers are operated at combustion temperatures of 1500 F
which are lower than conventional boilers which have flames well above
3000°F, thus inhibiting the formation of nitrogen oxides from combustion
air nitrogen.  This lower combustion temperature greatly reduces the
tendency of slag to build-up on boiler tubes.  Western coals, with high
alkali content, and which foul conventional boilers rapidly, have been
burned in fluidized bed research combustors without fouling.  An atmos-
pheric pressure coal-fired fluidized-bed boiler having a capacity of
5,000 pounds of steam per hour has been successfully operated for over
12,000 hours at furnace temperatures of approximately 1,600°F.  It demon-
strated that all types of coal, char and coal wastes can be burned in
an  environmentally acceptable manner.

     The ERDA strategy to accomplish the development of FBC of coal is
to  have the majority of the technology development done in industry,
to  work on the technologies promising the earliest significant payoff,
and to utilize experience gained on early development of small-scale
equipment to benefit the development of the following larger-scale units.
This reduces cost and accelerates commercial implementation.  An example
of  the relative costs for FBC vs. conventional systems is shown in
Table I for large utility applications.

     The minimum risk and the best schedule of progress are achieved by
performing parallel engineering development activities in several modest
size and cost systems and by testing various applications at the same
time.  The Federal Government will expedite industrial development by
providing the flexible, costly test facilities for common use that industry
finds to be too expensive.  By having development work performed by in-
dustrial firms with the use of these test facilities, the program can
proceed rapidly to commercial implementation when the development is
completed.

     The user's requirements are recognized and included from the start.
The industrial-institutional scale units are of of a size large enough
to  realize the economies of using coal, sorbent and ash handling systems,
which are not economical in very small systems.  Utility systems require
much larger plants with very firm technical specifications, and longer
construction times.  Improvements which come from developments in the
intermediate size industrial units will benefit the utility size systems.

-------
                                        TABLE I

                          CHARACTERISTICS OF COAL COMBUSTION
PERFORMANCE AND
COST CHARACTERISTICS
Capital Cost ($/kW)*
O&M Cost (Mills/kWh)
Efficiency (%)
(Heat Rate - Btu/kWh)
Max. Capacity Factor
Average Capacity Factor
Initial Capacity Factor
Years to Reach Max.
Capacity Factor
(End of Year)
Construction Time (Years)
Earliest Commercialization
Date
Steam with
AFB (3500 psi;)
1000°F/1000°F
450
(350)**
2.2
36
(9,481)
.70
.60
.50
3
5
1984
Steam with
PFB (3500 psi;)
1000°F/1000°F
514
(400)**
2.3
39
(8,751)
.70
.60
.50
3
5
1988
Conventional
Steam with FGD
578
(450)**
2.8
35
(9,751)
.70
.60
.50
3
5

                                                                                                       o
                                                                                                        o
                                                                                                        o
                                                                                                        CO
                                                                                                        i-3
                                                                                                        H
                                                                                                        O
                                                                                                        o
                                                                                                        I
 * 1975 Dollars with Allowance for Funds During Construction.

** 1975 Dollars without Allowance for Funds During Construction.

-------
                     TECHNICAL AND ECONOMIC PROBLEMS                  13


     The development of modular, transportable systems and components
capable of factory production is to be emphasized where economically
competitive with field constructed systems.  This strategy will afford
the program the benefits of rapid production, standardization and quality
control, but of more importance from the viewpoint of national objectives,
will provide an industrial production system capable of rapid expansion
in the event of a political curtailment of alternative fuels.

DIRECT COMBUSTION PROGRAM

     Direct combustion is the most efficient and economic method of
utilizing coal.

     Fluidized-bed combustion of coal has been a topic of research since
the early 1960's, both in the U.S. and in Europe; in the U.S., ERDA is
the manager of the Federal RD&D program for FBC.  Consequently, a large
data base exists to identify the technology gaps and to point out areas
which require additional research.  To address the relevant technical
issues through the commercial implementation phase, ERDA has developed
a comprehensive RD&D program for fluidized-bed combustion.

     An operational pilot plant, 30-megawatt in size and designed for
atmospheric pressure, represents both the largest size of fluidized-bed
combustion industrial boilers and is of sufficient size to be representa-
tive of modules for electric utility boiler applications.  It has super-
heater tubes, delivering the steam at 1325 psig and 930°F; consequently,
it is more typical of a utility application.  Because the unit is located
in a private power company's boiler house, it will be evaluated for its
capability for integrated operation in an actual operational environment.

     This pilot plant is the largest fluidized-bed boiler in the world
and will address scale-up problems never previously explored on a scale
this large.  The unit consists of four cells, three primary cells and
a carbon burn-up cell.  A 13 MWe pilot plant is being designed to explore
the pressurized fluidized-bed concept on a large scale.

     For potential industrial applications, ERDA has developed a multi-
project program to evaluate fluidized-bed combustion for both boiler and
process heat applications.  Several units are being built to demonstrate
the technological, operational and economic viability of using fluidized-
bed combustion in industrial applications.  Demonstrating the viability
of such systems is essential for the commercial implementation by
industry.

     An additional and important element in the fluidized-bed combustion
program is the design and construction of component test and integration
units (CTIU) for both atmospheric pressure and elevated pressure combustor
operation.  The CTIU's will be flexible facilities, available for use by
all of industry and the research community, that will permit optimization
of the combustion process/steam or heat regeneration cycle, testing of
components in an integrated operating mode, and evaluation of concepts
as yet untested such as the vertically stacked bed concept.

     The following are descriptions of several projects being conducted
by ERDA which are directly applicable to the industrial sector, where

-------
  14                     CLEM COMBUSTION OF COAL
 RD&D  efforts  can be expected to produce near-term implementation of the
 technology, which will be in compliance with existing and projected emis-
 sion  criteria.

 PROJECT:   30  MWg AFBC BOILER

      Description;  Represents large-scale industrial boiler.  Three
 parallel  primary cells, each 10' x 12'.  Cells operate up to 12 fps
 fluidizing velocity with a nominal 24" static bed in each cell  (one
 steam generating cell, two superheat cells, plus one carbon burn-up
 cell).  Installed in a utility generating plant.  Steam conditions are
 1325  psig and 930°F.  Coal feed of 15 tons per hour with sorbent feed
 dependent on  sulfur in coal.

      Contractor:  Pope, Evans and Robbins, Inc.

      Site:  Rivesville, WV, Monongahela Power Company Station

      Objective;  Early demonstration of FBC technology in utility environ-
 ment  using high-sulfur coal.  Evaluate scale-up of heat transfer, com-
 bustion efficiency, emission control.  Demonstrate automatic combustion
 control in central load dispatch utility system.

      Status:  Boiler installation completed.  Successful ignition of
 coal  in the carbon burn-up cell 7 December 1976.  Successful start-up
 of  superheater cell 20 April 1977.

      Plans:   Full load steam generation expected to begin by the end of
 CY  1977.   A full spectrum test program will then be initiated to meet
 the unit/program objectives.

 PROJECT:   INSTITUTIONAL SIZE INDUSTRIAL FLUIDIZED-BED BOILER

      Description;  This fluidized-bed boiler will generate 100,000 Ib
 per hr of  saturated steam at 650 psig.  The boiler will furnish this
 steam to a university and hospital for space heating.  The boiler plant
will  burn high-sulfur coal while attaining low levels of pollutant
 emissions.  The steam will be used to drive a small turbine for demon-
 stration as a co-generation system.

      Contractor:  Georgetown Hospital

      Site;  Georgetown University, Washington, D.C.

     Objective:   To demonstrate an institutional saturated steam boiler
capable of complete shop assembly can be built in capacities up to
150,000 Ib per hr and that materials handling systems for this  type of
application can be optimized.

      Status:   Letter of intent to negotiate contract issued.

     Plans:  To be operational sometime in 1979.  Three years operational
test program and data dissemination plan under final development.

-------
                     TECHNICAL AND ECONOMIC PROBLEMS                  15
PROJECT:  INDUSTRIAL APPLICATION FLUIDIZED-BED COMBUSTION PROCESS - TWO
          UNITS

     Description;  The work plan is divided into two phases, subscale
testing of natural circulation design capability and design and testing
of a demonstration natural circulation fluidized bed boiler.  This demon-
stration unit will generate 50,000 Ibs of steam per hour at 500 psig
and 600°F, or about 120°F of superheat.

     Contractor;  Combustion Engineering, Inc.

     Site;  Great Lakes Naval Training Center

     Objective;  Confirm the subscale test data design parameters and
procedures subsequently developed.  Provide endurance test data on
materials.  Provide design information for possible improvements to
demonstration and subsequent units.

     Status;  Letter of intent to negotiate contract issued.

     Plans;  Subscale design through testing for the natural circulation
unit will take 1-3/4 years from contract initiation.  Demonstration design
through testing will take two years.

PROJECT:  INDUSTRIAL BOILER

     Description;  Develop a demonstration multi-solids fluidized-bed
boiler.  Basically, a multi-solids FBC can be regarded as a recirculating
or entrained fluidized-bed superimposed on a conventional dense fluidized
bed region using fluidizing velocities as high as 35-40 fps.  Heat trans-
fer advantages are expected.  The demonstration unit for this process is
to generate 25,000 Ib/hr of saturated steam at 100 psig.

     Contractor;  Battelle Memorial Institute's Columbus Laboratories

     Site;  Columbus, Ohio

     Objective;  Development of the multi-solids FBC system to the point
where industry will accept and use it for commercial industrial steam
generation.

     Status;  Contract signed.

     Plans;  Phase I - R&D in support of demonstration plant (25 months
duration); Phase II - Construction and start-up of demonstration plant
(14 months duration); Phase III - Demonstration plant operation (36
months duration).

PROJECT:  FBC PROCESS HEATER/TUBE STILL

     Description:  Evaluate the FBC concept for fluid process heaters
such as those used in refineries.  A Process Development Unit (PDU)
will be designed and constructed to test a fired heater in the 10-15MM
Btu per hour range for application to a tube still.

-------
  16                     CLEAN COMBUSTION OF COAL


      Contractor:  EXXON Research and Engineering Company

      Site:  Linden, New Jersey (Engineering)
            PDU Location at Refinery

      Objective:  Conduct R&D to obtain engineering design data for coal
 fired fluid bed process heaters.  Demonstrate operation of the PDU.
 Prepare  design specifications and cost estimate for a commercial FBC
 process  heater.

      Status;  Contract signed.

      Plans:  Phase I - R&D test work (18 months duration) ; Phase II -
 PDU  design  (6 months duration); Phase III - PDU evaluation and commercial
 design (12 months duration starting two years after contract initiation).

 PROJECT:  FBC PROCESS HEATER/AIR HEATER

      Description;  The FBC heater will supply 39 x 10  Btu per hour to
 heat clean air to 900°F in the heat exchanger.  The total load will be
 served by three beds with a thermal storage unit in the system to pro-
 vide heat during intermittent operation of each bed to achieve load
 matching.  Bituminous coal, lignite, refinery coke, peat as primary
 fuels and such things as wood and paper waste plus cornstalks as secondary
 fuels will be burned.

      Contractor;  Fluidyne Engineering Corporation

      Site;  Owatonna, Minnesota

      Objective:  Demonstrate FBC heating of clean, low-pressure air to
 900°F.  Provide long-term data on life of air tubes in and above the bed
 at different temperatures.  Prove application to manufacturing processes
 and  space heating.

      Status;  Contract signed.

     Plans:  Construct by late 1978 or early 1979.

     Another facet of the ERDA direct combustion program that is accel-
erating in importance is the use of western coals in FBC units.  The
future of the western coals as compliance coals could heavily depend
on FBC.

     The ash of lignites and western subbituminous coals contains a high
percentage of the alkali constituents, calcium, magnesium, and sodium.
These coals, when burned in a conventional pc-fired furnace, will retain
from 10 to 40 percent of the sulfur on the ash.  These same coals, when
burned in a fluidized-bed combustor, should retain a greater percentage
of sulfur on the ash because of:  (1) The increased contact time between
the sulfur dioxide produced during the combustion process and the inherent
ash alkali, and (2)  The lower combustion temperatures for the fluidized
combustion process.   The longer contact time allows for the reaction
between coal ash alkali, oxygen, and S02 to be carried further towards
completion.  The lower temperatures allow for greater alkali utilization

-------
                     TECHNICAL AND ECONOMIC PROBLEMS                  17
because the temperatures are below the disassociation temperature of the
sulfates and below the ash fusion temperatures so that the alkali is not
tied up in the glassy agglomerate.  The sulfur retention percentages ob-
tained in preliminary tests on one North Dakota lignite at the Grand
Forks Energy Research Center range from 38 to 58 percent without ash
reinjection.

     The role of the Grand Forks Energy Research Center of ERDA in the
fluidized-bed combustion program is to develop data on sulfur retention
on the alkaline coal ash from western United States coals, when burned
in a fluidized-bed combustor without a sorbent.  FBC will result in
substantially lower S02 emissions than from conventional combustion sys-
tems, and may therefore meet the Federal New Source Performance Standard
of 1.2 Ib S02/10  Btu or even new lower federal or local emission
level.  The effects of operating conditions and coal ash composition on
the retention of sulfur dioxide during fluidized-bed combustion of western
coals is being thoroughly evaluated to provide a design base for the use
of these coals.

     Six coals from the western United States other than the lignite
previously mentioned have been tested in the fluidized-bed combustor at
Grand Forks.  Preliminary results are presented in Table II.  Recycle
was not used for these tests.  The sulfur retention would be expected
to improve significantly with recycle.

     It must be emphasized that this work with western coals was done
without a sorbent addition or a sorbent bed.  The importance here is
certainly both in improving the economics of FBC and in making the
auxiliaries less complicated.  The bottom line, however, is in the
significant impact of making currently non-compliance western coals
compliance coals in FBC without the use of a sorbent.  As a bonus,
the low temperatures used with FBC will minimize or eliminate the foul-
ing/slagging experienced in conventional units with these fuels.

Coal Conversion Processes

     The mandatory increase in the use of coal over the next quarter to
half of a century mandates the widespread application of control tech-
nologies, especially for S02, singly and in combination, to prevent
extensive degradation of the environment.  An SOo control program which
permits timely and economic attainment of the regulatory requirements
is essential.  Such a program will be comprised of various control options
including naturally occurring low-sulfur coal, coal cleaning and flue
gas desulfurization separately or in combination, coal gasification,
coal liquefaction, fluidized-bed combustion of coal, and others.  Coal
gasification and liquefaction are viewed as SOo control technologies
for the overall program.

     The choice of appropriate 862 control strategy for a specific site
is often complex due to the interaction of such factors as commercial
availability of competing systems, reliability and operability of the
systems in other applications, capital costs and annual revenue require-
ments, fluctuation in the availability and cost of low-sulfur fuels,
and uncertainty about pollutant emission regulations.

-------
                                                  TABLE II
                                                                                                                  oo
                                PRELIMINARY SULFUR DIOXIDE AND NITRIC  OXIDE
                              EMISSION DATA FOR VARIOUS  WESTERN COALS AND  LIGNITES
Rank
I±!
S.B.2/
L
S.B.
B^
S.B.
S.B.
Coal Tested
Beulah, N. Dak.
Naughton, Wyo.
Velva, N. Dak.
Cols trip, Mont.
Browning, Utah
Sarpy Creek, Mont.
Decker, Mont.
Calcium-to-
Sulfur Mole
Ratio
1.27
0.67
5.09
1.29
0.81
1.56
1.38
Total Alkali-
to-Sulfur
Mole Ratio
1.99
1.09
7.54
1.87
1.23
2.18
2.48
Percent
Coal
Sulfur
1.01
0.34
0.23
0.62
0.88
0.71
0.33
Ash
7.49
5.04
6.36
8.16
8.07
8.98
4.25
Coal
Nitrogen
0.45
1.1*'
0.7*'
0.6*7
1.28^
0.68*/
1.0*'
Emission, ppm @
45% excess air
Sulfur NQ
Dioxide x
480 340
307 400
10 390
.377 360
588 400
412 339
121 350
                                                                                                                  o
                                                                                                                  IT"
                                                                                                                  o
                                                                                                                  Q
                                                                                                                  C
                                                                                                                  CO
                                                                                                                  H3
                                                                                                                  H
                                                                                                                  O
                                                                                                                  £=!
                                                                                                                   O
                                                                                                                   o
-  Lignite.
2/
—  Subbituminous.

3/
—  Bituminous.

—  Percent of nitrogen not available  for  specific sample burned in FBC.
   sample of same coal.
Data taken from earlier

-------
                     TECHNICAL AND ECONOMIC PROBLEMS                   19
     Depending on application and end product, specific gasification
processes can be described by one or more of the following terms:  low-
Btu, intermediate Btu, high Btu, fixed bed, entrained bed, fluidized bed,
slagging, air blown, oxygen blown, atmospheric and pressurized.

     Gasifier systems like Lurgi, Winkler, Wellman-Galusha, Koppers-Totzek,
HyGas, Bi-Gas, Synthane, C02-Acceptor, Texaco, Riley-Morgan, U-Gas, and
COGAS are familiar names to those who have followed the history of coal
gasification.  Sulfur clean-up systems such as Stretford, iron oxide,
Benfield, fluidized bed dolomite, Trencor, Selexol, Sulfinol, and Rectisol
are partner systems well known to the industry.

     Considerable research, development, and engineering have already
taken place in coal gasification and some systems are commercially avail-
able and are used around the world.  Installations of the small sized
Lurgi, Winkler, Koppers-Totzek, and Wellman-Galusha gasifiers are too
numerous for mention here and these are definitely considered commercial.
New combinations of or scale-ups of these gasifiers, along with I^S and
particulate matter removal systems need to be coupled in practice.

     Economics of the leading coal gasification systems are available
from a number of studies completed during recent years.  TVA has conducted
considerable work in evaluating the economics of coal gasification and
H2S removal systems.  Economic analyses of 500 MW power units with sulfur
production of about 218 tons per day, indicated the following direct
capital cost (in 1975 dollars) and annual revenue requirements for 7,000
hours/year of operation:

                      Direct Capital Investment
                                                       Annual Revenue
     Process          Without Glaus  With Glaus         Requirements

     Stretford        $18,000,000    Not needed         $3,400,000
     Benfield           5,000,000    $10,000,000         3,600,000
     Trencor                          10,000,000         3,800,000
     Selexol           12,000,000     17,000,000         3,500,000

Total project cost would be somewhat higher than the direct costs indicated.

     Economics for new combined cycle plants depend a great deal on the
state-of-the-art of gas turbine operation.  For existing gas turbine
technology where the maximum temperature is around 1950 F, power genera-
tion via a coal gasification unit integrated into a combined cycle plant
is more expensive than power generation with conventional boilers with
FGD supplying steam to a conventional steam turbine plant.  If the advance-
ment of the gas turbine technology permits turbine temperatures up to
2500°F, the advantage could possibly switch to the gasification-combined
cycle units.

     Liquid fuels can be produced from coal by four general methods.
First, the coal can be pyrolyzed to break it down into a synthetic crude
oil, a by-product gas, and a char residue.  The char residue must be
used in some way since it represents a significant fraction of the total
heating value of the coal.  Second, the coal can be dissolved in an
organic solvent, treated with hydrogen, and filtered to remove the non-
combustible minerals.  The combustible fraction filtered out with the

-------
 20                     CLEAN COMBUSTION OF COAL


non-combustibles must be reclaimed and used if the process is to be
economically competitive with alternative technologies.  Third, the
coal  can be gasified to produce carbon monoxide and hydrogen which can
subsequently be converted to methanol or other liquid hydrocarbon fuels.
Fourth, pulverized coal can be contacted with hydrogen in the presence
of catalysts to form fractions of liquid and gaseous hydrocarbons.

      The economic aspects of coal liquefaction are not well defined
since the technology is still being developed.  Some studies were made
earlier in the development of the technology, but these estimates were
optimistic and are generally not regarded as accurate.  The preparation
of additional estimates is now underway on many of the systems.  In
general, the cost of coal liquefaction is comparatively high; base load
power generation using fuels from these processes is probably not econ-
omically competitive with low-Btu coal gasification, fluidized-bed com-
bustion, or conventional boilers with FGD.   Coal liquefaction will un-
doubtedly find application for intermediate duty & peaking service, where
it is much more competitive with other technologies.

Conclusion

      The accelerated use of coal in our national energy picture brings
along with it some interesting points.  It  is going to be expensive.
It is a definite technology challenge; a surmountable one,  but definitely
complex.  The environment can be protected.   Currently, the direct com-
bustion of coal in a fluidized-bed would appear to have the overall edge
when the technical and economic impacts are evaluated against some base-
line environmental compliance criteria; however,  it would appear that
we will have a continuing need for both liquid and gaseous conversion
fuels to meet our diverse utility/industrial energy consumer fuel needs.

-------
                                                                      21
             COAL COMBUSTION AND FUTURE EMISSION REGULATIONS

                                   by

                             Thomas Schrader
                    Office of Planning and Evaluation
                   U.S. Environmental Protection Agency
     It is a pleasure for me to be here this morning and have the oppor-
tunity to talk with a group of professionals who are involved with many
of the same problems that are important to my work at the Environmental
Protection Agency.

     As Dr. Freedman highlighted earlier this morning, the concern for
clean air has had and will l:.ave a considerable impact on the pattern and
techniques of burning coal.  My intent this morning is to talk primarily
about environmental policy and future emission regulations for coal
burning sources.  I will discuss three major areas which have consider-
able effect on the implementation of clean coal technologies.  These are:

     - emissions standards which are to be met by new facilities,

     - regulations to prevent the significant deterioration of air
       quality, and

     - limitations necessary to attain air quality standards and protect
       public health.

     As you appreciate, each of these areas will have an effect on the
demand, or market, for clean coal combustion technologies.  Moreover, as
you might expect the technical capability of clean technologies will
greatly influence the emission regulations in each of these cases.

     Over the next 15 years coal will become an increasing energy
resource.  The shift to coal is already occurring due to the existing
scarcity and price of oil and natural gas.  The President's proposed
National Energy Plan is to encourage the substitution of coal for gas
and oil through taxes on oil and gas and through investment tax credits
for the purchase of coal handling and burning equipment, including
pollution control.

     With increased coal use comes increases in atmosphere emissions of
a number of pollutants—sulfur oxides, particulate matter, nitrogen
oxides, trace metals, radioactive materials, and organic compounds.  The
potential damages are very real; they include:

     - Impaired health and premature death from respiratory ailments;

     - Stunted growth of crops and other plants;

     - Reduced resistance of plants and animals to disease.

-------
 22                    CLEAN COMBUSTION OF COAL


 Air quality standards and pollution emission regulations are intended to
 protect against these damages.

 New Source  Performance Standards

      The application of clean combustion technologies at new facilities
 will TDe important to minimize these effects.  In this regard, the Envi-
 ronmental Protection Agency prescribes nationally applicable standards
 of performance for  large new sources of air pollution.  The 1970 amend-
 ments to the Clean  Air Act specified that the new source performance
 standard, NSPS for  short, was to reflect the degree of emission limita-
 tion achievable through the application of the best, adequately demon-
 strated system of emission reduction taking the costs of achieving the
 reduction into account.

      Under  this authority, EPA promulgated new source performance stand-
 ards for large coal-fired boilers in 19T1-  Considering the regulation
 for sulfur  dioxide  presently in effect, the specific regulation of
 1.2 pounds  of sulfur dioxide per million Btu of heat input can be met by
 burning low sulfur  coals or by burning high sulfur coal with pollution
 control equipment.  Although such a regulation may have been appropriate
 for the early 1970's when commercial use of pollution control equipment
 and clean combustion technologies was just emerging, Congress has stated
 that it considers a standard which permits the use of untreated fuels to
 be inadequate and in direct conflict with the purposes of new source
 standards.   Accordingly, in the Clean Air Act Amendments considered in
 1976 and proposed again this year, the House of Representatives and the
 Conference  Committee have adopted legislative language which would
 require that the new source standard for new fuel-burning stationary
 sources require a specified reduction in emissions which is achievable
 when applying the best control technology.  In establishing the standard,
 EPA would be required to determine which technologies were adequately
 demonstrated and to consider cost impacts and the energy and non-air
 quality impacts before setting a regulation.

      Anticipating the passage of the amendments this year and considering
 recent  advances in  control technology development, EPA has already
 initiated a review  of the new source standards for electric power plants.
 In what  is  one of the most rigorous analyses performed by the Agency,
 EPA is  evaluating the technical performance (primarily reliability and
 removal  efficiency) of recent flue gas desulfurization systems.  In
 addition, the Agency is assessing potential coal production shifts and
 estimating  the economic impact on the electric utility industry and its
 customers.   While several alternative emission regulations are being
 considered,  one guess at the new standards would be a requirement for
 90  percent  removal  of sulfur dioxide with an emission ceiling somewhere
 between  0.5  and 1.2 pounds of sulfur dioxide per million Btu, a regula-
 tion of  0.05 pounds per million Btu (one-half the present standards) for
 suspended particulates, and an incremental tightening to 0.6 pounds per
million  Btu  for NOX.  The new standards for power plants are expected to
be  proposed  in early 1978.

     The implications of these standards, especially for sulfur dioxide,
for developing clean combustion technologies are significant.  They set
both a performance  and a cost target with which other technologies must

-------
                      FUTURE EMISSION REGULATIONS                    23
compete.  For example, to "be allowed as a substitute for a flue gas
desulfurization (FGD) system, a sulfur removal technology would have to
achieve comparable or better sulfur removal efficiency.  Use of coal
washing alone would not be suitable unless it were combined with another
technique to achieve the required overall reduction in emissions.  In
effect, this could mean that those technologies which are not capable of
high removal and are not economically competitive when used in combina-
tion with other technologies would be foreclosed from use on new coal-
fired power plants.

     While requiring the maximum available degree of emission reduction
from new sources, it is hoped that development of innovative means of
achieving equal or better degrees of emission control would also be
encouraged.  In this regard, it has been EPA policy to allow special
consideration for developing technologies.  To encourage innovative
technology, Congress has also proposed a formal variance for new tech-
nologies for up to 10 years from the federal new source performance
standards.  Only those technologies with promise of achieving equivalent
or greater emission control than that required under the NSPS would be
eligible for a variance.

     Continued progress in control technology development for each
pollutant is expected to lead to competition in both performance and
costs.  At some date in the future, with the demonstration of greater
emission removal efficiency at a reasonable cost, EPA would again revise
the new source performance standards to reflect new technical capabili-
ties.

Prevention of Significant Deterioration

     At this point I would like to discuss the regulations to prevent
the significant deterioration (PSD) of air quality in areas of the
country which are presently enjoying clean air.  As you all probably
know, the Supreme Court upheld a lower court decision that required EPA
to promulgate regulations to prevent the significant deterioration of
air quality.  These regulations were promulgated in December 197^ after
extensive technical analyses and public participation and provide air
quality designations which States may use to manage air resources.

     Briefly, the EPA regulations establish three classifications of air
quality levels based on how much increase will be permitted in ambient
concentrations of particulate matter and sulfur dioxide.

     Class I - Pristine areas where deterioration in any air quality
deterioration would be considered significant.

     Class II - Areas where deterioration in air quality that would
normally accompany moderate growth would not be considered significant.

     Class III - Areas where intensive and concentrated major industrial
growth is desired, thus permitting levels of air quality to rise to the
national ambient air quality standards.

     EPA's regulations initially designated all areas of the country as
Class II, allowing states to redesignate areas to either Class I

-------
24                     CLEAN COMBUSTION OF  COAL


(pristine) or Class III (intensive economic growth).   The regulations
passed, by the Senate and House are similar  in general approach, although
numeric limits, Class I designations,-and sources covered vary.

     Briefly, the regulations are intimately tied to  the capability to
control emissions of pollutants.   Limits on size of facility or possible
constraints on siting a large facility within the requirements of each
air quality designation depends upon the ambient ground-level concentra-
tions produced by the facility.  Since ground-level concentrations are
roughly proportional to emission  rates, any constraints on a facility
translate into a need for greater emission  control.

     For example, consider a national park  or wilderness area which is
protected as a Class I, or pristine, area.   In order  to preserve the
clean air values of this area, a  1000 Mw coal-fired power plant meeting
the present NSPS for sulfur dioxide would have to locate approximately
60 miles from the Class I boundary.  Depending upon the emission rate
and the specific terrain in the Class I area, a 1000  Mw plant could
locate less than 10 miles from the park or  wilderness area.   A signifi-
cant portion of this distance reduction would be accomplished by the
planned revision in the NSPS for  sulfur dioxide.  And, in reality, the
demand for additional emission control due  to PSD requirements may be
small.  It will, however, be one  option considered by large power plants,
synthetic fuel plants, and other  facilities which wish to locate near
Class I areas or near mountainous terrain.

Attainment of Air Quality Standards

     The last topic I would like  to address concerns  the measures neces-
sary to protect public health in  areas which presently have poor air
quality.

     The 1970 Clean Air Act required each State to develop and implement
a State Implementation Plan (SIP) to attain national, health-related air
quality standards.  The Act provided for attainment of the air quality
standards by mid-1975 •  1975 has  come and gone and there remain a number
of areas, primarily metropolitan  areas, which are not attaining these
health standards.  Further, there are other areas which may have diffi-
culty maintaining air quality standards as  industry and population grow.

     These problems are cause for considerable concern from the stand-
point of public health.  If we are to progress toward attaining our
health standards we are faced with either requiring greater emission
control on existing sources or limiting the construction of new sources
in these problem areas.

     Although the nonattainment problem is  most widespread with respect
to photochemical oxidants and particulates, NOX and sulfur dioxide are
also problems in specific cases.   To manage new growth in these areas,
EPA adopted an interpretative ruling in December 1976.  Referred to as
an "emission offset policy," the  ruling allows new sources to locate in
areas not attaining health standards if the new facility meets the
lowest achievable emission rate (comparable to best controls on fuels
of moderate or low pollution potential) and a decrease in emissions from
existing sources more than offsetting the emission increase from the new

-------
                      FUTURE EMISSION REGULATIONS                    25


facility.  The emission decreases can be from other resources operated
by the owner of the new facility or from existing facilities operated
by others.

     The EPA policy is one approach which will lead to attainment of air
quality standards.  In effect, the policy creates an incentive for
States and local communities to adopt and enforce emission regulations
which will attain the air quality standards.

     EPA has issued calls to a number of States to revise their SIPs to
assure attainment of air quality standards as soon as possible.  In
addition, the Agency has designated at a number of other areas as air
quality maintenance areas'.  These areas may require some adjustment in
emission regulations to assure maintenance of air quality due to antici-
pated growth in population and industrial activity between now and 1985.

     In formulating SIP emission limits or in meeting the requirements
of the emission offset policy a central question is the technical feasi-
bility and cost of different levels of control on different types of
sources.  Thus, here, as with NSPS, the technical feasibility and afford-
ability of clean combustion technologies will determine the nature of
emission limits.  The best technologies for new sources would be relied
upon to achieve the lowest possible emissions and thereby minimize the
air quality effects of new facilities.  Moreover, the available technol-
ogies will establish the bounds for the emission regulations which could
apply to existing and smaller facilities.  In this latter case, the pre-
combustion treatment of fuels and those technologies that are economic
to retrofit will be particularly important.

Conclusion

     In closing, I would like to read a paragraph from the National
Academy of Science publication on Man, Materials, and Environment,
(pp. 8-9):

     Many foreseeable problems cannot now be solved by available
     technology.  Even if we control 99-5 percent of some pollut-
     ants, the remaining one-half of 1 percent, because of the
     large absolute amounts projected by the year 2000, can create
     environmental problems for which a workable remedy has not
     yet emerged from the laboratory.

-------
26                     CLEAN COMBUSTION OF COAL

-------
                                                                     27
            RECENT DEVELOPMENTS IN COAL COMBUSTION TECHNOLOGY

                                   by
                            James I. Joubert
                   Pittsburgh Energy Research Center
          U. S. Energy Research and Development Administration
                                ABSTRACT

     This paper considers three distinct areas of research and develop-
ment directed toward increasing the utilization of coal by industry and
electrical utilities:  coal-oil slurry combustion; solvent-refined-coal
combustion; and coal-based magnetohydrodynamic pover generation.  Current
research programs at the Pittsburgh Energy Research Center are discussed,
and an assessment of technical problems that require solution is presented.
Finally, an attempt is made to predict the potential impact of each of
these technologies on the U. S. energy supply situation.

                           I.  INTRODUCTION

     The increasing dependence of the U. S. on foreign supplies of
petroleum and the growing scarcity of domestic natural gas have placed
increased emphasis on developing technology to utilize our vast coal
resources in an environmentally acceptable manner.  Three approaches to
accomplishing this objective are discussed in this paper.

     The first approach considered involves retrofitting existing
utility and industrial boilers to burn coal-oil slurry.  The use of
coal mixed with heavy fuel oil could conserve 25 to 35 percent of the
oil now used for steam raising, and could be implemented almost immediately.

     The advent of low-sulfur, low-ash fuels derived from coal, such as
solvent refined coal (SRC), could make an impact on our energy supply
situtation in the mid-term (1985-2000).  Use of such fuels would be
preferable to the use of fuel oil or natural gas for power generation,
and may provide an attractive alternative to flue gas scrubbing.

     The development of coal-based magnetohydrodynamic (MHD) power
generation holds great promise for the period 1990 and beyond.  This
approach to utilizing coal offers a highly efficient means for gen-
erating electricity with minimal degradation of the environment.

     The combustion-related aspects of each of these approaches to coal
utilization will be considered.  Current research programs in coal-oil
slurry combustion, SRC combustion, and MHD combustor development at the
Pittsburgh Energy Research Center (PERC) are reviewed, and the status
of each of these technologies is assessed.

-------
28                     CLEAN COMBUSTION OF COAL
                         II.  COAL-OIL SLURRY

     The concept of burning coal and oil together as a slurry is not
novel.  Numerous reports dealing with the handling and combustion
characteristics of slurries have been published, the earliest appearing
nearly 100 years ago.  Research on coal-oil slurries was particularly
active during the period between World War I and World War II*.

     The main incentive for utilizing coal-oil slurries in lieu of fuel
oil alone is simply that a significant reduction in oil usage can be
achieved by substituting coal for oil.  Systems designed for feeding
and burning liquid fuels can still be employed, with some modifications.

     For a slurry containing 40 weight percent coal, a fuel oil savings
of 26 to 34 percent is possible for coals ranging in heating value from
10,000 to 14,000 Btu/lb.  Based on earlier studies by the U. S. Bureau
of Mines and Kansas State College, the maximum allowable^qncentration
of coal appears to fall in the range of 40 to 50 percent ' .

     For a 1 percent sulfur coal, a slurry containing about 50 percent
coal could be used without violating EPA emission standards for SO^,
assuming a 13,000 Btu/lb coal.  With a 2 percent sulfur coal, however,
stack gas scrubbing would be necessary if the slurry coal-content exceeded
about 17 percent.

     A number of potential problems may arise in attempting to convert an
existing oil-fired facility to coal-oil slurry firing.  These include:  1)
erosion and/or corrosion of piping, valves, flow meters, pumps, and burners;
2) erosion or fouling of boiler tubes; 3) deposition of solids in various
system components; and 4) particulate and gaseous emissions.  An additional
concern results from the tendency of the coal to settle out of the slurry
when stored; efforts to develop an effective and inexpensive "stabilizer"
have been only partially successful ' .   While a fair amount of success
was achieved in early short-duration combustion tests with slurries (see
Reference 1), not all of the problem areas listed above have been addressed.

     A coal-oil slurry combustion research program was initiated at PERC
in June 1975 .  The initial objectives of the program were to delineate
problems associated with firing coal-oil slurry in a package boiler designed
to burn oil or gas, and to determine the effect of coal ash on the boiler
components and feed system over an extended period (1000 hours).  The
boiler used was a 100 HP Cleaver-Brooks firetube type with four passes.
The fuel used throughout the test period consisted of a 20 percent by
weight concentration of Pittsburgh seam coal in No. 6 fuel oil.  The coal
size consist was nominally 90 percent minus 200 mesh.

     The major difficulty that persisted throughout the test period was
formation of carbonaceous deposits in the refractory-lined burner zone in
the combustion chamber, resulting from impingement of the flame on the
refractory.  It was necessary to shut the boiler down frequently to remove
the deposits.
*An extensive bibliography of literature relating to coal-oil slurries
is given in Reference 1.

-------
                      COAL COMBUSTION TECHNOLOGY                     29
     Increasing atomizing air pressure and excess air level was only
partially successful in limiting the rate of deposition.  Reduction of
load to about 80 percent of capacity also increased operating time.
Further improvement was achieved by modifying the nozzle to provide a
narrower flame pattern which reduced impingement.  However, deterioration
of the spray pattern occurred due to erosion of the orifices in the original
brass nozzle, and in some instances the nozzle had to be replaced after
100 hours of operating time.

     After completion of the 1000 hour test, the boiler was opened and the
internal tube surfaces were inspected.  Only a light coating was evident,
and there was no indication of slag deposition on any of the surfaces.
The test tubes installed in the second, third, and fourth passes were
removed from the boiler and sent to an independent testing laboratory for
metallographic examinination.  The analysis indicated that there was no
evidence of intergranular or subscale corrosion attack.  Estimated cor-
rosion rates were 0.94, 0.68, and 1.2 mils per year in the second, third,
and fourth passes, respectively; these are considered to be acceptable
rates.

     Subsequent to the 1000-hour test, the 100 HP test facility was
upgraded with respect to instrumentation and controls.  Problems due to
formation of carbonaceous deposits in the burner zone have now been
completely overcome as a result of modifications to the burner diffuser.
The unit was recently operated for 65 continuous hours burning a slurry
containing 30 percent coal.  The test was considered satisfactory in all
respects, and an efficiency of 79.9 percent was achieved at full boiler
load.

     A comprehensive research program has now been initiated at PERC
directed toward evaluation of the flame characteristics of coal-oil
slurries.  Combustion aerodynamics and heat transfer in the combustion
zone of the 100 HP unit  (first pass) will be examined in detail and
existing mathematical models will be modified and applied to the system to
aid in interpretation of data.  Coal concentration, coal size-consist, and
boiler load will be varied systematically and an attempt will be made to
develop fundamental design criteria for slurry-fired systems.  In addition,
laboratory studies of slurry rheological properties will be conducted, and
various slurry stabilizers will be tested.

     Much of the experience gained to date with the 100 HP boiler has
aided in the design of a larger coal-oil slurry test facility now under
construction at PERC.  This facility will include a 700 HP watertube
boiler which is more representative of the type of boiler employed in
industry.  The 700 HP combustion test facility will be highly instrumented
and equipped with a data acquisition system.  Shakedown tests are expected
to begin in March 1978.

     In addition to the program at PERC, ERDA is sponsoring several
industrial organizations in large-scale testing of coal-oil slurry  .
Studies have also been conducted by a consortium of companies headed by
General Motors, with support from ERDA .

-------
30                     CLEM COMBUSTION OF COAL
                      III.  SOLVENT REFINED COAL

     Solvent Refined Coal (SRC) is a Ipwjrsulfur, low-ash solid with a
 solidification point of 300°F to 400°F   .   The technology for producing
 SRC has been successfully piloted at a 6 ton/day plant in Wilsonville,
 Alabama, and at a 50 ton/day plant in Fort Lewis (Tacoma) Washington.

     The properties of SRC are remarkably unaffected by the quality of the
 original feedstock.  Typically, the final product contains 0.5 to 0.8
 percent sulfur, less than 0.25 percent ash, and has a heating value of
 15,500 to  16,000 Btu/lb.  A fuel with these specifications would meet
 federal standards for SO  emissions from new stationary sources (1.2 Ib
 SO-/10  Btu for a coal-fired station).  Utilization of SRC in industrial
 or utility boilers would eliminate, or at least minimize, the need for
 elaborate  flue-gas cleanup systems.

     Early studies (1964-1965) of the handling and combustion charac-
 teristics  of SRC by Babcock and Wilcox, Combustion Engineering, and the
 U. S. Bureau of Mines were inconclusive  ~  .  At that time, only limited
 quantities of SRC were available and the testing programs were too brief
 to resolve difficulties encountered.  However, all investigators reported
 problems with respect to pulverizing and conveying the fuel, and burner
 .fouling, when attempting to burn SRC as a solid.  To fire SRC as a liquid,
 it was necessary to heat the fuel to 700-800 F, which resulted in evolu-
 tion of vapors at about 350 F and severe foaming at about 650 F.

     Research on the combustion and handling of characteristics of SRC
 were initiated at PERC in 1974.  Analysis of the earlier studies of SRC
 combustion indicated that firing of the fuel as a solid would present
 fewer difficulties than firing as a liquid.  With solid SRC, no external
 heating of pumps, transport lines, or fittings is required, and problems
 with respect to devolatilization or foaming of the liquid are avoided.

     The SRC used in this investigation was obtained from the Southern
 Services,  Inc. pilot plant in Wilsonville,  Alabama and from the Fort
 Lewis, Washington, pilot plant operated by Pittsburg and Midway Coal Co.
 for ERDA.

     In contrast to earlier investigations of SRC properties, no problems
were encountered in pulverizing or transporting the fuels used in this
study.  An impact mill was used rather than a hammer mill or ball-and-race
mill; attempts to use the latter two types had resulted in mill plugging
due to softening and agglomeration of the SRC.  The grind obtained was
about 90%  through 200 mesh, somewhat finer than that usually obtained when
pulverizing coal in the same mill.

     Initial attempts to burn SRC with the burner usually used in coal
combustion studies resulted in clogging of burner passages due to SRC
melting upon contact with hot burner surfaces.  Water cooling was provided
to all surfaces which contacted the fuel prior to entry into the furnace;
a water-cooled deflector was added to protect the oncoming fuel stream
from flame radiation; and port velocity was increased to 100 ft/sec.
These modifications permitted trouble-free operation.

-------
                       COAL COMBUSTION TECHNOLOGY                     31


     Most of the SRC combustion tests* were conducted at fuel rates of  425
Ib/hr (=6.5 million Btu/hr) at stoichiometric air levels ranging from  115
to 140%, and with air preheat temperatures of 540°F.  In almost all tests,
carbon combustion efficiencies in excess of 99% were achieved (conversion
was 98^5% at 115% S.A.).  NO  emissions ranged from about 0.45 to 0.72  Ib
NO^/IO  Btu, increasing with increasing stoichiometric air levels.  Com-
pliance with EPA emission standards of 0.7 Ib N0_/10  Btu could easily  be
achieved by operation at less than 130 percent of stoichiometric air.

     Recent studies of SRC  combustion have also been carried out by
Babcock and Wilcox  , and Combustion Engineering  .  B&W reported that  SRC
could be pulverized easily  with an impact mill but that extreme diffi-
culties were encountered with a ball-and-race mill; combustion tests were
successfully conducted using a water-cooled burner.  Combustion Engineering
carried out a 100 hour combustion test with SRC with no major problems.
It was concluded that SRC  could be pulverized readily using a C-E bowl
mill, and that it could be  fired using conventional fuel admission assem-
blies provided with water  cooling.

     A large-scale combustion test**using SRC has recently been completed
at Plant Mitchell owned by  the Georgia Power Co.  The 22.5 MW B&W boiler
used was equipped with modified B&W dual-register burners.  The test was
considered a success in all respects with no problems encountered in
pulverizing, transporting,  or burning of the SRC.  SO  and NO  emissions
were substantially less than EPA standards for solid fuel firing.

             IV.  COAL COMBUSTORS FOR MHD POWER GENERATION

     A promising technique  for improving the efficiency in converting
thermal energy into electricity involves the principles of magnetohydro-
dynamics (MHD).  In an MHD  power plant, electricity is generated by
passing a high-temperature, electrically conducting fluid (or plasma)
through a magnetic field.   The interaction of the conducting fluid with
the magnetic field results  in a flow of electrons which can be collected
by electrodes on the walls  of the MHD generator.

     To enhance the electrical conductivity of the plasma, a "seed" mate-
rial, such as potassium carbonate, is injected into the high-temperature
gases produced in the combustion system.  In coal-fired systems, it
has been shown experimentally that the potassium reacts with SO- to
produce K^SO, which can be  collected as a solid in downstream components.
The K9SO, can be regenerated to K-CO,,, thereby recovering the seed and  pro-
ducing elemental sulfur.   The economic requirement for recycling seed also
results in a built-in pollution control technique.  Nearly quantitative
removal of sulfur oxides can be achieved.

     The combustion products leaving the MHD generator are at a suffi-
ciently high temperature to generate additional power in a "bottoming"
steam-turbine plant. -The  recently completed Energy Conversion Alter-
natives Study (ECAS)   concluded that a combined open-cycle MHD-steam
 it
  Detailed experimental results are reported in References 13 and  lU.
 f4&
  Sponsored by ERDA in cooperation with Southern Company Services, EPRI,
  Pittsburg and Midway Coal Company,  and Babcock and Wilcox.

-------
32                     CLEAN COMBUSTION OF COAL


power plant offers overall efficiencies greater than 50 percent as compared
to the predicted performance of advanced "conventional" steam plants of
about 40 percent.

     There-is now an active ERDA-sponsored national MHD development
program '  .  The goal of the program is to place in operation a com-
mercial scale demonstration plant, fueled by coal, by the 1990's.

     The discussion here will be confined to MHDRcoal combustion systems.
The reader is referred to recent publications '  '   for a description of
the overall national MHD program and development activities involving
other MHD system components.

     The development of high-temperature combustion systems for MHD appli-
cations was begun in the early 1960's  .  Most of the progress to date has
resulted from programs in Great Britain, the USSR, and the USA.  The work
in Great Britain, discontinued in 1968, was geared to the use of coal and,
to a lesser extent, oil.  Efforts in the USSR have been directed primarily
toward development of natural gas combustors, although some research in
coal combustion has also been conducted.  In the United States, current
research is oriented toward development of coal-fired systems exclusively.

     A.  Combustor Design Considerations

          Combustors for open-cycle MHD power plants are actually plasma
generators.  Although similar to conventional combustion systems in some
respects, there are certain aspects of MHD combustors that are unique.   An
MHD combustor must be able to meet the following design requirements:

          1.    Burn natural gas, oil, or coal with high efficiency (ap-
               proaching 100 percent).

          2.    Produce a working fluid with a temperature of 2700-3100 K
               (4400-5100°F).

          3.    Operate at pressures ranging from 5 to 10 atmospheres.

          4.    Accommodate highly preheated oxidizer  at a temperature of
               1800-2000 K (_2780-3140°F) ,  and possibly up to 2300 K
               (3680°F)  in future plants .

          5.    Exhibit low heat losses (preferably 5 percent or less of
               the fuel thermal input).

          6.    Incur low pressure losses (preferably less than 10 percent
               of the oxidizer inlet pressure).

          7.    Accommodate injection of seed in dry or aqueous form
               (preferably dry).

          8.    Provide a high degree of seed ionization (approaching 100
               percent).

          9.    Produce a uniform plasma.

-------
                      COAL COMBUSTION  TECHNOLOGY                      33


         10.   Operate with low levels of instabilities.

         11.   Have a long operating life (many thousands of hours).

         12.   Operate under conditions conducive to control of nitrogen
               oxides; the formation of nitrogen oxides must be suppressed
               sufficiently to allow their decomposition as the gases pass
               through the MHD plant.

     In addition to meeting these general requirements, combustors oper-
ating on coal must be designed to cope with liquid slag.  Materials of
construction must be able to withstand slag attack.  If slag is to be
rejected from the combustor proper, a suitable means for tapping the
molten material must be incorporated into the design.  Also, the inter-
action of seed and slag must be minimized to prevent excessive losses of
seed.

     Due to the electrical generating nature of the MHD channel, the
combustor will have a high potential, which in the case of commercial-
scale units may be 25,000 to 35,000 volts.  This results in the necessity
of electrically insulating the combustor from all supports and feed lines
(fuel, oxidizer, seed, cooling water), as well as from the slag rejection
system.

     Combustors designed for MHD applications may consist of one or more
stages.  When burning natural gas or fuel oil, there is no incentive to
use other than a single-stage configuration.  With these fuels, satis-
factory performance has been obtained in experimental single-stage com-
bustors operating under predominantly plug flow conditions.  When burning
coal, the selection of a combustor configuration will be dictated by
technical and economic considerations with respect to the level of slag
carryover that is permissible in downstream portions of the power plant.

     The presence of slag in an MHD plasma has a negative effect on
plasma conductivity, results in loss of seed material, and may seriously
affect the operation and integrity of downstream components.  While avail-
able evidence suggests that low slag carryover from the combustor is
desirable, it is not possible at present to say what constitutes a toler-
able level of carryover.  Depending on the combustor design chosen, as
much as 100 percent or as little as 10 percent of the mineral matter in
the coal will pass through the MHD generator and be deposited or collected
in downstream equipment.  The fundamental differences in combustor design
that most strongly affect the degree of slag rejection achieved are dis-
cussed below.

     Single-Stage Coal Combustors

     Single-stage combustors offer the advantage of simplicity of design
and operation, and lower cost relative to multistage combustors; also,
heat losses will probably be lower in single-stage units.  They may oper-
ate with 100 percent slag carryover, or with partial removal of molten
ash from the combustor proper.

-------
34                     CLEM COMBUSTION OP COAL


     If no slag is rejected from the combustor, a method must be developed
to remove slag from the combustion gases downstream of the generator to
avoid damage to other plant components.  To achieve this without signifi-
cant seed loss represents a major technical problem.  An advantage of this
mode of operation, however, is that the difficulty of tapping slag from a
combustor at high pressure, temperature, and voltage is avoided.

     In a single-stage combustor operating at temperatures in excess of
2700 K (4400°F), a major portion of the coal mineral matter will appear as
vapor in the plasma*.   Hence, the amount of liquid slag that can be tapped
from the combustor proper will be limited.  Also, there will be some
degree of seed loss to the tapped slag in addition to that lost by reac-
tion with slag carried over from the combustor.  Contact of seed and slag
in the combustor could be avoided by injecting the seed in a separate
chamber following the combustor.  However, this would lead to additional
system heat losses.

     Two-Stage Coal Combustors

     In a two-stage coal combustor, slag rejection levels approaching 90%
are possible.  In such a system, the first stage is operated as a coal
gasifier at temperatures ranging from 2200 to 2300 K (3500-3700°F).  At
these temperatures, very little of the coal slag is vaporized, and the
bulk of the liquid slag can be separated from the gas using a first stage
designed similar to a conventional cyclone combustor.  The gas produced in
the first stage is burned in the second state where seed is injected.

     Operation of a two-stage combustor is more complex than in the case
of a single stage unit.  Additional high-temperature air lines, controls,
and instrumentation are required.  It is also likely that thermal losses
will be somewhat higher in a two-stage system because of added system
surface area.

     Multistage Coal Combustors

     Many conceptual studies of-multistage MHD combustion systems have
been reported in the literature     .  Several of these systems offer
potential slag rejection levels in excess of 95 percent.  In general, the
systems are highly complex, and represent novel technology that must be
further developed before their feasibility is ascertained.

     B.  Current MHD Coal Combustor Development Programs

     Pittsburgh Energy Research Center

     Combustor development work at PERC is presently focusing on a two-
stage design.  Studies are being conducted with an atmospheric-pressure
vertical cyclone that represents the first stage of a two-stage combustion
system.  Major objectives of the program include:  1) evaluation of
cyclone combustors operating under fuel rich conditions with coals of
*For example, in burning Pittsburgh seam coal  (10% ash) with 100% of
stoichiometric air at 8 atm, thermodynamic calculations indicate that
about 40% of the slag is vaporized at 2800 K,  and 75% is vaporized  at
3000 K.

-------
                      COAL COMBUSTION TECHNOLOGY                     35


different ranks; 2) testing of high-temperature plasma diagnostic tech-
niques; 31 studies of seed-slag interactions; 4) evaluation of materials;
and 5) development of techniques for controlling nitrogen oxides emissions.
The size of the facility, and its atmospheric-pressure operation, allow
rapid and inexpensive changes to be made in hardware design thereby per-
mitting several design options to be tested during short periods.

     During the past year, three different cyclone configurations were
evaluated.  Tests were carried out at coal feed rates ranging from 0.01 to
0.03 kg/s, oxidizer/fuel ratios ranging from 55 to 110 percent of stoi-
chiometric, and (vitiated) oxidizer temperatures of 1100-1360 K  (1500-
2000 F).  Primary emphasis has been placed on optimizing cyclone perform-
ance with respect to slag rejection and carbon conversion.  A Montana
Rosebud seam coal has been used in most of the tests.

     The tests conducted thus far indicate that coal size consist plays a
major role in determining slag rejection and carbon conversion.  Slag
rejection rates as high as 90 percent and carbon conversions greater than
98 percent have been achieved under substoichiometric conditions.

     A separate facility has been constructed at PERC to test MHD com-
bustors  (nominal 5 MW thermal input) at pressures up to 8 atmospheres.
The facility is capable of continuous operation and includes equipment
for coal pulverizing, drying, conveying, and feeding under pressure.

     A two-stage combustor has been installed in the facility.   The
first stage, a vertical cyclone, will be operated as a coal gasifier.
The second stage is a gaseous fuel combustor.

     The maximum total flow rate attainable in the facility is 1.7
kg/s.  An air preheat temperature of 1200 K (1700 F) will be possible,
with provisions available for vitiated air (or 0«-enriched air)  at
temperatures up to 1866 K (2900°F).  Initial tests will be conducted
at 6 atm pressure, a coal thermal input of 2-3 MW, and a total mass
flow of 1.2 kg/s.

     The objectives of the test program are similar to those indicated
for the atmospheric-pressure facility.  However, the operation is more
complex due to the high pressures involved and the two-stage configura-
tion.  Also, an instrumented test section, designed and supplied by
Avco Everett Research Laboratory, will be installed downstream of the
second-stage combustor to permit measurements of plasma properties in
time and space including temperature, pressure, velocity, and electrical
conductivity.

     Component Development and Integration Facility  (CDIF)

     The Energy Research and Development Administration is presently
constructing a major MHD test facility in Butte, Montana.  The Component
Development and Integration Facility (CDIF) is designed to permit long
duration testing of major components required in commercial MHD  plants
of the future.  The test program will allow resolution of many technical
problems associated with direct use of coal, and will provide valuable
scale-up data for the design of a complete prototype MHD power plant.
Initial testing at the CDIF is expected to begin in  late  1978.

-------
36                     CLEM COMBUSTION OF COAL


      The  facility will have  a  thermal  rating of  50 MW;  the  fuel  thermal
 input will  be  38 MW and  the  balance will  be due  to sensible heat of
 the oxidizer.   Total mass  flow rate will  be 9.6  kg/s.   Initial tests
 will be conducted using  vitiated air at 1866 K  (2900°F); high-temperature
 regenerative air heaters will  be added at a later date.  The test
 train will  be  operated at  pressures ranging from 3 to 10 atmospheres.

      The  first coal combustor  tested will be supplied by the Pittsburgh
 Energy Research Center.  The combustor will be a two-stage  design
 similar to  the unit presently  being tested at PERC.

      University of  Tennessee Space Institute (UTSI)

      A new  MHD test facility is being  constructed at UTSI in Tullahoma,
 Tennessee,  and is expected to  be operational in  1979.   The  purpose of
 the facility,  in addition  to generator development, is  to provide
 design data for single-stage coal-fired combustors.  Problems associated
 with coal handling, downstream slag removal, and seed-slag  separation
 will be investigated.  A major objective  is to determine the effects of
 long-duration  operation  on component performance.

      Two  plug  flow  combustors  will be  tested.  One unit, with a  mass
 throughput  of  3.6 kg/s,  will be capable of continuous operation.  Total
 thermal input  will  be  23.7 MW  (21.1 MW coal).  Oxygen and vitiated air
 will be used as oxidizers.   The combustor will operate  at a pressure of
 8 atm and a temperature  of approximately  3000 K.

      Short  duration tests  (several hours)  will be conducted with a
 larger combustor having  a  mass throughput of 13.6 kg/s.  Coal will
 provide 56.2 MW of  the total thermal input of 66.8 MW.  This combustor
 will also operate at a pressure of 8 atm  and will generate  a plasma at
 about 2800  K.

                    V.   DISCUSSION AND CONCLUSIONS

      Three  different approaches to utilizing coal to reduce our  depend-
 ence on natural gas and  foreign oil have  been discussed.  An attempt
 will now  be made to assess the status  of  each of these  technologies, and
 to determine their  potential impact on our national energy  supply
 situation.

      Coal-Oil  Slurry

      There  appears  to be no  technical  reason why coal-oil slurry could
 not be utilized now in existing oil and/or gas fired boilers.  Consid-
 erable success has  been  achieved in firing slurries in  industrial units
 for short periods.  Tests  at PERC indicate that  erosion and corrosion
 would not appear to be a problem in retrofitting industrial gas/oil
 boilers to  fire a slurry.

      Some additional research  is required to better define  the effects
 of coal rank,  coal  concentration, and  coal particle size consist on  com-
 bustion characteristics  of slurries.   The development of an effective,

-------
                       COAL COMBUSTION TECHNOLOGY                     37


 inexpensive stabilizer would greatly enhance the attractiveness  of  slur-
 ries;  coal-oil mixtures could then be prepared in one  location and
 transported as other liquid fuels  are,  thereby obviating  on-site prepara-
 tion of  coal by the user.

      In  general,  utilization of slurries  in lieu of  low-sulfur oil  or
 natural  gas will produce air pollution concerns.   Particulate  emissions
 will have to be dealt with, as will SO emissions unless  a  low-sulfur
 coal is  used.   However, adequate particulate removal equipment is avail-
 able,  and effective flue gas desulfurization technology is  being developed.

      Wide spread utilization of coal-oil  slurry by utilities could
 potentially produce a savings of about 500,000 bbl/day of fuel oil, based
 on 1975  oil consumption data*  .   Considering only the fuel oil  used  by
 industry to generate steam in boilers  ,  an additional 400,000 bbl/day
 could be conserved.   If one also considers  direct heating operations, it
 is likely that a total savings of  fuel oil  of well over one million
 bbl/day  could be achieved.

      It  appears that the incentive for a  particular  industrial user to
 convert  to coal-oil slurry will simply be to conserve  oil supplies, or
 to have  fuel available in  the event of  natural gas curtailments.  At
 present  the estimated cost of slurry per  million  Btu is nearly equiva-
 lent to  that for fuel oil.   However,  should costs  for  foreign  oil
 escalate (and assuming coal costs  remain  stable),  use  of  coal-oil slurry
 could also provide  a significant savings  in fuel  costs.

      Solvent Refined Coal

      SRC appears  to  be a very attractive  fuel for  utility and  industrial
 applications.   Problems in handling and burning SRC  have  been  overcome,
 and the  technology  for producing SRC  is close to  being demonstrated on  a
 commercial scale.

      The use of SRC  provides  an alternative  to  flue-gas desulfurization
 which can be a significant  consideration with respect  to  capital  costs
 for a  new coal fired power  plant.   Since  the  properties of  SRC appear to
 be  uniform regardless  of the  coal  feedstock,  economies in design  and
 construction of new  power  plants can  be achieved.

      SRC could be used directly in  many existing power plants which were
 designed for coal but  are now burning low-sulfur oil to avoid  the use of
 flue-gas  scrubbers   However,  the major impact  of SRC will be through its
 use in future  (1985  and beyond) power plants which will undoubtedly be
 prevented  from using natural  gas or low-sulfur oil.

     While  the  cost  for producing SRC, or any  coal-derived  fuel is
 uncertain,  it may well be competitive with  fuel oil  .   Costs,  however,
will likely  become a secondary  concern in future years when the choice
 is between an assured  domestic  supply of fuel on the one hand,  and an
uncertain foreign supply on the other.
*Assuming 40% coal/60% oil slurry.

-------
38                     CLEAN COMBUSTION OF COAL
     Coal Fired MHD Combustors

     The attractiveness of MHD as an efficient, pollution-free process
for converting coal to electricity provides a strong incentive for
further development of this technology.  Although some degree of success
has been attained in the development of MHD combustors, a considerable
amount of additional development work must be carried out before com-
mercial-size units are a reality.  Many uncertainties exist with respect
to scale-up from present experimental information.

     A commercial MHD power plant of, say, 2000 MW (thermal) capacity
will have a combustion products mass flow rate of about 700 kg/s, and
will operate at a peak pressure of about 8-10 atm.  However, experi-
mental combustors have operated at mass flow rates in the range of 0.2
to 3 kg/s.  In general, experiments have been conducted at pressures
considerably below that required in commercial applications.

     In a base-load MHD power plant, a single combustor (with some
amount of turndown capability) would probably be most desirable.  Addi-
tional data must therefore be acquired to formulate reliable scaling
laws; this can only be accomplished by construction and operation of
combustors considerably larger than those tested to date, yet still much
smaller than a commercial unit.  Key questions that must be answered
include the effect of scale-up on combustion stability, efficiency, and
heat losses.

     It may be feasible to limit the magnitude of scaling required if
one considers the use of multiple combustors in a commercial plant.
Such an arrangement provides an added degree of flexibility in terms of
turndown  capability.  However, it is likely that the use of multiple
combustors will result in greater heat losses than those associated with
a  single  large combustor.  In addition, the complexity of the total
combustion system is increased substantially due to the need for addi-
tional controls, hot-air piping, and fuel feed systems; and problems
related to combustion stability may be magnified.  Even if the use of a
multiple-combustor system were practicable, the size of the individual
units would be substantially greater than combustors that have been
operated  to date, and the need for further testing on a larger scale
would still be necessary.

     Aside from concerns directly related to the combustion process,
there are additional issues that must be addressed in the development of
commercial MHD combustion systems:

     1)   reliable, long-term operation of MHD combustors must be demon-
strated.  Typical continuous tests conducted thus far have been of short
duration  (several hours or less);

     2)   reliable and durable systems for feeding large quantities of
high-temperature air, fuel, and seed into pressurized combustors must be
developed.  The uniform injection of coal at rates of about 65 kg/s
(2000 MW  thermal) represents a particularly challenging problem;

-------
                      COAL COMBUSTION TECHNOLOGY                     39


     3)  techniques for electrically isolating a high-temperature com-
bustor from feed lines, cooling water lines, slag rejection system, and
supports must be developed;

     4)  optimum coal-fired combustor configurations must be established;
this will be governed primarily by the percentage of ash deemed desirable
to be carried over through the generator and the downstream portions of
the system.  Single-stage combustors are likely to provide relatively
poor slag rejection, but have lower heat losses than multistage com-
bustors, whereas two- and three-stage combustors can probably operate
with slag rejection rates approaching 90%;

     5)  methods for minimizing heat losses from large-scale combustors
should be explored, such as air-cooling of combustor walls or transpira-
tion cooling of combustor refractories.

     Many of the items discussed above are being addressed in programs
currently underway in the United States and the Soviet Union, or will be
as new test facilities become operational in the near future.

-------
   40                     CLEAN COMBUSTION OF COAL
                              REFERENCES
  1.  Demeter, J. J., C. R. McCann, G. T. Bellas, J. M. Ekmann, and
      D. Bienstock, "Combustion of Coal-Oil Slurry in a 100 HP Fire-
      tube Boiler", Pittsburgh Energy Research Center, PERC/RI-77/8,
      May, 1977.

  2.  Barkley, J. F., A. B. Hersberger, and L. R. Burdick, "Laboratory
      and Field Tests on Coal-in-Oil Fuels", Trans. ASME, 66, 185 (1944)

  3.  Jonnard, A., "Colloidal Fuel Development for Industrial Use",
      Bulletin No.  48, Kansas State College, Engineering Experiment
      Station, Manhattan, Kansas (1946).

  4.  Anon., "Coal-Oil Slurry Combustion", Energy Research and Develop-
      ment Administration, FE-COSC-1, April 1976.

  5.  Cook, T. D., "Problems with Burning Coal and Oil Slurry in a
      Packaged Oil-Fired Boiler", Fall Meeting, Amer. Flame Res. Comm.,
      Philadelphia, November 17, 1976.

  6.  Harrison, W. B., "The Solvent Refined Coal Process:  Potential-
      ities and Problems", Short Course on Coal Characteristics and
      Coal Conversion Processes, Pennsylvania State University,
      May 23, 1974.

  7.  Schmid, B. K., "The Solvent Refined Coal Process", Symp. on Coal
      Gasification  and Liquefaction, University of Pittsburgh, August
      6-8, 1974.

  8.  Anon., "Fossil Energy Program Report", ERDA 76-10, Energy Research
      and Development Administration (1976).

  9.  Anon., "Coal Liquefaction", Quarterly Report July-September 1976,
      Energy Research and Development Administration (1976).

10.  Sage, W. L., "Combustion Tests on a Specially Processed Low-Ash
      Low-Sulphur Coal", Babcock and Wilcox Report No. 4439, Prepared
      for Office of Coal Research, July 1964.

11.  Frey, D.  J., "De-Ashed Coal Study", Interim Report to Office of
      Coal Research, Contract No. 14-01-001, Prepared by Combustion
      Engineering, Inc., September, 1964.

12.  McCann,  C.  R.,  J.  J.  Pfeiffer, A. A. Orning, and W. H. Oppelt,
      "Combustion Trials Spencer Low-Ash Coal", Pittsburgh Coal
      Research Center,  January, 1965.

13.  McCann,  C.  R.,  J.  J.  Demeter, and D. Bienstock, "Combustion of
      Pulverized, Solvent-Refined Coal", Presentation at Spring
      Meeting of Combustion Institute Central States Section, Battelle
      Columbus Laboratories, April 5-6, 1976.

-------
                        COAL  COMBUSTION TECHNOLOGY                      41
14.  McCann, C. R., J. J. Demeter, and D. Bienstock, "Combustion of
      Pulverized Solvent-Refined Coal", ASME Paper No. 76-WA/Fu-6
      (1976).

15.  Wagoner, C. L., et al., "Investigating the Storage, Handling,
      and Combustion Characteristics of Solvent Refined Coal", Babcock
      and Wilcox Monthly Reports to EPRI, RP-1235-1, January 6, 1975 -
      June 16, 1975.

16.  Borio, R. W., Z. J. Fink, G. J. Goetz, and J. C. Haas, "Solvent
      Refined Coal Evaluation", Technical Report 2, Prepared for EPRI
      by Combustion Engineering, Inc., June, 1976.

17.  "Comparative Evaluation of Phase 1 Results from the Energy Conver-
      sion Alternatives Study  (EGAS)" Prepared by NASA for ERDA and NSF,
      NASA TM X-71855, February 1976;  "Evaluation of Phase 2 Conceptual
      Designs and Implementation Assessment Resulting from the Energy
      Conversion Alternatives Study  (EGAS)", Prepared by NASA for ERDA
      and NSF, NASA TM X-73515, April 1977.

18.  Anon., "Fossil Energy Research Program of the Energy Research
      and Development Administration, FY 1978", ERDA 77-33, April 1977.

19.  Proceedings  of Symposia, Engineering Aspects of Magnetohydrody-
      namics, available from Dr. John Fox, Dept. of Mech. Eng.,
      University  of Mississippi, University, Mississippi 38677.

20.  "Fuel and Combustion", Chapter  10 in Joint US-USSR Status Report
      on Open Cycle MHD Power Generation, Energy Research and Develop-
      ment Administration  (in press).

21.  Carrasse, J., "Chemical Recovery of Energy in a Combined MHD-Steam
      Power Station", Proc. Int. Symp. on MHD Electrical Power Genera-
      tion Salzburg, Vol. Ill, p 883  (1966).

22.  Way, S., "Char Burning MHD Systems", Trans. ASME, J. Eng. Power,
      p 345, July 1971.

23.  Lacey, J. J., J. J. Demeter, and D. Bienstock, "Production of a
      Clean Working Fluid for Coal-Burning, Open-Cycle MHD Power Genera-
      tion", Proc. 12th Symp. on Eng. Aspects of MHD, Argonne, Illinois,
      p VI.2.1  (1972).

24.  Gannon, R. E., D. B. Stickler,  and H. Kobayashi, "Coal Processing
      Employing Rapid Devolatilization of Reactions in an MHD Power
      Cycle" Proc. 14th Symp. on Eng. Aspects of MHD, Tullahoma,
      Tennessee, p II.2.1  (1974).

25.  Zinko, H., S. Linder, and J. Raunsborg, "A New Scheme for a Coal
      Gasification-MHD Power Plant", Proc. 6th Int. Conf. on MHD
      Electrical Power Generation, Washington, Vol. 1, p 105  (1975).

26.  Brzozowski, W. S., J. Dul, and W. Pudlik, "New Concepts of Coal
      Burning MHD Plants", Ibid., p  137.

-------
  42                     CLEM COMBUSTION OF COAL
27.  Hoy, H. R.,  A. G. Roberts, and D. M.  Wilkins, Chapter 4 in "Open-
      cycle MHD Power Generation",  J.  B.  Heywood and G.  J. Womack,
      Eds., Pergamon Press, Oxford (1969).

28.  Zelinski, J. J., J. Teno, and L.  F.  Westra, "A Coal Combustion
      System for MHD Generators",  Proceedings 5th Intersociety Energy
      Conversion Engineering Conference,  Las Vegas, Vol. 1, p. 7-41
      (1970).

29.  Shanklin, R. V., L. W. Crawford,  J.  F.  Martin, J.  B.  Dicks, W. D.
      Jackson, C. R. Gamblin, and  C.  H. Tsai, "The UTSI Coal Burning
      MHD Program", Proceedings 13th Symp. on Eng. Aspects of MHD,
      Stanford University, p.II.8.1 (1973).

30.  Dicks, J. B., L. W. Crawford,  J.  W.  Muehlhauser, J. F. Martin,
      N. L. Loeffler, and B. S. Arora, "The Direct-Coal-Fired MHD
      Generator System", Proceedings 14th Symp.  on Eng.  Aspects of
      MHD, University of Tennessee Space Inst.,  p.II.1.1 (1974).

31.  Tager, S. A., E. V. Samiulov,  I.  B.  Rozhestvensky,  R. U. Talumaa,
      and F. M. lakilevich, "Development and Investigation of High-
      Temperature Combustor to be  Used for a Solid Fuel MHD Generator
      and Thermodynamic Analysis of Combustion Conditions", Proc.
      Fifth Int.  Conf. on MHD Power Generation,  Munich,  Vol. 1, p. 471
      (1971).

32.  Bienstock, D., R. C. Kurtzrock,  R. J. Demski, and J.  H. Field,
      "Experimental Unit for Study of High-Temperature Combustion of
      Coal for MHD Power Generation",  Paper 62-WA-147,  ASME Winter
      Annual Meeting, New York (1962).

33.  Bienstock, D. , R. J. Demski,  and R.  C.  Kurtzrock,  "High-Temperature
      Combustion of Coal Seeded with Potassium Carbonate in the MHD
      Generation of Electric Power",  Bureau of Mines Rep.  of Investiga-
      tions 7361 (1970).

34.  Bienstock, D., P. D. Bergman,  J.  M.  Henry,  R. J. Demski, J. J.
      Demeter, and K. D. Plants,  "Air Pollution Aspects of MHD Power
      Generation", Proceedings 13th Symp.  on Eng. Aspects of MHD,
      Stanford University, p.VII.1.1 (1973).

35.  Anon., "Quarterly Report of Foreign and Domestic Developments
      Affecting Energy", Energy Research and Development Administration,
      Planning and Analysis, May 21,  1976.

36.  Locklin, D.  W. , H.  H. Krause,  A.  A.  Putnam, E. L.  Kropp, W. T. Reid,
      and M. A. Duffy, "Design Trends and Operating Problems in Com-
      bustion Modifications of Industrial Boilers", EPA-650/2-74-032,
      April 1974.

37.  Montgomery,  J., "De-Polluting of Coal Before It Is Burned Is
      Tested as an Alternative to  Scrubbers", Wall Street Journal,
      June 14, 1977.

-------
                                                                      43
          SESSION  IB -  STRATEGY  AMD  APPROACH  TO  SPONSORED R&D

               SESSION  CHAIRMAN:  ANDREJ  MACEK,  U.S. ERDA
      While it  is  realized that  the massive  projected increase  in utili-
 zation of coal (more than 50% between 1976  and 19&5)  cannot  be accom-
•plished merely by infusion of federal funds,  federally  sponsored R&D
 will clearly be instrumental for this attainment.   This session provides
 information from top officials   of the three  principal  federal agencies
 sponsoring R&D in this area: ERDA, Bureau of  Mines, and EPA.   These
 papers cover what the Government is doing,  and what it  intends to be
 doing in the future, to implement clean combustion of coal.

      A significant amount of research is also being sponsored  by the
 Electric Power Research Institute, but they were not able  to send a
 representative to this conference.

-------
44                     CLEAN COMBUSTION OF COAL

-------
                                                                      45
             STRATEGY  IN  COAL PREPARATION RESEARCH  PLANNING
                              W.  E. Warnke
                    Coal  Preparation Research Manager
                          U.S.  Bureau  of Mines
     Good evening to  all of you.   Because we've just eaten dinner there
is some danger of being lulled to  sleep.  As  it happens, I don't have
any  visual aids or funny stories to tell as an aid in maintaining your
attention but I will  cite  some percentage numbers that may pique your
interest.  Those of you who are familiar with RI 8118 on sulfur removal
potential may be pleasantly surprised  by these numbers tonight.

     Let's not belabor the issue of increasing the utilization of coal.
The  necessity of replacing gas and petroleum  with coal has been
eloquently discussed  by earlier speakers.  So, let's talk about the
four options as I perceive them in burning more coal.  Then, I'll out-
line the Bureau of Mines strategies in planning a five year research
program for coal preparation.

     The electrical utility industry and other industries requiring
large amounts of fossil fuel have four options for burning coal in
compliance with sulfur dioxide emission standards.  The first is the
use  of flue gas scrubbers.  Costs of scrubbing sulfur oxides from flue
gases range from $10  per ton of coal burned to $19 per ton depending on
how  the capital costs are  amortized and who does the cost evaluating.
For  lack of a better  number, let's assume these costs are $15.  Flue
gas  scrubbing also is an energy user because  limestone must be mined
and  transported to the utility plant and perhaps as much as 5 percent
of the energy derived from burning coal is used to drive the scrubbing
system.

     Transporting low sulfur subbituminous coal from Wyoming and
Montana to Eastern and Southern utilities is  the second option.  Unit
train costs are about one  cent per ton mile so a haul of 1200 miles
adds $12 to the costs of a low grade fuel averaging less than 20,000,000
Btu's per ton.  Unfortunately, much of the subbituminous and lignite
coals contain one percent  or more sodium which causes the ash minerals
to slag at temperatures of about 2000  degrees F.  Boilers designed to
burn the bituminous coals  of the East  need retrofitting to avoid slag
buildup on heat transfer surfaces.  The retrofitting is expensive and
decreases boiler efficiencies.  I've not considered coal slurry pipe-
lining as a means of moving coal from  mine to utility for the simple
reason that pipelines may  not be used  for reasons other than economics.
If costs of installing a coal slurry pipeline escalate like the Alaskan
oil line,  the pipelines may not be competitive with unit trains.

-------
46
                       CLEM COMBUSTION OF COAL
     The third option is cleaning high sulfur Northern Appalachian  and
Midwestern coals to acceptable sulfur levels.  This option is restrict-
ed to cleaning coals from these two regions because the Mississippi
Basin and Northeastern regions of this country represent areas of high
demand for steam coal.  According to a study conducted by the Bureau of
Mines on the sulfur removal potential of 227 samples taken from major
bituminous seams in the states of Pennsylvania, Northern West Virginia,
Ohio, and Maryland, 100 samples could be beneficiated or burned without
any beneficiation as compliance coal.  The mean Btu recovery was 75
percent.  Coals from the Midwestern region are generally higher in  both
pyritic and organic sulfur than Eastern or Southern coals.  Therefore,
it was not at all surprising to learn that only 11 of the coals sampled
in the Midwestern region could be beneficiated to compliance levels of
sulfur with a mean recovery of 64.5 percent.  The results of sink/float
tests on 455 samples of coal from throughout the United States show
that 42 percent can be burned as run-of-mine coal or can be benefici-
ated to meet compliance sulfur levels.  Mean Btu recovery of these  192
coals was 83.6 percent.  These washability tests simulated current
washing practices in which only pyritic sulfur is removed.

     Costs of cleaning coal by some form of physical separation to
remove the pyrite and ash varies from about $2 per ton for the more
simple flowsheets to $6 to $7 per ton for the more complex flowsheets.
The  latter includes equipment for fine grinding, flotation cells,
thickeners, filters and dryers.

     If we plan to use the 58 percent of the coals that are not amen-
able to conventional cleaning technology, then we must develop new
physical cleaning technology and in some cases, combine it with chemi-
cal desulfurization.  Bechtel Engineering Corporation recently complet-
ed an evaluation of six chemical cleaning processes.  Although several
of the six removed some organic sulfur, none of the six appeared
promising.  Costs for the entire process including crushing, grinding,
chemical processing and compaction of the fine clean coal, ranged from
$18 to $20 per ton.  The Bureau of Mines has reservations about these
processes for the simple reason that the costs equal or exceed flue gas
scrubbing costs.  We are investigating several other chemical processes
but Richard Killmeyer of the Bureau's Coal Preparation Laboratory will
discuss our plans later on another day.

     The fourth option mentioned earlier is a combination of coal
cleaning and scrubbing.  If a coal can be partially desulfurized by a
physical process costing $3 per ton or so and the coal burned with
about 1/3 of the flue gases going to a scrubber, the combined costs may
be less than scrubbing alone.  There is no hard and fast rule on the
amount of sulfur in the run-of-mine coal, the cost of partially desul-
furizing by physical cleaning and the ratio of unscrubbed flue gas  to
scrubbed flue gas.  Each coal would require testing to determine the
amount of sulfur removed by cleaning and the amount of the gases
requiring scrubbing to meet the 1.2 pounds of sulfur dioxide standard.
Hoffman-Muntner Associates prepared a study on combined cleaning-
scrubbing costs for the 3ureau.  This study which is being published
for public dissemination, outlines 12 combinations of coals, flowsheets
and scrubbing systems.  The costs of combined cleaning/scrubbing are
compared to scrubbing alone.

-------
                   COAL PREPARATION RESEARCH FLAMING                 47
     We in the Bureau of Mines believe that cleaning the coal prior to
combustion is the most attractive of the four options both economically
and environmentally.  When coal is cleaned to remove pyrite, a major
portion of the more objectionable trace elements such as cadmium, lead,
arsenic, and mercury are also removed.  These elements are generally
not removed by present day scrubbing processes which means they are
redistributed as fallout from the emitted flue gases.  Furthermore,
wastes from washing plants can be dewatered and used as stable landfill
material in an environmentally acceptable manner, whereas sludge from
scrubbing plants is more difficult to impound or store without jeopard-
izing the environment.  And, from an economic standpoint, if you'll
excuse reference to the obvious, a $6 or $7 per ton cost for sophisti-
cated cleaning processes is considerably less than the $15 scrubbing
costs.

     The Bureau recognizes that current coal washing practices are
inadequate in removing finely disseminated pyrite and remove no organic
sulfur.  Consequently, the Bureau's program is heavily oriented towards
fine grinding to obtain optimum liberation of pyrite and with emphasis
on flotation or high gradient magnetic separation to remove most of the
pyrite without sacrificing Btu recovery.  The Bureau is also concerned
with chemical desulfurization of coal as a means of removing organic
sulfur.  When coal was selling for 10-15 cents per million Btu's, the
type of projects the Bureau authorized was constrained by the funds
available for research and by the costs of the processes under
investigation.

     Our coal preparation research program in the past consisted of
small projects designed to achieve engineering advancements to the
state-of-the-art.  We didn't have funds for large projects and couldn't
afford much risk.  Virtually overnight the game changed from penny ante
to a sky-is-the-limit high roller game.  This is not to imply that
we're squandering funds on worthless ideas because the projects we fund
are based on bench scale data and evaluated for technical feasibility.
But the stakes are measured in billions of dollars annually to the
consumers of electrical energy and so because of the stakes we can
afford greater risks in the hope of developing a process for cleaning
Eastern coals.

     Perhaps I'm too optimistic but I believe that physical coal
cleaning technology will be developed so a much larger percentage of
the coals in the Eastern half of the United States can be cleaned to
compliance levels.  I'm also optimistic about increasing significantly
the Btu recovery in the clean coal product.  That, in essence, is the
Bureau's strategy.  We are convinced that both physical and chemical
coal cleaning deserve more consideration by policy and decisionmakers
in industry and government as very attractive alternatives to scrubbing
or transporting coal long distances!

-------
48                     CLEAN COMBUSTION OF COAL

-------
                                                                     49
             STRATEGY AND APPROACH TO  ERDA
                RESEARCH AND  DEVELOPMENT
              ON CLEAN COMBUSTION  OF COAL
                        Dr.  S.  William Gouse
           Deputy Assistant  Administrator for Fossil Energy
          U.S. Energy Research  and Development Administration
As  the  population  and economy of the United  States  continue  to  grow,
the  need  for energy  will  increase.   In  1970  the  U.S.  used  about 67
quads;  in 1976  about 74,  and projections  for the year  2000 range
from  less  than  100  to over  150  quads,  or more  than  twice  the  1976
level.

Although  conservation and  new end-use  technologies can  limit the
amount  of  energy we will need, society will  still need  large amounts
of  energy  to  support  our  growing  and  productive  economy.  This energy
can  come  from a variety  of forms,  including solar,  nuclear, and new
and conventional solid, liquid and gaseous fossil fuels.

The  main  issue  for RD&D planning is  not  how  many quads  we  will  need
in any  future year, but how we will  choose to provide  them.

Today about  75  percent  of  our energy  comes  from petroleum and natural
gas,  and  we  import some 40  percent of  our  liquid  fuels.   There  is  a
great deal  of uncertainty regarding  the  future  availability  and  cost
of  those  fuels.   Known reserves are being  depleted rapidly and new
discoveries have not  kept  pace with  rising demand.  Indeed,  there is
some  question as to how long  the earth's  total  recoverable  resources
could support historically rising demand  for  oil  and  gas.  All studies
indicate that synfuels in various  forms will be required  in  the future.

We  need appropriate  action  now to  insure adequate supplies  of energy
at  reasonable  prices  to  preserve  our  national  security and  provide
reasonable lifestyles.

There is no  magic  solution.   Any  specific approach to the problem has
its  own advantages  and  disadvantages and  these are often  perceived
differently by different groups.  There are many complicating factors,
each with  its own  degree  of uncertainty  and  each interacting with all
the others, often in unpredictable and counter-intuitive  ways.

As an example of this interaction,  consider new  home heating systems.
A perceived supply problem in natural  gas could push  electric heating,
even  though that  would require  considerably more primary energy.

-------
 50
                CLEAN COMBUSTION OF COAL
The  complexity  of  the  problem  makes  it  difficult  to determine  the
"best" solutions.   The first  part  of the fossil  energy strategy is to
define  alternative  ways  to meet future  energy  service requirements in
terms  of  social, institutional  and  economic consequences.   The second
is  to provide the  information to  stimulate  a productive national
debate, which  we hope  will lead  to a consensus  on acceptable  alterna-
tives.  Finally, we  seek  to conduct an R,R&D program that will clarify
issues and  insure  that an effective number  of  viable  alternatives is
available.

There  are  several  dimensions  to  the  energy  picture,  and   that  fact
accounts  for  much  of  the  complication.   Two  of the most  basic dimen-
sions  are the  energy resources  and  the  activities required to use them.
These  are effectively  related  by the  Reference Energy System.


  Reference Energy System, Year 1985 — With  the NEP
 Resource


 Nuclear Fuels
 U235 LJ238




 Hydropower
 G isothermal

 Solar

 Fossil Fuels
  Coal




 Crude Oil
Natural Gas
Total Resource
Consumption:
                Refining &
                Conversion
               Transport &
               Storage
Central Sta.
Conversion
Transmission,
Distribution &
Storage
Decentralized
Conversion

7.62 ( ( ^
Enrich & Fabricate Truck
*" Enrich & Fabricate Truck
3.11
Dam Hydroelectric 765 KVAC
Dpl8 Long Diitance
"0.06 *" *" *'
0.1 2 Thermal — — — _ _^ Q.18
0.03 B'mB
F^ion
-*•'" LMFBR
'-'1.34.
LWR '
HTGR *""
1.34) ^m
(.34)

f
L34)
^
"^%80
iQ^^b.
UG. Dili

°A if*.
• ?i fr*"
w! v
v/45 ¥ ^
       i.9as>~ ~ -. -«.
92.67 * 1015 BTU
The major energy  sources  are  listed  down the left-hand side and include
nuclear, hydro,  geothermal, solar,  and  the fossil fuels.   The activi-
ties  listed across  the  top  include extraction,  refining and conversion,
transport,  distribution  and  use.    Conversion  can be  both centralized
and decentralized.

The  Reference Energy  System,  by  identifying  both service  demand and
utilizing  device,  permits  the  implicit consideration  of  conservation
and efficiency improvements.   This system can help us consider alterna-
tive  pathways.    Once  the potential  demands are estimated and various
sets  of resources  and use technologies  identified,  we  can determine

-------
                         ERDA RESEARCH AND DEVELOPMENT
                                                                     51
various resource mix  requirements.   This forms  a standardized basis for
comparison in  these two dimensions.
                          Fossil  Pathways
                      Transport
                                    Transport
                              Process     ||
             '^^•LTconveyorU D^Lf  *"  T_ ;*"» "P-'-H.,

             K'Mi"8  J  I Truck  J I Crush J  Lp^meJ  ' dn-SSK. B,u,
                                  Convenion  Convertjon
                                     I      ||    Distribution
                                               Liquefy
                                             Gasify) Low Btul
  Natural.
  Gas
G*w«"  T	. _ ,  J" „ ^   ]- Gas Pipeline -,
      	Gas Pipeline -i- Pressurize J        V
Oil Well  J          I        rLNG Tanker J
,»i,«»4 r:«l          *- Liquefy  —*
           (Associated Gas!
         •-Underground Mi
         L  s,
  Oil Shale	1
             tin-Situ   -i
             In-Situ   J
         f Rail  -i
irground Mine -i I        Aboveground

.rfaceMine  H" ^ T~  Htl<"1  1
         L> Conveyor -*        ^> Pipeline

 i_ 0:...   1	tn-Situ Retort-l
           (Mine- Assisted)
            Oil Well
          Primary Recovery
          4rnmarv Hecovery -i
             Oil Well   J
          Secondary Recovery I
             nu Waii   -J
                        Oil
             Oil Well
          Tertiary Recovery
        • Option points                         Figure 2

 Within fossil energy there are four  major  resources—oil,  gas,  coal  and
 shale.   Those  resources  can,  via  alternative  pathways,   meet  end-use
 needs  for all solid, liquid and gaseous fuels.

 I  should point  out  that  exploration,  although  not  shown  here,  is  a
 major  activity,  particularly  for oil and gas.
                          Coal  Resources
      Legend

      Bituminous Coal
      Subbituminous Coal

      L'9nite
      Anthracite
                                                       Regions

                                                       1. Northern Appalachia
                                                  '•i-    2. Central Appalachia
                                              K conn.    3  Southern Appalachia
                                                       4. Midwest
                                                       5. Central West
                                                       6. Gulf
                                                       7. Eastern Northern
                                                         Great Plains
                                                       8. Western Northern
                                                         Great Plains
                                                       9. Rockies
                                                      10. Southwest
                                                      11. Northwest
                                                      12. Alaska (not shown)
                         Figure 3

-------
52
                         CLEAN COMBUSTION OF COAL
Within  each major category, there are  wide variations  as to the nature,
characteristics and  location of the  resources.  This  figure shows  where
the  nation's major  coal deposits  are  found,  as well as  the location of
coal  by  rank  from   peat  and  lignite  and  peat through bituminous  and
anthracite.

An  additional  complication concerns  the fact  that coal  is  not a single
material;  rather, it  is  a  family of  related materials,  differing widely
in  heat  content, contaminates  (such  as  sulfur  and  ash)  and physical
properties (such as  caking  and  hardness).   These characteristics
greatly affect their use.   Processes  for  coal conversion  do not work
equally well on  all coals.   It  is  not useful to  consider  liquefaction
of  anthracite.
                      Shale  Resources
       Green River Shales
       1.8 trillion bbl oil equivalent
       in shale over 15 gal/ton
                                                            Explanation


                                                            Tertiary deposits
                                                            Green River Formation
                                                            in Colorado, Utah, and
                                                            Wyoming; Monterey
                                                            Formation, California
                                                            middle Tertiary deposit
                                                            in Montana. Black areas
                                                            are known high-grade de-
                                                            posits
                                                            Mesozoic deposits
                                                            Marine shale in Alaska
                                                            Permian deposits
                                                            Phosphoria Formation,
                                                            Montana
                                                            Devonian and Mississippian
                                                            deposits (resource estimates
                                                            included for hachured areas
                                                            only). Boundary dashed
                                                            where concealed or where
                                                            location is uncertain
                                   Figure 4
This  figure  on  shale  resources  shows  the  great  variability  in  the
location  and  characteristics  of  the  shale  resource.   Western  shales
have  a high  potential  for  liquids  production  but  are far  from  prime
markets.   Eastern  shales have  promise for  both  gas and liquids  produc-
tion  and  are near demand  centers.   A similar  situation  exists  for  oil
and gas.

-------
                ERDA RESEARCH AND DEVELOPMENT
                                                             53
        Heavy  Oils And Tar Sands
      Legend
      Oil field
     3 Area of heavy oil accumulations
                               100 0 100 200 300
                               I  I  I  I  I
                                 Scale, miles
                            Figure 5
Low Permeability Sandstone Areas/
              Geopressured Zones
                          Geopressured
                          Aquifers
                          (Zones)
                           Figure 6
 1. Greater Green
  River Basin
 2. Northern Great
  Plains Province
 3. Piceance Basin
 4. Uinta Basin
 5. Anadarko Basin
 6. Arkoma Basin
 7. Big Horn Basin
 8. Cotton Valley Trend
 9. Denver-Julesburg
  Basin
10. Douglas Creek Arch
11. Ft. Worth Basin
12. Ovachita Mountains
  Province
13. Raton Basin
14. San Juan Basin
15. Snake River
  Downwarp
16. Sonora Basin
17. Wasatch Plateau
18. Western Gulf Basin
19. Williston Basin
20. Wind River Basin

-------
54                     CLEAN COMBUSTION OF COAL


Figure  1  can  be  examined  again  to  explain much of our program content.
For example,  the oil and gas program focuses on enhanced recovery tech-
niques;  shale focuses on  in situ conversion;  and coal conversion  on
producing  liquids and gases.

In  the  industrial  sector,  for example, we are  concerned with  the  need
for  process  steam,  feedstocks,  electrical  and mechanical energy,
indirect  heat and  direct  heat.   Each  of  these  represents a  demand  for
liquids,  gaseous and solid fuels.

Transportation,  with its  very   large  requirement for  liquids,  is  an
excellent  sector to  illustrate  the available  programs and  pathways.
The obvious way  to satisfy these needs  is through petroleum—off  shore,
enhanced  recovery,  etc.    But  coal and  shale  may also produce  liquid
fuels.    However,  coal  liquids  are  chemically  different  than  shale
liquids.   Coal could make gasoline while shale  could  produce  jet  fuels.

Boiler  fuels  from   all  four resources  can meet  electricity utility
demands,  but  the emphasis is on coal.

End  use devices can  offer  compromises  with  supply technologies.   For
example,  we  can heat homes  with coal  through  the  following  paths:

      o  Coal  to  SNG, conventional gas  furnaces

      o  Coal  to  SNG, gas heat pump

      o  Coal  to  synthetic liquids,  conventional oil furnaces

      o  Coal  to  electricity, scrubber,  resistence heat or heat  pump

      o  Coal  to  SNG to electricity, resistence  or heat pump
                                               /
      o  Coal  to  synthetic liquids to electricity, resistence  or heat
        pump

Because  the  energy  system is  so pervasive   and complex,  the two-
dimentional  matrix   is  not  sufficient.    There  is  another  important
dimension  to  the problem.   Another  dimension   involves constraints  to
applying each supply and end-use technology,  I  have listed eight  of  the
most  obvious  ones here,  representing  the areas  in which there must  be
compromises in selecting R&D activities for Federal support.

Each  of these must  be understood in terms of  its  consequences and  the
manner  in which it  interacts  with   all  others.   For example,  any
decision  on  environmental  controls  will  certainly  have  social   and
economic consequences.

The fourth dimension is  time.   We are facing  the  transition from  a
normal  oil  and  gas-based  economy  to  one based  on  renewable or non-
depletable  resources.   We  aim  to use  coal,   and  other conventional
fossil  resources  to get  us through  this  transition with  minimum
disturbance to society.

-------
                      ERDA RESEARCH AND DEVELOPMENT
55
           Constraints
                         Constraints








//










/&&&$3f$?m8t$vSW
Constraints
1.
2.
3.
4.
5.
6.
7.
8.
Scientific knowledge
Applied technology
Geographical
Legal & legislative
Political & social
Economics (capital, taxes)
Environment
Manpower & Materials
1
PlilSPi!










Coals
// Oils
/ Gases
Shales




                                                              Activities
                                   Figure 7
In  our  decisions, we  must recognize that we  are  being  pushed by the
decreasing  availability  and increasing  cost of  oil  and  gas.   We must
also  recognize  that  this situation will become more urgent with  time,
especially, if we do not act now.

Time  is  important because  all  technologies  are  not  in the same  state
of  readiness.   Because of several fossil energy options for each end-
use requirement plus other ERDA programs,  all technologies may  not have
an  implementation window.   Careful study of potential market  penetra-
tion must be used as  an  aid to setting  R&D  priorities  as well  as  other
things mentioned  thus far.

With  this  background  and  overview  let  us  concentrate on  the Fossil
Energy program, which in addition to clarifying social  issues,  seeks to
develop  the quantities  we  need  at  acceptable economic,  social and
environmental costs.

We  want  to increase  supplies  in the near  term  for  all markets  using
liquids and gaseous  fuels,  as  well as for electric power.  The appro-
priations and activities for coal, oil, gas  and  oil  shale differ.  The
reason concerns the fact that  because of  the unequal levels of private
R&D investment for the different fossil  fuels,  the  state of  development
of  conversion  and  extraction  technologies  varies considerably.   You
will note that  the  requested FY 78 Budget for Fossil  Energy is $656.9
million and coal receives almost 82 percent  of  that amount.

-------
56
CLEAN COMBUSTION OF COAL
 Fossil Energy  Budget  Estimates-Distribution of Funds
      Petroleum and  Advanced Research and
      Natural Gas    Supporting Technology
Oil Shale
and In Situ
Technology
        X  \  I l./7a 1     /
                        Coal
                        Utilization
                         15.4%
                         12.0%
   Modifications
   atERC'S
       _
     1.5%
                 Coal Conversion
                 Coal Utilization
                 Advanced Research
                 and Supporting
                 Technology
                 Demonstration Plants
                 Magnetohydrody-
                 namics (MHO)
                 Petroleum and Natural
                 Gas
                 Oil Shale and In Situ
                 Technology
                 Modifications at ERC'S
                      Total
              Percentage Distribution of Fossil Energy Budget
              Estimates in FY 1977 and FY 1978 Shown as Follows:
                         FY 1977%
                         FY 1978%
                                                           Budget Authority
                                                           (Dollars in Millions)
                                                                     Increase
                                                         FY77   FY78   Decrease
                                                         $150.3  $233.3  $+83.0
                                                          74.4   79.1   +4.7
                                                          37.1
                                                         100.3
                                                              40.3    +3.2
                                                             125.9   +25.6
                                                          40.0   50.5  +10.5

                                                          43.2   76.7  +33.5
                                                       31.0   41.5   +10.5
                                                      	6.9  	9.6   +2.7
                                                      $483.2  $656.9 $+173.7
                                    Figure 8
 Thus our commitment to Fossil Energy,  especially coal  is  great, because
 the  resources  are abundant.    In  total,   they can  supply  all  of  the
 Nation's additional needs  for  over a century—more than  enough time  to
 develop technologies  to exploit  "inexhaustible" resources economically.

 Our  objective  is  to   develop  technologies  that  will have  widespread
 applicaton  by private  industry.    How  do we  go about  this?   Is ERDA/FE
 the main  customer  for the  technology  it  helps create?  The  answer  is
 no.  At  ERDA we  feel   that  as  the main  producer or consumer of energy,
 the private sector  is primary  and  the  role  of the Federal Government
 should be the  supplementary one  of sharing risks with  private industry.
 Thus, the  national needs  we are charged to  satify are felt and satis-
 fied almost entirely  in the private sector—by industry,  commerce  and
 individuals acting through the marketplace.

 So  in  order  to  achieve  our missions  and  goals,   ERDA/FE seeks  to  (1)
 establish  the  appropriate  policy  and   technical   climate  (with  appro-
 priate  incentives,  if  needed)  for private  sector  action;  (2)  share
 risks with  private industry; and  (3) support a complementary RD&D  pro-
 gram to obtain necessary  and timely information  and  to  help stimulate
 the private sector.    Thus,  ERBA  does  not expect to  be  in  the energy
 business.   We  have  nothing  to   "sell."   Our  approach  is  to utilize
 federal funds   so  that the private sector will participate right  from
 the start  of  a project with its own know  how and financial resources.
 Indeed, we  aim for  industry to  be involved  in our programs every  step
 of the way—as  the major contributors  of technical  ideas  and approaches
 and as  the  major  beneficiaries of operating  experience.   Therefore,
 industry will  be  completely  familiar with  the problems as  well as  with

-------
                     ERDA RESEARCH AND DEVELOPMENT
                       57
the results  and data.   Such  involvement will  put our  industrial
partners  as  well  as  the  industries  themselves in  an  ideal  position to
decide whether  to implement  a  new  technology.   In turn, we will be in
an ideal position to  know  the obstacles—financial  and otherwise—to
overcome  in  order for the technology to be implemented.

This   figure shows that  industry receives the overwhelming percentage
of the Fossil Energy Budget.


        Distribution of Fossil Energy Funds
            Universitii
             3.8%
             435%
              National Labs
                7.3%
                5.2%
                         Percentage Distribution of Fossil Energy Budget
                         Estimates in FY 1977 and FY 1978 Shown as Follows
                                      FY 1977%
                              Figure 9
                                      FY 1978%
          Fossil Energy Budget Estimates Breakdown
          of Funds by R£rD Agency Budget Authority
                        (Dollars in  Millions)
      Energy Research Centers

      National Laboratories

      Universities

      Industry

      General Plant and Equipment,
        Construction, OSHA and
        Environment at Energy
        Research Centers

               Total
FY 1977
$47.0
35.2
18.2
375.9
(% of
Total)
(9.7)
(7.3)
(3.8)
(77.8)
FY 1978
$60.9
34.0
26.1
526.3
(% of
Total)
(9.3)
(5.2)
(4.0)
(80.1)
(1.4)
(1.4)

-------
58
                      CLEAN COMBUSTION  OF COAL
Industry  is  involved in the typical  development sequence  shown below.

  Typical  Development  Sequence

II



1

1


Private Industry
II 1

i

L 11 Jt
              Exploratory
              Research
              1-4 Years
 Process
Development
Unit (PDU)
 4-6 Years
 Pilot
 Plant
5-8 Years
                                             / Demonstra-  ' '
tion
Plant
/  I  Commercial  /
     Plant   /
                                               8-12 Years  I I
              rr     IT     IT   IT    "\i
                              Government
                              15 to 20 Years

                              Figure 11
Each phase in the  sequence involves  scaling up to larger units,  until
in  demonstration plants,  the  scale  is large enough to  provide  firm
data for cost estimates and design of  commercial scale  plants.   Cost-
sharing,  while  concentrated in the pilot  plant  and  later phases,  can
also take place earlier.  The time involved with  each phase of develop-
ment varies, depending  upon  the complexity  of  the  process,  project,
scope,  and resources applied.

Because  in our  accelerated program  the  phases  overlap, the  total
development time is  less than the sum  of  all  phases.  From laboratory
to  completion of demonstration plant operation  is typically  15  to  20
years,  and we evaluate  technical  feasibility  of  the  concepts  in each
phase.   Tentative economic  evaluations  start  early in process develop-
ment and  continue through pilot and demonstration phases.  We make more
extensive economic  evaluations with  demonstration plants,  as  well  as
tentative  environmental  acceptability evaluations,  water  resources
availability  assessments,  and  environmental  impact  statements,   as
required.

There has been much talk about  the possible need  to  start up whole new
industries  to implement  new technologies.  For some ERDA programs, such
as  solar energy, where  the  existing industrial  base  is  small or non-
existent, this  idea undoubtedly  has some  validity.   Such  is  not  the
case for Fossil  Energy where for years, a  large base of  existing com-
panies  has supplied the Nation's  oil, natural gas,  petrochemical  and
electricity needs.   And,  I might add, they  have  done  that job  well.
For FE we need,  not new industries,  but  technologies that either make
sense in  the marketplace, or which can  be made to make sense by govern-
ment initiatives, if there is sufficient public benefit from developing
such technologies.

-------
                      ERDA RESEARCH AND DEVELOPMENT                     59


An example would  be the synthetic  liquids  we  will need in the future,
derived  from  coal  or  from  oil  shale.   There  are barriers  to  their
commercialization.   For that  reason,  ERDA/FE has  a  commercial demon-
stration program, which  we  hope  will resolve the uncertainties related
to economic/environmental  feasibility,  socioeconomic  impacts,  resource
requiremments,  capital  cost,  financing and  regulations.   We  need to
resolve  them  in  order  to  bring about  the kind  of  broad-scale  plant
investment in the  mid  1980's  which we  will need  in  order  to achieve
significant production in the  1990's.

At  the  same  time  we  are  developing  Federal  initiatives  to  build a
synfuels industry,  we  are  also  working in  the  FE program to  overcome
the technical deficiencies  in  the current state-of-the-art.  We do this
through  heavy investments  in RD&D per  se.   We now support research in
universities  and  industrial labs,  and  process development in  corpora-
tions  and  laboratories.   Pilot  plant  scale operations are underway in
companies, in government and  in independent centers.   And ERDA/FE has
an  active and ambitious  program  to   demonstrate at  near  commercial
scale,  promising  second generation  coal  and shale processes developed
by both  industry  and government.

Such  activity pushes forward  the  frontiers of  knowledge  and  advances
the state-of-the-art.  Our  premises  are:  (1)  Because  of the regulatory
and economic  uncertainties  I mentioned before, the  private sector won't
push  as  fast or  as  extensively  as  the  national interest  warrants.
Thus,  government  provides  incentives and the means for faster  and more
extensive progress; (2) As  the state-of-the-art advances, a company can
finish  developing a commercially attractive process  more  easily.   For
patent  reasons,  a company may want  to  put  its own stamp on a process,
and thus, not adopt any we have sponsored  in  its  entirety.  But a com-
pany  will  have to  contend  with far less uncertainty  than  it  does now
(or did a year ago)  and should have in hand  the  ingredients  for suc-
cess;  (3)  Many  "systems"  problems  faced  by  new  technologies can be
understood  and resolved  only  by  actual trial  (and  some, by  not  too
much  error).    Environmental  impacts,  effects  of  regulations,  process
bottlenecks,  etc.,  are  still  most  clearly identified  and  best  dealt
with  in  actual  practice.   Hence, our demonstration program aims to put
the pieces together at  a near-commercial scale to  show potential  users
of the  technology what  applying  it  will involve;  (4)  The existence and
scale  of the  program  highlights an important  technological  area and
focuses  attention on it.  Companies  not  funded by  ERDA thus may also be
induced  to invest in a  particular  technology.   Companies having undis-
closed  processes  may  have to  advance  them  in  contention.   Researchers
may be stimulated to work on related topics, once  the  area, as a whole,
becomes  "fashionable."   For  such  spill-over  benefits,  the  government
need  not pay directly;  (5)  The  people involved  in the  program,  in
ERDA,  in our  Energy Research Centers, in industry,  and in the universi-
ties,  acquire  knowledge  and skills  indispensible  to the eventual
largescale commercialization of  new  technology.

In  our  program,  we also   address,  in  several  ways,  the  obstacle of
economic risk of the R&D.

-------
60
CLEAN COMBUSTION OF COAL
 Targeted Cost-Sharing Participation
          PDU's
          Pilot  Plants
          Demo  Plants
          Alternative  Fuels
          * Some Plants May Be Cost-Shared

                               Figure 12

We pay 100 percent of  the costs of basic and applied research and  all
the  costs  for the early  stages of process development.   We  pay only
about 2/3 of the costs  for pilot plants, and about 1/2 of the costs  for
demonstrations.  We do this partly because the costs rise  as one moves
along the  development  sequence.   But  more  importantly,  we do  it  to
begin commercialization.  Cost-sharing ensures that projects have merit
by private standards  and  that  industry will  make a later commitment
of other  resources to  protect  its  investment.

Because  our industrial  partners must put up a significant  share of  the
funds,  we  expect them to make  a  significant share  of the decisions
about what and  how to  test.   Thus,  they  bring their technical  and
marketing judgements to bear  early.   What  will they  learn  from  the
project?   Will the  product sell?   Will the process work?  What will
the  product cost?   We issue  flexible  Program Opportunity Notices  or
specific  requests for proposals in areas that appear   to warrant  that
scale of development.  Companies  then  respond with  detailed proposals
specifying  what  they prefer  to  do.   Because their stockholders' money
is on  the  line,  they  must  apply  their best commercial   insights  and
expertise.

As a result,  the process  development sequence ceases to be a straight
path.  Each stage,  merges, modifies or drops  components and adds  new
ones.  We  would be surprised  if  many  proposed  demonstration projects
were direct lineal? descendents of our process  development  units  or
pilot plants.

More generally,  ERDA  could  find  other ways  to  reduce economic  risk
besides  cost-sharing,  including  loan  guarantees,  price  supports  and
special  tax credits  for energy related  investment.

-------
                      ERDA RESEARCH AND DEVELOPMENT                      61


No one fully knows the  long-run  impacts  of  these  different  approaches.
We are  studying  some of them  now and we need  to look at others.   We
are  studying  one  of  them  now  and we need  to  look at others.  We  are
studying, for  example,  projected market demands  for  the  several  forms
of  fossil  energy  and  costs  of  advanced  technologies.   We  are  also
looking  at  contingency  plans  and  their  impact  on  overall  strategy
and  choice of  federal role.   We  consider too,  in selecting the Fossil
Energy  project  "portfolio," how to balance  the  need for  short-term,
relatively  sure  "blue  chips"  against  the  need  for  near-to-mid  term
improved  processes,   moderate  risks and  potential breakthrough  "high
flyers."

In conclusion, I  am optimistic that our Program  can  supply the  fossil
energy bridge  to  the next  century  if we can develop  a real government
private  industry  partnership.   Such a partnership will clear away  the
roadblocks--environmental,  price,  regulatory  and  legislative--to
bringing  promising  technologies   into  the  marketplace.    Conferences
such as  these  help to find the  path  to  innovation because we need  new
technology  to solve  our  energy  problems.    We  need  entrepreneurship
because whatever is developed  needs  to be  brought to market,  not
bought by the government.

-------
62                     CLEAN COMBUSTION OF  COAL

-------
                                                                     63
                      EPA R&D PROGRAM RELATING TO
                      CONVENTIONAL  COAL  COMBUSTION

                                  by
                          FrankT. Princiotta
                 Director, Energy Processes Division
               Office of Energy, Minerals and Industry
                 U.S. Environmental Protection Agency
     The Clean Air Act Amendments of 1970 are the major driving force for
control of air pollution from both new and existing combustion sources.
The Act has a statutory requirement to achieve acceptable ambient air
quality for the so-called criteria pollutants (Figure l).  Among these
pollutants are sulfur dioxide and total suspended particulates which are
essentially stationary source pollutants and nitrogen oxides which are
generated by both stationary and mobile sources in roughly equal quanti-
ties.  In addition, the Clean Air Act calls for the promulgation of New
Source Performance Standards (NSPS) for a variety of polluting industries
including coal-fired steam generators.  Presently standards are on the
books for control of sulfur dioxide, total suspended particulates and
nitrogen oxides from coal-fired, oil-fired and gas-fired steam genera-
tors.  The present NSPS for coal units, as well as typical uncontrolled
emissions are included in Figure 1.  As you can see, the present NSPS for
coal-fired steam generators calls for approximately 70-80% control of
sulfur oxide; for a high degree of particulate control (approximately
98$); and a moderate 30!? control for nitrogen dioxide.

     I must point out that there are certain changes which may occur both
in terms of the Clean Air Act  as well as some revised standards under
the present Act which could have important impact on control technology
requirements for fossil-fuel combustion units.  First, the House and
Senate have recently passed legislation which has gone to joint committee,
calling for a best available control technology approach (BACT) for new
coal-fired power plants insofar as sulfur dioxide and particulate pollu-
tion is concerned.  Although the implementation details of this approach
have not been worked out, and the final version of the Act has not yet
passed both Houses, it appears that this change would require best
available control technology for all new sources on both low- and high-
sulfur coal applications eliminating the low-sulfur control option to
meet sulfur oxide standards.

     The second change might involve more stringent NSPS for coal-fired
power plants than the existing standards (Figure l).  Revised standards
are under consideration'which could lead to more stringent control for
sulfur dioxide, nitrogen oxides and total suspended particulates.

-------
6A                     CLEAN COMBUSTION OF COAL
   STATUTORY REQUIREMENT  TO ACHIEVE ACCEPTABLE  AMBIENT  AIR QUALITY FOR:

        S02                      _
                                ~ STATIONARY SOURCES
        TSP

        NOX

        HYDROCARBONS
                                       \ MOBILE SOURCES
        CARBON MONOXIDE

        PHOTOCHEMICAL OXIDANTS
   STATUTORY  REQUIREMENT  TO MEET NSPS FOR COAL-FIRED  STEAM GENERATORS

               STANDARD                       UNCONTROLLED
        SO
         '2'
      1.2 LB/106 BTU                     5 LB/106 BTU

TSP:  0.1 LB/106 BTU                  6-10 LB/106 BTU

N02:  0.7 LB/106 BTU                     1 LB/106 BTU
     Figure 1.   Clean Air Act—Driving Force for Flue Gas Cleaning.

-------
                           EPA R&D PROGRAM                           65
     Third, the Agency is considering the possibility of NSPS for indus-
trial boilers which would be defined as being less than about 25 Mwe
(equivalent) for nitrogen oxide, total suspended particulate and possibly
sulfur oxides.

     The President's recent energy message highlights the need for
effective control technology on coal combustors.  The message calls for
the expansion of the annual coal production rate from the TOO million
tons presently produced to over 1 billion tons by 1985.  His policy also
calls for massive conversion of existing utility and industrial power
facilities from oil and gas to coal, and for essentially no new oil or
gas industrial or utility boilers.  Future options for these applications
would generally be coal or nuclear, or one of the emerging energy tech-
nologies.  Although the conservation aspects of the President's plan,
and the assumption of strict environmental control trends to minimize
environmental degradation, total emissions of nitrogen oxides and sulfur
oxides, are projected to rise above present levels by 1985.  The Presi-
dent's energy message also calls for accelerated research, development
and demonstration for the so-called clean coal technologies:  coal
cleaning, flue gas desulfurization, particulate control, fluidized bed
combustion, gasification, liquefaction and coal mining.  In fact, we, in
the EPA energy-environmental program, have been working with the Office
of Management and Budget at planning an accelerated program for certain
coal combustion control technologies.   We are hoping to get a relatively
large funding spike in fiscal year 1978 to accelerate our present
development/demonstration efforts in the following technology areas:
flue gas desulfurization, nitrogen oxide control, particulate control,
coal cleaning, and coal processing.

     So what we basically have is the possibility of more stringent
emission regulations, in light of possible Clean Air Act revisions and
upgrading of present NSPS, superimposed on an energy plan which calls
for increased burning of coal.  Clearly, effective and low-cost control
technology for utility and industrial sources are needed in the near
term.

     Figure 2 summarizes the critical problems associated with utility
and industrial conventional combustion.  This table attempts to briefly
describe:  the primary problems associated with such combustion; whether
or not there are existing standards; major near-term control technologies
available; the present status of these technologies; the secondary resid-
uals produced by these technologies; and finally the needed control
technology R&D in order to resolve some of the remaining problems.  As
you can see, the primary problems associated with industrial conventional
combustion include the primary air pollutants, nitrogen oxide, sulfur
oxide, particulates and potentially hazardous materials, as well as some
of the waste and water effluents associated with these control technology
and the power plant itself.

     Now that I have completed my background discussion, I'd like to
present some of the highlights of our ongoing research, development and
demonstration program.

-------
Figure  2
DESCRIPTION OF
PROBLEM


PRIMARY
POLLUTANTS
so0
2






NO-_
X






PARTICIPATES







POTENTIALLY
HAZARDOUS
MATERIALS
STANDARD TYPE OF FGC
PRESENTLY CONTROL
ESTABLISHED TECHNOLOGY




YES
NSPS &
AAQS






YES
NSPS &
AAQS







YES
NSPS &
AAQS


NO


COAL CLEANING



FGD





COMBUSTION
MODIFICATION


FLUE GAS TREAT-
MENT


ELECTROSTATIC
PRECIPITATORS

BAG HOUSES

WET SCRUBBERS

NOVEL DEVICES

UNDEFINED


PRESENT STATUS

1ST GENERATION
DEMO PLANNED


1ST GENERATION IN
FULL SCALE DEMO
2ND GENERATION IN
BENCH AND/OR
PILOT SCALE


COMMERCIAL FOR SOME
NEW UNITS


PILOT SCALE AND
DEMO IN JAPAN



COMMERCIAL

1ST GENERATION DEMO
1ST GEN. COMMERCIAL
2ND GEN. FULL SCALE
DEMO
BENCH OR PILOT SCALE

UNDEFINED

SECONDARY
RESIDUALS


HIGH-S REFUSE




SLUDGE,
PURGE STREAMS



POSSIBLY MORE
PARTIC. AND CO


VARIES WITH
PROCESS



FLY ASH







UNDEFINED

NEEDED CONTROL
TECHNOLOGY R&D
(INCLUDING ASSESSMENTS)

A) ELIMINATION OF SECONDARY
POLLUTANTS
B) DEMONSTRATE PRACTICABILITY
C) BROADEN APPLICABILITY

A) ELIMINATION OF SECONDARY
POLLUTANTS
B) IMPROVE RELIABILITY
C) BROADEN APPLICABILITY
D) IMPROVE ENERGY EFFICIENCY

A) BROADEN SOURCE
APPLICABILITY
B) IMPROVE ENERGY EFFICIENCY
C) IMPROVE NOX CONTROL
EFFICIENCY
D) MINIMIZE IMPACT OF
RESIDUAL POLLUTION

A) IMPROVE COST EFFECTIVE
FINE PARTICULATE CONTROL
B) BROADEN APPLICABILITY
C) DEVELOP NOVEL DEVICES
WITH IMPROVED CAPABILITY


PROBLEM REQUIRES
DEFINITION

                                                                            o
                                                                            o
                                                                            o
                                                                            c!
                                                                            CO
                                                                            i-3
                                                                            H
                                                                            O
                                                                            o
                                                                            g
                                                                            IT"

-------
                           EPA R&D PROGRAM                           67
Flue Gas Desulfurization
     I would first like to discuss our flue gas desulfurization (FGD)
program.  One can characterize these systems as producing either throw-
away (disposable) or saleable products.  In the "throwaway" FGD area,
one of our most important single projects is the Shawnee lime/limestone
prototype program which we've conducted with the aim of improving lime
and limestone scrubbing processes.  As many of you know, such processes
have been selected for many utility applications; approximately ^0,000 Mwe
or approximately 3 billion dollars worth of these systems are presently
in operation or on order at this time.

     The Shawnee program has been a cooperative effort of the EPA, the
TVA and the Bechtel Corporation.  Two 10 Mwe scrubbers have been operated
since 1972, and a 0.1 Mwe pilot scrubber has been operated in support of
the two larger facilities since about 1973.  This program has demonstrated
long-term reliable operation of both lime and limestone processes.  The
particularly troublesome mist eliminator plugging problem has been solved
by using a combination of careful operating conditions as well as care-
fully selected mist eliminator washing configurations.  During the course
of our program, we discovered the potential of unsaturated operation in
order to avoid gypsum scaling which had plagued earlier commercial sys-
tems.  Basically, this approach involves selecting operating parameters
so that scrubbing liquors never get saturated or super-saturated in cal-
cium sulfate thereby avoiding potential scaling problems.  Also, during
the course of the program, we've learned how to achieve high-alkali
utilization.  For example, limestone utilizations of over 90% have been
achieved.  This leads to lower alkali requirements and lower sludge pro-
duction rates, both of which yield lower operating costs.  We have also
achieved high sulfur oxide removal efficiencies.  Typically, efficiencies
for both lime and limestone systems in excess of 95% can be achieved
without an excessive economic penalty.  One of our most recent findings
is that minor process modification can allow for sludge oxidation by
gypsum producing a material capable of 90% dewatering by filtration.
Since last year, we've developed a design/cost computer model which acts
as a data base for all the information the EPA, TVA, Bechtel team knows
about lime/limestone scrubbing and allows a given utility or other FGD
user to input application parameters and output a conceptual design
along with some pretty good cost estimates.

     We have been working with the Air Force in operating and testing a
lime scrubber system on a 21 Mwe coal-fired industrial boiler at the
Rickenbacker Air Force Base.  This unit started up in March 1976, and,
although there were some boiler control and fan problems, the efficiency
and reliability of the scrubber have been good.  Also, it appears that
the economics of such scrubbing systems on industrial boilers of this
size do not appear prohibitive.

     As an alternative to lime and limestone nonregenerable FGD systems,
our program has been actively developing the double alkali scrubbing
process.  Although the double alkali system has basically the same
chemicals entering the process and leaving the process, they do have
several potential advantages over lime/limestone scrubbing processes.
These include:  less energy consumption; higher sulfur oxide removal
efficiency; lower maintenance; and, lower capital and operating costs.

-------
00                     CLEAN COMBUSTION OF COAL


The EPA program has actively developed double alkali technology at the
"bench and pilot levels.  We've worked with Southern Services at 20 Mwe
electric prototype system and, most recently, we've worked with the
General Motors Company at applying a double alkali process variation to
one of GM's industrial boilers.  In the past, this double alkali scrubber
had generally good operability on the 32 Mwe industrial boiler, although
some problems were encountered.  It should be noted that,as for the
previously mentioned Rickenbacker lime unit, the approximate capital
costs, on this size industrial boiler, of $100/Kwe do not appear exces-
sive.

     Recently, the EPA announced its plans to demonstrate the double
alkali process on a high-sulfur coal utility boiler.  The Louisville Gas
and Electric Company and its 270 Mwe electric Cane Run No. 6 unit was
selected for this demonstration.  Combustion Equipment Associates/Arthur
D. Little comprise the process supplier team.  I think it is noteworthy
that the cost of this unit, and these costs are fairly firm, are esti-
mated to be only $55 per Kwe capital costs; and 2.5-2.9 mills/Kwe hour
for total annual revenue requirements.  Also, only 1.1$ of the power
plant's energy output is required to run the fans and pumps for this
process.  This is roughly one-third to one-half the energy requirement
of similar lime or limestone systems.

     As an alternative to lime and limestone systems and their inherent
sludge production, we have been active at developing and demonstrating
regenerable (or saleable) product FGD systems.  For example, we have
demonstrated the promising magnesium oxide scrubbing -process at Boston
Edison's Mystic Station.  This scrubbing facility was tested on a 155 Mwe
oil-fired boiler and produced saleable sulfuric acid.  The test program was
initiated in April 1972 and completed in June 197^-  Although many early
problems were identified, particularly those associated with the various
solid handling operations, operability improved substantially toward the
end of the test program.  Unfortunately, no meaningful demonstration of
this process has been performed on the critical coal-fired combustion
units.

     Our major ongoing project in the regenerable area is the Wellman
Lord demonstration test program.  Wellman Lord technology involves scrub-
bing with a soluble reactant followed by thermal regeneration producing
concentrated sulfur dioxide which can yield either sulfuric acid or
elemental sulfur.  This process has been successfully demonstrated in
Japan on a variety of oil-fired facilities.  The EPA demonstration pro-
gram is on a 115 Mwe electric coal-fired facility at the Northern
Indiana Public Service Company and produces elemental sulfur.  This
facility started up in April 1976 but due to a boiler explosion we are
now in a restart-up mode.  Hopefully, definitive findings on this process
will be available at our next conference.

     Recently, the Atomic International's Aqueous Carbonate Process has
been selected for demonstration.  The process will be demonstrated at
Niagara Mohawk's Huntley Station and we are hoping that the system will
start up in early 1979-  It offers cost and other advantages over alter-
native regenerable processes, but must be considered a relatively high-
risk venture due to the relatively small scale of previous test experi-
ence.  Also note that the EPA is working with the U.S. Bureau of Mines in

-------
                           EPA R&D PROGRAM                           69
applying the regenerable Citrate Process to a large industrial facility.
Figure 3 summarizes the major EPA-sponsored demonstration programs for
both throwaway and saleable project FGD processes.

     In addition to our demonstration activities, TVA working for the EPA
has conducted a series of very relevant by-product marketability studies
which help put the sulfur, sulfuric acid, and other sulfur by-product sale
situation in perspective.  Also, studies have been performed at evaluating
alternatives to scarce natural gas, as the reductant materials to produce
elemental sulfur from the concentrated sulfur dioxide associated with many
regenerable FGD processes.

     Let me conclude my discussion of our FGD program by briefly mention-
ing three other important programs.  First, the TVA has prepared a number
of FGD process economic studies which in my mind are the most reliable
cost estimates for both throwaway and saleable FGD processes.  Secondly,
the EPA working with the Council on Environmental Quality is working on
a scrubber availability study which attempts to compare scrubber avail-
ability to that of other power plant components.  Preliminarily, it
appears that there is insufficient information to do a statistically
convincing comparative study.  However, it has become apparent that
several major power plant components appear to have reliabilities lower
than some of the more recent FGD systems.

     Finally, I'd like to mention the very active technology transfer
program in the FGD area.  Some of the outputs of this program include
the PEDCo Status Report -(published every other month) which summarizes
what's going on in the scrubber field both in terms of operating and
planned systems.  Also, we now publish quarterly RD&D status reports
which summarize the results of our ongoing research program.  We are
preparing lime and limestone data books and cost and reliability hand-
books which present potential users with important information on commer-
cial and noncommercial technologies.

Nitrogen Oxide Control

     Now, let us discuss our NOX control program.  I think a little back-
ground might be in order so I've added some background information that
I think is relevant.  As many of you know, there was an unreliability
problem discovered in the ambient air quality measurement technique for
NOv, back in 1972.  Prior to discovery of the problem, it was believed
that kj Air Quality Control Regions (AQCR) out of the total of 2kl AQCRs
for the country had an N02 ambient air problem.  What we've found out is
that due to an inherent measurement error, ambient levels of W02 were
measured too high.  Using more accurate techniques, only four areas
(AQCRs) for the country really seemed to have an NC>2 problem.
     However, since 1972 there hasn't been too much progress in
control from either stationary or mobile sources.  As a result, they are
now finding new AQCRs exceeding the TTO2 standard and, quite frankly, the
trend seems to be for further NC>2 ambient quality problems.  Therefore,
our present N02 control strategy does not appear very effective.  Our
present control strategy, basically, includes control of both mobile
(automobile) sources as well as stationary sources.

-------
                                           Figure 3
                   EPA-SPONSORED STACK GAS DESULFURIZATION DEMONSTRATION SYSTEMS
EPA-SPONSORED PROCESS
(BY PRODUCT)
NON-REGENERABLE
LIMESTONE SLURRY SCRUBBING
(SLUDGE)

LIME SLURRY SCRUBBING
(SLUDGE)

DOUBLE ALKALI SCRUBBING

PARTICIPATING
UTILITY
TVA


TVA

LOUISVILLE
G&E
PROCESS
DEVELOPER
BECHTEL AND
OTHERS

CHEMICO,
BECHTEL,
AND OTHERS
CEA/ADL


LOCATION
SHAWNEE
UNIT 10
PADUCAH, KY
SHAWNEE
UNIT 10
PADUCAH, KY
CANE RUN 6

UNIT SIZE EXPECTED
AND TYPE START UP
10 MW UNDER WAY
COAL

10 MW UNDER WAY
COAL
-
270 MW EARLY - 1979



CJ
i
o
Q
a
CO
t-3
H
O
^
REGENERABLE
  MAGNESIA SLURRY
   SCRUBBING - REGENERATION
   (98% SULFURIC ACID)

  SODIUM SCRUBBING
   REGENERATION (SULFUR)
  AQUEDOUS CARBONATE
  BOSTON EDISON
  CHEMICO-BASIC
NORTHERN INDIANA      DAVY
PUBLIC SERVICE CO   POWERGAS
                 ALLIED CHEMICAL
NIAGRA MOHAWK
   ATOMICS
INTERNATIONAL
    MYSTIC
  STATION 6
BOSTON, MASS

D.H. MITCHELL
  STATION 11
  GARY, IND

   HUNTLEY
   STATION
150 MW
  OIL
                                 115 MW
                                  COAL
100 M
COAL
COMPLETED
             LATE - 1975
EARLY - 1979
                                                                                   o
                                                                                   o

-------
                           EPA R&D PROGRAM                           71
     I should point out that almost one-half of the NOX emissions are
from mobile sources; a little more than one-half are associated with
stationary sources
air quality goals.
stationary sources.   So both are of interest if ve are to achieve NOX
     We have set so-called interim standards for automobiles.   They are
not very stringent standards, and are in fact substantially less strin-
gent than the Clean Air Act goal.  Due to economic, technological and
political considerations, it appears that in the near term relatively
moderate NOX emission standards will prevail for autos.

     Also, the new source performance standards philosophy is  based on
best available control technology.  Since control technology for strin-
gent control is not available, the NSPS for utility boilers is not very
restrictive and only requires, as I indicated earlier, on the  order of
30 or 35 percent control.  There's also an NSPS for nitric acid plants
which is a significant contributor to the NOX problem.

     So, our present strategy doesn't seem to be sufficient to get these
AQCRs back into line, and for that matter to turn around the trend toward
additional AQCRs getting out of standard.  Therefore, additional control
of stationary sources beyond the present level may be necessary.  Of
course, the NOX problem is more complex than just NOg; NO  is  a major
precursor for photochemical oxidant and nitrate production.

     There are three major categories of control for NOX control for
stationary sources.  The only near-term technology is combustion modifi-
cation, which the present EPA program is emphasizing.  This technology is
the basis for the present new source performance standard on large fossil
fuel generators.  Flue gas cleaning is the second technology.   It's
similar to flue gas desulfurization, and probably should be called flue
gas de-nitrification.  We have several embryonic programs in that area
which I will briefly describe.

     This combustion modification program aims at developing technology
capable of controlling emissions from the two major sources of NOX from
stationary sources; namely, "thermal" NOX and "fuel" NOX.  One hopes to
control "thermal" WOX by lowering the combustion temperature since the
equilibria relationships are such that lower temperatures retard NOX
formulation, i.e., higher temperatures favor the oxidation of  nitrogen.
And, one can control "fuel" NOX problems by lowering oxygen concentration.
One might note that the control approaches are generally common to both
thermal and fuel NOX, with the exception of flue gas recirculation, which
is really oriented primarily toward minimizing thermal NOX-

     One has to be careful implementing these techniques since it is
possible to aggravate other pollutant emission problems.  We could, for
example, increase emissions of carbon monoxide or particulate  matter, if
we drastically lower excess air.  We could also lower thermal  efficiency
with approaches such as flue gas recirculation.  These techniques can
also lead to operating problems since boilers will not be operated in
some cases under the design conditions that they are designed  for.  How-
ever, the results of our program to date indicate that each of these
potential problems are controllable if one is careful about the particular
control technology and the design parameters utilized.

-------
72                     CLEAN COMBUSTION OF COAL


     Let me now discuss the ongoing nitrogen oxide combustion modifica-
tion program.  I'll discuss this program by source area:   utility and
large industrial boilers; small industrial, commercial and residual
systems; stationary engines; and industrial processes and after-turner
equipment.

     In the utility and large industrial boiler area, the program's major
effort has focused on staged combustion approaches.   By utilization of
this combustion modification technology which involves adding combustion
air sequentially in more than one location to minimize total oxygen
requirements and combustion temperature, control levels of 0.^5 lb/10  Btu
appear achievable.  ¥e are presently planning corrosion tests to ascertain
whether or not boiler tube corrosion is an inherent problem.  It has been
postulated by some that the reducing environment associated with staged
combustion can remove the protective oxidative coatings from the tubes,
thereby accelerating corrosion.  Therefore, corrosion tests are considered
quite important.

     Perhaps our most encouraging activity in the NOX program is our work
in the low-NOx burner area.  By redesigning pulverized coal burners to
more carefully combust coal, experimental results have indicated that NOX
levels of 0.2 to 0.3 lb/10° Btu are achievable.  This represents overall
control of an uncontrolled coal-fired boiler of over 10%.  Since such
burner technology appears inherently inexpensive, this could be the
answer to low-cost nitrogen oxide control from both industrial and utility
pulverized coal boilers.  If we receive the incremental fiscal year 1978
funding that I mentioned earlier, we will be able to proceed on an orderly
demonstration program aimed at evaluating this approach at sizes culminat-
ing in an integrated full-scale demonstration program.

     In the small industrial, commercial and residual system area our
program's emphasis has focused on residual oil package boilers, residen-
tial furnaces and coal-stoker boilers.  We have demonstrated that staged
combustion is an effective control approach on oil-package boilers.  Pre-
liminary information indicates that optimum burners can achieve 10% NOX
control over conventional burners for residential distillate oil furnaces.
Our work in the coal stoker areas has been limited so far to the develop-
ment of emission factors for existing designs.  A combustion modification
technology program is presently under development for this increasingly
important class of coal boilers.

     Staionary internal combustion engines are a major source of nitrogen
oxide emissions, primarily because of their widespread use as the prime
mover for pipelines and gaslines around the country.  Preliminary work
indicates that it may be possible to lower nitrogen oxide concentrations
in combustion gases to 50-100 ppm.  This would represent an overall NOX
control of 75$> and a substantial improvement both in NOX efficiency and
overall energy conversion efficiency to the presently available control
technology, which involves introduction of water into the combustion zone.

     In the industrial processes and after-burner equipment area, we have
a relatively low level of activity due primarily to limitations in funds.
Our near-term goal in this area is to characterize NOX and other emissions
for these facilities.

-------
                           EPA R&D PROGRAM                           73
     Let me "briefly describe the NOX flue gas treatment research and
development program.  One might ask, in light of the comprehensive com-
bustion modification program I have just described, "Who needs flue gas
treatment?" and "What is the advantage of this technology over combustion
modification?"  Well, the only identifiable advantage in my mind is that
this approach has the potential for very high NOX removal efficiencies
which could conceivably be required if standards were to tighten in the
years ahead.

     However, you pay the price for high removal efficiency in terms of
high capital and operating costs and system complexity relative to com-
bustion modification technology.  These approaches are under active
development in Japan due to the stringent ambient and emission standards
for nitrogen oxides there.  Two basic approaches are being developed, dry
and wet processes.  The dry processes are the simpler of the two and
generally involve a chemical reduction reaction.  Typically, ammonia is
used as the reductant and a selective catalyst is needed to reduce
nitrogen oxide to elemental nitrogen and oxygen.  Problems with this
approach include uncertainty regarding the viability of this process for
coal-flue gas, the possibility of secondary emissions (such as ammonium
sulfate), and potentially high costs associated with capital expenses
and reagent needs.  Wet processes are generally more complex and can
involve oxidation to nitrogen dioxide followed by a scrubbing step;
others reduce the nitrogen oxides in solution.  For the oxidative/reduc-
tive approach, ozone is a typical oxidant.  The main advantage of this
class of processes is the potential for combined sulfur oxide and nitrogen
oxide removal.  Problems include large ozone and energy needs for the
oxidative/reductive process and the production of secondary wastes and
high costs for all versions of the wet processes.  The EPA program in
this area involves piloting promising processes treating flue gas from
coal-combustors, borrowing heavily from Japanese technology on oil-fired
boilers there.  Two pilot programs are in the final contractor selection
phase, one will involve nitrogen oxide removal only and the other will
investigate concurrent removal of nitrogen oxide and sulfur oxide.

Fine Farticulate Control

     First a little information in the way of background.  Fine particu-
lates are health hazards because they are airborne for extended time
periods, can penetrate deeply into the lung, and can act as transport
agents for other pollutants.  Our research, development and demonstration
program includes the following major areas:

     •  Improvement of characterization of present technology, e.g.,
        electrostatic precipitators, scrubbers, fabric filters

     •  New ideas/novel devices
     •  High temperature/high pressure control technology

     •  Collectability of dust
     •  Control from low sulfur coal combustors

     Our program is actively evaluating the potential for electrostatic
precipitators as high efficiency fine particulate control devices.  We
have characterized precipitators for seven particulate sources, and

-------
74                     CLEAN COMBUSTION OF COAL


developed a math model which characterizes precipitator performance as a
function of particulate characteristics.  We have concluded that precip-
itators offer the possibility of high fine particulate control when there
is no ash resistivity problem.  Precipitators are very sensitive to the
chemistry of the ash they must collect.  For example, for low sulfur coal
combustion facilities, the ash generally is of a low resistivity since
there is not sufficient sulfur trioxide to raise the ash conductivity
thereby leading to collectability problems.  Our program is actively
considering ways of upgrading precipitator performance when the ash has
a less-than-optimum resistivity.  We are actively evaluating fly ash con-
ditioning agents such as sulfur trioxide, ammonia, etc.  We are also
actively evaluating the possibility of precharging the flue gas upstream
of the precipitator in order to improve the collectability of the ash.
We are presently planning a pilot demonstration of an attractive pre-
charging approach.  We are carefully coordinating our program with the
Electric Power Research Institute who is also active in this area.

     We are also evaluating and developing various scrubber devices for
the removal of fine particulates from combustion facilities.  We have
evaluated ten devices on a variety of particulate sources, and currently
find a consistent pattern where fine particulate removal efficiency is
highly dependent on pressure drop (and, therefore, energy requirements)
of the scrubber device.  Generally, the higher the pressure drop the
better the fine particulate removal efficiency.  However, at least one
scrubber type has appeared uncharacteristically efficient in fine partic-
ulate removal, namely, the turbulent contact absorber (TCA).  Our studies
have indicated that a condensation mechanism is responsible for this good
performance.  We are evaluating this mechanism more carefully and hope to
be able to apply it to other scrubbers in other facilities.  We are also
piloting a flux force/condensation scrubber which uses a condensation
mechanism for efficient fine particulate removal.  Recent data indicate
that scrubbers may be limited in their fine particulate performance by
mist eliminators.  Mist eliminators are designed to avoid the entrainment
and carryover of scrubbing liquors from the scrubber into the existing
flue gas.  It appears that inefficient mist eliminators on commercial
units have allowed such entrainment thereby leading to carryover of par-
ticulates with subsequent degradation of fine particulate removal per-
formance.  Studies continue in this area.

     Fabric filtration is being evaluated as a fine particulate control
scheme.  We have tested filters applied commercially to three sources,
and find that fabric filtration is quite efficient down to 0.3 microns.
These devices have energy requirements between the low energy usage
precipitators and the high energy usage scrubbers.  Our present program
is aimed at increasing the applicability, operability and economic desir-
ability of these very promising devices.  We are presently planning to be
involved in a 350-megawatt demonstration program applying fabric filtra-
tion to a low-sulfur coal utility boiler.  This will be the first commer-
cial operation of a low-sulfur coal fabric filtration combination.  Our
program has also been active at evaluating new ideas and novel devices
for the removal of fine particulates from various sources.  Many of these
devices have been tested at bench scale and the most attractive devices
are scheduled for pilot scale testing pending availability of funds.

-------
                           EPA R&D PROGRAM                           75


     We have also actively pursued mobile particulate control devices
that enable us to collect data from a variety of particulate sources.
We presently have constructed and tested mobile units of electrostatic
precipitators, scrubbers, and fabric filters which can allow us to
determine what the best control device might be for a given application
and its associate dust problem.

     A final element of our program is the control of particulates from
low-sulfur coal combustors.  This is really not a new or separate part
of our program, but actually incorporates our ongoing work in other areas.
However, we find it helpful to focus in on this major problem afflicting
those utility and industrial sources who must meet particulate standards
and burn low sulfur coal.  Our present emphasis at this time is the
development of fabric filtration and modified electrostatic precipitators
for cost-effective particulate control from these sources.

     In conclusion, I believe the results of the conventional fossil fuel
control technology program I have discussed this evening will go a long
way in helping our nation achieve two of its most important goals com-
patibly:  energy availability and environmental protection.

-------
76                     CLEAN COMBUSTION OF COAL

-------
                                                                     77
                 SESSION II - PRECOMBUSTION PROCESSES

   SESSION CHAIRMAN:  P. STANLEY JACOBSEN, COLORADO SCHOOL OF MINES
                                           RESEARCH INSTITUTE
     The Air Quality Act of 1963 initiated at all levels of government
an effort to preserve the nation's air quality.  Because of the large
role that coal shares in providing the total fuel energy consumption of
the nation, the effects of coal combustion in electric power generation
were given special emphasis.  Coal consumption by utilities was
kOh million tons for 1975 and may be as high as 700 million tons by
1985, and as recommended by the President could be over 1200 million
tons in the year 2000.

     Control of sulfur oxide emissions from fossil-fuel-fired combustors
has for years been considered desirable, but with the promulgation of
air quality standards the control has become mandatory.  Typically, the
removal of sulfur has been accomplished by postcombustion processes such
as flue gas desulfurization.  But the advantage of utilizing precombus-
tion processes as well offers very definite advantages, such as:

     1.  Using high sulfur coals with little to medium physical cleaning.
     2.  Availability of a more uniform coal.
     3.  Lower effective fuel transportation costs.
     h.  Reduced maintenance costs.
     5.  Lower coal pulverizing and ash disposal costs.

     The overall advantage of combining precombustion processing with
postcombustion processing can be substantial up to a savings of lUo$
over using postcombustion processing alone.

     To bring into perspective the complete gamut of precombustion pro-
cesses, the session commences with papers covering the following four
areas:  an overview of coal preparation; the influence of mining prac-
tices on the quality of raw coal; research aspects of coal preparation;
and transportation of coal.  Three papers in the evening session cover
rather innovative approaches to coal preparation, namely:  chemical coal
cleaning, magnetic desulfurization, and a waterless coal cleaning method,
called The Otisca Process.

     The entire combustion process including precombustion, combustion,
and postcombustion surely must be considered as an integrated whole.  It
is anticipated that the papers on precombustion processes will serve to
refocus attention on the important role this sometimes-neglected segment
plays in the overall combustion process.

-------
78                     CLEAN COMBUSTION OF COAL

-------
                                                                      79
                           COAL PREPARATION
                         HISTORY AND  DIRECTION

                                  by
                         Robert L. Llewellyn
                            Vice President
                      Roberts & Schaefer Company
     Historically, coal has "been prepared in some fashion since the
beginning of mining over 150 years ago in the continent alone, and double
that figure in Europe.  Slotted hand shovels were used to leave the
untreated fines inside the mine; men, boys and even women sorted the
coal from impurities on picking tables.  Then, many ingenious men, too
numerous to mention, invented jigs, tables, launders and vessels; using
water and air, and the combination of both to separate the coal and
refuse.

     Later, sand, minerals (such as magnetite) and heavy liquids were
added to these various washers to increase their efficiencies of separa-
tion.  Fifty-five years ago oil agglomeration was patented here in the
States to clean ultra fines, but the use was considered uneconomical
because these fines (at the time) could be wasted.  Then, over 30 years
ago electrostatic coal cleaning was demonstrated in a test plant on
lUm x 0, and this system is still resurrected by some about every 10
years without success.

     Some confused Convertol with agglomeration, which was a process
using oil with the fines to dewater and improve the bulk density.
Agglomeration reduces the ash.  Electrostatic was a unique idea to dry
clean fines by surface separation; it worked in the laboratory, but not
in practice because the coal had to be practically bone dry and drying
coated the particles.

     We in the coal preparation field are very fortunate to have excel-
lent cleaning devices available for obtaining efficient physical sepa-
ration of the raw feed from the mines throughout our nation.  Chemical
processes are also being considered to make coal practically ash- and
sulphur-free and the activity of all this science is very exciting to
every man and woman connected with coal, which will be affecting the
energy field for the next decade.  Huge sums of money will be expended
by government and industry and it must be spent wisely; it must not be
wasted.

     If electric power is to be most economical, the fuel must be obtain-
able at a reasonable cost.  All of us have been made aware of this when
receiving our monthly or quarterly bills, showing the fuel cost adjust-
ment.  Therefore, it behooves all of us to investigate the lowest cost
mining and cleaning systems.  We have "mine mouth" operations, which were

-------
80                     CLEAN COMBUSTION OF COAL
recognized during the past 15 years as an excellent method to provide
cheap fuel "by locating the power plants at the coal mines.  Pipelines
from Cadiz to Cleveland transported coal in slurry form from the mine
to the power station, which was a pioneer to other pipelines in the
west for the same purpose.  Railroads obtained unit train load rates
to greatly stimulate coal transportation from mines to power plants by
rail.  Mining equipment for surface and deep mines have been manufactured
to improve the cost of coal to the ultimate user.

     Never before have there been so many people involved with the art
of coal preparation.  This will continue into the next century.  And
with such people whose backgrounds vary in education and talent, the
possibility for developing new cleaning techniques are bound to be fan-
tastic.  The interest in coal preparation has attracted numerous compan-
ies whose expertise were in different fields of engineering and construc-
tion.

     Many consultants have recognized the tremendous growth in the
immediate future in coal preparation and now there are industries outside
of coal mining who are engaged in a tremendous effort and with financial
aid in the coal preparation field.  Utility companies with their vast
networks in the U.S.A. have sought and secured large coal reserves for
energy resources.  Oil companies are making huge investments in coal
properties, which will be mined and the product beneficiated to suit the
use in the making of energy or steel.  Those people involved have learned
that many railroads and steel corporations are already owners of coal
lands since the turn of the century.

     It is predicted that great activity will be observed as we look
toward the year 2000.  As coal production doubles, or triples, the type
of coal will have to be prepared by the highest efficiency equipment
available.  These coals left unmined in the east and midwest by our
grandfathers or great-grandfathers will require the best cleaning methods
to obtain the ash, sulphur and Btu requirements.  Our ancestors had to
mine the cleanest seams in the area, because coal preparation was in its
infancy and the price of coal was relatively cheap.

     For your information, the first million dollar coal preparation
plant was built near Benton, Illinois, for U.S. Steel in 1918; the book,
Coal Washing  by Prochaska, described the plant to some extent.  I
visited the plant in 1938 and personally saw the Plato Concentrating
Tables and Feldspar Jigs for fines which were intended to reduce the
sulphur content for metallurgical purposes.

     Coal cleaning systems can be utilized economically, but only if
applied properly.  Coal Washers differ significantly; magnetic recovery
circuits have wide variations, and unless these factors are not dili-
gently scrutinized, preparation plants will be built without the best
coal beneficiation equipment or systems.

     Anyone who has studied flowsheets for years can readily observe the
various systems to prepare coal.  First, I believe that one should
thoroughly acquaint themselves with what is going on underground.  That
is where the coal and refuse are formed.  Just lately I have learned how

-------
                            COAL PREPARATION                         81
much "bottom and roof gets into the R.O.M.  The plastic bottles used for
water inside has "become a problem.  We all have known about wood, metal,
aluminum cans, and now plastic bottles have to be removed before any
cleaning system.

     Preparing the coal ahead of cleaning is most important; recognize
that breaking and/or crushing can help or hurt the raw product.  Often
times reducing the size improves the quality and recovery on some coals.

     Necessity breeds invention which brought on devices to prepare coal
in the past and some of us here will perhaps be involved in some way in
the future.  However, there is nothing wrong to apply proven equipment.
We do not have to invent the Edison light bulb again; the "bulb" has
been improved in our lifetime by many companies.

     We must learn by personal experiences and from experiences of
others.  It is very difficult to "weed" out good and bad information.
Some integrity has been lost in our life style and I believe that
presently you cannot believe everything you read or hear as you could
some years ago.

     Conferences like this one attract people who are very eager to
learn, otherwise they would not be attending.  The program input should
be rewarding to all who are interested in coal and its growth.

-------
82                     CLEAN COMBUSTION OF COAL

-------
                                                                      83
          THE INFLUENCES OF MINING PRACTICES ON COAL PREPARATION

                               C.A. Goode
                             Bureau of  Mines
                United States Department of the Interior
     There is no doubt that the coal reserves in the United States are
more than adequate to meet the energy needs of the country.  The reserve
base in the East is over 200 billion tons.  Of that total, it is esti-
mated that 160 billion tons will have to be extracted by underground
mining methods owing to the depth of cover, which averages over 600 feet
in the seams that are currently being exploited.  The maximum cover in
the East in active mines is about 2,300 feet.  Extraction by surface
methods seldom exceeds 100 feet in total depth, but this is more a func-
tion of seam thickness and quality than of equipment capability.  Recov-
erability averages from 50 to 60 percent for underground and from 80 to
90 percent for surface methods.  Again, coal quality and environmental
aspects tend to constrain the systems.

     Coal mining is a high-volume, high-production effort; it is highly
developed art.  Preparation of the product has its inception in explora-
tory stages prior to earthmoving.  The area to be mined is drilled on
5 mile centers or more, and based on an analysis of the cores, a decision
is made whether or not to proceed.  Favorable analyses lead to a second-
phase program whereby the drilling centers are tightened to anywhere from
1 to 3 miles, and again a decision point is reached before finally going
to 1/2" to 1/4-mile core hole centers.  Intelligence gathered on the de-
posit characteristics includes superjacent strata information as well as
coal character.  Selection of mining and preparation equipment, layout of
the mine, location of points of entry, areas to be mined and mining se-
quence, drainage of the workings, method of mining, and other mining and
marketing considerations are based largely on the analysis of the cores
and any outcrops.  High-resolution seismic reflection may also be used to
give definition to the deposit in terms of geologic anomalies (faults,
channel sands, washouts, etc.) that may influence extraction.

     Information gathering relative to deposit characteristics continues
throughout the life of the mine.  Isopach maps are prepared showing the
deposit properties of interest, be they sulfur, volatile matter, BTU, or
ash.   These are useful in establishing blends of coal from various sections
of the mine to produce a more marketable product.  Proper input in the
early stages of mine planning is of incalculable importance when consider-
ing the life of the property, which may be in excess of 25 years, and the
total planning effort which may take 5 years before production is realized.
Once a commitment to layout, equipment, and method has been made, any
changes could be economically disastrous.

-------
 84                     CLEAN COMBUSTION OF COAL


     The decision whether to surface mine or to attack the deposit by
underground methods is based on the extent, quality, and thickness of the
deposit and topography.  Basically, a ratio of 20 feet of overburden per
foot of coal is the maximum stripping that is currently undertaken.  The
size of the equipment is also based on the stripping ratio, the concern
being to match coal extraction rates and overburden removal rates over
the life of the property.  Fragmentation by blasting or by ripping equip-
ment and loading practices affect the size and quality of the coal ex-
tracted.  Proper choice of equipment for removing overburden, for clean-
ing the top of the exposed coal bench, and for loading the coal can en-
hance the marketability and/or the beneficiation character of the product.

     Augering mining is a technique that can be used to extend production
capability after surface mining methods are no longer economical.  It is
less efficient in terms of recovery.  Usually, the augers are sized
6 inches or more under the seam thickness to be taken in order to leave
coal on the top and/or on the bottom, thus minimizing rock contamination.
Research is continuing in the development of coal-rock interface detectors
which allow for vertical control of the auger to maintain its position
in the seam.  Auger coal is normally coarse and dry, but product size
decreases with depth of penetration.  Webs of coal are left between holes
to insure the integrity of support during mining.  Too thin a web may
lead to seizure of the unit, as floor and roof converge under stress.
Attempts to develop low-cost reliable sensing equipment to maintain
consistent and substantial web thicknesses are being made.

     Selective mining can be readily accomplished underground by produc-
ing coal from various sectionsof the mine and blending the product by
controlling section output.  Of the two generally used methods of mining,
room-and-pillar mining is more amenable to selectivity than longwall min-
ing.  Flexibility of the former in terms of the number of operating units
and relative ease of moving production units from one area of a mine to
another leads itself to planned development and scheduling.  Room-and-
pillar development can be accomplished by either conventional methods
(cutting, drilling, blasting, loading, and hauling) or continuous methods
where a single machine mechanically rips the coal from the face, thus
eliminating drilling and blasting.  Studies have been conducted on both
these systems in terms of quality and quantity of product, and they show
significant differences.

     While size of product is a function of the seam characteristics and
is governed by the hardness of the coal and fracture system, continuous
mining units generally produce finer sizes and introduce more extraneous
materials into the product than do conventional units.  The difference
is due more to operator judgment than to anything else.  Lower rotational
cutting speed and deeper bit penetration could reduce the size differen-
tial between the two systems.  The increased use of continuous miners at
the expense of conventional units attests to the fact that productivity
is the thrust of today's mining.  The burden is on the cleaning plants
to upgrade the product.

     Relief to overtaxed cleaning facilities can be accomplished by ex-
ercising face preparation techniques, and its importance should not be
overlooked or minimized.  Ash-forming or sulfur-bearing impurities can
be avoided in the mining process if they occur, as is often the case, in

-------
                     INFLUENCES OF MINING PRACTICES                  85
association with the bottom and top portions of the seam.  The coal it-
self is often a better room or floor material than the associated shales
and clays, and in some instances a 6-inch layer left in place provides
a better bearing surface or can give better protection against falls of
roof as well as yielding a better product.  Vertical distribution of im-
purities as well as seam thickness dictates the acceptability of this
practice.  Cutting and blasting in conventional mining can be used to
advantage to control banded impurities.  Soft bands can actually be cut
out, or hard bands can be fragmented for size control and eventual ease
of physical elimination in a screening operation.

     Longwall mining is touted as a safer, higher production system than
room-and-pillar.  There are approximately 80 active units in the United
States today, and they account for about 5 percent of the country's under-
ground production; in foreign countries, longwall mining is almost uni-
versally used.  There are opportunities in this system for both high
production and quality control.  The miner at the working face is in a
protective envelope of hydraulic supports (chocks or shields) and is in
better control of the environment.  Dust and methane are more easily
diluted by directed quantities of air.  Additionally, the size of the
panels, i.e. up to 600 feet wide and a mile long, allows for more con-
tinuous operation.  The mining unit (shearer or plow) is generally under
better control in terms of its attitude in the coalbed.  More importantly,
especially in the case of the shearing machine, there is an opportunity
for guidance control and remote operation with the potential for better
quality control and higher production.  The mining machine can be kept
within the confines of the seam, and a mining strategy to leave room and/
or floor coal can be readily established.  Probes, sensors, and controls
leading to full face control are being developed, and some are already
available.  The economic practicability of the control systems has yet
to be established.

     The preceding discussion is by no means all inclusive and is fraught
with generalities.  The intent here is only to give the nonminer a
glimpse of the mining processes.  The high degree of variability of coal
seams and associated strata must be recognized.  It is this fact that
makes each mine unique and product control difficult.

                             BIBLIOGRAPHY

Leonard, J. W. and D. R. Mitchell, ed., Coal Preparation, Third ed.,
     AIME, New York, 1968.

Stefanko, R., R. V. Ramani, and I. K. Chopra.  The Influence of Mining
     Techniques on Size Consist and Washability Characteristics of
     Coal Office of Coal Research and Development, Report 61, 1973,
     85 pp.

Thomson, P. D. and H. F. York.  The Reserve Base of U. S. Coals by
     Sulfur Content, BuMines 1C 8680, 1975, 537 pp.

-------
86                     CLEAN COMBUSTION OF COAL

-------
                                                                      87
           CURRENT COAL PREPARATION RESEARCH AND DEVELOPMENT
                       Richard P. Killmeyer, Jr.
                           Chemical Engineer
               Coal Preparation and Analysis Laboratory
                            Bureau of Mines
                    U.S. Department of the Interior
                        Pittsburgh, Pennsylvania
     The renewed interest in coal preparation and its rising importance
are evidenced by the many new and innovative research projects ongoing
around the country.  The purpose of this paper is to point out and de-
scribe some of this research.  Some of the Bureau of Mines new projects
and updates of the major ones will be covered, followed by a few others
from industry and other Government agencies.

     The Bureau of Mines has recently awarded three research contracts
to study chemical coal desulfurization, dewatering, and high-gradient
magnetic separation (HGMS).  Under one contract Jet Propulsion Labora-
tory will perform bench-scale desulfurization experiments with the aim
of optimizing their low-temperature chlorinolysis process.  This
process looks promising as an organic sulfur removal step following
physical removal of pyrite, because it reduces both pyritic and organic
sulfur (up to 80 and 50 percent, respectively).  At present, there are
several other chemical desulfurization processes being researched with
only one or two ready for demonstration units.

     The dewatering contract is with Dravo Lime Co. and is titled
"Management of Coal Preparation Fine Wastes Without Disposal Ponds."
Fine wastes will be sampled at 10 selected preparation plants and
analyzed for their physical, chemical, and engineering properties.
Bench-scale stabilization tests will be conducted using Calcilox and
other stabilizing agents.  The results will be evaluated for effec-
tiveness in hardening fine refuse.

     The third contract was awarded to the Naval Ordnance Station at
Indian Head, Maryland.  They will investigate the feasibility of mag-
netically separating pyrite from finely crushed coal dispersed in fuel
oil.  A Frantz Ferro Filter containing a stack of closely spaced grids
will be used for the experiments.  The Bureau is also in the process
of establishing in-house capability in high-gradient magnetic separa-
tion (HGMS).  A high-gradient magnetic separator has been purchased
from Sala Magnetics to treat finely ground coal.  The results will be
compared with representative fractions treated by the Bureau's two-
stage coal-pyrite flotation process.

-------
 88                     CLEAN COMBUSTION OF  COAL
     A continuing HGMS contract proposes an application different from
removing pyrite.  Researchers at MIT are studying the problem of re-
covering fine magnetite from dense-medium circuits.   Recent  advances
in the cleaning of finer sizes of coal require that  the entire coal-
magnetite streams be fed directly to magnetic separators.  Because of
their high capacity, fever HGMS units may be needed  to handle a given
flovrate than conventional drum-type separators,  and at a  higher re-
covery of fine magnetite particles.

     The Bureau's lignite development and utilization program is com-
posed of several projects.  In-house work is being done on the wash-
ability of Texas and Western lignites to assess ash  and sulfur reduc-
tion potentials.  Also, work on an ion exchange process for  sodium
reduction is in progress.  In conjunction with this  work,  the
Salt Lake City Metallurgy Research Center, through an interagency
agreement, is developing scale-up data for the design of a continuous
ion exchange pilot test facility.  Finally, a contract for the develop-
ment and demonstration of a lignite-pelletizing and  pellet-drying
process was awarded to Holley, Kenney, Schott, Inc.   The initial
research and development phase of the contract will  be carried out at
Michigan Technological University to determine such  things as optimum
size consist of the lignite, type of binder, and  pellet size.

     Two of the Bureau's fine-coal-cleaning projects are entering into
new phases, the dense-medium cyclone and the two-stage coal-pyrite
flotation.  A closed-loop dense-medium cyclone test  circuit  has been
installed with the objective of detailing and optimizing the perform-
ance of cyclones for fine-coal cleaning at lower-than-normal specific
gravities of separation.  The program is in cooperation with the owners
of the Homer City Coal Preparation Plant, the Environmental  Protection
Agency, and the Electric Power Research Institute.

     A coal-pyrite flotation demonstration unit has  been constructed
at Barnes and Tucker Lancashire No.  25 mine.  It  will process 12 tons
per hour in a pyrite flotation circuit to show the improvement in
sulfur reduction over conventional flotation.  Also, an instrumentation
circuit which will control reagent addition according to the mass flow
of the feed pulp will be added at Barnes and Tucker, after testing at
the Bureau.

     The Ames Laboratory—ERDA, located at the Iowa  State  University,
is doing work on the magnetic fraction of coal fly ash. The work
covers three areas:  (l) the recovery of the magnetic iron oxide from
the fly ash by magnetic separation,  (2) the evaluation of  the physical
and chemical properties of the two products of separation, and (3) the
utilization of the magnetic material.  Coal fly ash  is a product of
combustion and is being produced in increasing quantities  as the use
of coal increases.  It represents a considerable  source of iron and
aluminum oxide.

     Roberts and Schaeffer Co. has been experimenting with the
upside-down, or inverted, cyclone.  They recently started  up the first
commercial installation that uses dense-medium inverted cyclones.  It
is a 100-tph plant owned by the Wise Coal Co. in  Virginia, and it is

-------
                        COAL PREPARATION RESEARCH                     89


cleaning the full 1-1/2-inch by 0 size fraction in the cyclones.   Early
reports indicate good performance, although the plant operation is
still "being optimized.

     A major piece of washing equipment which has remained essentially
unchanged for many years is the Baum jig.  Now, however, McNally-
Pittsburg is developing an innovation which should reduce the amount
of misplaced float material in the refuse.  The float mechanism for
sensing the bed depth and the refuse gate for rejection usually work
the width of the jig, but McNally is dividing these units into three
separately working sections across the jig.  This accounts for varia-
tions in the bed.  They have been running tests at some preparation
plants with the revamped jigs, and results seem promising.

     Finally, FMC Corp. is researching two pieces of cleaning equip-
ment .  One, the dry table, has been in development for a few years.
It is a vibrating, wedge-shaped separator which is being touted for
cleaning coal in the West where water is scarce.  FMC is currently
testing a 12-foot pilot unit on bituminous and subbituminous coals for
ash and pyrite reduction and claim the cleaning performance is somewhat
like that of a Baum jig.

     The other FMC washer is a wet vibrating table, about the size and
shape of the conventional concentrating table.  Work on it has not been
as extensive as on the dry table, but FMC is beginning additional re-
search on it.  The company claims it has over twice the capacity of a
concentrating table, which would greatly increase the efficacy of wet
tables in preparation plants.

     These examples of current R & D projects show the wide variety of
ideas in washing, dewatering, and desulfurization for Btu recovery and
environmental protection.  Many of these ideas have become feasible
because of the high price of coal and the need to meet environmental
regulations.

-------
90                     CLEAN COMBUSTION OF COAL

-------
                                                                     91
                     COAL TRANSPORTATION IN 1985

                                 by
                          David J. Hoexter
                      Office of Energy Programs
                     U.S. Department of Commerce


                            Introduction

There are a number of basic questions concerning the transportation of
coal.  Some of these questions include:  What are the factors
affecting the modes by which coal will be shipped?  Can we supply the
required facilities and equipment?  And, will special incentives or
legislation be required to insure that the demand for transportation
services will be met?  A great deal of analysis, and a fairly large
literature, is part of the public record on this topic.  By presenting
and comparing some of these analyses, we can get a good handle on the
magnitude of the transportation services which will be required.

It is the stated intention of the current Administration, as it was
the goal of the previous Administration, that production and consump-
tion of coal in this country in 1985 should be slightly in excess of
one billion tons; the exact quantity is given in various places as
anywhere from 1.1 billion tons to a "doubling" of our use based on
1975 or 1976 consumption.  This same range of estimated use has been
suggested by private sector sources, both within and outside of the
coal and transportation industries.

                Factors Affecting Coal Delivery Mode

What are the factors affecting the modes by which coal will be
shipped?  First, there is the basic cost factor, expressed as dollars
per ton or cents per ton-mile.  Given the more rapid relative
development of large open-pit mining operations in the West and, to
a much lesser extent, underground western mining operations, a premium
is arising on the ability to move large quantities of coal long
distances.  The first candidates which come to mind when considering
very long distance hauling of coal are dedicated, or unit, trains and
slurry pipelines.  Campbell and Katell, in their 1975 report for the
Bureau of Mines, do provide some estimates of the comparable modal
costs.  They point out that, for similar volumes of coal, the cost per
ton mile varies between 3 and 7 mills for slurry pipelines, and 4 and
9 mills for unit trains.  This comparison only serves to tell us that
there will probably have to be features or conditions unique to each
case before a decision on which mode to select can be made.  Both
modes require long-term commitments on the part of coal users, yet
both have features which recommend them.  A slurry pipeline, once
installed, will prove to be highly resistant to inflation in operating
costs, simply because its operating costs are minimal.  The railroads
cannot make this claim.  Reliability of service provides another point
of distinction between the two modes.  The two existing slurry pipe-
lines — in Ohio and Arizona — have been available for operation over

-------
92                     CLEM COMBUSTION OF COAL
98% of the time.  No railroad has provided this degree of reliability.
Where flexibility of route and scheduling is desired, the railroad is
the mode of choice.

Not all coal will be shipped very long distances.  In fact, over
three-quarters of our coal will still be coming from eastern surface
and underground mines in 1985.  There is likely to be a greater
variety in the size of these mining operations, as well as a greater
variety in the distances between the mine and the final consumer.  In
this situation, the cost factors favor trucks, conveyors, non-unit
trains and barges.  In the same study by Campbell and Katell, the
costs for truck and conveyor belt for very short distances (less than
15 miles in the case of conveyors, and, not specified but quite likely
less than, 100 miles for trucks) were about equal to the costs for unit
trains and slurries on a cents per ton-mile basis.  The average cost
per ton-mile for barges was the lowest for any mode examined.  Data
obtained from the National Energy Transportation Study recently
completed by the Congressional Research Service indicated a modal cost
of just over 3 mills for an average trip of 480 miles.  Obviously, the
limiting factor in the use of barges is the availability of usable
rivers and canals.  The bulk of the coal, however, in 1985 will still
be carried by the railroads.  They are second only to trucks in terms
of flexibility, and they provide the best unit costs for the medium
tonnages and distances which will be encountered in bringing eastern
coal to eastern users..

There are other factors which affect the choice of coal transportation
modes.

There are political factors which affect the choice of mode.  I
believe the issue of eminent domain will be resolved at the national
level in favor of the slurry pipeline interests.  The question of
economics is not a stumbling block; the railroads have even agreed
that building a slurry pipeline is cheaper than laying a new rail
line.

The problem of water is at the heart of the controversy.  The use of
the West's limited water supply will be the last political issue to
be resolved in the matter of slurry pipelines.  A smaller parallel
line could carry the water back from Arkansas to Wyoming to be re-used.
This would minimize the amount of water used in the slurry, and solve
the problem of what to do with 20,000 acre feet of almost unreclaimable
water.

Another political factor concerns such environmental constraints as
sulfur emission standards and surface mine reclamation standards.
According to a number of studies, the impact of the surface mine
legislation will, among other things, reduce coal output in Appalachia
by tens of millions of tons below what it otherwise would have been.
The likely effect of this would be to increase the modal shares
accounted for by unit trains and slurry pipelines in the west.  To
the extent that demand for low-sulfur coal increases relative to the
demand for all coal, this, too,' bodes well for the owners and
operators of unit trains.  The impact on truck hauling and short-haul

-------
                           COAL  TRANSPORTATION                        93
rail operations of a drop in demand for medium to high sulfur coals
would not be insignificant.

There are also structural factors involved in determining the mode of
shipment.  One structural factor which should be mentioned is owner-
ship of coal reserves.  The railroads have the largest coal reserve
base in private ownership.  The oil and gas companies are estimated to
hold the next largest share, and coal-consuming industries such as the
steel, electric utility and chemical companies are next.  The implica-
tions of this pattern of ownership would tend to confirm an already
sound view that the railroads will continue to dominate the coal
hauling business.

        Transportation Equipment Requirements and Cost

Can we supply the facilities and equipment required to move the coal?

For estimates of the equipment requirements, we have several recent
reports to draw on.  The first report is "Coal Transportation Practices
and Equipment Requirements to 1985," written by Gary Larwood and David
Benson of the Bureau of Mines.  The authors assume that 1.2 billion
tons of coal will need to be moved in 1985.  According to their study,
the railroads will be required to carry between 894 million and 945
million tons of coal.  To do this would require between 126,000 and
142,000 hopper cars or about 1,260 to 1,420 unit trains in 1985.  A
report prepared by Beehtel Research & Engineering for the Office of
Energy Programs projected a requirement for the equivalent of about
700 unit trains in 1985.

The third report is the 'toal Transportation Capability of the Existing
Rail and Barge Network, 1985 and Beyond" prepared by Manalytics, Inc.,
for the Electric Power Research Institute in September of 1976.  Two
scenarios were prepared.  In the first, it was assumed that 1.5 billion
tons of coal would be moved by rail in 1985; in the second, it was
assumed that 1.24 billion tons would be shipped by rail.  The rolling
stock required for these amounts is projected to be the equivalent of
1,947 unit trains and 2,129 unit trains for the first and second
scenarios, respectively.  Manalytics estimated the number of locomo-
tives required to be 9,735 and 10,645 in the first and second scenarios,
respectively.  Larwood and Benson estimated the number of locomotives
required to be between 6,900 and 7,800.

Considering the number of barges which will be needed in 1985, Larwood
and Benson projected that between 1,800 and 3,400 barges of 1,400 ton
equivalent will be required to move the barge share of coal traffic.
The Office of Energy Programs report, converted to 1,400 ton equiva-
lent units, puts this figure at just over 3,000 barges.

How much is all of this equipment going to cost?  We have the estimates
of two companies.  The first is the Beehtel Corporation and the second
is the Bankers Trust Company.  Both estimates cover the years 1976-
1985, inclusive, and are expressed in 1974 dollars.

-------
 94                     CLEM COMBUSTION OF COAL
The Bechtel Corporation estimates that     approximately $8.8 billion
will be required to design, construct and start up transportation
facilities.  Bankers Trust Company projects that about $10.25 billion
will be required to provide the same services.  Both estimates show
expenditures beginning at around three quarters of a billion dollars
in the late 1970's, and peaking in 1985 at $1.1 billion       and $1.2
billion, respectively.

                         Modal Share Estimates

It is clear that there will not be any very large shifts in the share
of coal hauled by a particular mode.  Rail traffic, which in the first
quarter of 1977 moved half of the coal shipped, will likely increase
its share of the volume as western mining comes into its own and the
increased use of unit trains thus lowers unit shipping costs.  The
Federal Power Commission, in January of this year, released a .staff
study which estimated the modal shares for coal delivered to new coal-
fired electric utility units coming on stream between now and 1985.
The railroads' share of this traffic is estimated by the FPC to be
about 66% in 1980 and 62% in 1985.

River traffic, which accounted for 15% of coal moved in the first
quarter of 1977, is projected by the FPC to carry 9% of the coal
required by new coal-fired utilities in 1980, and just over 6% of
the coal required in 1985.

The FPC projects a large increase in the share of truck traffic
bringing coal to new generating facilities relative to its current
level.  In the first quarter of 1977, trucks hauled over 13% of all
coal; in 1980 and 1985, trucks are projected to account for about
22% of all shipments to new coal-fired facilities.  The single
greatest cause of this rise in modal share is probably the increase
in the number of mine-mouth electric generating plants which are
expected to come on stream.

The remainder of the coal-carrying modes — conveyors, pipelines,
etc. — will have their shares affected little, if at all, although
the actual absolute tonnages they carry may increase dramatically.

-------
                          COAL TRANSPORTATION                        95
                             REFERENCES
Battelle Columbus Laboratories.  "A Report to the Interagency Coal
  Task Force, Project Independence Blueprint, on the Manpower
  Requirements of Coal Transportation."  Washington, D. C.:   June,
  1974.

Battelle Columbus Laboratories.  "A Report to the Interagency Coal
  Task Force, Project Independence Blueprint, on the Modal Transpor-
  tation Costs for Coal in the United States."  Washington,  D. C.:
  May, 1974.

Bechtel Corporation.  "Capital, Manpower, Materials and Equipment
  Requirements for a Department of Commerce Energy Projection."  San
  Francisco, California and Washington, D. C.:  December, 1976.

Campbell, T.C., and Katell, Sidney.  "Long-Distance Coal Transport:
  Unit Trains or Slurry Pipelines."  Washington, D. C.:  Department
  of the Interior, 1975.

Congressional Research Service and U.S. Geological Survey.  "National
  Energy Transportation."  Committee Print of the U.S. Senate #95-15.
  Washington, D. C.:  May, 1977.

Electric Power Research Institute.  "Coal Transportation Capability  of
  the Existing Rail and Barge Network, 1985 and Beyond."  Palo Alto,
  California.  September, 1976.

Federal Power Commission.  "Status of Coal Supply Contracts for New
  Electric Generating Units 1976-1985."  Washington, D. C.:   January,
  1977.

Larwood, Gary M. and Benson, David C.  "Coal Transportation Practices
  and Equipment Requirements to 1985."  Washington, D. C.:  Department
  of the Interior, 1976.

Mutschler, P.H., et. al.  "Comparative Transportation Costs of
  Supplying  Low-Sulfur Fuels to Midwestern and Eastern Domestic
  Energy Markets."  Washington, D. C.:  Department of the Interior,
  1973.

Peat, Marwick, Mitchell and Co.  "Railroad Freight Car Requirements
  for Transporting Energy, 1974-1985."  Washington, D. C.:  Federal
  Energy Administration, November, 1974

-------
96                     CLEAN COMBUSTION OF COAL

-------
                                                                     97
          COAL DESULFURIZATION TEST PLANT STATUS - JULY 1977

      L.J. Van Nice, M.J. Santy, E.P. Koutsoukos, R.A. Orsini and
                               R.A. Meyers

                         TRW Systems and Energy
                         Redondo Beach, CA 90278

I.  INTRODUCTION

     An 8 metric ton/day process test plant for chemical desulfurization
of coal has just been built at TRW's Capistrano Test Site in California.
The plant, shown in Figures 1 and 2, was constructed under an Environ-
mental Protection Agency sponsored project for the development of the
Meyers  Process.  Current plans call for plant shakedown followed by
processing of 100-200 tons of American Electric Power Service Corpor-
ation's Martinka Mine coal.

     The Meyers Process removes up to 80 percent of the total sulfur
content of coal through chemical leaching of 90 to 95 percent of the
pyritic sulfur contained in the coal matrix with aqueous sulfate
solution at temperatures of 90° to 130°C.  The ferric sulfate content
of the leach solution is regenerated at similar temperatures using air
or oxygen, and elemental sulfur and iron sulfates are recovered as
reaction products or alternatively gypsum can replace all or a portion
of the iron sulfates as a product.  The physical form of the coal re-
mains unchanged; only pyrite and some inorganic materials are removed.

     Low and medium organic sulfur coal  can be desulfurized  prior to
combustion using the Meyers Process (1,2)  to meet governmental  require-
ments for sulfur oxide emissions.

     The Environmental Protection Agency estimates that 90 x 109 tons
(80 x 109 metric tons) of coal reserves in the U.S. Appalachian Coal
Basin can be reduced in sulfur content by the Meyers Process to levels
which will meet New Source Performance Standards.  Successful bench-
scale testing (3,4) and promising engineering analyses (3,5-7) together
with applicability testing (8,9), have led the Environmental Protection
Agency to sponsor the construction and operation of a test plant.

     Present estimated processing costs using utility financed depreci-
ation of capital, and including coal grinding and compaction of the
product (where necessary), are $8-12/ton (3,6).  The price varies with-
in this range according to factors such as offsite, coal pre-cleaning,
reaction rate of the particular coal and coal top-size utilized, i.e.,
3/8"processing and pre-cleaning of run-of-mine coal contribute to lower
cost.  Recent advances, which have not yet been fully designed, promise
to reduce the above costs.

     Process chemistry, and test plant design and operation will be
described below.

-------
                                                                            CO
                                                                            §
                                                                            g
Figure 1.  Test Plant - Front View

-------
                                                                                   I
                                                                                   I
                                                                                   M
                                                                                   i
Figure 2.  Test Plant - View Through Tank Farm

-------
100                    CLEAN COMBUSTION OF COAL
II.  PROCESS CHEMISTRY, KINETICS AND SCHEME

     The process is based on the oxidation of coal  pyrite with ferric
sulfate (Equation 1).  The leaching reaction is highly selective to
pyrite with 60 percent of the pyritic sulfur converted to sulfate
sulfur and 40 percent to elemental  sulfur.  The reduced ferric ion Is
regenerated by oxygen or air according to Equations 2 or 3.

                 FeS2 + 4,6 Fe2(S04)3 + 4.8 H20 -v                    1}

                   10.2 FeS04 + 4.8 H2S04 + 0.8S

                 2.4 02 + 9.6 FeS04 + 4.8 H2S04 +                    2j

                   4.8 Fe2(S04)3 + 4.8 H20

                 2.3 02 + 9-2 FeS04 + 4.6 H2S04 -»•                    3j

                   4.6 Fe2(S04)3 + 4.6 H20


Regeneration can be performed either concurrently with coal  pyrite
leaching in a single operation or separately.   The  net effect of the
process is the oxidation of pyrite with oxygen to yield recoverable iron,
sulfate sulfur, and elemental sulfur.  The form of  process products
varies to some extent with the degree of regeneration performed.  Thus,
Equations 1 and 2 lead to the overall process  chemistry indicated by
Equation 4 with products being a mixture of iron sulfates and elemental
sulfur.  Equations 1 and 3 yield ferrous sulfate, sulfuric acid, and
elemental sulfur as indicated by Equation 5.

          FeS2 + 2.4 02 * 0.6 FeS04 + 0.2 Fe2(S04)3 + 0.8S           4)

          FeS2 + 2.3 02 + 0.2 H20 -»• FeS04 + 0.2 H2$04 + 0.8S         5)

Several options exist in product recovery.  Iron sulfates may be recov-
ered as pure solids by stepwise evaporation of a spent reagent slip-
stream with ferrous sulfate being recovered first because of its lower
solubility.  Alternately, ferrous sulfate may be recovered by crystal-
lization, and ferric sulfate or sulfuric acid removed by liming spent
reagent or spent wash water slipstreams.  Iron sulfates may be stored
as solids for sale or may easily be converted to highly insoluble basic
iron sulfates (by air oxidation) or calcium sulfate (by low-temperature
solid phase reaction) for disposal.  Elemental sulfur may be recovered
from coal by vaporization with steam or by vacuum,  or it can be leached
out with organic solvents such as acetone.  Product marketability and
product recovery economics will dictate the choice.  Recovery economics
may be influenced by quantity and concentration of  product in the
process effluent streams which in turn are influenced by the pyrite
concentration in the coal and the desired extent of desulfurization.

     The process has been extensively studied at bench-scale.  Parameters
investigated included coal top-size, reagent composition, slurry concen-
tration, reaction temperature and pressure, and reaction time.

-------
                         COAL DESULFURIZATION                       101


Additional investigations completed or underway include concurrent coal
Teaching-reagent regeneration, product recovery, product stability,
and the effect of coal physical cleaning on process performance and
economics.  The process scheme depicted in Figure 3 is based on the
bench-scale testing.  Coal is a) crushed to the desired size for pro-
cessing, b) contacted with hot recycled reagent in the Mixer (90-100°C),
c) leached of pyrite in the Reactor(s) with simultaneous or separate
reagent regeneration, d) washed with hot water, and e) stripped of
elemental sulfur, dried and finally cooled.  The iron and sulfate sulfur
are recovered from spent reagent slipstreams prior to reagent recycle.
Figure 4  shows typical data on pyrite removal  rates from Appalachian
coal as a function of temperature.  Removal of 10-20 percent of  the
pyrite is obtained during slurry mixing and heat-up.

     Bench-scale data indicated that the pyrite leaching rate from coal
can be adequately represented by the empirical  rate expression (Equation
6).


                       'L ' -

where

     KL  = AL exp (-EL/RT),

     Wp  = wt percent pyrite in coal,

     Y   = ferric ion-to-total iron ratio in the reactor reagent, and

     A. and E. are constants for each coal and particle size at least
     over most of the reaction range.

     The leach rate is a function of coal type.  Pyrite extraction rates
vary considerably, as detailed in a study of the Meyers Process as
applied to U.S. coal (8), e.g., there was more than one order of magni-
tude difference between the fastest and slowest reacting coal.  The
reagent regeneration rate is governed by the rate expression (Equation
' ) •

                      _    dFe    = K  p   (Fe+2)2                   7)
                                    K  P   Ue  '                    n
                               _
                    R --- d        R  02

where

     KR  = AR exp (-ER/RT),

     Pn  = oxygen partial pressure,
      °2

     Fe+2 = ferrous ion concentration in the reagent solution, and

     A  and E  are constants.

-------
  102
CLEAN COMBUSTION OF  COAL
                                                                        COAL
                     Figure  3.   Process Flow Schematic
   90
   80
   70
   60
3?
I"
UJ
Of.

m  40
   30
   20
   10
                                        130"C
                                                  120»C
                            i
                                            110°C
               0.5
 1.0          1.5         2.0


      REACTION TIME, HOURS
2.5
                                                                         3.0
        Figure  4.   Temperature Effect on  Processing of 14 Mesh

                    Top-Size Lower Kittanning  Coal  (33% w/w Slurries)

-------
                         COAL DESULFURIZATION                       103


     Engineering evaluation of available data shows that it is prefer-
able to process fine coal (<2 mm top-size) under simultaneous Teaching-
regeneration conditions in the temperature range of 110-130°C until  the
majority of the pyrite 1s leached out.  Ambient pressure processing
(approximately 100°C) is indicated for the removal of the last few
tenths percent of pyrite since the low Wp value substantially reduces
the rate of ferric ion consumption and, therefore, the need for
simultaneous reagent regeneration.  Ambient pressure processing appears
to be indicated also for coarse coal (e.g., 10 millimeter top-size)  for
several reasons.  It is difficult to continuously feed a non-siurryable
coal into and remove it from a pressure vessel.  It is much easier and
less costly to drain leach solution from the coal and pump it into a
small pressure vessel for regeneration.  Also the slower reaction rate
with coarse coal would require much longer residence times and unreason-
ably large total volume for pressure vessels.  These engineering evalu-
ations were part of the data used to design the test plant.

III.  TEST PLANT DESIGN AND OPERATION

     A test plant sized to process up to 8 metric tons per day of coal
has been built, under the sponsorship of the Environmental Protection
Agency, at TRW's Capistrano Test Site.  A plant flow diagram is shown
in Figure 5.  The facility is capable of on-line evaluation of the
following critical process operations:

•  Pressure leaching of pyritic sulfur from 150 micron to 2 mm top-size
   coal at pressures up to 100 psig,

•  Regeneration of ferric sulfate both separately, for processing larger
   top-size coal or low pyrite coal, and in a single vessel with the
   leaching step for processing of suspendable coal,

•  Filtration of leach solution from reacted coal,

•  Washing of residual iron sulfate from the coal.

     Iron sulfate crystallization, elemental sulfur recovery and coal-
drying unit operations will be evaluated in an off-line mode in equip-
ment vendor pilot units.  Leaching of 10 mm top-size coal can be
evaluated in an off-line mode in an atmospheric pressure vessel installed
in the test plant.  Coarse coal processing (5-10 mm top-size) has been
very promising in laboratory tests (3).  If this approach proves out
in bench-scale evaluations, more extensive and on-line coal leaching
units can be readily added to the present test plant.  Processing fine
coal allows the highest rate of pyritic sulfur removal, while processing
coarse coal, although slower, allows lower cost coal dewatering units
and the direct shipping of desulfurized coal product without need for
pelletizing.

     The test plant constructed at the Capistrano Test Site is a highly
flexible facility capable of testing the numerous alternate processing
modes of potential interest in the Meyers Process.  The flow diagram
shown in Figure 3 presents an equipment train for continuous process
testing of slurried coal.  Fine coal ground to the desired size is

-------
104
CLEAN  COMBUSTION OF COAL
                                                     ATMOS.
                                                                   WATER
                 1      I

                                                                 Y}™
                                                                	1
                                                     ATMOS.
                                                         TO TRUCK FOR DISPOSAL
                      Figure 5.  Test Plant  Flow Diagram

-------
                         COAL DESULFURIZATION                       105
stored under nitrogen gas in 1.8 metric ton sealed bins.  As required,
bins are emptied into the feed tank (T-l).  Dry coal is continuously
fed by a live bottom feeder to a weigh belt which discharges through
a rotary valve to the three stage mixer (Stream 1).  The aqueous iron
sulfate leach solution (Stream 2) enters the mixer after first passing
through a foam breaker (T-2).  Steam is added (Stream 3) to raise the
slurry to its boiling point.  Foaming will occur in the early stages of
mixing,  but  will cease when  particle wetting  is complete.   It is
 believed that the mixing  time and conditions  necessary  to complete  the
wetting  and  defoaming of  the slurry will  depend on  the  coal type and
 size  and on  the  residual  moisture in the  feed coal.  To allow study of
 the mixing  parameters, the mixer stages have  variable volume with
 variable speed agitators, and   the feed flow  rates  for  coal, leach
 solution and steam  can be varied over wide  ranges.

      The defoamed slurry  (Stream 4) is pumped to a  five stage pressure
 vessel  (Reactor  1)  in which most of the pyrite removal  reaction occurs.
 Some  of  the  pyrite  reaction occurs during mixing,  but in the mixer  the
 reaction rate slows  rapidly  because the remaining  pyrite (Wp) decreases
 and because  the  ferric iron  is  rapidly being  converted  to ferrous iron
 (Y decreases).   The  pressure reactor overcomes the  decreased rate in
 two ways.   First, it increases  the temperature (and pressure) to
 increase the reaction rate constant.  Second, oxygen is introduced
 under pressure to regenerate ferric iron  and maintain a high solution
 Y. The  flow diagram shows that steam and oxygen can be added to any
 or all of the five  stages and that cooling  can be  provided  for any
 stage if necessary  to remove the excess heat  of reaction.   The unused
 oxygen saturated with steam  (Stream 7) is contacted in  a small pressure
 vessel  (T-3) with the feed leach solution (Stream  5) to provide heated
 leach solution for  the mixer  (Stream 2) and cooled  vent gas.  The vent
 gas from both T-2 and T-3 are scrubbed in T-4 to remove any traces  of
 acid  mist.   The  reaction  parameters of importance  have  already been
 well  studied at  laboratory and  bench-scale  in batch mode.   The test
 plant reactor will  accommodate  the necessary  studies of key parameters
 in a  continuous  reactor at coal  throughputs between 2 and 8 metric  tons
 per day.  Parameters which will  be studied  include  temperature,
 pressure,  oxygen purity,  slurry concentration, iron sulfate concentra-
 tion, iron  sulfate  concentration, acid concentration, residence time
 per stage,  number of stages, mixing energy, type of mixing, coal size
 and type.   The reactor can also be used to  study leach  solution regen-
 eration  in  the absence of coal.

      Reacted coal slurry  (Stream 8) at elevated temperature and
 pressure is  flashed  into  a gas-liquid separator vessel  (T-5).  The
 steam generated  (Stream 9) is condensed in  T-4 and  the  condensate plus
 any entrained acid mist is removed with the water.  The residual slurry
 (Stream  10)  is fed  to a belt filter.  The filtrate, which is regener-
ated  leach  solution, is removed from the  coal slurry through a vacuum
 receiver (T-9) and  pumped (Stream 12) to  a  large leach  solution storage
 tank  (T-6).  The coal on  the filter belt  is washed  with water (Stream
11) and  discharged  from the  filter belt.  The wash  water is removed
through  a  vacuum receiver (T-10) and sent to  a large liquid-waste
holding  tank (T-8)  for subsequent disposal.   The filter is  a highly
versatile  unit which should provide the data  necessary  for  scale-up.

-------
106                    CLEM COMBUSTION OF COAL


It has variable belt speed, variable belt areas assigned to washing,.
variable cake washing rates, belt sprays if needed to control blinding
of the pores in the belt, and steam nozzles to provide for partial cake
drying.

     As an alternate process step, the slurry from the flash tank (T-5)
can be passed into a secondary reaction vessel (Reactor 2).  At typical
coal feed rates, this vessel can be filled in about two hours and then
closed off, stirred and heated for any desired period of time before
being pumped to the filter.  Residence times up to about 10 hours are
available in the primary reactor, Reactor 1.  This secondary reactor
can be used to extend residence times to much longer times for examining
the removal of final traces of pyrite or examining any other long term
behavior.  The stirred vessel also can  serve to repulp the filter cake
for additional coal washing studies.

     The final item of major equipment in the test plant is the coarse
coal contact vessel (Reactor 3).  This insulated and heated tank will
hold a full bin (about 1.8 metric tons) of coarse coal (5 to 10 milli-
meter top-size).  The principal use for this vessel is to convert the
regenerated leach solution in storage tank T-6 to a more depleted solu-
tion in the process feed tank, T-7.  In general, the iron sulfate leach
solution in the filtrate going to tank T-6 will have a high Y because
no  secondary reactor was in use.  For some test conditions, the feed
to the process must be at a lower Y to simulate recycle leach solution
from a secondary reactor.  Passing all or some portion of the solution
through coal will lower the Y of the solution to the desired value.
This vessel is basically a coarse coal reactor and if appropriate
sampling ports and  possibly some flow distribution internals were
added, it could be used to obtain design data for coarse coal processing.

     Solution tanks are sized at about 50,000 liters to provide for
about a week of continuous operation on the same feed without recycle
or change.  It also provides for uniform leach solution and coal samples
of a large enough size for product recovery studies performed by equip-
ment vendors.  Operation at the scale of the test plant will provide
experience and data expected to be adequate for the design of a demon-
stration-size commercial plant.


                              REFERENCES

1.  R.  A.  Meyers, J. W.  Hamersma, J. S.  Land and M. L. Kraft, Science.
    177.  1187 (1972).

2.  R.  A.  Meyers, "Removal  of Pyritic Sulfur from Coal Using Solutions
    Containing Ferric  Ions," U.S. Patent 3768988 (1973).

3.  E.  P.  Koutsoukos,  M. L.  Kraft, R. A. Orsini, R. A. Meyers, M.  J.
    Santy and L.  J. Van  Nice (TRW Inc.), "Final  Report Program for
    Bench-Scale Development of Processes for the Chemical  Extraction of
    Sulfur from Coal,"  Environmental  Protection Agency Series
    EPA-600/2-76-143a  (May 1976)."	

-------
                         COAL DESULFURIZATION                       107
4.  J. W. Hamersma, E. P. Koutsoukos, M. L. Kraft, R. A. M.eyers, G.  J.
    Ogle, and L. J. Van Nice (TRW Inc.), "Chemical Desulfurization of
    Coal:  Report of Bench-Scale Developments, Volumes 1 and 2,
    "Environmental Protection Agency Series, EPA-R2-173a (February 1973).


5.  E. M. Magee  (Exxon Research and Engineering Co.), "Evaluation of
    Pollution Control in Fossil Fuel Conversion Processes,  Coal Treat-
    ment:  Section 1.  Meyers Process," Environmental Protection
    Technology  Series, EPA-650/2-74-009-k  (September 1975).

6.  W. F. Nekervis and E. F. Hensley (Dow  Chemical, U.S.A.), "Conceptual
    Design of a  Commercial Scale Plant for Chemical Desulfurization of
    Coal," Environmental Protection Technology Series. EPA-600/2-75-051
    (September  1975).

7.  M. Rasin Tek (U.  of  Michigan),  "Coal Beneficiation," in Evaluation
    of Coal Conversion Processes, PB-234202 (1974).

8.  J. W. Hamersma and M. L. Kraft  (TRW Inc.), "Applicability of the
    Meyers Process for Chemical Desulfurization of Coal:  Survey of 35
    Coal Mines,"  Environmental Protection Technology Series,
    EPA-650/2-74-025-a (September 1975).

9.  U.S. Environmental Protection Agency,  Office of Research and Devel-
    opment, Washington,  DC,  "Applicability of the Meyers Process for
    Chemical Desulfurization of Coal:   Initial Survey of Fifteen Coals,"
    by J. W. Hamersma, et al., Systems Group of TRW, Inc., Redondo
    Beach, CA,  Report No. EPA-650/2-74-025 (April 1974) Contract No.
    68-02-0647.

-------
108                    CLEAN COMBUSTION OF COAL

-------
                                                                    109
              STATUS AND PROBLEMS IN THE DEVELOPMENT OF
          HIGH GRADIENT MAGNETIC SEPARATION  (HGMS) PROCESSES
                    APPLIED TO COAL BENEFICIATION*

                        Y. A. Liu and C. J. Lin
         Department of Chemical Engineering, Auburn University
                        Auburn, Alabama  36830

ABSTRACT

     An overall discussion of the status and problems in the develop-
ment of HGMS processes applied to coal beneficiation is presented.  In-
cluded in the discussion are such topics as:  bench-scale and pilot-
scale experimental studies; quantitative modeling of experimental
results; conceptual process design and cost estimation; and comparison
with alternate technologies for coal beneficiation.  The needs and
opportunities in the future research and development work related to
magnetic beneficiation of coal are also suggested.

INTRODUCTION

     High gradient magnetic separation (HGMS) is a new technology which
provides a practical means for separating micron-size, feebly magnetic
materials on a large scale and at much faster flow rates than are pos-
sible in ordinary filtration.  The technology is also applicable to
separating nonmagnetic materials which can be made to associate with
magnetic seeding materials.  It was developed in 1969 for the wet sepa-
ration of weakly magnetic contaminants from kaolin clayl~6.  A typical
HGMS unit for this wet application is shown schematically as Figure 1.
The electromagnet structure consists of energizing coils and a sur-
rounding iron enclosure.  The coils in turn enclose a cylindrical work-
ing  volume packed with fine strands of strongly ferromagnetic packing
materials such as ferritic stainless steel wool.  With this design,
a strong field intensity up to 20 kOe can be generated and distributed
uniformly throughout the working volume.  Furthermore, by placing in
the uniform field the ferromagnetic packing materials which increase
and distort the field in their vicinity, large field gradients of the
order of 1-10 kOe/ym can be produced.  In the wet beneficiation of
kaolin clay, the HGMS unit is used in a batch or cyclically operated
process like a filter.   The kaolin feed containing the feebly magnetic


*This work was supported fay the National Science Foundation  (grant no,
GI-38701), by the Energy Research and Development Administration
(contract no. W-7450-eng-26 ORNL/Sub-7315), by the Gulf Oil Foundation
and by Auburn University. The authors wish to thank Messrs, A. W.
Deurbrouck and R. E. Hucko of the Coal Preparation and Analysis Labor-
atory of Bureau, of Mines for their continued assistance in providing
valuable technical information and coal samples for the work reported
herein.

-------
110
                       CLEAN COMBUSTION 07 COAL
                               CLEANED
                              "EFFLUENT
                                                IRON
                                                ENCLOSURE
                                                ELECTROMAGNET
       BED
                                    STAIN LESS STEE
                                    WOOL STRANDS
                                       NON MAGNETIC
                                          PARTICLES
                           FEED

                                   MAGNETIC PARTICLES
           Figure 1.  Cyclic High Gradient Magnetic Separator^

contaminants of low concentrations is pumped  through the stainless
steel wool packing or matrix of the separator from the bottom while the
magnet is on.  The magnetic materials (mags)  are captured and retained
inside the separator matrix, and the nonmagnetic components (tails)
pass through the separator matrix and are collected as the beneficiated
products from the top of the magnet.  After some time of operation, the
separator matrix is filled to its loading capacity.  The feed is then
stopped, and the separator matrix is rinsed with water.   Finally,  the
magnet is turned off, and the mags retained inside the separator matrix
are backwashed with water and collected.   The whole procedure is re-
peated in a cyclic fashion.  The significance of HGMS, and its latest
engineering and commercial development have been adequately discussed
elsewhere*-"".
     Because of its very low costs and outstanding technical perform-
ance  demonstrated in the kaolin application, HGMS was recently adapted
to solving many separation problems related to minerals and chemical
processing industries.  An important and promising application of HGMS
is the magnetic removal of inorganic sulfur and ash from coal.  Pre-
vious experimental investigations have indicated clearly that most of
the mineral impurities in coal which contribute to its pyritic sulfur,
the sulfate sulfur and the ash content are paramagnetic.  These sulfur-
bearing and ash-forming minerals, if sufficiently liberated as discrete
particles, can be separated normally from the pulverized diamagnetic
coal by magnetic means?"**,  indeed, the technical feasibility of apply-
ing HGMS to the beneficiation of pulverized coal suspended in water has
been demonstrated in a number of recent studies, with substantial
amounts of sulfur and ash removal reported""-^.  Recently, a bench-
scale feasibility study of utilizing a novel air-fluidized separator
matrix in HGMS applied to the beneficiation of the dry pulverized  coal
was conducted in the author^' laboratory in cooperation with Oak Ridge
National Laboratory.  The available results from this study were quite
encouraging, indicating that a dry HGMS process for coal  beneficiation

-------
                  HIGH GRADIENT MAGNETIC SEPARATION
                                                                 111
with performance comparable to the wet  magnetic approach could be devel-
oped at much  cheaper cost.-"
     In this  paper,  an overall discussion  of  the recent development and
current status  of HGMS processes applied to coal beneficiation is pre-
sented.  A  quantitative assessment of the  technical and economical fea-
sibility of applying HGMS to remove sulfur and ash from both dry and
wet coals is  given.   In particular, because of its more advanced state
of development,  the magnetic beneficiation of water slurries of pul-
verized coals is discussed in some detail.   The needs and opportuni-
ties in the future research and development work related to magnetic
beneficiation of coal are also suggested.

MAGNETIC BENEFICIATION OF COAL/WATER SLURRY
              "D&>CAJJp£Lon. ^  A conceptual  process for the magnetic
beneficiation of pulverized coal suspended in water by HGMS is shown
schematically in Figure 2.
                            Water Supply
                          1.653 million gal/day
                                              Wash Water
                                              1440 gal/cycle
                                        HGMS 7'D, 20" L.
                                        2.61 cm/sec Velocity
                                         480 gal. Canister
                                             Volume
                                        0.96 Tons S.S. Wool
                                         94% Void Volume
                                         38.46 GPM/Ft2
                  Cycle Time 6.lmin.
                  Duty Cycle 67.4%
                                                 Water
                                              Mags
                            Refuse 9.92 TPH-
                                                  Toils
                                                          Water
                                                    •Vacuum Filter


Thermal Dryer
                                                         Clean Dry Coal
                                                           56.21 TPH
      Figure 2.  Magnetic Beneficiation of Coal/Water Slurry by HGMS

 A coal slurry of a fixed concentration is prepared by mixing known
 amounts of pulverized coal, water and a dispersant (wetting agent)  like
 Alconox.   The HGMS unit used  here is the largest commercial unit  now
 in use for producing high-quality paper coating clays.  It is operated
 at a fixed field intensity of 20 kOe generated in an open volume  7  ft
 in diameter and 20 in. long.   A stainless steel wool separator matrix
 having 94% voids is placed in the open volume.  The coal slurry is
 pumped through the energized  separator matrix at a fixed retention
 time (flow velocity) until the matrix reaches its loading capacity.
 After rinsing with water, the mags are sent to a settling pond or a
 clarifier to recover water for re-use.  The tails are collected,  dewa-
 tered,  and dried.  The typical specifications of process streams  are
 also included in Figure 2, and the detailed operating conditions  for
 such a case are given in the  Appendix.
Exp&ujne.nta£. Studio  oft ?n.o
-------
112                    CLEM COMBUSTION OF COAL
adapted in a bench-scale exploratory study to remove sulfur and ash
from a finely pulverized Brazilian coal suspended in water.-'--'-  Since
then, other investigators have utilized pilot-scale HGMS units to
desulfurize and deash water slurries of some pulverized Eastern U.S.
coals.  For example, results of pilot-scale studies that demonstrated
the technical feasibility of the magnetic separation of sulfur and ash
from water slurries of pulverized Illinois No. 6, Indiana No. 5 and No.
6, Kentucky No. 9/14, and Pennsylvania Upper Freeport and Lower Kittan-
ning coals have been reported.9'10'12'13*15  Depending upon the types
of coal and the separation conditions used, the existing bench- and
pilot-scale results have shown that the use of single-pass HGMS was
effective in reducing the total sulfur by 40-55%, the ash by 30-45%,
and the pyritic sulfur by 75-90%, while achieving a maximum recovery of
the beneficiated coal of about 95%.8>16  These available results have
indicated also that both the grade and recovery of the separation can
be enhanced generally with the use of a larger separator matrix or by
the recycle of the tail products.  Although the existing data have not
yet established total deashing by magnetic means, there is some indica-
tion that by enhancing the magnetism of ash-forming minerals and by op-
timizing the separation conditions, etc., the effectiveness of magnetic
separation of ash from coal can be improved.  Further discussion of
the reported experimental results of the magnetic beneficiation of
water slurries of pulverized coals can be found in the literature.
Also, the published Proceedings of Magnetic Desulfurization of Coal
Symposium held at Auburn University in March, 1976 can be used for
ready reference on HGMS and its experimental application to coal bene-
ficiation.

     Quantitative, ttod&ting and Pizdiction o^ Se.pa/iation PeA^o^once.  An
important and latest breakthrough in applying HGMS to the beneficiation
of coal/water slurry is the successful development of the conceptual
understanding and the quantitative model required for predicting the
technical performance of pilot-scale separation.  In particular, the use
of an experimentally verified mathematical model developed in the authors'
laboratory ' ^ has now allowed one to quantitatively identify the trade-
off of separation variables so as to optimize the magnetic removal of
sulfur and ash from coal.  Also, the model can be used to assess the
technical and economical feasibility of the magnetic beneficiation of
coal without extensive trial-and-error testing.  In order to stimulate
a more systematic approach to the future development of HGMS processes
applied to coal beneficiation, some new insights on the quantitative
aspects of magnetic beneficiation of pulverized coal in water are
briefly discussed below.
     The technical performance of a HGMS is generally characterized
by the grade and recovery of the separated product, and by the capacity
of the separator.  When applying HGMS to beneficiating a fixed amount
of pulverized feed coal of known sulfur and ash contents, the grade
may be represented by the sulfur and ash contents in the beneficiated
coal; while the recovery may be specified by the amount of the benefi-
ciated coal.   The capacity of a separator in the magnetic beneficiation
of coal may be characterized by the amount of coal processed from the
start of the separation until the instantaneous sulfur and ash contents
of the beneficiated coal reach a prespecified percentage of the known
sulfur and ash contents of the feed.  In particular, the instantaneous

-------
                  HIGH GRADIENT MAGNETIC SEPARATION                 113
change in effluent sulfur and ash contents with time, the so-called
breakthrough curve, can be used to determine the processing time after
which a separator is loaded.  Here, a separator is said to be loaded
if the effluent sulfur and ash contents of the coal to be beneficiated
remain practically unchanged.
     There are a number of important considerations involved in the
quantitative modeling and prediction of the technical performance of
HGMS applied to coal beneficiation.  The first one is related to the
particle trajectory and buildup.   Specifically, the technical perform-
ance of an HGMS depends on how efficiently it captures magnetic parti-
cles, and how much of the captured particles can be retained in the
separator matrix.  The capture of magnetic particles by HGMS has recent-
ly  been studied theoretically by using the equations of motion of
magnetic particles flowing around a single magnetized, ferromagnetic
collecting wire, and the performance of an ideal, unloaded HGMS com-
posed of many such wires in the separator matrix is then related to the
particle trajectories computed from the specified separation condi-
tions. 2»17  Preliminary efforts have also been made to examine both
theoretically and experimentally the effects of particle buildup or
matrix loading on the performance of a nonideal, partially loaded HGMS.
However, none of the existing models based on the recent analyses of
particle trajectory and buildup has been shown to be applicable to
quantitatively predicting the effects of separation variables on the
technical performance observed in pilot-scale experimental studies of
HGMS.  The next important consideration is the characteristic of the
feed stream to be magnetically beneficiated.  In the literature, most
of the reported modeling and experimental studies of HGMS have been
limited only to the feed streams containing either pure magnetic par-
ticles or simple mixtures of magnetic and nonmagnetic particles of
approximately monodispersed or narrowly distributed sizes.  However,
little attention has been devoted to relating the technical perform-
ance  of HGMS to the characteristics of the feed stream containing
particles of a wide range of sizes, densities and magnetic susceptibil-
ities as found in the magnetic beneficiation of coal.  The third, but
relatively less important, consideration is related to the mechanical
entrapment or filtration of particles at low or zero field intensity.
For separations at high field intensity, however, the effects of me-
chanical entrapment on the technical performance of HGMS are often con-
sidered to be negligible.  In the recent work from the authors' labora-
tory, 6,15 the technical performance of HGMS has been quantitatively
examined with reference to the above major considerations.  Two practi-
cal observations of importance in the modeling of HGMS applied to
coal beneficiation resulted from this work can be summarized as follows.
     (1)  The technical performance of HGMS is quantitatively related
to the particle trajectory  (or capture) and buildup.  Under the pres-
ently used or proposed conditions for wet HGMS processes, however, the
capability of the wire matrix to capture magnetic particles remains
practically unchanged and essentially all magnetic particles are cap-
tured before the matrix is saturated or loaded with the buildup of
particles.  As a result, there is practically no need to calculate the
trajectories of particles.  The main factor in determining the techni-
cal performance of a pilot-scale or an industrial HGMS appears to be
the particle buildup, but not the particle trajectory (or capture).
     (2)  Both the particle trajectory (or capture) and buildup are

-------
                       CLEAN COMBUSTION OF COAL
highly dependent upon the ratio of the so-called magnetic velocity Vm
of the particle to the free stream fluid velocity V^.  In particular,
the maximum amount of particle buildup on the wire increases with in-
creasing value of VHI/VK,; and there exists a minimum value of Vm/V^, for
the captured particles to remain sticking to the wire matrix.  Here,
the magnetic velocity Vm is essentially a terminal velocity of a par-
ticle in a magnetic field, and it is defined by:

                           V  = 2y° M H° * R2                   (1)
                            m       9n a

Except for the length L and packing void volume e of the separator ma-
trix, the ratio V^/V^ contains almost all the major independent vari-
ables in HGMS, namely:  (a) particle properties - radius R and magnetic
susceptibility x> (b) flow field - fluid viscosity n and free stream
fluid velocity V^; (c) magnetic field - field intensity Ho and magne-
tization of wire M; and (d) separator matrix packing characteristics -
wire radius a  (and magnetization M) .  Under conditions of constant mag-
netic and flow fields as well as fixed separator matrix packing charac-
teristics, the magnetic velocity is mainly a function of particle prop-
erties.  In order to quantitatively relate the technical performance
of HGMS to the characteristics of the feed stream containing particles
of a wide range of sizes, densities and magnetic susceptibilities, it
is necessary to determine the magnetic velocities of particles in the
feed stream.
      Based on the above findings, a simple particle buildup model in-
corporating the feed characteristics for predicting the technical per-
formance of HGMS under the presently used or proposed conditions for
wet HGMS processes has been developed by first calculating the minimum
magnetic velocity for particle buildup, denoted by Vm m^n   The latter
is uniquely determined from the knowledge of the following process
variables in HGMS:  (a) (LOAD) - the total weight of captured magnetic
particles per unit cross section of the separator matrix, (b) L - the
length of the separator matrix, (c) F - the packed fraction of the
separator matrix (F = 1-e) , (d) d - the apparent density of the par-
ticle buildup, and (e) V^ - the free stream fluid velocity.  The
specific details for calculating VmjIQin from these variables can be
found in the literature  "»15.  Next, a so-called magnetic velocity
distribution function for characterizing the feed stream, denoted by
F(Vm) , can be introduced.  Specifically, F(Vm) represents the cumula-
tive weight fraction of magnetic particles in the feed stream having
a magnetic velocity less  than Vm.  Thus, if Vm = Vm)min, F(Vm)  cor-
responds to the cumulative weight fraction of magnetic particles  in
the feed stream which will not be captured inside the separator matrix.
This result suggests that a simple particle buildup model for HGMS can
be written as

                    COUt _ -C./TT \         TT
                     •— -              Vm =
Here, COut and Cin refer to the concentrations of the specific magnetic
particles of interest, such as pyritic sulfur content in the magnetic
beneficiation of coal, at the outlet and inlet of the separator matrix,
respectively.  Furthermore, a simple material balance of particles sug-
gests that (LOAD) can be calculated from the total weight of feed

-------
                  HIGH GRADIENT MAGNETIC SEPARATION                 115
particles per unit cross section of the. separator matrix, (FEED), by
integrating the following equation

                d(LOAD)
                                                               (4)
The magnetic velocity distribution, F(Vm) is to be quantitatively de-
termined by the experimental characterization of the feed stream.  In
the magnetic beneficiation of coal, for example, such a characteriza-
tion requires the determination of particle size distribution, a stan-
dard float-and-sink separation, and measurements of the magnetic sus-
ceptibility along with the ash, sulfur and pyritic sulfur contents of
each separated fraction.  Specifically, suppose that x(W) and Ai(W)
(i=l,2, and 3) are the magnetic susceptibility, and the ash, sulfur,
and pyritic sulfur contents, respectively, of the pulverized feed coal
at a cumulative weight percent W; and U(R) is the cumulative weight
percent of particles having a radius smaller than R.  Under conditions
of constant magnetic and flow fields as well as fixed separator matrix
packing characteristics, R can be related to the magnetic velocity of
the feed stream Vm according to (1).  Thus, in order for particles to
have a magnetic velocity less than Vm, the radius of particles with a
magnetic susceptibility X should be smaller than
                              9 Vm n a      h
                        R = ( - )
                             2  y0MHo*

The magnetic velocity distribution of the feed stream, F(V ) , can then
be obtained by

                        100                 l
       F(Vm)  =   1    {   /  U{(9VmTiav  )*}   dW + Wo}      (5)
                100     ^o       2 y0 M H0X             u '

Here,  W  is  the  weight percent of particles  in  the feed stream having
negative  magnetic susceptibilities.  Note  that  implicitly included  in
 (5)  is the practical approximation  that  except  by mechanical  entrapment
or  filtration, no diamagnetic particles will be attracted magnetically
by  the wire  matrix.   Based on  (5) and  the measurements of ash, sulfur
and pyritic  sulfur  contents of the pulverized  feed coal, the  specific
magnetic  velocity distributions of ash,  sulfur  and pyritic  sulfur
 (i=l,2, and  3, respectively), denoted  by F.j_(Vm) , can be approximated
by:

                  ,     100   A,(w)           9 Vm  1 a   *
        F,(Vm) = J_ {  /    (  lV  ) U{  (— - ) }  dW +  A.  W  }
                 T±    W0      100        2 y0 MH0 X(W)          ±o °
                                                                (6)

In  (6), T±  (i=l,2,  and 3) are  the average  ash,  sulfur and pyritic  sul-
fur  contents, respectively, of the pulverized  feed coal; and  Aio
(i=l,2, and  3) are  the average ash, sulfur and  pyritic sulfur contents,
respectively, of the fraction WQ  with  negative  magnetic  susceptibili-
ties.
      (1)  -  (6) represent the new  mathematical model developed in the
authors'  laboratory for predicting  the technical performance  of  HGMS

-------
116
CLEAN COMBUSTION OF COAL
applied to coal beneficiation &>1^.  The model has been successfully
applied to predict satisfactorily the grade, recovery, concentration
breakthrough, etc., observed in pilot-scale experimental studies of
HGMS applied to the beneficiation of water slurries of several pul-
verized coals such as Illinois No. 6, and Pennsylvania Upper Freeport
and Lower Kittanning coals, etc.  For example, Figure 3 illustrates
the magnetic velocity distributions for coal, ash, sulfur and pyritic
sulfur for characterizing the pulverized Illinois No. 6 coal under
conditions of HQ = 107/2u ampere-turn/meter (20 kOe), M = 45 x 106 am-
pere-turn/meter (1.4 kOe) , y0 = 4ir x 107 henry/meter, a = 45 x 10~6 m
and n = 10~3 kg/m-sec.  These conditions correspond to the typical
separation conditions for the experimental results reported in Ref. 9.
            Figure  3.  Magnetic Velocity Distributions of
                       Pulverized Illinois No. 6 Coal

      Figure 4  shows the typical comparison between the theoretical
 and  experimental  concentration ratios, cou^/^±n> °f pulverized Illinois
 No.  6 coal particles in tail products obtained at a field intensity
 HQ = 20  kOe, a superficial slurry velocity Vo = 1.86 cm/sec, and dif-
 ferent total feed rates per unit cross section of a pilot-scale HGMS.
 The  comparison indicates  that concentration ratios predicted by
 the model are in a reasonable agreement with experimental results.   In
 Figure 5, a typical comparison between theoretical and experimental
 concentration breakthrough curves  is illustrated.  The curves in the
 figure are obtained by numerically integrating (3) based on the same
 feed  characteristics and separation conditions as in Figure 4.  It can
 be seen that there is also a reasonable agreement between theoretical
 and experimental concentration breakthrough curves.  Note that there
 is no  free parameter included in the new model given by (1) to (6) .
 Also,  the distributions of magnetic susceptibility, particle size and
 density along with ash, sulfur and pyritic sulfur contents of each sep-
 arated fraction of the pulverized feed coal after the standard float-
 and-sink separation can all be easily measured experimentally prior to

-------
                   HIGH GRADIENT MAGNETIC  SEPARATION
117
                     100
                                     Model Prediction:	

                                     Experimental Data: O

                                      VQ  1.86 cm/sec
H0 20 We
I 1 i i I I 1 1
0 40 80 120 160

20
                                TOTAL FEED ( g coal/cm2 )

    Figure 4.  Theoretical and Experimental Concentrations of  Coal
                Particles  in Tails as a  Function of Total Feed
                                         Model Prediction:  	

                                         Experimental Data: ago
                                           VB  1.86 on/sec
                                     FEED TOTAL SULFUR
                                   PRODUCT TOTAL SULFUR
                                 40     BO    120

                                   TOTAL FEED ( g coal/cm2 )
         Figure 5.  Theoretical and Experimental Ash and Sulfur
                    Concentrations of Coal  in Tails as a Funtion
                    of Total Feed

actual magnetic separation testing.  Thus, based on the above compar-
ison between model predictions and experimental results for Illinois
No. 6 coal,  and the successful application of the model to several other
Eastern  U.S. coals described elsewhere6»ij>li), it can be  suggested
that thesie. U now an e.x.peAvne-ntMy veAlfced,  &
-------
118                    CLEAN  COMBUSTION OF COAL
developed, it is possible to quantitatively determine  the  optimum sep-
aration conditions so as to optimize the magnetic removal  of  sulfur and
ash, while achieving an economically acceptable recovery of the  bene-
ficiated coal.  The new model can also be used to provide  the needed
information such as the proper separation duty cycle under the selected
process conditions for the engineering design and cost estimation of
wet HGMS processes applied to coal beneficiation  .  Further  discussion
on the practical applications of the modeling results  can  be  found in
the literature^.
     P-toce/6.6 PotuntiaJt and EconomJicA  .  By removing 75-90% of the
pyritic sulfur magnetically and achieving 85-90% recovery  as  was demon-
strated by the results of reported studies of magnetic beneficiation
of pulverized coals in water slurries 8-15 the process illustrated in
Figure 1 may be used to clean about one-fifth of the recoverable U.S.
coals with a low organic sulfur content of 0.7-0.9 wt % to produce an
environmentally acceptable fuel.  A detailed documentation of the re-
serve and production of U.S. coals which may be magnetically  cleanable
to 1 wt% total sulfur according to the seam, district, and  county in
each state, along with the total and organic sulfur contents  can be
found in the literature*°.  A technical survey of such magnetically
cleanable coals from Pennsylvania is currently being conducted in the
authors' group in cooperation with the Coal Preparation and Analysis
Laboratory of the U.S. Bureau of Mines.
     Here, a reasonable range of add-on costs (excluding those for
grinding, dewatering, drying and refuse disposal)  can be estimated for
the magnetic beneficiation of water slurries of pulverized coals  which
have desulfurization characteristics similar to those reported in the
recent studies °~l->.  -phe method used to estimate the costs of magnetic
desulfurization is based on the technique used by the Federal Power
Commission Synthetic Gas-Coal Force to estimate the cost of synthesis
gas*"'  .  The investor capitalization method used in this approach is
the discount cash flow (DCF) financing method with assumed DCF rates  of
return such as 15% after taxes.  This method essentially determines  the
annual revenue during the plant life which will generate a DCF equal  to
the total capital investment for the plant.  Several assumptions  are
included in the methodl9,20:
     (1)  The plant life is assumed to be 20 years with no cash  value
          at the end of life.
     (2)   A straight-line method is used to calculate the annual depre-
          ciation.
     (3)   Operating costs and working capital requirements are assumed
          to be constant during the construction period, and  100% equi-
          ty capital is assumed.
     (4)   Total plant investment, return on investment during the plant
          life, and working capital are treated as capital costs in
          year zero (the year ending with the completion of start-up
          operations).
     (5)   Start-up costs are treated as an expense in year zero.
     (6)   48% federal income tax is assumed.
Based on these assumptions, equations for calculating  the  unit costs
($ per ton of coal processed) are suggested by the published  docu-
ments-^ »20.  They are summarized in the Appendix, in which the de-
tailed operating conditions and estimated costs for a  typical example
shown in Figure 1 are also given.  The costs of major  installed  equip-
ment and the unit costs listed in the Appendix are based on values of

-------
                   HIGH GRADIENT MAGNETIC SEPARATION                 119
J-jne 1976.  For instance,  the  costs  of  pump  and  tank used were  estima^
ted first according  to Ref .  21  and then brought  up  to date by multiply-
ing by the CE plant  cost index ratio of (205/113.6),  while the  cost of
the installed HGMS unit with a separator matrix  of  7  ft  in diameter
and 20 in. long was estimated to be 1.936 million23.
     The estimated capital investment and  unit costs  for four typical
cases, designated as A-D,  are  summarized in  Table I.   Slurry veloci-
ties of 2.61 and 4.0 cm/sec, slurry  concentrations  of 15, 25 and 35
wt%, as well as separation duty cycles  of  59.0-77.9%  are considered.
These separation conditions  illustrate  clearly the  effects of slurry
velocity and concentration as  well as separation duty cycle.  For in-
stance, comparison of cases  A  to C shows that at the  same slurry veloc-
ity  and similar magnetic  desulfurization  characteristics, the higher
the slurry concentration,  the  cheaper will be the investment and unit
costs.  While this observation is to be expected, there  have been pilot-
scale tests which indicate that increasing the slurry concentration of
pulverized Illinois  No. 6  coal, from  2.57 to  28.4 wt%  did not appreci-
ably change the grade and  recovery of the  separation9.   Further effects
of processing conditions,  and  other  cost factors on the  unit costs, are
illustrated in Table II.   It is seen from  the table that by doubling
the amount of coal processed per cycle  relative  to a  fixed amount of
stainless steel wool packed  in  the separator matrix,  the unit cost can
be reduced by about  15%.   This  result shows  the  importance of the sep-
arator matrix loading characteristics on the costs of magnetic desul-
furization.  Another factor which affects  the unit costs considerably
is the washing time  required in a complete separation cycle.  This is
seen by comparing items 4  and  6 in Table II.  In particular, the com-
puted results indicated that doubling the  amount of wash water required
only leads to a negligible increase  (0.27  -  0.60%)  in unit costs.
However, if both the amount  of  wash  water  and the washing time are
doubled, the unit costs are  increased by about 15%.   The above obser-
vations clearly suggest an important economic incentive  for further
pilot-scale investigation  of the separator matrix loading and washing
characteristics in the magnetic beneficiation of coal/water slurry.
Finally, item 7 of Table II  shows that  labor cost seems  to be a signif-
icant fraction of the unit cost.   Fortunately,  it is not expected that
the labor requirement will be  doubled in actual  commercial practice
from the nominal case in Table  I.  This follows  because  existing ex-
perience in the commercial cleaning  of  kaolin clays by HGMS indicates
that the labor requirements  for both operation and maintenance are
     A comparison of approximate estimated capital and unit costs of
different pyritic sulfur removal processes currently under develop-
ment30'31 is given in Table III.  With the exception of the MAGNEX pro-
cess,30 all the processes listed in Table III are wet cleaning methods
and require grinding of feed coal, thus requiring relatively comparable
grinding, dewatering, and drying costs.  This table indicates that the
costs of coal beneficiation by HGMS are attractive when compared with
those of other approaches, even after adding the necessary costs of
grinding, dewatering and drying.  However, the above comparison is only
an approximate one, because of the difference in the methods used to
estimate the costs and in the desulfurization characteristics reported,
etc.   Based on the available cost information for these pyritic sulfur
removal processes, it is not yet possible to carry out a rigorous com-
parison.

-------
120
                                CLEAN COMBUSTION OF COAL
                    Table I.    Cost of Desulfurization of  Coal/Water  Slurry
                                 by  HGMS Using  Separator Matrix of  7-Ft
                                 Diameter  and 20-In Length
                                                   Case A
          Case B
Case C    Case D
1.
2.
3.
4.
5.
6.

7.

Slurry velocity (cm/sec)
Slurry concentration (wt %)
Coal feed rate (ton/hr)
Cycle time (min)
Duty cycle (%)
Tons of coal processed
per cycle
Unit costs ($ per ton
coal processed)
2.61
15
44.77
9.00
77.9

6.7


2.61
25
66.13
6.10
67.4

6.7


2.61
35
83.07
4.85
59.0

6.7


4.0
25
89.61
4.50
59.6

6.7


                    Mo
          8.
Capital investment per
ton  coal processed
per  year ($)
2.083      2.369

1.802      1.063

3.676      2.412
                                                    6.93
           4.69
 1.109     1.067

 0.858     0.829

 1.967     1.880
 3.73
3.53
                   Table II.  Sensitivity Analysis of Unit Costs ($ Per Ton Coal Processed)
                             of Desulfurization of Coal/Water Slurry by HGMS
                   1.  Basis: 2.61 cm/sec, 25 wt %
                      slurry, and other conditions  1.0628     0.00     2.4117    0.00
                      in the Appendix and Table I
                   2.  Amount of  coal processed
                      per cycle  doubled           0.9004   -15.23     2.0341  -15.66
                   3.  25% reduction in capital
                      investment                 0.9389   -11.28     2.0341  -19.12
                   4.  Amount of  wash water
                      required doubled            1.0691    +0.60     2.4181   +0.27
                      (washing time unchanged)
                   5.  Coast of water increased
                      5/3 times  (5C/1000 gal)      1.0835    +1.95     2.4324   +0.86
                   6.  Both amounts of wash
                      water and  washing time       1.2256   +15.32     2.7883  +15.62
                      doubled
                   7.  Labor requirement doubled    1.3587   +27.82     2.7077  +19.12

-------
                 HIGH GRADIENT MAGNETIC SEPARATION                  121
           Table III.  Comparison of Estimated Approximate Capital and
                     Unit Costs of Different Pyrite Removal Processes


               Process             U a    *    U *       Capital Investmentb
1.

2.

3.

4.

MAGNEX-Hazen ..
£Q
Research, Inc.
Froth flotation-
Bureau of Mines 24
Meyers-TRW
Systems and Energy 25
Ledgemont oxygen
leaching-Kennecott
5.83 7.05

2.77 4.47

6.0-14.0

comparable
to Meyers
4.17

5.71

13.80
(leaching only)
11.30
(leaching only)
       Copper Corp.2
    5.  HGMS-This work,            0.83-1.06   1.88-2.41        3.53-4.69
       see Table I

    °- Unit costs expressed in $ per ton coal processed.
    " Capital investment expressed in $ per ton coal processed per year.


MAGNETIC BENEFICIATION OF DRY PULVERIZED COAL

    The  application of HGMS to remove  sulfur and ash from dry pulver-
ized coal  has  just been initiated recently.  Much  of the  work reported
thus far has been limited to  investigating the technical feasibility
of magnetic beneficiation of  dry pulverized coal via either gravity
feeding  or air-entrained fluidization.  Although some degree of mag-
netic  removal  of sulfur and ash was  observed in using these approaches,
the experimental results reported had  not  yet achieved separation
performance  comparable to those observed  in magnetic beneficiation of
coal/water slurry.  In fact,  the weight percent of sulfur removed _from
dry pulverized coal was rarely more  than  10 to 20% even with multiple-
pass separation.  There was also no  definitive indication in the
existing reports10'12 as to why dry  beneficiation with either gravity
feeding  or air-entrained fluidization  was less effective compared to
wet  separation.  Quite recently, a joint  experimental program on  the
high  gradient magnetic beneficiation of  dry pulverized coal has been
initiated at Auburn University  and Oak Ridge National Laboratory.  Its
objectives are to develop a  novel recirculating air-fluidized separa-
tor  matrix for use  in HGMS  applied  to  dry pulverized coal, and  to com-
pare the performance of both dry and wet HGMS processes  for  coal  bene-

 studies are briefly discussed below.   '

-------
 122                     CLEAN COMBUSTION OF COAL
              Fe.zdU.ng Approach.,  The simplest way  for feeding dry pul-
verized coal to a magnetic separator is by  gravity.   Preliminary re-
sults on the desulfurization of dry pulverized  Indiana Nos,  5 and 6
coals by HGMS via gravity feeding reported  in 1976 are summarized in
Table IV.


               Table IV.  Magnetic Desulfurization of Dry  and Wet Pulverized
                       Indiana Coals via Gravity Feeding Approach^


                            Sulfur Removal, %          Sulfur Removal, %
                              No. 5 Coal                Ho. 6. Coal
                             Drya       Wet
1
2
3
a
b
. l-Passc
. 2-Pass<1
. 3-PassC
99% below 200 mesh,
99% below 200 mesh,
void, 30 wt% solids
9.
17.
18.
3
9
4
20 kOe,
20 kOe,
slurry.
18
33
35
.8
.0
.2
Frantz screens
stainless steel
13
18
20
wool
.7
.5
.6
packing ,
24
42
44
94%
.5
.4
.8

              retention time per pass:  30 seconds


 Also included in the table are their comparison with the  results  ob-
 tained by wet magnetic beneficiation.12  There was no explanation given
 in Ref.  12 as to why wet beneficiation was more effective in  sulfur re-
 moval compared to dry separation with gravity feeding as  shown in the
 table.  Some insights to answering the latter question  can be provided
 by the recent experimental results obtained in the authors' labora-
 tory.12  Thus, by studying the magnetic desulfurization characteristics
 of dry pulverized coal of different particle size ranges,  it was  found
 that the use of feed coal  of  small particle sizes  would often lead to
 the agglomeration of the pulverized coal and its mineral  impurities. In
 order to minimize the occurrence of fine coal agglomeration which
 could impede magnetic beneficiation, it would not be desirable to use
 the feed coal ground to, for example, less than 100 to  200 mesh.   This
 point may be illustrated by the experimental results given in Table V,
 in which the effect of particle size on magnetic desulfurization  of
 Pennsylvania Lower Kittanning coal via gravity feeding  is shown.^
 Essentially no sulfur was removed from the pulverized Lower Kittanning
 coal of  particle sizes between 100 and 200 mesh; and actual agglomer-
 ation of coal and its mineral impurities was visually observed from the
 separated samples.  In contrast, over 15% of the sulfur could be
 separated magnetically from the coal of particle sizes  between 60 and
 100 mesh.   A major problem in using a moderately ground coal  in dry
 separation, however, is that its sulfur-bearing and ash-forming miner-
 als may  not be sufficiently liberated; and there is a trade-off be-
 tween avoiding fine coal agglomeration and grinding to  liberate the
 mineral  impurities in coal.  In order to effectively apply HGMS for
 beneficiation of dry pulverized coal, it is necessary to  closely  exam-
 ine the  proper grinding level for a specific coal chosen  at given
 separation conditions, while taking into account the desired  libera-
 tion characteristics of its sulfur-bearing and ash-forming minerals.

-------
                   HIGH GRADIENT MAGNETIC SEPARATION                   123
              Table V.  Effect of Particle Size (Fine Coal Agglomeration) on Magnetic
                     Desulfurization of Dry Pulverized Lower Kittanning Coal,
                     Jefferson County, Pennsylvania via Gravity Feeding
                     Approach: 100-gm Feed Coal, 5-minute Feeding Time
                            Case A       Case B      Case C

1. Particle size, mesh -60 + 100a -100 + 200fa
2. Field intensity, kOe 10 10
-60 + 100a
20
-100 + 200b
20
         3.  Mags
weight, gm
sulfur, wt%
4. Sulfur removal, %
5.3
10.54
8.89
7.4
6.34
3.70
8.14
11.42
16.05
9.24
8.43
3.90
         a Feed coal contains 1-2% moisture, 4.05% total sulfur, 3.73% pyritic sulfur and 18.8%
           ash.
         ° Feed coal contains 1% moisture, 4.335% total sulfur, 3.354% pyritic sulfur and 13.70%
       The second  problem which  seems to contribute to the reported poor
performance of magnetic beneficiation of dry  pulverized coal  is the
limited separator matrix loading capacity in  using such packing mate-
rials  as stainless steel screen, expanded metal,  etc.  Because of the
possible excessive pressure drop, the low viscosity of the  carrier gas,
etc.  in dry separation, the latter packing materials are generally more
appropriate compared to stainless steel wool.  Unfortunately, the in-
sufficient sharp  edges and surface areas associated with these packing
materials limit their capabilities for producing  large field  gradient
and  for capturing magnetic particles. For example, an approximate  in-
dication of such  a limited matrix loading capacity in dry separation
with stainless steel screens is  illustrated in  Table VI.

                 Tatle VI.  Illustration of the Limited Matrix Loading Capacity
                         in Dry Magnetic Separation with Stainless Steel
                         Screens13
Pyrite , -60 + 200 mesh, 24 gm feed, 50 screens, gravity feeding
1.
2.


Field intensity,
kOe
Mags
weight , gm
wt%
0 5 15 20

3.8 6.3 7.0 8.6
15.7 26.3 29.3 35.7
For the  experiments  shown in the  table, fifty  Frantz screens of 1-in. I.D.
and 0.3125-in. thickness constructed from stainless steel strips were
stacked  together along with forty-nine spacers of 0.1875-in.  thickness
as separator matrix  packing materials.  By calculating the  active
surface  area of a screen and by assuming a monolayer particle capture
and buildup on both  sides of a screen, the maximum amount of  spherical
pyrite particles which can be collected as mags can be found  to be
about 8.0 gm for 60-mesh particles,  5.68 gm  for 100-mesh particles,
and 2.80 gm for 200-mesh particles.   These estimated amounts  are fairly
consistent with those collected experimentally as shown in  the table.

-------
 124                     CLEAN COMBUSTION OF  COAL
      Despite  the  problems of fine coal  agglomeration and  limited sep-
arator matrix  loading capacity in dry  separation via gravity feeding as
as illustrated in  the above, two important findings resulted from
the recent work with Pennsylvania Lower  Kittanning and  Upper Freeport
coals in  the authors' laboratoryM   First, it was found that the coal
feeding rate had essentially no effect on the performance  of magnetic
beneficiation  of dry pulverized coal via gravity feeding.  Secondly,
under comparable experimental conditions, the sulfur and ash contents
of the mags  resulted from dry separation via gravity feeding were found
to be almost  the same as those obtained  from wet beneficiation. This
point may be  illustrated by the approximate comparison  between dry and
wet magnetic beneficiation of Lower  Kittanning coal shown  in Table Vli.
                    Table VII.  An Approximate Comparison Between Dry and Wet Magneti
                           Beneficiation of Lower Kittanning Coal, Jefferson
                           County, Pennsylvania^

p
Grams Ash (Z)
6 g
Sulfur (%)
	 —*~
Pyrite (%)

            1. Dry, gravity feeding,       100  9.40   13.07  30.12   4.49  12.96   2.19  10.23
              5 minute feeding time,
              -100 + 200 mesh, stainless
              steel screens, 19 kOe
            2. Dry, air-entrained        100  5.29   14.68  30.60   4.42  12.96   2.59  12.92
              f luidization, upward flow,
              28.3 cm/sec, stainless steel
              screens, 19 kOe
            3. Wet, 2.5 vtZ solids slurry,    100  18.02   10.66  29.28   3.93  11.22   2.52   9.11
              2.3 cm/ sec, stainless steel
              wool packing, -200 + 325
              raesh, 19 kOe



 It can be seen from item nos.  1 and 3 of  the  table that the weight  per-
 cents of ash, sulfur  and pyrite in the mags resulted from dry separation
 via gravity feeding and wet beneficiation are of the same order  of
 magnitude, having  the ranges of 29.28-30.60,  11.22-12.96, and 9.11-
 12.92%, respectively.   The only difference between dry and wet benefi-
 ciation is the amount of mags collected due to their different separator
 matrix loading capacities. This difference has thus led to the generally
 poor performance  of dry separation via gravity feeding compared  to  wet
 beneficiation based on the ash, sulfur and pyrite contents of the sep-
 arated product  coal.   Further implications of this difference between
 dry and wet beneficiation  are discussed below along with some results
 obtained on the  magnetic  beneficiation of dry pulverized coal in a

 fluidized-bed  separator matrix.
       xL-.    ftuAdization Approach.   Prior to the work in the au-
 thors' laboratory}3 there was only  one reported study on  the magnetic
 beneficiation  of  dry pulverized coal via an air-entrained fluidization
 approach.10  In that study, a screw or vibratory feeder was used to inject
 dry pulverized Delmont coal  from Westmoreland County, Pennsylvania, into
 an air stream  which would carry it  doWW)OJid through the  separator matrix
 of 1-in.  I.D.  and 6 in. long.  The  preliminary tests with a screw feeder
 and an expanded metal matrix showed that the presence of  fines in the
 typically 60-mesh feed coal  would make the feeder  inoperative; and no
 significant  reduction in the sulfur and ash of the product coal was ob-
 served.   Subsequent tests conducted using  a vibratory feeder with  a
 field intensity up to 50 kOe and  air velocities of 59-1019 cm/sec indi-
 cated that the weight percent  of  sulfur removed from  the feed  coal was
 very small and irregular, rarely  more than 10%. In contrast, experiments

-------
                  HIGH GRADIENT MAGNETIC SEPARATION                 125
carried out with the same pulverized coal in a water  slurry  showed  that
HGMS was effective in reducing the total sulfur of  the  feed  coal by 50%,
the ash by 50%, and the pyrite by 60-80%.  No explanation regarding this
significant difference in the performance of dry and  wet beneficiation,
however, was given in Ref. 10.
     An extensive, pilot-scale experimental study of  magnetic beneficia-
tion of dry pulverized Pennsylvania Lower Kittanning  and Upper Freeport
coals via an air-entrained fluidization approach was  conducted recently
in the authors' laboratory, and the results have been described else-
where.    One major difference of this study compared to that described
in Ref. 10 was that an upuXVid. air-entrained fluidization of  the pulver-
ized feed coal through a separator matrix of 5-in.  I.D. and  20 in.  long
was examined.  The experiments revealed the same problems of fine coal
agglomeration and limited separator matrix loading  capacity  as observed
in dry separation via the gravity feeding approach.   Further, it was
found that under comparable experimental conditions,  the sulfur and ash
contents of the mags resulted from air-entrained fluidization were  of
the same orders of magnitude as those found from dry  separation via
gravity feeding or wet beneficiation.  This point has already been  il-
lustrated earlier in Table VII as item no. 2.  In addition,  the exper-
iments indicated that due to the high velocity of the air stream re-
quired to achieve a good entrained fluidization of  the  pulverized feed
coal, the retention time of magnetic particles in coal  in the separator
matrix was generally very short.  As a result, the  efficiency of the sep-
arator matrix in capturing and retaining magnetic particles  in coal was
very low.  For example, item no. 2 in Table VII shows that for 100  gm of
pulverized Lower Kittanning coal entrained in an air  stream  of velocity
28.3 cm/sec, the amount of mags collected in the separator matrix was
only 5.29 gm.  The latter was also less than that collected  in dry  sepa-
ration via gravity feeding or wet beneficiation as  shown in  the table.
     Recx>LCu£a£oi<2 AsiA-P&LU.dU.za£ion Approach..  Based  on the  preceding
discussion of  experimental results and their implications, it is evi-
dent  that an effective HGMS process to be developed for the  beneficia-
tion  of dry pulverized coal in a fluidized-bed separator matrix must
include at least two desirable features.  Thus, it  must have a simple
means  to reduce the presence of fines in the fluidized  coal  stream
and to avoid their possible agglomeration.  It must also provide a  su-
ficient retention time for the fluidized coal stream  to promote the
contact between the magnetic particles in coal and  the  active surface
areas available on the separator matrix packing materials.   The latter
is of importance in increasing the capacity of the  separator matrix for
capturing and  retaining magnetic particles in coal.   As a result of a
joint research program initiated at Auburn University and Oak Ridge
National Laboratory on magnetic beneficiation of dry  pulverized coal,
a recirculating air-fluidization approach possessing  the above desir-
able features has been developed recently; and several  novel fluidized-
bed separator matrices for use in HGMS applied to dry pulverized coal
are currently being tested experimentally.^ In fact, by using one  of
the recirculating fluidized-bed separator matrices  developed, the per-
formance of magnetic separation of sulfur and ash from  dry pulverized
Pennsylvania Upper Freeport coal was repeatedly found to be  be^tte/i  than
that from coal/water slurry.  This separator matrix was made of three
primary sections of increasing inside diameters arranged from the

-------
 126
                         CLEAN  COMBUSTION OP COAL
bottom to  the top of the matrix.   These primary  sections are of  5-in.  I.D.
and 5 in.  long (section A),  3.5-in.  I.D. and 10  in.  long (section  B) ,
and 0.75-in.  I.D. and 7 in.  long  (section C) ,  respectively.  Connecting
sections A and B, and sections B  and C are two expanded sections of^3  in.
and 4 in.  long, respectively.   In the experiments conducted with this
matrix, packing materials  such as stainless  steel screens were placed  in
section B, and the pulverized  feed coal placed in an auxiliary fluidized
bed outside of the separator was  air-fluidized through the separator.
By properly controlling the  flow  velocity of the air stream during the
whole separation period, it  was possible to  collect  most of the  fines
in the pulverized coal stream  as  top tail product of low sulfur  and ash
contents.  At the same time,  because of the unique expanded sections from
the bottom to the top of the separator matrix  and the resulting  gradual
decrease  in the upward fluidization velocity of  the  pulverized coal
stream,  the majority of the  pulverized coal  particles of medium  and large
sizes would tend to recirculate inside the expanded  sections.  As  a re-
sult, a  sufficient retention time inside the separator matrix could be
provided  to the bulk of the  fluidized coal stream without the presence
of fines,  thus allowing the  magnetic particles in coal to be captured
and  retained by  the matrix.  Toward the end of  the desired separation
period,  the flow velocity  of the  air stream  was  reduced, and the magnet-
ically  beneficiated coal of  low sulfur and ash content was collected as
bottom  tail products.
      Table VIII shows the typical experimental  results obtained with
Pennsylvania Upper Freeport  coal  of particle sizes between 100 and 200
mesh.

                Table VIII.  Magnetic  Beneficiation of Dry Pulverized Upper Feeport
                         Coal,  Jefferson County, Pennsylvania via Recirculating
                         Air Fluidization: Air Velocity - 17.7 cm/sec13
                          Grams
                                  Sulfur (%)
Pyrite (%)
Ash (%)
          1.  Feed         100.00      2.123

          2.  Mags
             (20 kOe)

              1-pass        9.05      12.80

              2-pass        5.99      4.68

              3-pass        5.12      2.25

              total         20.20      7.71

          3.  Total tails
             (20 kOe)       79.80      0.68

          4.  Feed separated
             as mags        20.20      68.16 a

          5.  Mags
             (0 kOe)        5.03      1.35
 1.519




11.79

 4.18

 1.93

 7.02


 0.130


86.80a


 0.85
 6.320




24.87

13.26

10.00

17.65


 3.45


52.43a


 4.66
          a Weight percent of sulfur, pyrite or ash separated from feed as mags.
 It can be  seen that by using  the recirculating  air-fluidization  approach
 and  three-pass separation,  HGMS was able to reduce the total sulfur of
 the  Upper  Freeport coal by  68.16%, the pyrite by 86.8%, and the  ash by

-------
                  HIGH GRADIENT MAGNETIC  SEPARATION                 127
52.43%.   Also, the total sulfur content of the tail product was suffi-
ciently low (0.68 wt%) that the magnetically beneficiated Upper Free-
port coal could be used immediately as an environmentally acceptable,
low sulfur fuel.
     The experimental results illustrated in Table VIII have clearly
suggested that HGMS with recirculating air-fluidization appears to hold
much promise as an effective physical method for cleaning coal. In par-
ticular, these results along with those obtained from wet beneficiation
described elsewhere13,15 have shown that the performance of magnetic
separation of sulfur and ash from dry pulverized coal via recirculating
air-fluidization can be even better than that from coal/water slurry.
Obviously, further research and development work related to the quanti-
tative modeling and prediction of separation performance, equipment and
process design, etc., in the magnetic beneficiation of dry pulverized
is justified.2'
APPENDIX:  BASIS FOR ESTIMATING THE UNIT COSTS OF MAGNETIC BENEFICIA-
TION OF COAL/WATER SLURRY

     The detailed operating conditions and estimated unit costs for a
typical conceptual process for the magnetic beneficiation of coal/
water slurry (case B in Table I) are illustrated as follows.

Operating Conditions

  (1)  Concentration of coal/water slurry = 25 wt%
  (2)  Superficial flow velocity = 2.61 cm/sec
  (3)  Stainless steel wool packing density = 6 wt%
  (4)  Amount of coal processed per cycle = 7 times weight of stainless
      steel wool
  (5)  Amount of washing water required per cycle = 3 times volume of
      separator matrix
  (6)  Amount of rinse water required per cycle =1.5 times volume of
      separator matrix
  (7)  Flow velocity of rinse water = flow velocity of coal slurry
  (8)  Washing time per cycle = 1 min
  (9)  Time of energizing the magnet per cycle =0.5 min
(10)  Labor required = 2 men per shift
(11)  Amount of dispersant = 10 ppm

Investment Costs

(1)  Costs of major installed equipment ($)
     one HGMS unit                                          $1,936,000
     pump                                                       38,480
     tank                                                       24.370
                                                             1,998,850
(2)  Add 20% Contingency                                       399.770
     Total Investment (!_•$)                                 $2,398,670

Operating Costs ($ per year)

(1)  Dispersant (57c/lb)                                    $   24,120
(2)  Electric power (2c/KW, 650KW)                             102,960

-------
128                    CLEAN COMBUSTION OF COAL
(3)  Water (30/1000 gal)                                   $   16,440
(4)  Operating labor
        (2 men/shift X 8304 man-hr/yr X 6.5$/man-hr)          107,960
(5)  Maintenance labor (1.5% of operating investment cost)     35,980
(6)  Supervision (15% of operating and maintenance labor
     costs)                                                    21,590
(7)  Operating supplied (30% of operating labor costs)         32,390
(8)  Maintenance supplied (1.5% of investment cost)            35,980
(9)  Local taxes and insurance (2.7% of investment cost)       64.760
     Annual Net Operating Cost N $                         $  442,180
     Coal Processed Annually Q tons                           528,900

Unit Costs ($ per ton coal processed)
     See Ref. 19 and 20 for the cost equations used below.

(1)  Based on 0% DCF rate of return:
     TJ0 =  (N + 0.05 JO/G = 1.063 $/ton
(2)  Based on 15% DCF rate of return:
     U15 = (N + 0.34749 I)/G = 2.412 $/ton
(3)  Based on capital amortization over 20 years at 1Q% interest
     rate:
     U = 
-------
                   HIGH GRADIENT MAGNETIC SEPARATION                 129
11.   Trindade, S.C., "Studies on the Magnetic Demineralization of
      Coal", Ph.D. Thesis, Department of Chemical Engineering, Massa-
      chusetts Institute of Technology, Cambridge, MA (1973).
12.   Murray, H.H., "High Intensity Magnetic Cleaning of Bituminous
      Coals", National Coal Association and Bituminous Coal Research,
      Inc., Coal Conference and Expo III, Lexington, KY (1976).
13.   Liu, Y.A., "A Feasibilty Study of High Gradient Magnetic Benefi-
      ciation of Coal in a Fluidized Bed", Progress Report, issued by
      Auburn University to Oak Ridge National Laboratory, Energy Re-
      search and Development Administration, under contract no.
      W-7450-eng-26 ORNL/sub-7315, September (1977).
14.   Lin, C.J., and Liu, Y.A. , "Desulfurization of Coals by High-Inten-
      sity, High-Gradient Magnetic Separation:   Conceptual Process De-
      sign and Cost Estimation", Paper presented at ACS National Meet-
      ing, New Orleans, LA, March, 1977; accepted for publication in
      ACS Symp. Ser., "Coal Desulfurization", December (1977).
15.   Liu, Y.A., Oak, M.J., and Lin, C.J., "Modeling and Experimental
      Study of High Gradient Magnetic Separation Applied to Coal Bene-
      ficiation", Symposium on novel separation technique, AIChE Annual
      Meeting, New York, NY, November  (1977).
16.   Liu, Y.A., Editor, "Proceedings of Magnetic Desulfurization of
      Coal Symposium:  A Symposium on the Theory and Applications of
      Magnetic Separation", IEEE Trans. Magn. MAG-12 (5), 423-551
      (1977).
17.   Watson, J.H.P., "Magnetic Filtration", J. App. Phys., 44. 4209
      (1973).
18.   Hoffman, L., "The Physical Desulfurization of Coal:  Major Con-
      sideration of S02  Emission  Control", Special Report, Mitre Corp.,
      McLean, VA, November (1970).
19.   Federal Power Commission, "Final Report:   The Supply-Technical
      Advisory Task Force on Synthetic Gas-Coal", April (1973).
20.   Batchelor,J.D., and Shih, C., "Solid-Liquid Separation in'Coal
      Liquefaction Processes", AIChE National Meeting, Los Angeles, CA,
      November (1975).
21.   Guthrie, D.M., "Capital Cost Estimating", Chem. Eng., 76. 114-
      142, March 24 (1969).
22.   lannicelli, J., "Assessment of High Extraction Magnetic Filtra-
      tion", Special Report to the National Science Foundation, avail-
      able as document No. Pb240-880/5 from the National Technical
      Information Service, Springfield, VA (1976)
23.   lannicelli, J., personal communication, Aquafine Corp., Bruns-
      wick, GA, October (1976).
24.   Kindig, J.K., Turner, R.L.,  "Dry Chemical Process to Magnetize
      Pyrite and Ash for Removal from Coal," Preprint No.  76-F-366,
      SME-AIME Fall Meeting,  Denver, September  (1976).
25.   Van Nice, L.J., Santy,  M.J., Meyers, R.A., "Meyers Process:
      Plant Design, Economics and Energy Balance", National Coal
      Association and Bituminuous Coal Research, Inc., Coal Conference
      and Expo, III, Lexington, KY, October (1976).
26.   Agarwal, J.C., Gilberti, R.A., Irminger,  P.F., Sareen, S.S.,
     "Chemical Desulfurization of Coal", Min. Congr. J., 70 (3),
      40-43 (1975).
27.   Liu, Y.A.,  and Lin, C.J., "Research Needs and Opportunities in
      High Gradient Magnetic Separation of Particulate-Gas Systems",

-------
13°                    CLEAN COMBUSTION OF COAL
     invited paper, to appear in Proceedings of National Science
     Foundation-Environmental Protection Agency Research Workshop on
     Novel Concepts, Methods and Advanced Technology in Particulate-
     Gas Separation, University of Notre Dame Press, Notre Dame,
     IN (1977).

-------
                                                                     131
              A THEORETICAL APPROACH TO WASHABILITY CURVES
                    IS COMPARED TO THE OTISCA PROCESS
                         SEPARATION OF PINE COAL

                                   by

                            D. V. Keller, Jr.
                         Otisca Industries, Ltd.
                   (On Leave Prom Syracuse University)
ABSTRACT

     An approach to the theoretical washability curve was developed
based on the variables of raw coal size distribution, ash and pyritic
sulfur concentration.  The size distribution of iron pyrite and mineral
matter in that coal seam is also required.  The analytical treatment
was compared to the washability curves developed independently in an-
other laboratory on a pulverized coal sample and on the data developed
from the separation of the same coal by the Otisca Process.  Although
numerous assumptions were required for a complete solution at various
gravities, the theoretical curve for pyritic sulfur was positioned
about 10 percent higher than the observed washability or Otisca Process
data.
INTRODUCTION

     Coal preparation technologists have long recognized that their
immediate task is to desulfurize and deash coal with a maximum yield of
coal product; and this task is complicated by extreme variations in
coal seam chemistry and morphology which include the coal itself as well
as iron pyrites, mineral matter, and moisture.  Furthermore, they also
face severe variations in the coal preparation procedure such as the de-
gree and techniques of size reduction, separation specific gravity and
procedure, and other significant factors.  Some generalities have
emerged from this complexity over the years.  For example, it is gener-
ally recognized that increasing size reduction permits the release of
more mineral matter and pyritic sulfur for separation.  Verification of\
this is made evident in numerous reports from U. S. Bureau of Mines*1"^"'
as is the observation that at a constant size distribution, a reduction
in separation specific gravity will produce a coal product also reduced
in ash and sulfur.  The extreme complexity arising from the presence of
these interdependent variables, together with the additional problem
that each coal seam is quite different and that there are also varia-
tions within one seam of coal from mine to mine, have presented a coal
preparation task which is usually solved by experience of the engineer
in charge or good old-fashioned intuition.

-------
 132                     CLEAN  COMBUSTION OF  COAL


     This approach has served the industry well for many years.  However,
with the demand for large quantities of cheap coal with an absolute min-
imum of sulfur and mineral matter we are now faced with the problem of
separating these variables in order to achieve the maximum possible
reduction of mineral matter.  The purpose of this paper is to examine a
theoretical approach to the determination of the ultimate limits of ash
and pyrite removal from coal, given the specific gravity of separation
together with the concentration and size distribution of the raw coal,
mineral matter, and pyrite.  Some of the experimental data used to test
this theoretical analysis were obtained from several other sources and
were accumulated for other purposes.  As a consequence, the precision
suffers.  However, it was presumed that with the establishment of this
theoretical approach, more accurate experimental data will be developed,
leading to an improvement in the correlation.  The application of this
approach to mineral matter reduction other than pyrite was also devel-
oped but could not be tested as the necessary size distribution data
were not available.
THE EFFECT OF PARTICLE SPECIFIC GRAVITY

     The following analysis is based on the rather straightforward as-
sumption that in the ideal gravity separation of a coal particle from
refuse, a very small mineral matter particle buried in the center of a
relatively massive coal particle will be recovered as coal product.
That is, a theoretical lower limit of retained ash and pyritic sulfur
in the coal product is calculated for an ideal separation of the raw
coal in an ideal bath.  In the simplest case, consider a two-phase sys-
tem of coal with an unique specific gravity of 1.30 and iron pyrite
with a specific gravity of 5.0 where the iron pyrite is embedded in the
coal

                                 TABLE I

             Variation Of Sulfur Concentration In An Ideal
                Coal~fi'onPyrite System As A Function Of
                     Volume Fraction Of Iron Pyrite

   Volume Fraction            Combined            Weight Percent
     Iron Pyrite          Specific Gravity        Pyritic Sulfur
        0.1                    1.6?                   15.99
        0.0?                   1.559                  11.49
        0.05                   1.485                   8.99
        0.04                   1.448                   7.38
        0.03                   l.ioi                   5.68
        0.02                   1.3?4                   3.89
        0.01                   1.337                   2.00

such that the apparent specific gravity of the two-phase particle in-
creases with the volume fraction of pyrite as would, the concentration
of the pyritic sulfur in the particle.  This relationship is important
because in an ideal gravity separation of coal and refuse, free pyrite
with a specific gravity of five will obviously separate as a reject
instantaneously; however, it is now evident that coal product with a

-------
                           WASHABILITY CURVES                         133


substantial amount of pyrite could also be recovered from the separation
bath with the coal product at gravities from 1.60 to 1.30.  One should
observe that there is no regard for particle size in this relationship
since only the volume fraction and specific gravity are involved.  This
relationship will be used later to control the effect of separation
bath specific gravity on the concentration of ash or iron pyrite in the
coal product.  A similar argument can be developed for the three-phase
system coal, mineral matter and iron pyrite if we assume an average
specific gravity for the coal and the mineral matter.
ANALYTICAL APPROACH

     The analysis used in this discussion is based on the assumption
that in a given narrow size range of raw coal, if that system were sep-
arated in an ideal parting liquid of a particular specific gravity,
the coal product floated from the bath will also include some fraction
of mineral matter and iron pyrite which was included in the coal parti-
cles as a three-phase mixture.

     The reason that these high density mineral matter particles were
included in the coal product was the fact that they had a particle
volume fraction small enough that the apparent specific gravity of the
particle was less than that of the separation bath, and that volume
fraction of mineral matter was embedded in the coal matrix.  The rest
of the mineral matter (including iron pyrite) is rejected due to the
fact that its specific gravity exceeds that of the separation bath.  It
should be recognized that this line of argument also implies that a
small volume fraction of coal is included in the mineral matter such
that the combined specific gravity never becomes less than that of the
bath.

     Consider, as an example, a raw coal crushed to 1 cm x 0 from a
coal seam which had a six inch (15 cm) slate parting.  If we examine
only that fraction of raw coal in the size range of 1 cm x 0.5 cm, we
can assume that all of the mineral matter from the mineral parting not
in the interfacial zone in this size range has a specific gravity which
exceeds the separation bath specific gravity of, say, 1.50; and thus,
will be rejected.  In fact, we can also assume that all mineral matter
particles larger than that raw coal size range will be rejected.  The
only recoverable mineral matter will be that which is small enough so
that its volume fraction does not cause the specific gravity of the
multiphase particle to exceed that of the separation bath.  The re-
quired information for this solution is the concentration of the raw
coal and mineral matter in the given size range and the bath specific
gravity, if we can assume an average specific gravity of coal and min-
eral matter.  After examining all size ranges and accumulating the re-
sults, we can theoretically ascertain limits of the coal product yield,
ash and pyritic sulfur concentration and heat content, if we are pro-
vided with the BTU/lb-ash relationship for that particular coal.  Fur-
thermore, we can also examine how these functions vary with various
methods and limits of raw coal size reduction and separation bath grav-
ity.

-------
  134                     CLEAN COMBUSTION OF  COAL
RAW COAL SIZE DISTRIBUTION

     According to Evans, et al.,  coal fractures into a particle size
distribution which can be represented by a Rosin-Rammler distribution
function of the form

                            R = e-b (x)t                     (1)

where R is the weight fraction of coal retained on a sieve of opening
size (x) in millimeters.  The constants b and t. are a function of the
fracture characteristics of the coal and fracturing technique.  For ex-
ample, different top sizes of raw coal prepared by different comminu-
tion techniques result in different slope t and intercept b values for
one particular coal when the data are presented on Log-Log R versus
Log x coordinates.  Mechanically crushed coals usually follow this rela-
tionship (linear) for raw coal top sizes well above one inch to sizes
well below 0.037/^m (400 mesh).  Chemically comminuted coal^6-', on the
other hand, shows a distinct break in the curve in the range of 595_//m
(28 mesh), giving two lines of different slope and intercept.  The
Rosin-Rammler distribution was accepted as a reasonable representation
of the raw coal size distribution in the following analysis.

     In order to examine a narrow size range distribution of raw coal,
say from 30 mesh to 60 mesh, the weight fraction of the raw coal within
this range with a particular set of parameters (b, t) is given by

                               (R! - R2)                     (2)

where R,, R? are the fractions retained and are given by equation 1.


MINERAL MATTER SIZE DISTRIBUTION

     To the author's knowledge, the particle size distribution of min-
eral matter in raw coal, disregarding iron pyrite, has not as yet been
quantified.  However, we are all aware that mineral matter occurs in
various seams not only as partings 6 to 18 inches (15 to 4-5 cm) thick
and larger, but also in thicknesses in the millimeter range as well as
individual sand and/or clay particles in the micrometer and submicro-
meter size range.  Let us presume for the moment that we can describe
the particle size distribution of mineral matter in the raw coal seam
using a Rosin-Rammler function described above, even though any other
suitable function representing that size distribution would also be
applicable as long as the key variables were fraction retained (R) on
a sieve of opening size (x).  The Rosin-Rammler parameters for this
mineral matter are taken as R1, b1, t', and x'.  Consider a. raw coal
sample from this particular seam which is crushed to some top size
where the raw coal size distribution is given by R, b, t, and x in
equation 1.  In a narrow size range, x + dx, the fraction of raw coal
in that range is given by (RI - R2) and we can assume that the average
particle diameter (x + ^r) aiso represents an average particle diameter
on the mineral matter size distribution curve.  All of the mineral mat-
ter larger than this average size can be regarded as having been re-
duced in size by the crushing procedure to reduce the raw coal sample

-------
                           WASHABILITY  CURVES                        135
to its size consist; and as such, has the specific gravity of mineral
matter, say, 2.2 as an average specific gravity.  A fraction of the
remainder of the mineral matter, i.e., smaller than the average particle
size, will be recovered with the coal product.  That fraction is deter-
mined by the volume fraction of mineral matter with an average specific
gravity of 2.2 which will just cause the coal product to float and be
recovered as coal product.  All of the mineral matter less than the
average particle diameter (x ) is given by
                            cL

                                1 - Ba                       (3)

where the fraction with diameters larger than xa is given by Ra which is
defined by V , t1, and xa (=x + =£).  The fraction of two-phase, ash-
coal particles which meet the specific gravity requirements is provided
by a factor (m); that is, equation 3 becomes

                       1- exp (-b'CmxJ*')                  (4)
                                      CL

where m is the ratio of the diameter of the mineral matter particle (da)
embedded in the coal matrix as a two-phase system to the diameter of the
whole system (dm).  The specific gravity of the mineral matter-coal com-
bined particle is given by (G) which is equivalent to the specific grav-
ity of the separation bath.  More specifically, the density of the two-
phase particle is given by the volume fraction of mineral matter /^a.\
                                                                  ^
times its average specific gravity (fa) plus the volume fraction of coal

(y~) times its average specific gravity (f^) such that

                     v  P + (-'  (v -v ) - G v                (5)
                      a  a   s c v m  a'      m               N^'

if we assume a two-component system where Va + Vo = Vm and Vra is the
volume of the two-phase particle.  Since we are examining the density of
any two-phase particle which will separate from the bath if its density
is greater than that of the bath and relating this to the apparent diam-
eter of the mineral matter (da) and the two-phase particle (dra) we can
then define (m) as                      r
                                        |G -

               m = da/dm


where G is the specific gravity of the separation bath.  The effect of
m on the exponential equation 4 is to cause more mineral matter to re-
port to the rejects as m becomes a smaller fraction, i.e., as the grav-
ity of the bath is increased, increasing m, more mineral matter will
report to the product coal as we all recognize from practical experience.

     The weight percent concentration of mineral matter (M) in the coal
product for any raw coal is then given by
                          n=o
where the first terra Gm is the concentration of mineral matter which is

-------
 136                    CLEM COMBUSTION OF COAL


approximately 1.1 times the weight percent ash concentration (dry basis)
in the raw coal(3).  The second term represents that fraction of raw
coal in a particular size range under examination and the third term
represents that fraction of mineral matter which can be recovered based
on the limitations of specific gravity.  The summation is taken over
all size ranges of the raw coal size consist such that M represents all
of the ash reporting to the coal product as a two-phase system.  In
achieving this end we have made (inferred) some rather significant as-
sumptions which, in fact, are not necessarily true but may not seriously
effect the results.  Firstly, one has to assume that the mineral matter
is homogeneously distributed which is probably true only for very small
sizes, I ^, less than 0,1 SOL, but is quite unlikely for particles greater
than 1 cm.  This suggests that the smallest error will be incurred in
the analysis of raw coals with a top size below 0.5 cm.  The specific
gravity of the pure coal was taken as an unique value which is, of
course, untrue due not only to the coal itself, but also due to the
mineral matter dispersed therein.  The latter problem was accounted for
in the analysis of iron pyrite.  However, there must be a cross term
which is complex due to the large number of exponential terms.  In fact,
an ultimate analysis should contain each component of mineral matter
with its respective size and density accounted for and this is clearly
a complex mathematical problem.


IRON PYRITE (PYRITIC SULFUR) SIZE DISTRIBUTION

     With the determination of the ideal mineral matter concentration
in the coal product, the concentration of the iron pyrite (specific
gravity (pp) of five), or the concentration pyritic sulfur (cp), can
be calculated following a similar procedure, recognizing that the
specific gravity of coal should be modified to account for the ash con-
tent.

     The relationship between the iron pyrite concentration, Cip, and
that of the pyritic sulfur (Cp) is

                            0.53^ Gip=  Gp                   (8)

and the density of coal with M weight percent mineral matter is given
by
                                   - c
                      ~ /k + 0.01 M (Pc-tO)


where * a is the average density of mineral matter (assumed 2.2) and Pc
is the average density of coal (assumed 1.30).

     Due to the lack of extensive data describing the size distribution
of iron pyrite in coal, a distribution function has not been estab-
lished; however, it was interesting to observe that data recently accu-
mulated could be roughly represented by a.Rosin-Rammler plot.  Figure 1
shows data points for the observed values^?' of the iron pyrite parti-
cles as plotted on Rosin-Rammler coordinates for the upper, middle and

-------
                          WASHABILITY CURVES
                                                          137
   5
   10

  20
  30
  40
  50
§70'
S  80
Lu
DC
oc
LU
90--


95--

97--

98--
   99
                                         LOWER  BENCH
                                         MIDDLE BENCH
                                         UPPER  BENCH
                                         AVERAGE CURVE
            345      10      20    40   60 80 100    200

                      PARTICLE  SIZE  mm  x I03
              Figure 1. Particle Size Distributions for Iron Pyrite

-------
 138                    CLEM COMBUSTION  OF  COAL
lower benches of one particular coal seam.  For a precise Rosin-Raramler
correlation, all of the points should lie on one straight line which is
evidently not the case for these particular pyrite particles.  The fol-
lowing analysis will incorporate an approximate Rosin-Rammler fit to the
olsserved data for mathematical simplicity.  The distribution used here
could be replaced with almost any other function which more closely
represents the data when it is developed as long as that function in-
volves the variables applicable to the Rosin-Rammler relationship.  The
error of the approximation used in this case should not be significant.
RETAINED FYRITIC SULFUR

     The concentration of pyritic sulfur retained (Sp) in a coal product
after separation of raw coal is given by
                                                                   , t*
        _ /     . / _n \         i / -.n^"j_ \ \ / •«       t jt '\cun/
        G (exp -b(a2 )  - exp -b(a2   ) ,) (1 -exp -b* i-r-(
 F  ~Q  P                                            (^

n = 0,1,2
                                                               (10)
where 0  is the concentration of pyritic sulfur in the raw coal system;
the second term represents the weight fraction retained between succes-
sive screen openings, with the size distribution having Rosin-Rammler
parameters b and £; and the last term represents that fraction of pyri-
tic sulfur which will be recovered with that coal.  The series estab-
lished by (a2n) allows a continuous examination of each size range of
coal beginning at some minimum value (a) in mm and progressing to the
coal top size limit where we must set (n+l) =ODas there is no coal
beyond this size.  The term | (2n+2n 1) in the third term establishes
the average size of the iron pyrite particle in the raw coal size range
under investigation.

     The iron pyrite size distribution is established with new distribu-
tion parameters b* and t* and m  based on the ash modified density of
coal (^k).  Due to the lack of an exact distribution function for the
pyrite under investigation, it was presumed that the dashed lines shown
in Figure 1 represented a Rosin-Rammler fit for the data, giving a solu-
tion which must be considered only a first approximation.
RESULTS

     A raw coal sample pulverized commercially to a nominal 200 mesh
      ) x 0 size consist from a coal mine in which the iron pyrite size
distribution was recently measured was used to test equation 10.  Raw
coal samples were carefully separated at different specific gravities
by the Otisca Process and also separated at a different laboratory by
means of the accepted washability procedures.  The complete sulfur form
analysis of all of the products of separation and the raw coal were con-
ducted by all parties as a cross-check.

-------
                           WASHABILITY CURVES                        139
     The observed size distribution of the iron pyrite in the three
benches of that coal seam were given in Figure 1.  In order to simplify
the mathematics, each curve was approximated by a straight line, or
Rosin-Rammler equivalent, to represent all of the observed data points.
The approximation is given by the dashed line in Figure 1.  The total
pyritic sulfur concentration in the raw coal was about 2 percent (dry
basis) which was distributed through the three benches in approximately
the following order t

                          BENCH
                  Percent             Mass
              Pyritic Sulfur        Fraction       Pyritic Sulfur (C-p)

Top Bench          3.02              0.339                1.022
Mid-Bench          1.81              0.39?                O.?l8
Bottom Bench       1.21              0.264                0.31?

Due to different size distributions and concentrations of pyritic sulfur
in the three benches, equation  10 was solved for each bench separately
and then the data accumulated to ascertain the total pyritic sulfur that
would be recovered.  The variables necessary for the solution of equation
10 for various separation gravities are shown in Table 2 where it was
assumed that the coal retained  8 percent ash as indicated by the wash-
ability data.  The raw coal size ranges tested, i.e., for n=o, n=l, n=2,
etc., when  (a=5xlO~-3mm) were 5/(m-10/'m, 10/'m-20/tm, 20An-40./iii, and
so on; and  there was no coal larger than 0.32 mm.

     Identical samples of the pulverized coal sample were separated by
an outside  laboratory experienced in the standard washability technique
and independently on a riffled  quarter of that sample of raw coal by the
Otisca Process at our laboratory.  The products of the Otisca Process
separation  were then sent to the former laboratory for complete proxi-
mate and sulfur forms analysis.  The results for the pyritic sulfur in
the raw coal cited before and those for the coal products separated at
various gravities are shown in  Figure 2.  A comparison of the results
from these  two tests for coal product yield, total sulfur and ash are
shown in Figures 3 and ^ respectively.
 DISCUSSION  OF RESDLTS

      The  experimental  separation data of the pulverized  coal at  various
 gravities shown in Figures 3 and b can be considered  quite  reliable due
 to the  fact that the products and raw coal were analyzed independently
 by several  different groups.  The agreement of the theoretical analysis
 to within 10 percent of the experimental values for the  retention of
 pyritic sulfur  in the  product coal was considered somewhat  fortuitous
 considering the large  number of assumptions that were required to com-
 plete the analysis and the fact that the analytical data necessary for
 the solution of the equations was received by chance  rather than by a
 directed  effort.  Clearly, the next step is to test this analytical
 procedure further for  other coals and coal seams in order to prove
 whether or  not  it is indeed a direction to a theoretical determination
 of the  washability curves.

-------
 140
                        CLEAN COMBUSTION OF COAL
                                 TABLE 2

            SOLUTIONS FOR THE DETERMINATION OF PYRITIG SULFUR
                        RETAINED IN GOAL PRODUCT
                                                     BENCH
                                         Upper
                                                      Mid
            Lower
Variables For Equation 10
b
t
b*
t*
CP
41
1.30
14.2
1.33
1.033
41
1.30
17.6
1.1
0.?18
41
1.30
11.2
0.72
0.317
Solutions For m=0.345 (1.50 Sp.Gr.) a=5xlO"^)

                                         0.0003      o.ooi       o.o°3
                                         0.002       0.0047      0.0089
                                         0.0079      0.0187      0.0259
                                         0.026       0.0505      0.0505
                                         0.0368      0.0567      0.0412
                                         0.0095      0.0112      0.0062
                                         0.0825      0.1428      0.1357
         3
         4
         5
     TOTAL
           Total Summation 0.361 Percent Pyritic Sulfur Recovery
             At 1.50 Specific Gravity
Solutions For m=0.300 (1.46 Sp.Gr.) a=5xlO"3
     n = o
         1
         2
         3
         4
         5
     TOTAL
                                         0.0
                                         0.001
                                         0.00?
                                         0.022
                                         0.031
                                         0.008
                                         "67559
0.001
0.004
0.016
0.044
0.050
0.011
0.126
           Total Summation 0.321 Percent Pyritic Sulfur Recovery
             At 1.45 Specific Gravity
Solutions For m=0.239 (1.40 Sp.Gr.) a=5xlO"3
     n = o
         1
         2
         3
       •  4
         5
     TOTAL.
                                         0.0001
                                         0.001
                                         0.005
                                         0.016
                                         0.024
                                         0.006
                                         0.052
0.0006
0.003
0.013
0.035
0.041
0.009
0.107
           Total Summation 0.272 Percent Pyritic Sulfur Recovery
             At 1.40 Specific Gravity
0.002
0.007
0.021
0.042
0.035
0.006
0.113

-------
                        WASHABILITY CURVES
                                                          141
(ft
or
>-
a.

Q
UJ
UJ

QC
UJ
o
cc
LU
CL
    0,40
2  0.30
0.20
                                        THEORETICAL

                                        EQUATION  10
                                           INDEPENDENT

                                           WASHABILITY
             OTISCA  PROCESS
0.15-
1.50
i i
1.40
\
V

1
1.30
                  SEPARATION  SPECIFIC  GRAVITY  -
        Figure 2.  Percent Retained Pyritic Sulfur as a Function

                 of Separation Gravity for 200 Mesh X 0 Coal

                 and Observed Data

-------
        142
   2.0 -
Q




CL
CL

D
ID

CO
O
OL
LJJ

£L
    0.8
       1.30
          CLEAN COMBUSTION OF COAL
                                    PULVERIZED COAL

                                       200m x 0
                                  x OTISCA  PROCESS


                                  • WASHABILITY
1.40                1.60               1.80



 SEPARATION  SPECIFIC  GRAVITY



   Figure 3.  Washability Results for Pulverized Coal
                                                                '•80
                                                                •-70
                                                                --60
                                               --50
                                                 40  u
                                                     D
                                                     O

                                                     s
                                                     Q.
                                                                --30
                                                     o
                                                     u
                                                                --20
                                                                •• 10

-------
                         WASHABILITY CURVES
                                                      143
    14 +
    12 +
D
O

2
0.
O
u
10 I
^   8 +
x
CO
26 +
LU

O

QC
LU

CL
    4 +
                               WASHABILITY
                   OTISCA  PROCESS
                                         PULVERIZED COAL

                                             200m x 0
                                                      +
 1.30      1.40                1.60               1.80

              SEPARATION  SPECIFIC GRAVITY
                                                               1.90
          Figure 4.  Observed Washabllity Results for Pulverized Coal

-------
 144                    CLEAN COMBUSTION OF COAL
     Since there is virtually no mineral matter size distributions
available, let us turn immediately to the pyritic sulfur analysis cited.
If we assume that current coal preparation techniques have little effect
on the organic sulfur in coal and the concentration of sulfate sulfur
in eastern coals is small enough to be inconsequential, then the key to
desulfurization of eastern coals lies in our ability to eliminate iron
pyrite from the raw coal matrix.  The variables required to express a
quantitative analysis of iron pyrite extraction from raw coal include
raw coal size distribution, pyrite size distribution and the specific
gravity of the separation bath.  Only the pyrite size distribution is
outside the control of the coal preparation engineer unless he has a
choice of coal seams that have significant differences in pyrite size
distribution from which he can extract raw coal.  In this case, the
choice, if the size distribution is Rosin-Rammler would be that seam
with the smallest positive slope (t*) and the smallest intercept value
(b*).  This would present the greatest release of pyrite with the least
amount of raw coal size reduction.

     Whether or not various techniques of raw coal size reduction, i.e.,
Hardinge mill versus hammer mill versus other impact methods or even
chemical techniques, promote greater iron pyrite release in a given
situation, i.e., fixed iron pyrite distribution, has not been demon-
strated unequivocally.  Quantitatively what is implied is that for a
given iron pyrite size distribution, a given raw coal particle size
distribution generated by the two different comminution techniques and
a given bath separation specific gravity, one form of comminution might
produce coal product with less retained iron pyrite.  Such might well be
the case, if one of our initial assumptions, which was purely arbitrary,
was incorrect and that the very small iron pyrite was not homogeneously
dispersed throughout the coal system, but lay on preferred sites which
during fracture were preferentially freed from the coal particle.
Clearly, pyrite embedded within the coal particle will remain there al-
ways.  Only very careful experimentation with strict control over the
variables will provide the final answers.

     The analytical expression given in equation 10 is only a first ap-
proximation as the effect of certain variables known to exist in the raw
coal system have not been considered in their entirety and their effect
can only be surmised at this time.  Firstly, it was assumed in the util-
ization of the term (m) which accounted for the variation in bath spe-
cific gravity that coal was of only one specific gravity (1.30), and, of
course, this is just not true.  If, however, this value were an appro-
priate average specific gravity for all of the coal in that particular
seam, one might suspect that the coal of lower specific gravity and its
recovery of iron pyrite would just about cancel the lost iron pyrite
with the coal fraction of higher specific gravity.  The average value
of the specific gravity of the coal might be obtained from a standard
experimental washability curve which was developed to include the calo-
rific value of that particular coal.  After accounting for the BTU
units in the pure rejects, i.e., 1.60 gravity or higher, then find that
gravity where there are equivalent coal heat units in the rejects and
products.

     The mathematical simplification of first separating the mineral

-------
                           WASHABILITY CURVES                        145
matter and then the pyritic sulfur neglected the fact that in the real
case a three-phase system is involved, i.e., the limiting specific
gravity (m) for mineral matter retention should also include the effect
of the^retained iron pyrite.  Although this correction is rather small,
i.e., Tk *" 1»35Q versus a correct value of Cfc« = 1.353t it would tend
to depress the theoretical values shown in Figure 2 into the range of
the observed data.  Most encouraging, however, is the general slope
change of the theoretical curve with the variation of specific gravity
and its correlation with the experimental curves.  That is, the retained
pyrite versus specific gravity curve tends towards negative infinity as
the separation bath specific gravity approaches the specific gravity of
the coal, cf. equation 6.

     The development of a washability curve based on the data determined
from the above equations is relatively straightforward, that is, the ash
in the coal product versus separation specific gravity curve is obtained
from solutions for 0.9 M in equation 7i as the G value in equation 6 is
varied.  The total sulfur in the coal product is determined through
equation 10 which provides the pyritic sulfur content plus the sulfur
contributions due to organic sulfur as most sulfate sulfur is lost in
the rejects.  Again equation 10 is varied over the entire specific grav-
ity ranges.

     Since each of the above calculations represents that fraction of
the mineral matter and pyrite recovered in the coal product, the residue
of the concentration in the raw coal identifies that quantity of mater-
ial rejected in the separation and that difference from one represents
the coal recovered, i.e., yield, in each separation specific gravity
case.  A theoretical washability curve has been generated provided our
assumptions are not too severe.

     The difference between the standard washability curve and the
Otisca Process data was due to the different procedures that were used.
The standard washability tests used a standard commercial solvent system
with a separation time of about twenty-four hours.  The Otisca Process
used a proprietary liquid and an additive package which allowed a sep-
aration time of not more than twenty minutes for size consists of less
than  100 mesh (0.1 mm) x 0.  Figures 3 and k represent the first head-
on tests using the Otisca Process and standard washability on a common
sample.  The results were rather encouraging.
CONCLUSIONS

     A first approximation for the theoretical determination of wash-
ability curves was presented and partially tested using a nominal 200
mesh x 0 raw coal separated by classical washability techniques and the
Otisca Process.  The results indicated that the pyritic sulfur retained
in the product coal could be determined within 10 percent for raw coal
separations in an ideal bath at specific gravities of 1.50t 1.^5. and
1.40.  The key variables required for this analysis include for the raw
coal the size distribution of the iron pyrite, mineral matter, and the
crushed raw coal and the concentration of the ash and iron pyrite.  The
specific gravity of the coal, mineral matter, iron pyrite, and separation

-------
 146                     CLEM COMBUSTION  OP  COAL
bath are also required.  One of the key utilizations of this analytical
approach is that for a given seam, i.e., fixed mineral matter and iron
pyrite size distribution, the optimum comminution schedule can be de-
termined to permit the maximum reduction of ash and pyritic sulfur in
the product coal.  Furthermore, once the theoretical limits for this
seam are available, then one has a fixed reference point for comparison
with practical coal preparation procedures.
ACKNOWLEDGMENT

     The author would like to acknowledge the valuable comments and in-
sights provided by Dr. Andrew Rainis during his review of this paper.
REFERENCES

1.  Cavallaro, J.A., Johnson, M.T., and Deurbrouck, A.W., "Sulfur Re-
    duction Potential of the Coals of the United States", U. S. Bureau
    of Mines, Report of Investigation 8118, 1976.

2.  Deurbrouck, A.W., "Sulfur Reduction Potential of the Coals of the
    United States", U.S. Bureau of Mines, Report of Investigation 7633,
    1972.

3.  Lowry, H.H., «i, "Chemistry of Coal Utilization", John Wiley and
    Sons, Inc., New York, Chapter 8, 1963.

Ij-.  Leonard, J.W.,  Mitchell, D.R. ed, "Coal Preparation", Amer. Inst. of
    Mining, Met. and Pet. Eng., Inc., Chapter**, 1968.

5.  Evans, I. and Pomeroy, C.D., "Strength, Fracture and Workability of
    Coal", Pergamon Press, New York, Chapter 7_» 1966.

6.  Keller, Jr., D.V. and Smith, C.D., "Spontaneous Fracture of Coal",
    Fuel, &, 273-80, 1976.

7.  Private Communication.

-------
                                                                    147
                  SESSION III - COMBUSTION TECHNOLOGY

             SESSION CHAIRMAN:   G.  BLAIR MARTIN, U.S.  EPA
     In achieving the projected increased use of coal in the industrial
and utility sectors, a major emphasis must be placed on the environmental
impacts of the combustion systems.   The pollutant species to "be consid-
ered include not only the criteria pollutants (SOX, NOX, CO and particu-
late) but also trace species (POM,  organics and trace metals).   To control
emissions in the most energy-efficient and cost-effective manner,  the
combustion system should be considered from an overall point of view and
may include provisions for precleaning of the fuel, proper design  of the
combustion device and postcleaning of the flue gas.  The combustor design
is determined by the choice of process for control of SOX emissions.  The
sulfur may be removed from the fuel (coal cleaning or gasification),
removed during the fuel combustion process (fluid bed combustion), or
scrubbed from the flue gas (pulverized coal boiler).

     Coal cleaning and flue gas treatment are covered in other  sections
of the report; therefore, this session concentrates on the combustion
processes or equipment that may be used in conjunction with the various
SOX and particulate removal schemes.  The pollutants that may be con-
trolled by proper design of the combustion process include NOX, CO, POM,
organics and carbon particulate.  Modification of the combustion process
has significant potential for NOX control while maintaining low levels
of the other pollutants.  Since the currently available technology is
the conventional pulverized or stoker coal-fired boiler with low sulfur
fuel or flue gas desulfurization, emphasis is placed on this type  of
equipment.  Some potential near-term technologies are also being devel-
oped and two of these are also represented.

     The overall session is split up into three types of papers.  There
are two papers covering general aspects of combustion technology:
(l) the basic design and operating principles of combustion systems,
and (2) the basic aspects of pollutant formation and control in the fuel
combustion process.  There are three papers dealing with current tech-
nologies for combustion of coal:  (l) low emission burner design for
pulverized coal-fired utility boilers, (2) coal oil mixture combustion
in industrial and utility boilers,  and (3) stoker coal-fired industrial
boilers.  Finally, there are two presentations on potential near-term
alternatives:  (l) fluid bed combustion of coal for industrial  steam
generation, and (2) pulverized coal burners for industrial process
furnaces.

-------
148                    CLEAN COMBUSTION OF COAL

-------
                                                                    149
              SOME CHARACTERISTICS OF COAL COMBUSTION SYSTEMS
                             Janos M.  Beer
                  Massachusetts Institute of Technology
                        Cambridge, Massachusetts
INTRODUCTION

     Coal is burned in a large number of industrial processes:   in
furnaces, kilns and boilers.   The recent development of coal  combustion
systems is, however, closely tied to the development of steam raising
plants.  Stoker firing has reached a high level of sophistication but
could not follow the steady increases in unit capacity demanded by the
economy of scale in power station development  and gave way to pulver-
ized coal combustion which in its various forms is the presently pre-
vailing mode of industrial coal combustion.   As a result of environmen-
tal concerns and also due to the plans for significant expansion in the
use of coal, interest turned toward a new combustion system,  the fluid-
ized combustion of coal, which holds out promise for sulphur  retention
without flue gas desulfurization, and low NOX emission at low capital
and operating costs.  The scope of this paper does not permit  detailed
discussion of these processes; it is intended instead to draw attention
to some characteristics of these combustion systems, seen through the
physical-chemical processes that coal undergoes during combustion.

     In coal combustion processes the engineering task is to  insure

     •  stable ignition of the coal

     •  complete combustion of both the volatiles and the residue of
        char

     •  low pollutant emission

     •  good availability of plant (free of excessive deposit formation
        and corrosion)

     •  all these achieved with a minimum of excess air at an acceptable
        cost in pressure energy.

It is of special advantage if a combustion system can be easily inte-
grated into energy conversion cycles of high thermodynamic efficiency.
In the following we shall look at combustion systems from the point of
view of how far they satisfy the above criteria.

-------
150                    CLEAN COMBUSTION OF COAL
PYROLYSIS, IGNITION AND COMBUSTION OF COAL

     When coal is heated, the moisture is expelled first, followed by
the thermal decomposition of the coal.  This latter process is usually
called devolatilization.  The amount of the volatile matter driven off
depends upon the type of coal and also upon the rate of heating and the
final temperature of pyrolysis.  The higher the final temperature and
the rate at which the coal is heated, the larger the proportion of the
mass lost during devolatilization.  At sufficiently high temperature
the total solid combustible mass of the coal can be volatilized.   The
process of volatile evolution can be quantitatively described in terms
of a large number of simultaneous irreversible, first-order reactions .

     Ignition usually occurs during the thermal decomposition of the
coal.  It is likely that some of the lower ignition-temperature high
molecular weight hydrocarbons ignite first.  Their ignition may be
catalyzed by the solid surface which may also be ignited during this
process2.  In large particles, where the volatile evolution and burning
goes on concurrently with the combustion of the solid matter, it  is
difficult to separate these processes in time.  (This is reflected in
the need to provide sufficient space for volatile-combustion in stoker-
fired combustors almost above the whole length of the grate.)

     The combustion process of the residual char can be considered to
consist of steps such as (a) the diffusion of the oxidant to the particle
surface in counter-diffusion with the products of combustion away from
the surface, (b) the diffusion into the pores of the particle, and
(c) the chemisorption reaction at the surface, consisting of the acti-
vated adsorption of the oxidant and the desorption of the compounds
formed at the surface.  The relevance of these individual steps to the
overall rate of burning of the particle depends upon parameters such as
particle size and reactivity of the char, temperature, and the nature
of the oxidant (COg, H20, OH or 02).  For example, when the particles
are large and the temperature is high (stoker firing), the rate deter-
mining step is external diffusion; when the particles are large and the
temperature is low, diffusion and/or chemical reaction can be the rate
limiting steps (fluidized combustion); and when the particles are small
and the temperature is high (pulverized coal combustion), diffusion and/
or chemical reaction at the surface determine the overall rate of the
oxidation reaction .

STOKER FIRING

Ignition on the Traveling Grate

     The top layer of coal on the grate is heated by radiation from the
flame, and by convection from hot combustion products.  The combustion
air preheat temperature is usually low (<200°C) in order that the air
can cool the grate.  When coal is ignited from below by air preheated
to higher temperatures, high relative velocities between coal and gas
will reduce ignition delay.  Figure 1 illustrates how ignition delay-
times depend upon the air velocity for the case of ignition from the top
by radiation and/or from below by air preheated to H25°C; it can be seen
that there is an "ignition gap" in the range of gas velocities of 0.11-
0.27 Nm/s in which neither of the above-mentioned ignition modes can

-------
                         COAL COMBUSTION SYSTEMS
                                                                     151
produce ignition by themselves.  Ignition can, however, be obtained by
the combination of the two methods .
     Figure 2 represents an optimized form of the refractory arches
determined by modeling radiative transfer between the burning coal on
the grate and a refractory arch on one hand, and between this refractory
arch and the green coal on the other5.  Care had to be taken to ensure
that the ignition arches serve their purpose without causing slagging
at the high temperature end of the grate.

     Another engineering solution to improve the ignition of a high
moisture coal is shown in Figure 36.  The lower part of the coal hopper
is altered to enable hot combustion products to be drawn through the
coal layer before it reaches the grate.  In this arrangement only the
fixed grate, at the bottom of the hopper, has to stand up to high gas
temperatures and the beneficial effect of combined ignition can be
obtained without the flow of highly preheated air through the traveling
grate.
     20
   min
     is

     K

     ft
                IGNITION
   C3
From Above
  1170
                       From Below
                       Coal Layer
                  Ignition
              — -- First Inflam.
     0       
-------
152
CLEM COMBUSTION OF COAL
is  shown for three temperatures7.  While the  combustion rates  can  be
seen to level off at low air velocities (laminar flow), they are increas-
ing again,  once the flow becomes turbulent  at higher air velocities.
Another significant finding of these researchers is that the composition
of  the combustion products shifts toward higher CO/C02 ratios  as the
blast velocity increases.  This has been confirmed to hold also for a
fixed bed of particles8 as shown in Figure  5-

     On the traveling  grate steady state conditions prevail:   the  events
that happen sequentially on the fixed grate can be seen spatially  sepa-
rated, but  occurring concurrently on the traveling grate.  Figure  6
illustrates this point:  the composition of gaseous species above  a
traveling grate are shown on the top, and their respective volume  flow
rates through the burning layer of coal, on the lower part of  Figure 69.

     The operation of  practical fuel beds is  hampered by nonuniform
pressure drop across the bed due mainly to  the nonuniform distribution
of  the fine particles  on the grate.  Fine particles tend to segregate
during their passage through storage bunkers  and in the hopper above the
grate and will cover the grate in patches.
                                           25

                                           2.0

                                           1.5

                                           1.0

                                           0.5
                          \fltXf
                                        X
                                                10   20   30   40   50  60 l/min
                                         Fig. U.  Variation of Combustion
                                                  Rate of a Carbon Channel
                                                  with Air Velocity (After
                                                  Tsukhanova7).
Fig. 3.  Ignition by Hot Recirculated
         Gas.
     Figure 7 represents results of experimental studies10 showing the
effect of the proportion of fines (<5mm) in the feed upon boiler per-
formance and efficiency.

-------
                          COAL COMBUSTION SYSTEMS
                         153
              10  XI  JO  IO  10
            Distance into layer, mm
                          II
                           t*
                           ISOO
         u " t  JJfojo  ta40
            Distance into layer, mm
         **r        	"I
           e  a  10  la  to so
            Distance Into layer, nun
                                                 m
Fig. 5-   Gas Formation in  a Layer   Fig. 6.   Species Concentration and
         of Particles of Electrode
         Carbon: I) Pate of  Blast
         0.11 m/sec; II) Rate of
         Blast O.U9 m/sec; III)  Rate
         of Blast 1.50 m/sec (After
         Khitrin8).
Flow Rate  Distribution Above
a Travelling Grate (After
Werkmeister ).
                          u
                           SO       (O       TO       tto
                               fCRCCNTMC MCATER 7 HAN 5mm.
               Fig. 7-  Effect  on Efficiency  and Performance
                        of Removing Fines from a Fuel Bed
                        (After  Zagon in Ref.  6).

-------
154
              CLEAN COMBUSTION OF COAL
      One of the engineering solutions of this problem was the develop-
ment of spreader stokers  (Figure 8):  the coal, unclassified, is thrown
or blown into the combustion chamber and due to the  inertial and drag
forces acting o'n the particles in flight, the smaller particles land on
top of the large particles  on the grate.  The finest particles burn in
suspension above the grate.   This in turn makes it necessary to raise
the height of the combustion chamber and then to  cool walls by screens
of steam generating tubes to avoid slagging of the refractory surfaces
by the fly-ash.  Because  of the high dust loading of the flue gas,
spreader stokers normally require mechanical particle precipitators.

      Another solution of  the problem caused by fine  particles, a combi-
nation of coal classification, stoker and pulverized coal combustion
is shown schematically  in Figure 911-   The coal is fed into a flash
Fig.  8.
Traveling Grate Stoker.   Section Through Rotograte Stoker In-
                       stallation.
  1.—Raw Ceil.
  2.—Feeder.
  3.—Classifier Tube.
  4.—Fine Coal.
  5.—Coarse Coal.
  6.—-Vessel.
  7.—Rotary Seal.
  8.—Hopper.
  9.—Combustion Chamber.
 10.—Bypass.
 11.—Transport Duet.
 12.—Pulverizer and Fan.
 13.—P.F. Duet.
 14.—Branch tor Temp. Control.
  Fig.  9-   Diagrammatic Arrangement  of the Combined Firing System.
                                 (Beer11)

-------
                        COAL COMBUSTION SYSTEMS                     155
drier tube and drops in counter-flow with hot combustion products drawn
from the combustion chamber.  The fines are lifted from the surface of
large particles, carried by the gas stream into a pulverizer, and blown
through burners into the combustion chamber above the grate.  Due to
the interaction of the combustion of classified coal on the grate and
the pulverized coal flame above the grate, high combustion efficiencies
with low excess air can be obtained even when burning low grade coal.
When existing stoker-fired boilers are fitted with this system, the
boiler performance can also be significantly raised11.

PULVERIZED COAL COMBUSTION

     As the unit capacity of boilers rose significantly, during the
period between the two World Wars, and there was a requirement for
improved efficiency and better automatic control, pulverized combustion
of coal has steadily replaced stoker firing in units larger than
250,000 Ib/hr steam, and therefore in utility application almost entirely.

     The combustion efficiency in pulverized coal combustion is high
because particles dried and ground to below 200 ym can be intimately
mixed with the combustion air, and burned completely with a minimum of
excess air.  The combustion air can be preheated to high temperatures
unlike in stokers where the combustion air has to cool the grate.  The
use of higher air-preheat means that the flue gas leaving the boiler
can be cooled by the much larger air preheater and hence advantage can
be taken of regenerative feed water preheat with steam bled from the
turbine without any deterioration in boiler efficiency.

     Due to the relatively short residence time of the burning coal in
the combustion chamber (<2 sec) load-following is only limited by the
rate at which the coal grinding-feeding system can respond to command.
This varies with different coal preparation systems but the response is
generally fast, the time constant being about one minute.  The fast
response characteristics are compatible with requirements for flexible
automatic control and good load following of the boiler plant.

     Last but not least, the nature of the flame:  capable of filling
the combustion chamber with a strongly and uniformly radiating medium
made it possible to adapt to this combustion system the water walled,
"radiant" combustion chamber design, and this has ensured the scaling
up of pulverized coal combustion to the sizes of the largest utility
units (>1000 MWe).

Ignition of Pulverized Coal

     When a cloud of pulverized coal particles is injected into a
furnace, it is heated up as it approaches the flame front partly by
radiation and partly by mixing with hot recirculated combustion products.

     Nusselt12in a classifical paper developed the theory of unidimen-
sional laminar flame propagation in a coal dust cloud and solved his
equations for the flame speed as a function of particle  size, dust
loading of the gas and temperature of the flame, assuming solely
radiative heat transfer.  Essenhigh and Csaba13 have used Nusselt's

-------
156                    CLEM COMBUSTION OF COAL
theory for the interpretation of Csaba's experiments.  Figure 10 shows
results calculated from their study for "ignition temperatures" of
650°K and 900°K, respectively.  The prediction is that particles of
30 um diameter will have ignition delays of about 100 ms and flame
speeds lower than 1 m/s for the above ignition temperatures of the dust
cloud and assuming a maximum radiance of the flame at 100 kcal/m^sec.

     Rates of flame propagation in practical pulverized coal flames are,
however, about an order of magnitude higher than those shown in
Figure 10, due mainly to the effect of hot gas entrainment into the
fuel-jet as it approaches the flame front (Figure ISa)1"*.  The respective
contributions of radiation and hot gas entrainment to the ignition of a
pulverized anthracite as a function of distance from the burner and for
different coal finenesses are shown in Figure 12, and ignition distances
for the same anthracites and for two burner types in Figure II15.

     The heat required for raising the combustion air temperature to
ignition is several times that necessary for heating up the coal parti-
cles in the coal-air mixture.  This is one of the reasons for injecting
the coal dust with only a fraction of the total combustion air, the so-
called primary air.  The rest of the combustion air is normally mixed
in downstream of the flame front.

     The primary air fraction is usually proportional to the volatile
matter of the coal so that sufficient air is available for the combus-
tion of the volatiles as they evolve.  Figure 13a and b show the effects
of the primary air fraction and the volatile and ash content of the coal
upon the flame propagation rate.

     Results of Badzioch's experiments on isothermal decomposition of
coal16 are shown in Figure Ik.  The family of curves illustrates the
point that volatile yield is a strong function of pyrolysis temperature.

     In recent studies carried out by Anthony and Howard17 and by
Kobayashi, et al. ,18 it was shown that an increasing proportion of the
combustible solid mass can be volatilized as the pyrolysis temperature
increases and that the volatile yield so produced can far exceed the
volatile matter determined by ASTM standard test (Figure 15).  In staged
combustion processes it will be of advantage to volatilize a large pro-
portion of the coal and thus provide conditions for the conversion of
the maximum amounts of fuel-nitrogen to N2 in the first, fuel-rich
stage of the process19'20.

     The solid residue burns out in the second stage or the tail-end of
the flame.  The combustion of "large" particles (>100 ym) is normally
controlled by external diffusion and hence their burning times are
proportional to the square of the initial particle diameter** (Figure l6).
For smaller particles the burning time becomes longer than that predicted
from diffusional transport because the rate determining step in their
reaction mechanism is the activated adsorption of the oxidant on active
sites at the surface or in the pores of the solid.  Since the rate of
mass transfer varies inversely with particle diameter and the rate of
chemical reaction is independent of particle size, the temperature at
which the transition from mass transfer to chemical rate control occurs
must depend on the particle size.  The nature of this dependence is

-------
                          COAL COMBUSTION  SYSTEMS
                                                              157
                         O-2   0-4    O-6   0-8   1-0
                           Input velocityJt/0 i m/s
                         0-1    0-2   O-3   0-4   O-5   0-6   O-7
                                Input velocity, Ua : m/s

                               Particle diameter
                               Inlet temperature = 350"K
                               If = 100kcal/mzs

Fig.  10.  Theoretical Variation of  Ignition Time with Input  Velocity,

           Coal Concentration and Ignition  Temperature in Plug Flow.
                    4-Qt	1	1	1	1	
                    3-0
                   S.
                   "1-0
                          \
                         X S
                        Low-volatile coal
                        O
                       • \
                                            X Burner A
                                            O Burner B
   Fig.  11.
               20OO      4OOO      600O
           Specific surface of fuel ,SW : cm2/g
The Variation of Ignition  Distance in Anthracite Flames

as a  Function of Fineness15.

-------
  158
            CLEM COMBUSTION OF COAL
 200
               100         200
            distance from burner
Fig. 12.   Heat Transferred to the Pulver-
ized-Fuel Jet Upstream the Flame Front  by
Recirculation and Radiation.
ft
It
It
10
\>
s,
f
i
Quantify of ttr — *•
*






1

1
/
/
1

1





1


>,



\


•rf.



N



^






%


\
x<





'
R





1
sf-






^







^




\

"*+






^




                                                t  -a 20 x w so so 70 so so 100 tin so oar. Kt
                                                 frimary air-coal ratio for 'a'     *
                                                       Bituminous coal 'a'l
                                                           Votatilm 30%
                                                           Ash  : SY,
                                                           Moisture: 3V..
                                                           H, : 7530 Heat/kg
                                                           LU-- i.25Noi/Kg
                                              Fig.  13a.  Relation Between Pri
                                              mary-Air Fraction and  Ignition
                                              Velocity of Pulverised-Coal
                                              Mixture.
Wl
^
f"
«e ^


-<.
V

V
A


X

                                           »   ^   30  V. W
                                           	 Votatilf    .
                                           ^"^ matter    -
                                           	Ash 	-
                                          Influence  of Volatile
                                              Fig. 13b.
                                              Matter  Content and Ash Content on
                                              Ignition Velocity of  Pulverised-
                                              Coal Mixture (After Dolezal1*).
                         20      4O      60      60
                            Isothermal decomposition time,/: ms
                                         100
                                                120
    Fig.
Variation of Weight  Loss with Time at Various Temperatures
For a  Coal of Low Rank (After Badzioch16).

-------
         COAL COMBUSTION SYSTEMS
                                         159
                                 O2100 K
                                 01940
                                 O174O
                                 0151O
                                 V126O
                                 A 1000
SO        100       150
   RESIDENCE  TIME (ms)
                                          200
                 Figure 15.
   (After Kobayashi-Howard and Sarofim18).

iv i
1.4
1.2
1.0
0.8
<
\
\
V
\





1











\











\











V










<^











**•«
JO 1.2 1.4 1.6
excess air


b







,j


It








If

rr







^
n

In
c




n
*s

1

Ol
•)i



/t
//



'T

JS
r

/


/



al


t
f
/
/
/









/
I,
j
' f
/ 1
S

















/
/
J
i
/
' S
**t
















I
/
i
f
f
yS
/








J
I
'
/
f

f
f
t,
cc

r6vy
















/
/
/
i
I


f/
t
f


>K

rn











/
/


/
J
^
/



e

cc






1
'

i
f


f
j
,
f







a







/
i


j
}

)
















) 0.1 0.2 0.
Particle Diameter
sec
2.0
1.5
u
E
en
1.0 <=
c
X!
0.5
O
m.m
Fig. 16.  Burning times of Coal Particles.
              (After Gumz1*).

-------
160
                        CLEM COMBUSTION OF  COAL
illustrated by the  logarithmic plot of burning  time against particle
diameter given in Figure IT21-  It follows that  the rate of burning of
small particles  (<50 ym) depends more on the temperature and the partial
pressure of the  oxygen,  than on the particle size.

     In practice there are many methods for aiding ignition or combus-
tion of difficult-to-burn coals.  The environmental constraints of low
NOX emission, however, make their application more difficult.  Two such
methods are mentioned in the following with examples of application to
low grade coal combustion in power station boilers:   (a) improving
ignition and combustion  by means of changes in  the coal preparation
system, and (b)  the use  of pulverized coal pilot flames to ensure stable
ignition when burning low grade coals.
                   K>
                                10          10'
                           Initial particle diameter 6»m)
                      10
                    A
                    B
                    C
                    D
                    E
                    F
                    G
                    H
                               T OK)  E (kcal mole-i) P (atm.)
1000
2000
1000
2000
1000
2000
2000
2000
20
20
10
10
 5
 5
10
10
 10
100
                   Solid lines indicate mass transfer control, all pres-
                   sures; a, 1000-K; b. 1600"K; c, 2000'K
     Fig. 17.  Theoretical Particle Burning Times Showing  Effects of
               Particle Diameter,  Temperature and Pressure.  (Ref. 21)

-------
                        COAL  COMBUSTION SYSTEMS
                        161
     Unlike in other  coal burning systems, in pulverized coal combustion
the coal preparation  (drying,  grinding,  transport of fuel to the burners)
is an integral part of the  combustion system.  The coal preparation can
be "direct" or "bin and  feeder."   Direct coal preparation (Figures l8a,
b) means that the drying medium—preheated air and/or flue gas—is
moving the fuel through  the grinding mill, and the particle size classi-
fier, transports the  ground coal  to the  burners and serves also to
provide the immediate environment for the coal during pyrolysis, igni-
tion and the initial  stages of burning.   Direct coal preparation plants
are relatively simple, but  the task to satisfy the many, often conflict-
ing requirements for  drying, classification, transport and primary air
during ignition and combustion is difficult, particularly for low grade
coals22.
                                      Burner.
                   Feeder.
   Fig..I8a.
     Direct-system
with  i/ie pu/vemer
mill under pressure.
                    Fig.  l8b.
                      Direct-system
                 with  ihe pulveriser
                 mill under suction.
   Mill-Fan.
                                         Mill.
                                 Preheated air and/or
                                 Furnace Coses.

                                 Figure 18.
      The "bin and feeder" systems  (Figures  19a, b) permit the  separation
 of the drying-grinding circuit from that of transporting the fuel  to  the
 burner, so that the coal can be injected into the flame by  a prescribed
 proportion of the combustion air.  "Bin and feeder"  systems are more
 complex in their operation, require more precaution  particularly when
 the dried pulverized coal is stored.

      An example of how the suitable choice  of the coal preparation
 system can be used for improving plant efficiency is shown  in  Figure 2023.
 Here the total amount of flue gas  is recirculated and used  as  drying
 medium for a high moisture lignite.  The flue gas can, in this way, be
 cooled down to below 100°C instead of the usual 200°C when  expensive
 heat exchangers in the boiler have to be protected from low temperature
 corrosion.

-------
162
CLEAN COMBUSTION OF COAL
                    Fig.  19a.
                        Open-circuit
                    with storage.
                                 Furnace
                                  gas.
                           Tottie
                         atmosphere
                      Exhausf  '
                       tan._

                        -RF.precipitator
                                                     T0rfhe
                                                     PFburner
                    Fig. 19b.
                       C/osed-c/rcuit
                    with storage.
                                              Cyclone
                                        Mill.
                                   Furnace
                                    Coses
                                              Prtmoru
                                              Air Fon
                              VFan.

                               Effluent
                               Burner

                             \f. Burner
                           Preheated
                             Air
                     Fig.  19.   "Bin and Feeder"  Systems.
                                        A/ff
                                        COMBUSTION CAS.
            Fig.  20.  Applied  Open-Circuit  System with Storage.
                        (After P.N. Kendys, World Pover Conference,
                       1956.  Report No. 2^862/lJ.)

-------
                        COAL COMBUSTION SYSTEMS                     163
     In the other example the combination of a "bin and feeder" coal
preparation system and a regenerative-type burner has been applied to
solve the load following problem when burning a low grade coal2".
Figure 21 shows the arrangement of the pilot burner in relation to the
main burners in a corner-fired combustion chamber and Figure 22 illus-
trates the schematic of the recuperative-type burner.  The high air
preheat (650°C), the specially fine grinding (90% 
-------
164
CLEM COMBUSTION OF COAL
                          nir
                                      coal
                                      dust
                            . .. .^T"
                                ^ ji.t.
   *v-
    ^c-^v
                    t'*_V!.\J  V
                   - j_a    .
                      i secondary
                              mm
Fig. 22.  Schematic of Pilot Burner for Low Grade Pulverized Coal (Ref.
               Ash 4
               removal
    Grit
    retiring
                                                 Economiser
                                                 PRATT
                                                -DANIEL
                                                 Precipitator
                          Circulating
                      Air  pumps
        Fig.  23, BCSL Fluidized Combustion Water-Tube Boiler.

-------
                        COAL COMBUSTION SYSTEMS                     155
fluidizing air will rise through the bed in the form of bubbles27.   The
coal particles are larger than in pulverized combustion (mm size range)
and their combustion is, normally, external-diffusion limited.  At the
low temperature limit of operation, however, at around 650°C, the slow-
est step in the overall reaction mechanism is most likely the desorption
step in the chemisorption reaction at the coal surface.  The oxygen
concentration in the dense phase is low (l-3%), and an important factor
in the rate of burning of the coal particle is the rate at which oxygen
will diffuse from the bubble to the dense phase.  The basic considera-
tions and methods of calculation of combustion of coal particles have
been established by Avedesian and Davidson28.  Extensive theoretical and
experimental research to develop further this theory is in progress at
MIT.

     Figure 2\ shows the reduction of sulphur emission that can be
achieved in fluidized combustion by the use of sorbents, limestone of
dolomite in the bed.  The NO emission determined in a number of pilot
plant experiments and plotted as a function of bed temperature is given
in Figure 25-  Recent studies of the formation and destruction of WO in
fluidized combustion processes have shown significant potential for the
further reduction of WO emission29.

     There are several fluidized combustors in industrial operation and
their present development can be considered to have reached the stage
for application in industrial process heat or power generation systems.
The scaling up of fluidized combustion to utility size boilers requires
some further research and development:  an improved coal feed system
would have to be developed which does not require one feed point for
every MW thermal input, and further studies would have to be carried out
to ensure that superheater tubes immersed in the fluidized bed have com-
parable life time to steam generating tubes in the bed.  Figure 26 shows
the variation of the enthalpy composition of steam (i.e., liquid heat,
latent heat and superheat) as a function of steam pressure.  It can be
seen that as the steam pressure rises, the latent heat proportion of the
total enthalpy decreases.  As the steam pressure rises beyond a value
(>1500 psi), it becomes difficult to maintain a balance between the heat
that has to be extracted from the bed to maintain its temperature below
900°C and the latent heat proportion of the steam generated in the bed;
additional bed cooling by superheater tubes becomes necessary.  This
explains the need for further research on the corrosion-erosion of
superheater tubes immersed in the bed, before fluidized combustion can
be safely applied to large utility boilers.

CONCLUSIONS

     An attempt has been made to present coal combustion systems from
the point of view of the physical-chemical process that the coal and
accompanying mineral matter undergo during combustion and by considering
the requirements of plant and power generating cycle efficiency.  It was
considered that the development of coal combustion systems was closely
tied to that of steam raising plant, that the transition from stoker
firing to pulverized coal combustion occurred because the latter could
be scaled up more easily to larger units, produced higher combustion
efficiencies at lower excess air and could make use of higher air preheat
which in turn was thermodynamically advantageous in power generating cycles.

-------
166
CLEM COMBUSTION OF  COAL
IUO
9O
00
7O
o5 6O
•^-*
.1 5O
5 4O
61 30
i; 100
H- 9O
•3
in
BO
7O
An
	 1 — • i ~^5^
o /b^V-"-
/S*-x
X /Pressurised -
O Unpressunsed
/
' / 0 DOLOMITE'

I i i i
LIMESTONE ^--
^x
Unpressurised''
^X/ rf
/ "reSSUr^/^-
,/ , ^JL .
IUU

9O

^ 8O
^100
.0
S 90
L.
^ 8O
_, i-><^
00

70

•______' ..^i__ — Q-
Pressurised
Unpressurised
• ""
-------
                        COAL COMBUSTION SYSTEMS
167
                                           \
                                  a   a    §
                                  1"   I    1
                                  8   e
k cc
i
il/kg
800
600
400
200
ft w — —
Superheating

•~—



/






x

.

•.
Eva


*^*

.
—~.
oratio

--


•^


,-•

Preheating




• — ..


•^"*












N
y







-




C.R
                             50
                                  100   ISO
                                            200
             Fig. 26.  Enthalpy Pressure Diagram of Water.
     Fluidized coal combustion, a new advanced system, has the added
advantage of being capable of retaining sulphur in the bed and it prom-
ises also significantly lower emission of NOX.  Due to its lower operat-
ing temperature, fewer or no submicron particulates are emitted from
fluidized combustors.  This system which is ready for application for
smaller industrial combustion systems requires further concerted research
for its application to utility-size power generators and for combined gas
turbine-steam cycle operation.

REFERENCES

1.   Anthony, D.B., J.B. Howard, H.C. Hottel, and H.P. Meissner: "Rapid
     Devolatilization of Pulverized Coal," Fifteenth Symposium (Int'l)
     on Combustion, p. 1303, Combustion Institute, Pittsburgh, PA (1975).

2.   Howard, J.B., and R.H. Essenhigh: "Pyrolysis of Coal Particles in
     Pulverized Fuel Flames," Ind. Eng. Chem. Process Design Dev. , 6_,
     7^ (1967).

3.   Beer, J.M., and R.H. Essenhigh: Nature, 187, 1106 (i960).

k.   Gumz, W. :  Kurzes Handbuch der Brennstoff und Feuerungstechnik
     Springer - Berlin, p. U92 (1962).

5.   Beer, J.M.: "Combustion Chamber Design for Stoker Firing and Low
     Grade Coals Based on Both Laboratory Light Model Experiments and
     Large Scale Trials," (in Hungarian), Magyar Energiagazdasag, No. 9,
     pp. 306-315 (1951).

6.   Beer, J.M.: "Combustion Research for Industrial and Power Station
     Boilers,"  (in Hungarian), Magyar Technika. No. U, pp. 210-216 (195M

-------
16g                    CLEAN COMBUSTION OF COAL


7.   Tzukhanova, O.A.: J.  Techn.  Physics 9 (1939) Vol. U, pp. 295-30**,
     Leningrad, USSR.

8.   Chukhanov, Z.F.,  and M.K.  Grodzovskii,  as cited in L.N. Khitrin:
     Physics of Combustion and  Explosion Israel Program for Scientific
     Translations, Jerusalem (1962).

9.   Wertaneister, H. :   VDI Berichtsheft , Trier (193*0, pp. 7^-79-

10.  Zagon, P. , as cited by Beer, in Ref.  6,  this paper.

11.  Beer, J.M. :  "Some Current Trends in Combustion Research in
     Hungary," University of Sheffield Fuel  Soc . , J. pp. 1-12 (1958).

12.  Nusselt, W. :  Die Ferbrennung in der Kohlenstaubfeverung Z. V.D.I.
     68, No. 6, pp. 12*1-128 (192*0.

13.  Essenhigh, R.H. , and J. Csaba:  Ninth Symposium (int'l) on Combus-
     tion, Combustion Institute, Pittsburgh,  PA, pp. 111-125 (1963).

1U.  Dolezal, R. :  Large Boiler Furnaces, Fuel and Energy Sci. Series,
     Ed. J.M. Beer, Elsevier Publishing Company, England (1968).

15.  Beer, J.M.:  "The Effect of Fineness and Recirculation on the Com-
     bustion of Low-Volatile Pulverized Coal," J. Inst. F. , pp. 286-313
     (July
 16.   Badzioch, S. :  "Thermal Decomposition," Chapter U in M.A. Field,
      et al. :  Combustion of Pulverized Coal, BCURA, England (1967).

 17.   Anthony, D.B., and J.B. Howard:  "Coal Devolatilization and Hydro-
      gasification," AIChE Journal, 22_, 625 (1976).

 18.   Kobayashi , H. , J.B. Howard, and A.F. Sarofim:  "Coal Devolatiliza-
      tion at High Temperatures," Sixteenth Symposium (int'l) on Combus-
      tion,  The Combustion Institute, Pittsburgh, PA, pp. U11-U25 (1977).

 19.   Pohl,  J.H. , and A.F. Sarofim:  "Devolatilization and Oxidation of
      Coal Nitrogen," ibid, pp. ^91-501.

 20.   Song,  Y.H., J.M. Beer, and A.F. Sarofim:  Fate of Fuel Nitrogen Dur-
      ing Pyrolysis and Oxidation, Second Symposium on Stationary Source
      Combustion, EPA, New Orleans (August 1977).

 21.   Mulcahy, M.F.R., and I.W. Smith:  Kinetics of Combustion of Pulver-
      ized Fuel:  A Review of Theory and Experiment, Rev. Pure and Appl.
      Chem., 19_, 8l (1969).

 22.   Beer,  J.M. :  "Choice of Coal Preparation Systems for Pulverized
      Fuel Firing," Engineering and Boiler House Review  (September 1-6,
      1959).

 23.   Beer,  J.M. :  "Some Current Trends in Combustion Research in
      Hungary," University of Sheffield Fuel Soc., Journal, pp. 1-12  (1958).

-------
                        COAL COMBUSTION SYSTEMS                     169
2U.   Beer, J.M., T. Csorba, and J. Csaba:  "Pulverized Fuel Burner for
     Burning Low Grade Coal," (in Hungarian), Magyar Energiagazdasag,
     No. 2, pp. 61-68 (1956).  Translated into English by the Ministry
     of Pover.

25-   Skinner, D.J.:  Fluidized Combustion of Coal, Mills and Bonn Mono-
     graph CE/3 (1971).

26.   Robinson, E.B., R.D. Glenn, S. Ehrlich, J.W. Bishop, and J.S.
     Gordon:  EPA Contract CPA 70-10, PB210 828 (February 1972).

27.   Davidson, J.F., and D. Harrison:  Fluidized Particles, Cambridge
     University Press (1973).

28.   Avedesian, M.M., and J.F. Davidson:  Trans. Inst. Chem. Eng.,
     £L, 121  (1973) London.

29.  Beer, J.M.:   "Fluidized Combustion of Coal, A Review" (invited Paper)
     Ibid., Paper No. 33.

-------
170                    CLEAN COMBUSTION OF COAL

-------
                                                                     171
              POLLUTANT FORMATION DURING COAL COMBUSTION

                      M. P. Heap and R. Gershman
             Energy and  Environmental Research  Corporation
                          Irvine, California
1.0       INTRODUCTION

     Increased coal utilization is undoubtedly the key to decreased
reliance upon imported crude oil and energy independence.  However, the
methods which will be used in the next three decades to increase con-
siderably the amount of energy being supplied from coal are open to
question.  Economic and environmental considerations will be factors in
the choice of several available energy conversion alternatives since
there is a need -to utilize coal in an efficient, but environmentally
acceptable manner.  Coal contains constituents which have the potential
to form pollutants which might be emitted into the atmosphere during the
energy conversion process.  These constituents are:

     •    Sulfur which forms SO  and SO., under oxidizing conditions.

     •    Mineral matter which results from the biological, chemical
          and environmental conditions existing during the formation
          of coal and is vaporized or converted to ash or slag during
          combustion.  In a study of 65 Illinois coals Gluskoter(l)
          found that the mineral content ranged from 9.4 to 22.3 per-
          cent.  Coal contains almost all the elements in the periodic
          table.

     •    Nitrogen which can also be oxidized to give nitrogen oxides
          (NOX).

The above compounds, together with the products of incomplete combustion
(CO, hydrocarbons, carbonaceous particulates and large molecular weight
organic molecules), are potential pollutants resulting from coal utili-
zation.  Economics of pollution control are extremely important with
regard to the increased utilization of coal as a fuel.

     For one of the major pollutants produced from coal, sulfur, the
choice is one of removal from the fuel or from the products of combus-
tion.  Removal from the fuel involves the production of a liquid or
gaseous fuel or a clean, solid fuel.  Coal gasification converts the
majority of the sulfur to hydrogen sulfide which is then scrubbed from
the fuel gas.  Neglecting absorbant efficiency, the reduced gas volume
ensures that the task of removing ^S from the fuel gas is easier than
the removal of S02 from the combustion products.  Fluidized bed combus-
tion of coal offers the possibility to burn coal as mined and yet con-
trol sulfur emissions by in-bed absorption of S02.  Although coal gasi-
fication is not a new technology, its application to today's needs is

-------
172                     CLEM COMBUSTION OF COAL
far from straightforward and there are no operational  fluidized  bed
combustors on a commercial scale.  It appears that direct  coal combus-
tion  in conventional plants provides the greatest opportunity for  a
rapid increase in coal utilization.  The major problem associated  with
direct coal combustion is the control of three atmospheric pollutants —
SOX,  NOX and particulate matter.

      This paper is primarily concerned with the formation  of and control
of nitrogen oxides since this is one pollutant which can be controlled
by modifying the combustion process.  Coal-fired utility and industrial
boilers account for 37 percent of the stationary source emissions  of NOjj,
and gas- and oil-fired boilers contribute a further 16  percent to the
total. The existing New Source Performance Standard (NSPS) for  large
coal-fired boilers is 0.7 Ibs NOX/106 Btu which is 3.5 and 2.3 times as
high  as the NSPS for gas- and oil-firing respectively.  The increased
emphasis on use of coal in large boilers will increase total NOX emis-
sions unless appropriate control technology is applied.

      This paper does not provide a complete literature survey on the
formation of nitrogen oxides in coal-fired systems.  Only  those  refer-
ences have been given which are necessary to illustrate a  point  or which
contain controversial data.  Before the fundamental aspects of NOX for-
mation are discussed, the state-of-the-art of NOX control  for coal-
fired boilers will be reviewed.  Finally, the prospects for advances in
NOX control technology for pulverized coal-fired boilers are discussed.

2.0       REVIEW OF NOX CONTROL TECHNOLOGY

      The reader is referred to several reports in the  literature(2>3,4,5)
which discuss the application of NOX control technology to coal-fired
boilers.  The existing state-of-the-art is reviewed below:

      •   Firing Type.  Coal can be burned crushed in  cyclone furnaces,
          pulverized in tangential or wall-fired boilers,  or spread as
          lump coal in stoker-fired boilers.  NOX emissions are  normally
          highest for cyclone-firing, and wall-fired boilers characteris-
          tically emit more NOX than tangential firing.  Spreader
          stokers, which might have 50 percent of the  heat released in
          the suspension phase, generally give higher  emissions  than
          underfeed stokers.

      •   Operational Parameters.  Reduced load or reduced excess  air
          operation tends to reduce emissions from all boiler types.

      •   Flue Gas Recirculation.  The addition of cooled  combustion
          products to the combustion air is not an effective method of
          controlling NOX emissions from coal-fed boilers.  This can
          be attributed to the predominance of fuel NOX in the total
          emission from coal-fired units.

      •   Staged Heat Release.  Staged heat release accomplished either
          by the use of overfire air ports or removing burners from
          service is an effective method of NOX control.

-------
                   POLLUTANT FORMATION AND CONTROL                 173
     •    Burner Redesign.  Redesign of the burner to decease the rate
          of fuel/air mixing for wall-fired burners reduces NOX forma-
          tion in pulverized coal flames.

3.0       NOX FORMATION DURING COAL COMBUSTION

     The nitrogen oxides emitted by coal-fired boilers are formed by the
oxidation of molecular nitrogen (thermal NOX) and nitrogen which is
chemically bound in the fuel (fuel NOX) .  Small-scale studies carried
out by Pershing and Wendt(^) indicate that for pulverized coal-fired
combustors the major portion of the emission can be attributed to fuel
NOX.  However, under controlled conditions it might be expected that
the thermal NOX fraction will become more significant.  The major por-
tion of this discussion will be restricted to NOX formation in pulver-
ized coal combustion since this is the predominant method of coal
utilization by direct combustion.

3. 1       Coal Chemistry

     In the restricted sense of this review coal chemistry refers to
those changes which occur during the period of particle heating.  Parti-
cles in pulverized coal flames are subjected to heating rates between
10,000 K/sec and 100,000 K/sec causing volatiles to be driven from the
coal.  The particles might swell and off gas as a jet, they might explode
or  the volatilization process might occur with little physical change
in  the particles.  Blair et al(?) and Pohl and Sarofim(S) have shown
that the total quantity of volatiles produced is a strong function of
pyrolysis temperature.

     Blair et al(?) carried out a series of controlled pyrolysis experi-
ments with single particles and found that:

     •    Total nitrogen volatilized is a more sensitive function of
          pyrolysis temperature than is total mass pyrolyzed.

     •    Only 20 percent of the volatile fuel nitrogen appears to be
          released from the coal as light gases (HCN,
This latter conclusion may have less significance because even though
the initial nitrogen fragments may be heavy molecular weight compounds
it is probable that these compounds undergo pyrolysis and that the
nitrogen is converted to either HCN or NO depending upon the availa-
bility of oxygen.  Of much more significance is the fact that the
initial volatile fractions are low in nitrogen content.  Thus these
volatile gases have the opportunity to deplete the available oxygen
before the major portion of the nitrogen-containing volatiles are
evolved .

     Studies (8) in a laminar flow furnace indicate that residence times
of approximately one second are necessary at temperatures in excess of
1500K to reduce the nitrogen content of the char to less than 40 per-
cent of the value of the original coal.  Oxidation experiments in the
same furnace clearly indicated that the conversion of char nitrogen to
NO occurs with a lower efficiency than the conversion of coal nitrogen
under comparable conditions of temperature and equivalence ratio.

-------
174                     CLEAN COMBUSTION OF COAL


     In summary it appears that:

     •    At flame temperatures pyrolysis kinetics are  sufficiently
          fast to suggest that in the  absence of oxygen  HCN would be
          the primary fuel nitrogen product.

     •    Increased temperatures decrease char nitrogen content.

     •    Initial volatile fractions are low in nitrogen species.

     •    The efficiency of char nitrogen conversion to NO  is  lower than
          that of the parent coal nitrogen to NO.

 3.2       Homogeneous Gas Phase Reactions

     The  details of fuel NO kinetics are outside the scope  of  this
 paper other  than to qualitatively describe those conditions which
 minimize  fuel NO formation.  Reaction  zone stoichiometry appears to
 have the  strongest influence on fuel NO formation, the  conversion  of
 fuel nitrogen to NO decreases rapidly  as the mixture becomes fuel-rich.
 This can  be  accounted for by considering two competing  reaction paths:

     XN   +   Ox  ->  NO  +  . . . .                                    (1)

 which predominates in oxygen-rich mixtures and

     XN   +   YN  -*•  N2  +  ....                                    (2)

 which is  faster in fuel-rich mixtures.  There is also considerable
 evidence  to  suggest that NO is a necessary nitrogen specie  for N£  pro-
 duction.  Another type of reaction which may be very significant in
 pulverized coal flames is the reduction of NO by hydrocarbon specie,
 e.g.,

     NO   +   CH  ->  XN  +  . . . .                                    (3)

 thus allowing the production of nitrogen via reaction (2).

     Premixed gaseous studies indicate that reaction zone temperature is
 not  a significant factor in fuel NO production.  However, the  conversion
 of fuel nitrogen to NO appears to be greater in hydrogen flames than in
 hydrocarbon  flames.  The presence of sulfur appears to  enhance the pro-
 duction of fuel NO in rich, premixed gaseous flames and two-phase
 turbulent diffusion flames suggesting  that sulfur specie might inter-
 fere in the  competing paths represented by reactions (1)  and (2).

-------
                   POLLUTANT FORMATION AND CONTROL                  175


3.3       Coal Char Reactions

     Char is that solid which remains after the volatile fractions have
been driven from the coal particles.  Char contains nitrogen; however,
as noted earlier, the efficiency of char nitrogen conversion is less
than that of the parent coal nitrogen.  This can be attributed to the
fact that the stagnant boundary layer surrounding the char particles is
probably reducing.  Thus even if the char nitrogen were to produce
nitric oxide at the particle surface there is the possibility that it
could be reduced to N2 as it diffuses through the boundary layer.  Sur-
prisingly little concrete information is available concerning the pro-
duction of NO from char nitrogen.  Theoretical calculations carried out
by Wendt and Shulze(9) suggest that the production of NO from char
nitrogen is most sensitive to free stream oxygen concentration and
temperature.

     In a series of experiments designed to define probe conditions
necessary for the measurement of NO in pulverized coal flames, Heap
et al'lO) showed that NO in nitrogen could be reduced by coal particles
at temperatures above 200°C.  Beer and co-workers (H) have shown that
under fluidized bed conditions NO can also be reduced by coal char.
Sarofim(12) has recently reported that coal char suspensions will also
reduce NO to N2 in the absence of oxygen.  Consequently, it appears that
NO produced in the early stages of heat release could be reduced by char
in the burnout regions.  Further research effort is necessary to ascer-
tain whether heterogeneous NO reduction by char is an important phenom-
enon in pulverized coal flames.

3.4       Fuel/Air Contacting

     Pulverized coal is burned in an air suspension.  Fuel/air mixing,
entrainment of recirculated combustion products, particle heating,
volatile evolution, volatile combustion and char oxidation occur almost
simultaneously.  The major factor controlling the emission of nitrogen
oxides is the fuel/air contacting process since this will affect:

          the rate of particle heating and the final particle tempera-
          ture which will determine the proportion of the bound nitrogen
          remaining in the char;

          the atmosphere under which the volatile compounds react.  If
          this is oxygen-rich then it would be expected to maximize
          NO production;

          the residence time in any rich zone.  The degree of conversion
          of fuel nitrogen fragments to N2 or the reduction of NO to N£
          will depend upon residence time in the rich zone; and

          the rate of heat release and reaction zone quench rate which
          will affect the production of thermal NO.
     Heap et al(' carried out a series of subscale experiments to esta-
blish the influence of fuel/air contacting on NOX production in pulver-
ized coal flames by varying burner parameters.  These investigations
demonstrated that NOX emissions were very sensitive to the fuel/air

-------
176                     CLEAN COMBUSTION OF COAL
mixing rates.  Combinations of fuel injection method and axial and
tangential velocity distributions produced the two major flame  types
sketched in Figure 1.  The top sketch represents the conditions in a
near-field dominated flame in which the coal and air are rapidly mixed.
Maximum particle heating rates and stable ignition are provided by the
entrainment of hot combustion products from an axial recirculation zone.
These conditions are typical of flames in uncontrolled wall-fired boilers
and maximize fuel NOX formation since ample oxygen is available in the
regions of coal volatilization.  With the same coal, excess air level
and air preheat temperature NO levels are reduced from approximately
800 ppm to 200 ppm by producing a high fineness ratio diffusion flame
with a minimum oxygen level associated with the fuel jet.  Provided
there is a stable ignition front around the fuel jet the volatile frac-
tions will remain in an oxygen-deficient atmosphere providing the oppor-
tunity for fuel nitrogen fragments to form N2-  The type of fuel/air
contacting producing the high  fineness  ratio diffusion flame has many
similarities with the processes occurring in corner-fired boilers whose
NOX emission levels are interestingly lower than uncontrolled emissions
from wall-fired units.

     Figure 2 taken from the work of Pershing and Wendt^") provides
experimental evidence to support the hypothesis presented above.  These
workers substituted argon for molecular'nitrogen in the combustion "air,"
and thus the total emission can be attributed to fuel nitrogen conver-
sion.  Two different injectors were used which produced a near-field
dominated flame (divergent injector) and a high fineness ratio diffusion
flame (axial injector) and it can be seen that the fuel NO production is
much less in the case of the high fineness ratio flame.

4.0       ADVANCED CONTROL TECHNOLOGY DEVELOPMENT

     The stated goal of the U.S. Environmental Protection Agency is the
development and demonstration of maximum NOX control for stationary
sources.  This strategy is based upon the premise that stringent con-
trols cannot be applied to mobile sources without incurring serious
economic penalties.  Consequently, if air quality is not to be degraded,
more restrictive control must be imposed upon stationary sources.  Con-
version to coal or coal-derived liquids will tend to emphasize the
significance of stationary source emissions, thus providing impetus for
the development of advanced NOX control technology.

     The use of off-stoichiometric firing in combination with reduced
load, low excess air operation and a modified burner design reduced
NOX levels to 200 ppm in an opposed wall-fired 270 MW boiler(14).
Gershman et al^5' have reported on the development of a low NOX proto-
type coal burner which is being sponsored by the EPA.  This burner con-
cept was based upon the subscale work of Heap et al(16) and relies upon
distributed air addition which is explained schematically in the sketch
shown in Figure 3.  The air supply is divided into three streams.  The
distribution of axial and tangential velocity at the burner throat
creates a high residence time, hot recirculation zone into which the
coal is injected.  The third air stream is injected at the periphery of
the burner exit to provide an oxidizing envelope to prevent impingement
of corrosive reducing gases on the furnace walls.  The distributed air

-------
 POLLUTANT FORMATION AND CONTROL
                       177
          Axial  Recircul
             Zone
      Hot Products - 1950 K
~~^9
ation^/\
            Near Field Dominated
                 Air + Recirculated Products
Early Volatiles Deplete  CU
       High Fineness Ratio  Diffusion  Flame
         Figure  1.  Major Flame Types

-------
                                                                                              —I
                                                                                              CO
   1200
   1000
    800
o
4J
o   600
    400
    200
  I    I    I    I        I    I
Divergent Injector
                          Thermal NO
               Fuel  NO
  O  Air
  A  Ar/02/C02
  I    I    I    I
                                I	I
 \    \    I   I   I    I
Axial Injector
                                             Thermal  NO -^  _
                                                              Fuel  NO
 I    I    I    I    I    I	I
       .0     1.1     1.2      1.3     1.0      1.1
                              Stoichiometric Ratio
                                             1.2
                     1.3
                                                    o
                                                    tr1
                                                    £
                                                                                     O
                                                                                     i
                                                                                              en
                                                    O
                                                    23
                                                    O
                                                    g
                                                    f
     Figure 2.  Comparison of Thermal and Fuel NO Production for
                Two Basic Flame Types (After Pershing and Wendt^/)

-------
                    Tertiary Air
   Divided Secondary

   Air Stream
Coal + Primary Air
Progressive

Leaning Out
                                                                                              Burnout Zone
                                                                                                                  O
                                                                                                                  tr>
                                                                                                                  f
o


I
(-3
M
O
                                                                                                                 O
                                                                                                                 O
                             Figure 3.  Distributed Fuel/Air Mixing Concept

-------
180                     CLEAN COMBUSTION OF COAL
addition ensures that the coal first reacts .in a rich region  (
-------
             POLLUTANT FORMATION AND CONTROL
                           181
   800
   600
 CVJ
o
o

4->
to
   400
Q.


O
   200
                                    Small-scale Hot Tunnel

                                        5  x  106 Btu/hr


                                 O  Low Swirl

                                 O  Medium  Swirl

                                 A  High  Swirl
Water-Cooled Simulator

    50  x  106  Btu/hr


  Medium  Swirl

— 35%  Excess Air
                               	  55% Excess Air
   100
                                             I
     0.8      1.0       1.2       1.4       1.6

                 Primary Zone Equivalence Ratio
                  1.8
                   Figure  4.   Test Results

-------
182                     CLEAN COMBUSTION OF COAL
                               REFERENCES
  1.   Gluskoter,  H.J.,  Occurrence  of  Mineral Matter in Coal:   An Intro-
      duction,  Paper presented  at  The International Conference on Ash
      Deposits  and Corrosion from  Impurities in Combustion Gases,
      Henniker,  New Hampshire,  July  1977.

  2.   Martin, G.B.  and  Bowen, J.S., Development of  Combustion Modifica-
      tion Technology for Stationary  Source NOX Control:   Health,
      Environmental Effects, and Control  Technology of Energy Use, EPA
      Report 600/7-76-002.

  3.   Crawford,  A.R., et al, Field Testing:   Application  of Combustion
      Modifications to  Control  NOX Emissions from Utility Boilers, EPA
      Report 650/2-74-066,  June 1974.

  4.   The Proceedings of the NOX Control  Technology Seminar,  EPRI Special
      Report, February  1976.

  5.   Pershing,  D.W., Ph.D.  Dissertation,  Department of Chemical Engi-
      neering University of Arizona,  Tucson, 1976.

  6.   Pershing,  D.W. and Wendt, J.O.L., Pulverized  Coal Combustion:   The
      Influence of Flame Temperature  and  Fuel Composition on  Thermal and
      Fuel HOX,  presented to the 16th Symposium on  Combustion,  Cambridge,
      Massachusetts, August, 1976.

  7.   Blair, D.W.,  Wendt, J.O.L. and  Bartok, W.,  Evolution of Nitrogen
      and Other Species During  Controlled Pyrolysis of Coal.   Presented
      at 16th Symposium (International) on Combustion, Cambridge, Mass.,
      August 1976 (to be published).

  8.   Pohl, J.H.  and Sarofim, A.F., Devolatilization and  Oxidation of
      Coal Nitrogen. Presented at 16th Symposium (International) on
      Combustion, Cambridge, Mass., August 1976 (to be published).

  9.   Wendt, J.O.L.  and Schulze, O.E., On the Fate  of Fuel Nitrogen
      During Coal Char  Combustion, AIChE  Journal,  22, 102 (1976).

 10.   Heap, M.-P.,et  al,  The Development of Combustion System Design
      Criteria  for the  Control  of  Nitrogen Oxide Emissions from Heavy
      Oil and Coal-Fired Furnaces, Vol. II,  EPA Report 600/2-76-061B.

 11.   Gibbs, B.M.,  Pereira,  F.J. and  Beer, J.M.,  The Influence of Air
      Staging on  NO  Emission from  a Fluidized Bed Coal Combustor, pre-
      sented to  the  16th Symposium (International)  on Combustion,
      Cambridge,  Mass.,  August  1976.

-------
                   POLLUTANT FORMATION AND  CONTROL                  183
13.   Heap,  M.P.,  et al,  Burner Criteria for NOX Control, Vol. I, EPA
     Report 600/2-76-061A, March 1976.

14.   Barsin, J.A., Dual Register Burner as NOX Control Device, Proceed-
     ings of the EPRI NOX Control Technology Seminar, San Francisco,
     February 1976.

15.   Gershman, R., Heap, M.P. and Tyson, T.J., Design and Scale-Up of
     Low Emission Burners for Industrial and Utility Boilers, Pro-
     ceedings of the Second Stationary Source Combustion Symposium,
     Vol.,V, EPA Report 66/7-77-073E.

16.   Heap,  M.P., et al, The Optimization of Burner Design Parameters to
     Control NOX Formation in Pulverized Coal Flames, Proceedings of
     the Stationary Source Combustion Symposium, Vol. II, EPA Report
     600/1-76-152B, June 1976.

17.   Brown, R.A., et al, Investigation of Staging Parameter for NOX
     Control in Both Wall and Tangentially Coal-Fired Boilers, Pro-
     ceedings of the Second Stationary Source Combustion Sumposium,
     Vol. Ill, EPA Report 600/7-77-073C, July 1977.

-------
184                     CLEAN COMBUSTION OF COAL

-------
                                                                    185
    THE DUAL REGISTER PULVERIZED COAL BURNER:   FIELD TEST RESULTS

                                  by

                         E.  J.  Campobenedetto
                         Development Engineer
                          Combustion Systems
                   Fossil Power Generation Division
                       Babcock & Wilcox Company
                        Power Generation Group
                           Barberton, Ohio


ABSTRACT

      The Federal Environmental Protection Agency's "Standards of
Performance for New Stationary Sources" limits the emissions of oxides
of nitrogen for pulverized coals (other than lignite) to a maximum
level of 0.7 Ibs N0£ per million Btu heat input to the boiler.  To
comply with the above standard, it was necessary to develop a new pul-
verized coal burner because the existing Babcock & Wilcox (B&W) stan-
dard high turbulence burner was not capable of producing NOX emissions
within the above limit.

      B&W's judgmental criteria for the success of the new burner
development was that the burner/furnace system be capable of reducing
NOX generation to levels below the EPA NOX limit while maintaining
carbon utilization equal to or higher than the former standard burner.

      After extensive laboratory and field testing, the B&W Dual
Register Burner was developed.  The data obtained to date has shown
that this burner is capable of operating below the existing NOX limit
without increasing unburned carbon and excess air levels over those
experienced with the previous burner.

      This paper discusses the NOX results obtained to date utilizing
the Dual Register Burner.
THE DUAL REGISTER BURNER

      The Dual Register Pulverized Coal Fired Burner (Figure 1) is a
limited turbulence, controlled diffusion flame burner designed to mini-
mize the amount of fuel and air mixed at the burner to that required
to obtain ignition and sustain stable combustion of the coal.  A ven-
turi mixing device is located in the coal nozzle to provide a homo-
geneous coal/primary air mixture at the burner.  The remainder of the
combustion air (secondary air) is introduced through two concentric
air zones which surround the coal nozzle.  The air flow to each air
zone is independently controlled through inner and outer air zone
registers.  Adjustable spin vanes are located in the inner air zone to
provide varying degrees of swirl to the inner air to control coal/air
mixing during the combustion process.

-------
186
                        CLEAN COMBUSTION OP COAL
          Primary
         air & coal
         Adjustable air vane
          and register drives
«^-*
1 	
1 	 1
t=


Hr-
~"~"~— — ~ ~ — ___ — , i


s

>fk
econdary a
/ i
nj
L
ir i
*r

                                               - Windbox -
                       Dual register burner
                                  Figure 1
       By controlling the mixing of the coal and air,  the combustion
 process is initiated at the burner throat  and the zone of completion
 can be varied in the furnace chamber.   This method of delayed combus-
 tion reduces the high temperature zones at each burner.   Thus the
 peak temperature in the furnace is lowered, minimizing the thermal
 conversion of combustion air nitrogen  to HOX.   Also,  through controlled
 fuel and air mixing, the oxygen availability is minimized during the
 process reducing fuel nitrogen conversion.

       The Dual Register Burner lowers  NOX  by delaying combustion and is
 not a staging device.  Previous work on pulverized coal firing has
 shown that two-stage combustion is the most effective method of NOX
 reduction.  However, in reviewing overall  unit performance, the Dual
 Register Burner has the following benefits over staging techniques:

       l)  The furnace is maintained in an  oxidizing environment to
 minimize slagging and reduce the potential for furnace wall corrosion
 when burning high sulfur bituminous coal.

       2)  More complete carbon utilization through better air/coal
 mixing in the furnace.
       3)  Lower oxygen levels required with total combustion air ad
 mitted through the burners rather than above the burner zone.

-------
                        PULVERIZED COAL BURNER
187
     In addition to the Dual Register Burner, the burner windbox
design was changed to provide air flow control on a per pulverizer basis
(Figure 2).  This compartmented windbox permits careful control of fuel
    Compartmented
       windbox
                                                              Furnace
                                                           observation doors
           Burner secondary
           air control dampers
                       Burner secondary
                          air foils
                  Pulverizer-burner system
                                 Figure 2
and air flows to each burner group.  This permits operation with lower
total excess air while maintaining an oxidizing atmosphere around each
burner.  The results are lower UOX emissions and increased flexibility
in the introduction of excess air to the burner zone for both slagging
and NOX control.
FIELD TEST DATA
      To date, three (3) Babcock & Wilcox pulverized coal-fired boilers
equipped with the Dual Register Burner have been tested to determine
TTOX emissions.  The program for each unit consisted of a series of WOX
tests performed at various operating conditions over the load range.
When operating the units at normal excess air required for combustion,
the NOX emissions were well below the present EPA limit of 0.7 Ibs/lOo
Btu.  Figure 3 is a summary of the full load NOX emissions under normal

-------
188
     CLEAN COMBUSTION OF COAL
  NOx-lbN02/106 Btu

    1.0 |-
    0.9


    0.8


    0.7


    0.6


    0.5


    0.4


    0.3

    0.2


    0.1
Circular burner
                                250
                            Unit capacity, MW
                                   EPA NOx
                                 emission unit
                               700
             Comparison of NOx emissions
         circular burner vs dual register burner
                            Figure 3

-------
                        PULVERIZED  COAL  BURNER                       189


conditions (15 - 17$ excess air to  the burners).   The  emissions  range
from 0.33 to O.H7 lias/million Btu for the three  (3) units, which vary
in capacity from 90MW to TOO MW.  Also,  depicted in Figure 3 are the
NOX emissions obtained from similar units equipped with the high tur-
bulence Circular Burner previously  supplied by Babcock & Wilcox.

      The direct data comparison indicates that  a k^ to 6d% reduction
in NOX is achievable through burner design.  The majority of the reduc-
tion is attributable to controlled  air/coal mixing in  the furnace  cham-
ber.  Through controlled mixing at  the burner, the peak flame tempera-
ture is reduced thus minimizing the thermal NOX  formation at the burner.
In addition, the controlled mixing  decreases the oxygen available  to
react with the nitrogen chemically  bound in the  fuel.

      The type of coal utilized also affects KOX formation.  The 90 and
TOO MW units burn a sub-bituminous  type  coal while the 250MW unit  burns
a bituminous coal.  A comparison of typical analysis for these two (2)
coals (Figure k] indicates why WOX  emissions are lower when firing
                                       Bituminous      Sub-Bituminous
      Total moisture, %                      8.5            30.6

      Proximate analysis, % (dry)
          Volatile matter                   30.8            45.5
          Fixed carbon                     54.8            47.5
          Ash                            14.4             7.0

      Ultimate analysis, % (dry)
          Carbon                         71.8            69.2
          Hydrogen                        4.9             5.0
          Nitrogen                         1.0             0.8
          Sulfur                           1.6             0.7
          Ash                            14.4             7.0
          Oxygen                          6.3            17.3

      HHV, Btu/lb (dry)                    13,060           11,950

      Nitrogen, lb/106 Btu                   0.77            0.67

                    Typical coal  analyses

                                 Figure 4


 sub-bituminous coal.   The higher moisture  content of the sub-bituminous
 fuel  (30.6$ versus  8.5$) provides  for  lower thermal NOX generation be-
 cause of reduced  furnace zone temperatures  due to the greater heat of
 vaporization required.  The nitrogen content of the coal also affects
 NOX generation.   Comparing  the nitrogen content on a Ibs.  nitrogen/
 million Btu heat  input basis (thereby  eliminating variations in heating
 value), the fuel-contained  nitrogen introduced to the furnace is 13%
 lower with sub-bituminous coal (0.67 versus 0.77 Ibs N/million Btu).

-------
190
CLEAN COMBUSTION OF COAL
 With less nitrogen  available, the contribution of fuel bound nitrogen
 conversion is reduced.

       In addition to variations in coal supply, the two principal  param-
 eters  affecting NOX formation are combustion zone temperature  and oxy-
 gen availability.   NOX versus heat release  rate in the burner zone of
 the furnace  (HA/SC) for various excess air  levels is plotted in Figure
 5.  The correlations were based on actual field data from the 250MW
   NOx - ppm v at 3% 02

     800
     700
     600
     500
     400
     300
     200
     100
                                        Circular burner
                                         base line data
                                         at 115% air'
                                         to burners
                          I
          I
I
                                      High excess
                                       air, 45%


                                      Moderate excess
                                         air, 30%

                                      Normal excess
                                         air, 15%

                                      Low excess
                                        air, 7%
                100       200      300      400
                             HA/Sc - MBtu/hr-ft2
                           500
                 600
                          Field test results
                                   Figure 5

-------
                        PULVERIZED  COAL BURNER                      191


bituminous fired unit.  Statistical analysis was used to develop the
linear best fit curves from the test data.

      The effect of furnace zone temperature is illustrated by the in-
crease in NOx for higher heat release rates.  The WOX increases 80 ppm
per 100 MBtu/hr-ft2-  For a given unit, the larger the HA/SC, the higher
the temperature will be in the burner zone of the furnace.  NOX genera-
tion attributable to thermal fixation will increase per the relationship.
Since the thermal WOx formation is extremely temperature sensitive at
high temperatures, the shaded areas on Figure 5 represent the degree
of uncertainty at the higher heat release rates.  As the burner zone
temperature approaches the threshold temperature, the rate of thermal
WOX fixation rapidly increases.

      The formation of HOX also requires oxygen availability and the
greater the excess oxygen, the higher the KOX generated.  NOX emissions
versus excess air quantities ranging from 7% to ^5% is plotted in
Figure 5-  Actual field data indicates NOX emissions increase by 20 -
25 ppm per 5% increase in excess air.  Utilization of the Dual Register
Burner coupled with the compartmented windbox has provided proper con-
trol of air and fuel mixing which minimizes air requirements thus
limiting NOX formation.
 SUMMARY

      Results from the three  (3) pulverized coal units tested to date
 show the Dual Register Burner to be an effective tool for NOX emission
 control.  NOX levels ^0 to 50% lower than those achievable with the
 high turbulence Circular Burner are attained through limited turbulence
 Combustion.  At the same time, carbon utilization has been maintained
 at levels comparable to those obtained utilizing a high turbulence
 Circular Burner.  The Dual Register Burner provides low NOX emission
 without sacrificing unit efficiency.

      The data acquisition and analysis program as reported herein will
 continue to provide data to  update and improve future units.  It is
 expected that by the end of 1977 KOX and performance data will be
 available from an additional four  (U) units now in the final stages of
 start-up.  These tests will include units firing both bituminous and
 sub-bituminous grade coals.
REFERENCE
      1.  Barsin, J.A. and Brackett, C.E., i:The Dual Register Pulverized
Coal Burner", presented at the Electric Power Research Institute NOX
Control Technology Seminar, San Francisco, California, February, 1976.

-------
192                     CLEAN COMBUSTION OF COAL

-------
                                                                    193
          COAL-OIL MIXTURE COMBUSTION IN BOILERS - AN UPDATE
                                  By
                             Sushil K. Batra
Manager of Energy Conversion Research, New England Power Service Co.

                            1.  INTRODUCTION
     As a result of unavailability and increased cost of fuel oil for
power plants, the nation faces large capital outlays in converting the
existing oil-fired steam generating capacities to coal burning.  Also,
power plants that have been converted from coal to oil firing would be
converted "back to coal burning.  At some plants, however, it may not be
practical to convert to coal burning due to lack of coal storage/coal
handling facilities or high capital cost and short remaining life of
the units.  An alternative to the above could be to burn pulverized
coal-oil mixture (COM).  This will be a near-term alternative to mini-
mize the expense and complexities of total conversion, while at the
same time reducing the dependency on foreign oil.

     In general, by making use of coal-oil mixture, a utility will be
able to take advantage of relatively low prices and more stable supply
of this country's coal reserves.  Coal-oil mixture will provide an in-
terim solution for short-term coal utilization, before new technologies
such as coal gasification, coal liquefaction, and solvent refined coal
become commercially available.  Another extremely attractive feature
of the coal-oil mixture combustion concept is that an extensive new
technology doesn't need to be developed before application of coal-oil
mixture.  The fuel could be made commercially available within two to
four years, with a minimum requirement for new equipment and technology.

     The paper presented is devoted to:  l) discussion of past studies
on coal-oil mixture stability; 2) a brief discussion on experimental
programs of combustion of coal-oil mixture at Pittsburgh Energy Re-
search Center (PERC) and Institute of Gas Technology (IGT); and 3} a
review of coal-oil mixture development programs at General Motors and
New England Power Service Company.

                         2'.  HISTORICAL BACKGROUND

     Since 18791, several attempts have been made to burn coal-oil
mixture .   Combustion of coal-oil slurry was first tried in ocean-
going vessels and locomotives to save bunker space and minimize the
risk of fire in the storage of coal.  In 19322, the Cunard liner
Scythia steamed from Liverpool to New York and back with one boiler
fired on a coal-oil mixture.  However, the principal disadvantage of
such a fuel was the difficulty of maintaining the coal in suspension.
During World War II, the research in coal-oil mixture was again re-
sumed in Japan, Germany, and in this country, and burning of coal-oil
mixture was tried on a small scale.  However, there has been no large-
scale application of this technology.  More recently, several programs

-------
194
CLEM COMBUSTION OF COAL
have been initiated in this country funded by ERDA or by private funds
for development of coal-oil mixture technology.

                  3.  COAL-OIL MIXTURE PREPARATION

     Coal-oil mixture can be prepared by simply mixing finely ground
coal in No. 6 residual oil and providing continuous agitation to pre-
vent settling.  Alternatively, a stabilized coal-oil mixture may be
prepared by adding a chemical additive to a mixture of pulverized coal
and No. 6 residual fuel oil with or without water.  A stable mixture is
essential for good flame stability and also to keep the coal from set-
tling in tanks, heat exchangers, valves and pipelines.  Evidently, both
of these systems have their own merits.  Whereas it may be possible  to
eliminate continuous agitation by use of chemical additives, current
cost of these additives is very high.  Further research and development
needs to be done to investigate possibilities of less expensive  addi-
tives that can be used for stabilization of coal-oil mixture.

            4.  STABILITY AND VISCOSITY OF COAL-OIL MIXTURE

     The stability of coal-oil mixture  has been studied by various  re-
searchers in the past.  In the laboratory work reported in 1944  by
Barkley^, et al., it was observed that the stability of the mixture  in-
creased considerably with increase in percentage of coal in coal-oil
mixture.  Also, the effect of temperature, as well as particle size  on
stability of the mixture, was studied.  Figure 1 shows the effect of
percentage of coal in mixture on stability at 77°F.  It was observed
         80
         100
                                  6>O      80

                                AGE, DAYS
                          too
120
    Figure 1.  Effect of percentage of  coal in mixture  on  stability
               at 77F, Ref. 3.

-------
                     COAL-OIL MIXTURE COMBUSTION
195
that the mixtures are relatively stable when the maximum particle size
does not exceed 230 mesh.  Furthermore, the distribution of the size
of coal affected the stability of the mixture considerably.  An addi-
tion of even small quantities of coarser materials in the coal-oil mix-
ture adversely affected the stability of the mixture.

     The viscosity of the mixture increases markedly in the region of
40 to 45 percent of coal by weight (Figure 2).  Also, the rapid rise
                               20    30   40   50
                             COAL IN SUSPENSION
              Figure 2.  Effect of percentage of coal in mix-
                         ture on viscosity at 77F,  Ref.  3-

in viscosity in this range was found to be relatively independent  of
the viscosity of the oil and temperature of the mixture.  A rapid  in-
crease in viscosity was also observed when the coal size was reduced
from 230 mesh to 230/325 mesh.

     In view of the results reported above, from the point  of  view of
stability and viscosity of the mixture,  the amount of coal in the mix-
ture should be limited to 40 percent, and the coal size to  98  percent
passing through 230 mesh screen.  For practical purposes and to reduce
grinding and pumping costs, however, it is desirable to limit  the  per-
centage of coal in the coal-oil mixture to 30 percent, and  coal size

-------
196
CLEAN COMBUSTION OF COAL
 to 80 percent passing through 200 mesh.

              5.   COAL-OIL MIXTURE COMBUSTION DEMONSTRATIONS

      The combustion of coal-oil mixture  has recently been investigated
 at Pittsburgh Energy Research Center and at General Motors facility in
 Saginaw,  Michigan.   The later research was sponsored by ERDA, EPRI^,
 General Motors,  and more than 15 companies including New England Power
 Service Company.  Besides these investigations for application of coal-
 oil mixture to package boilers, ERDA is  partially funding the develop-
 ment and demonstration of coal-oil mixture combustion at two utility
 plants:   namely,  New England Power Company's 80 MW unit at Salem Harbor
 Station;  and City Public Service of San  Antonio's 69 MW unit.

               6.   EXPERIMENTAL PROGRAM  AT PERC

      Coal-oil mixture combustion studies were conducted at Pittsburgh
 Energy Research  Center5 of ERDA in June  1975.   The principal objectives
 of these studies  were to determine major problem areas while firing 20%
 coal-oil mixture  in a package boiler designed to burn oil or gas and to
 determine corrosion and deposits after 1,000 hours of operation.

      A new burner designed for No.  6 fuel oil was installed in a sur-
 plus 100 hp 4 pass  Cleaver-Brooks fire tube boiler and a coal-oil mix-
 ture preparation  and feed system was assembled.   Figure 3 shows a sim-
                                           STEAM   x*-V*

                                                T^}
                                                /^v
                                                SLURRY
                                              TRANSFER PUMP
                                               FEED WATER PUMP
    Figure 3.  Simplified flow diagram of the 100 H.P. coal-oil mix-
               ture combustion test facility, Ref. 5-

plified flow diagram of the 100 H.P. unit.  This system was intention-
ally assembled utilizing existing or readily available equipment  to as-

-------
                     COAL-OIL MIXTURE COMBUSTION                    197


certain the minimum equipment changes necessary to adapt an oil-fired
boiler to coal-oil mixture combustion.  Modifications and/or improve-
ments were made in control and instrumentation as required during the
experimental program.

     In general, the program concluded that no slagging was experienced
in the "boiler, and carbon combustion efficiency was more than 99%.
Boiler efficiency tests indicated that the efficiency obtained with
coal-oil mixture combustion was essentially the same as that obtained
with fuel oil.  Some char deposition was experienced when firing at a
high rate due to an excessively wide spray pattern that produced flame
impingement on the relatively narrow combustion chamber.  This became
progressively worse with time, as the orifices in the brass nozzle be-
came enlarged due to erosion and the atomization pattern was changed.
This problem was eliminated by modifying the combustion air diffuser to
obtain the desired flow pattern and fabricating nozzles of stainless
steel with a narrower spray pattern.  Inspection of the boiler tubes
after the completion of the 1,000-hour test revealed only a light de-
posit on the internal tube surfaces.  The carbon monoxide concentration
in the flue gas was slightly higher than that obtained with oil (260 vs.
160 ppm) but was acceptably low.  Similarly, the NOX emissions were
also slightly higher reflecting the higher fuel bound nitrogen content
of the coal-oil slurry.  Sections of tubes were taken from within the
passes of the boiler to determine the effect of corrosion/erosion.  No
evidence of severe corrosion attack was observed.  Also, there was no
evidence of inter-granular or subscale corrosion attack on any of the
test specimens.

               7.  COMBUSTION TESTS AT GENERAL MOTORS FACILITY

     In 197-4, General Motors  had a three-day, 10,000-gallon pilot test
using a mixture of No. 2 fuel oil and pulverized coal delivered to
their plant in Saginaw, Michigan.  Some of the technical problems en-
countered in that demonstration were settling of the coal out of the
mixture, increased viscosity of the mixture, and problems associated
with coal-oil mixture preparation techniques.  A development program
was initiated to solve these problems, and to burn coal-oil mixture
with No. 6 oil in existing oil-fired package boilers.  In March of
1975, a consortium of companies and organizations was established.
This included U. S. Energy Research and Development Administration,
Electric Power Research Institute, several utilities including New
England Power Service Company, coal suppliers, and chemical manufactur-
ers.   Preliminary work was conducted to gain specific information in
regard to fuel-oil and coal selection, fuel blending, fuel characteri-
zation, fuel atomization, and optimization of combustion before pro-
ceeding to extensive tests.  The results in general indicated that
coal-oil mixture  should perform as well as residual fuel oil using
conventional equipment.

                      8.  COMBUSTION TESTS AT IGT

     Detailed combustion tests were carried out at the Institute of Gas
Technology,  Chicago,  to identify possible operational problem areas and
define combustion characteristics for coal-oil mixture.   These tests
involved the use of steam and air atomized burners on coal-oil mixture

-------
198                    CLEM COMBUSTION OF COAL
in a rectangular furnace to investigate flame stability, geometry,
emissivity and NOX emissions.  The results of these tests indicated
that the combustion flame produced from coal-oil mixture  (30-50 weight
percent coal) using conventional atomizing nozzles was stable with good
geometry.  The flame was also very similar to the flame produced by
straight No. 6 fuel oil.  Good carbon burnout within the test chamber
was noted.  As expected, the particulate emissions were higher than
for oil only.  The coal-oil mixture's flame emissivities were the same
as or higher than the emissivity values obtained from residual fuel oil.
Furthermore, it was observed that the presence of 5$ water in the fuel
did not degrade combustion, and in fact appeared to enhance the atomi-
zation.

            9.  DEMONSTRATION AT GENERAL MOTORS FACILITY

     Tests were carried on combustion of coal-oil mixture in a packaged
oil-fired boiler rated at 120,000 pounds per hour steam capacity
installed at the Chevrolet Nodular Iron Plant Power House in Saginaw,
Michigan.  A summary of the boiler specifications is given in Table 1.

                                TABLE 1

                      Test Boiler Specifications

     Plant Boiler No.	5
     Year Installed	1966
     Boiler Manufacturer	CE-Wickes
     Type	A
     Pressure Standard	250
     Rated Capacity (PPH)	120,000
     No. of Drums	3
     Heating Surface (Ft.2)	8,897
     Furnace Heating Surface (Ft.2)	1,073
     Furnace Volume (Cu. Ft.)	2,180
     Furnace Width	7' 8-1/2"
     Furnace Length	30'  8-3/4"
     Gas Passes	Horiz.
     Temp. Flue Gas Leaving Boiler (°F)	505
     Number of Burners	2
     Air Preheater	Yes
       Gas Temp. Entering (°F)	505
       Gas Temp. Leaving (°F)	353
       Air Temp. Entering (°F)	134.
       Air Temp. Leaving (°F)	309
       Heating Surface (Ft.2)	1,400
     Stack Control Devices	None

This work has been carried out in two phases:  Phase I,  March to June
1976;  Phase II, March to May 1977.  The test boiler has a natural draft
with no flue gas cleaning equipment.   Also, the boiler has no provision
for ash removal.  A lance-type traveling soot-blower was mounted in
the boiler floor to cover the entire length of the boiler, for blowing
steam to either re-entrain deposited ash or push ash to the rear of the
boiler for removal.  The ash collected in the ash hopper at the rear

-------
                     COAL-OIL MIXTURE COMBUSTION                    199
of the boiler could be sluiced out.  (This system was, however, not  ,
used in actual operation, since there was very little ash deposition
in the radiative section of the boiler.)

                            9.1  Phase I

     During the period May to June 1976, the boiler was operated for
250 hours, burning a mixture of 35 weight percent coal, 59 weight per-
cent oil, and 6 weight percent emulsifier solution.  (3/4 of 1% emulsi-
fier and 5 1/4$ water).  The steam load ranged from 45,000 to 70,000
Ibs. per hour, with occasional loads of 96,000 Ibs. per hour.  Two
types of burners were tested during this period:  namely steam atomized
burner and air atomized burner.  With both types of burners, the flame
envelop was as good or better than with No. 6 oil.  Although the fire
appeared less brilliant with coal-oil mixture, no sparklers were no-
ticeable.  In general, the air atomized burners gave a better perfor-
mance than the steam atomized burners, and both burners could be
switched from coal-oil mixture to No. 6 oil, with minor air register
adjustments.  The only major problem associated with the steam atomized
burners was the deposition of carbonacous materials on the tip of the
burners.  The burners had, therefore, greater tendency to plug than the
air atomized burners and had to be cleaned more frequently.

     Stack testing was conducted to measure NOX, CO, and opacity over
a wide range of oxygen levels.  The stack had a gray smoke with 32-35$
average opacity during the test.  The opacity increased considerably
during soot blowing periods and was unacceptable.  Fly ash participates
samples were drawn and analyzed.  The fly ash size was very small and
almost all of the material was less than 20 microns in size.  The NOX
levels increased approximately 100-150 ppm for coal-oil mixture com-
bustion over burning No. 6 oil.  This can be somewhat attributed to
the higher nitrogen level in coal.  Carbon monoxide levels were, how-
ever, almost the same in both cases (Table 2).

                                TABLE 2

                      Air Atomized Burner Performance
                                                         Coal-Oil
                                        No. 6 Oil        Mixture

       Steam flow rate (ib/h)           40-45,000        40-45,000
       Fuel pressure (psig)                116              140
       Oxygen (%}                       5.2-8.0          4-1-7.6
       NO (ppm)                         155-198          265-328
       CO (ppm)                           6-33             4-25
       Particulate emissions (gr/acf)   .0153-.0188      .373-.472
       Opacity (%}                        0-8             32-35

                              9.2  Phase II

       During this phase, which ended April 30, 1977, a mixture con-
taining 50 weight percent coal, 43.3 weight percent oil, 6.5 weight
percent water, and 0.2 weight percent additive was prepared and burned.
Tests were carried out to determine emissions, combustion conditions,
and coal-oil mixture stability characteristics.  The boiler fired coal-

-------
200
CLEAN COMBUSTION OF COAL
 oil  mixture for 494 hours at a range of loads, almost twenty percent
 of this  time was at steaming rates over 70,000 Ibs. per hour, with the
 highest  being 100,000 Ibs.  per hour.  As in the case with Phase I, it
 was  observed that coal-oil  mixtures burn in a manner very similar to
 No.  6 fuel  oil.  Also,  it verified many of the observations that had
 been made earlier in regard to flame characterization, boiler perfor-
 mance, ash  build-up,  slagging and trouble-free operation of the air
 atomizing burners.

       >In this phase of the program, a new additive was used to improve
 the  stability of coal-oil mixtures.   A high pressure homogenizer was
 used to  make an oil-additive-water emulsion.  The oil-water emulsion
 thus obtained was mixed with pulverized coal in a Marion mixer.

       During the period of the test,  there was some settling of coal
 from the coal-oil mixture in the storage tank.  This was caused mainly
 by high  temperature of the  coal-oil mixture.  This affected the vis-
 cosity as well as the stability of the coal-oil mixture.  This problem
 can, however,  be alleviated by controlling the temperature of the
 mixture  in  the storage tank to below 150°F.

       As in Phase  I of the program, the steam atomized nozzle was
 installed initially.   This  system performed satisfactorily as the per-
 centage  of  coal in the mixture was increased gradually from 10$ to 40$
 coal.  However, at  50$ coal in the mixture,  the nozzle plugged, and
 the  remaining tests were carried out with air-atomized burners manu-
 factured by Forney  Engineering Company.   A schematic of the burner head
 is given in Figure  4.   In general,  the air-atomized burner performed
 satisfactorily.
                              SECOND-STAGE
                  OUTER CONE.  SWIRL CHAMBER
               INNER CONE
                              FIRST-STAGE
                              SWIRL CHAMBER
                                                           AIRFLOW
                    SPIRAL ATOMIZER

                             CONTROL SLEEVE
                                                           AIRFLOW
                             SWIRL CONTROL
                              MECHANISM
    Figure  4.   Burner head of air atomized Verloop burner.

-------
                    COAL-OIL MIXTURE COMBUSTION
201
              10.   DEMONSTRATION AT SALEM HARBOR STATION

       New England Power Service Company has been awarded a cost-
sharing contract by the Energy Research and Development Administration
(ERDA)  to develop and demonstrate a coal-oil mixture as a primary fuel
for generating electricity.  In this program, the overall feasibility
of coal-oil mixture combustion will be demonstrated in an 80 MW unit
at Salem Harbor Station in Salem, Massachusetts.  This boiler was ori-
ginally designed for burning pulverized coal, but has been converted
to and is presently burning residual oil.  The specific objectives of
the project are to achieve a stable combustion with coal-oil mixture
                                   R
                                   111
                            SECONDARY
                            SUPERHEATER
   Figure 5.  Sectional view  of  80 MW Babcoek and Wilcox boiler.

-------
202
GLEAM COMBUSTION OF COAL
fuel and to demonstrate satisfactory performance of all systems and
sub-systems, while meeting state and federal environmental regulations.
The target concentration' of coal in coal-oil mixture is 30 weight per-
cent.   Initial tests in the program are planned to be conducted with
straight coal-oil mixture without the use of a chemical additive for
stabilization.  Concurrently, however, research and development is
being conducted for development of an effective and economical stabi-
lizing  additive.  Tests will be carried out with the use of this ad-
ditive  to determine its effect on coal-oil mixture combustion.

        The test boiler (Figure 5) is a Babcock and Wilcox front-fired
radiant boiler rated at a steam flow of 625,000 Ibs./hour with a super-
heat/reheat temperature of 1,OOOF and a design pressure of 1,675 psi.
Presently, there are 12 mechanical atomizing high pressure burners that
supply  a full load heat input of 880 million Btu/hr.  Particulate re-
moval from the flue gases is accomplished by means of a Research Cot-
trell electrostatic precipitator.  The design collection efficiency of
the precipitator is 97% when burning coal.

        Several modifications and additions are planned for this demon-
stration program.  Figure 6 shows a simplified flow diagram for the
                                         NO. 6 OIL

                                         t
COAL
UNLOADING



HOPPER*
FEEDER
»
PULVERIZER
•
PREWETTING
4
WATER
«••
V
BLENDING

H EMULSIFIER |
                         COM FEED
        COOLING AIR FAN A
   Figure 6.  Simplified flow diagram of coal-oil mixture combustion
              demonstration.
various systems and their interfaces.  Coal will be delivered to the
dock by barge and unloaded into a pile adjacent to an existing conveyor
belt.  Coal will be pushed into hoppers which feed the enclosed conveyor
to carry coal into storage bins, inside the power plant.  The bins will
gradually be emptied into coal pulverizing equipment and the finely
ground coal will be delivered to a storage silo to be constructed ad-
jacent to the south wall of the power plant.  The pulverized coal will
then be blended with oiland pumped to an existing fuel storage tank
for distribution to the burners.  The existing high pressure burners
and control system will be replaced with low pressure air atomized
Verloop burners and the associated burner control and logic system.

-------
                    COAL-OIL MIXTURE COMBUSTION                    203
As indicated in the schematic diagram, provision will be made for
prewetting and addition of emulsifier additive to improve the stability
of the coal-oil mixture, if necessary.

       The test procedure is divided into two phases, each containing
incremental test steps, in order to debug the various systems and to
determine the effects of coal-oil mixture combustion on the boiler,
precipitator, and ash system, as well as on stack emissions.

                     10.1  Phase I - Feasibility

       The purpose of this phase is to develop combustion stability and
evaluate the effects of coal-oil mixture combustion on the boiler, pre-
cipitator, auxiliary systems and stack emissions.

       The initial operation will involve baseline testing of the Ver-
loop burners on No. 6 oil.  This testing will be used to debug the new
system and to obtain baseline data for later comparison with the coal-
oil mixture combustion results.  Initially, coal-oil mixture will be
burned in a single row (three burners) with remaining nine burners
burning No. 6 oil.  Tests will be done with mixtures of 10$, 20$, and
30$ by weight coal concentration and will evaluate flame stability,
emissivity, opacity, and burner performance, etc.  Once the combustion
stability is demonstrated, the next step in this test sequence will be
a six-burner test, supplying half of the boiler heat input from coal-
oil mixture fuel.  At this time the effects of coal-oil mixture on the
equipment, pumps, piping systems, instrumentation and burner system
will be closely monitored.  Potential problem areas that will be moni-
tored  are slagging in the furnace and the convection pass, plugging of
the air preheater, effects of larger quantities of coal ash on the ash
water  recycle system and stack emissions.

                        10.2  Phase II - Demonstration

       In this phase of the program, all twelve burners will burn coal-
oil mixture, thus 100$ of the boiler heat input will be supplied by
the coal-oil mixture.  Throughout the test period, close attention will
be given to environmental aspects of burning coal-oil mixture  as well
as the effects of coal-oil mixture combustion on erosion, corrosion,
and boiler performance will be monitored.  Tests will be carried out to
optimize performance of various systems and subsystems as well as to
determine the efficiency of boiler and precipitators, while burning
coal-oil mixture.

                            10.3  Schedule

       It is presently anticipated to complete the installation and
construction of the coal-oil mixture preparation, blending and com-
bustion facility in the middle of 1978.  The Phase I of the program is
scheduled for six months duration.  After the Phase I results have
been evaluated, a decision will be made to proceed into Phase II of the
program.  It is anticipated that this phase will begin in early 1979
and run through the conclusion of the project scheduled for the end of
1979-

-------
204                    CLEM COMBUSTION OF COAL
                          11.  POTENTIAL  BENEFITS

       There are a number of potential benefits  to  the  United States if
 coal-oil mixture combustion feasibility can be demonstrated.   The most
 obvious advantage is a reduced dependency on  imported oil  if  30-40$ of
 the  fuel burned becomes  domestically produced in the coal-oil mixture.
 Combustion  of  coal-oil mixture will, evidently,  have less  impact on
 air  quality than 100% coal burning.  It has potentially a  wide area of
 application on boilers originally designed to burn  coal or oil but can-
 not  burn coal  in view of changed circumstances.  Although  some further
 technical development is desirable, coal-oil  mixture as a  fuel has a
 potential for  a near-term alternative for utilities and industrial boi-
 ler  applications.

       Furthermore, a central coal-oil mixture preparation plant can
 be visualized  that prepares and ships a stabilized  coal-oil mixture to
 outlying power plants.   Small users could buy such  fuel and presumably
 interchange coal-oil mixture and oil with a minimum of  turn-around and
 (•rpense.  Large users of coal could probably  justify their own prepara-
 tion facilities.

                                ACKNOWLEDGEMENT
        The  author  gratefully  acknowledges the assistance of Mr. Andrew
 Brown,  Jr.,  Project Manager,  General Motors Corporation, for providing
 the results  of  the General Motors demonstration tests; and Mr. Richard
 M.  Dunn,  for his contribution related to the coal-oil mixture project
 at  Salem  Harbor Station.

                                REFERENCES

 1.   Smith, H.R. and H.M. Hunsell, Liquid Fuel, U.S. Patent 219,181
      (February 24, 1879)

 2.   Manning,  A.B.  and R.A.A.  Taylor, "Colloidal Fuel", Trans. Inst.
      Chem.  Engrg., 14, 45 (1936).

 3.   Barkley,  J.F., et al., "Laboratory and Field Tests on Coal-in-Oil
      Fuels", Trans. ASME, 66, 185 (1944).

 4.   Giiman,H.H., "Coal-oil Emulsions for Boiler Fuel", EPRI Journal 2,
      56, April 1977.                                               ~

 5.   Demeter,  J.J., et al., "Combustion of Coal-oil Slurry in a 100 H.P.
      Fire Tube Boiler" Pittsburgh Energy Research Center.  Report
      PERC/RI-77/8, May 1977.

 6.   Brown, A. Jr., "General Motors Powdered Coal-oil Project", pre-
      sented  at Fuel Switching Forum U.S. ERDA, June 6-7, 1977.

 7.  Annon, "Proceedings of the Coal-oil Mixture Combustion Technology
      Exchange Workshop", U.S. ERDA Washington, D.C., CONF. 767019 -
      M77-8,  February 1977.

-------
                                                                    205
                   STOKERS FOR INDUSTRIAL BOILERS:
                 ASSESSMENT OF TECHNICAL, ECONOMIC,
                     AND ENVIRONMENTAL FACTORS
                          Robert D. Giammar*
                              Battelle,
                        Columbus Laboratories
                        Columbus, Ohio  ^3201
                             INTRODUCTION
          The firing of boilers for process steam, space heating, and on-
site power generation accounts for about half of all the fuel used by
American industry—nearly as much as that consumed in the generation of
electricity by utilities.  Coal once was the dominant fuel for such in-
dustrial boilers; it now provides only about one-quarter of their fuel.

          Economic and environmental factors over the past 30 years have
made coal relatively unattractive for industrial boilers.  Both the capital
and operating costs of coal-fired boiler installations are inherently
higher than those of equivalent facilities designed for gas and/or oil
firing.  Thus, with the widespread availability of oil and gas in the
late 1940's and early 1950's, there was a drastic decline in the demand
for the industrial stoker.  Only in some of the largest boiler installa-
tions could the savings resulting from the somewhat lower cost of coal
offset the greater capital and operating costs.  As a consequence, there
has been little incentive for manufacturers to improve stoker technology
through research and development.  However, a resurgence of demand can
be forecast for the near term in view of the fuel supply situation and
possible legislative action.

          Stoker-fired boilers are now at a further disadvantage with
respect to environmental considerations, as their emissions of particulate
and SO- are high in comparison to gas- and oil-fired boilers.  The estab-
lishment of stringent air-pollution regulations in recent years (with
national standards for large new installations and local standards for
many other installations) has accentuated the economic disadvantages of
stoker boilers.  Costs of installing and operating the downstream con-
trol equipment that is needed to control emissions of fly ash and S02
for industrial boilers are relatively higher on a Btu basis than for
larger utility power plants.

          The potential market for stokers is dependent on many factors,
some of which are difficult to assess because they involve both public

* The author wishes to acknowledge the guidance and assistance given by
  Battelle staff—R. E. Barrett, D. W. Locklin, and R. B. Engdahl.

-------
206                    CLEAN COMBUSTION OF COAL


and industry acceptance of stoker firing.   The industrial boiler
(typically with rates steam-generating capacity between 10,000 and 500,000
pph*) is primarily utilized to generate steam for process and space
heating and often is subjected to widely varying loads.  Unlike the
utility boiler, the industrial boiler is usually a secondary consideration
in an industrial complex.  The characteristics sought include:

            Low first costs
            Low maintenance
            High reliability
            Ease of operation
            Minor auxiliary equipment requirements
            Wide turndown
            Quick response.

          Most of these features are more easily obtained with the oil-
or gas-fired boiler.  Recently, however, the availability of oil and gas
has become an increasingly important factor in the selection of boilers
and may outweigh all other factors if the shortage of these fuels becomes
as severe as expected.  Thus, the stoker-fired boiler will become more
popular in the near term until coal conversion processes such as coal
desulfurization, gasification, or liquefaction are commonly available
or advanced coal-burning techniques such as fluidized-bed combustion
reach the economically competitive stage.
                   CHARACTERIZATION OF STOKER TYPES
          Stokers of various types are designed to mechanically feed coal
uniformly onto a grate or tuyeres within a furnace, to supply combustion
air to the fuel bed, and to remove ash from the zone of combustion.  The
development of the mechanical stoker has advanced through the years,
and the modern stoker is still considered an important element in the
industrial combustion of coal as well as other types of solid fuel.  The
modern stoker-boiler system incorporates controls to coordinate air and
fuel supply with changing loads, dust-collecting equipment to minimize
emissions, and, in many cases, fly-carbon return systems to increase
efficiency.

          Stokers are classified according to the method of feeding fuel
to the furnace, namely:

          • Spreader
          • Underfeed
          • Chain grate or traveling grate
          • Vibrating grate.

          The type of stoker used in a given installation depends upon
*  pph = pounds steam generated per hour; one Ib/hr - approximately
   1000 Btu/hr.

-------
                     STOKERS FOR INDUSTRIAL BOILERS                 207
the general system design, the capacity required, and the type of fuel
burned.  In general, the spreader stoker is the most widely used in the
capacity range of 75,000 to 400,000 pph because it responds quickly to
load changes and can burn a wide range of coals.  The underfeed stokers
are principally utilized with small industrial boilers of less than
30,000 pph.  In the intermediate range the large underfeed units, as well
as the chain- and traveling-grate stokers, are being displaced by spreader
and vibrating-grate stokers.  A brief description of each class of stoker
is given below.  The major features of each are summarized in Table 1.
Detailed discussions of the various stoker types can be found in the
37th and 38th editions of Steam, Its Generation and Use, published by
the Bab cock & Wilcox Company, New
                 TRENDS IN COAL FIRING AND STOKER TYPES
          Recent Battelle-Columbus studies directed to the characteriza-
 tion of the industrial boiler population provide some insight as to trends
 in  coal-firing methods by boiler size range.  Information on the indus-
 trial boiler population and design trends was developed in a special
 survey of boiler manufacturers conducted jointly by Battelle and the
 American Boiler Manufacturer's Association (ABMA) and reported in
 Reference (2).  These results were later refined on the basis of an
 analysis of recent sales data for industrial-size water-tube boilers'^).

          Firing methods are boiler-size dependent.  Thus, different size
 categories were defined as follows for the presentation of trend information:

                Size Category        Size Range, 1000 pph

                      A                    10-16
                      B                    17 - 100
                      C                   101 - 250
                      D                   251 - 500

 Smaller boilers are placed in the "commercial" class and larger boilers
 are considered to be "utility boilers".

                    Trends in Coal-Firing Capability

          Figure 1 shows trends in coal-firing capability estimated for
 boilers in the four size categories  installed in 1930, 1950, and 1970
 and forecasted for 1990^'.  The data apply to boilers designed to fire
 coal or having a capability to fire  coal as a secondary fuel.  This dis-
 tribution is shown as a percentage of all industrial boilers.  The pro-
 jection for 1990 was revised by Battelle from the earlier survey on the
 basis of the following broad assumptions:

          • Oil and gas supplies will be limited.
          • Oil and gas will be utilized in smaller equipment and
            for high-priority uses (but not fully by mandatory
            allocation).
          • Coal will dominate new installations in the larger sizes.

-------
                    TABLE 1.  CHARACTERISTICS OF VARIOUS TYPES OF STOKERS
                                                                                                   to
                                                                                                   o
                                                                                                   00
  Stoker Type
  and Subclass
     Typical
 Capacity Range,    Maximum Burning Rate,
     pph(a)                Btu/hr-ft2
                                Characteristics
Spreader
  Stationary and
   dumping grate
  Traveling grate

  Vibrating grate

Underfeed
  Single or double
   retort
  Multiple retort

Chain grate and
  traveling grate
Vibrating grate
 20, 000 to 80, 000

100,000 to 400, 000
 20,000 to 100, 000


 20,000 to 30, 000

 30,000 to 500,000

 20,000 to 100,000
 30, 000 to  150,000
 450,000

 750,000
 400,000


 400,000

 600,000

 500,000
400, 000
Capable of burning a wide range of coals,
 best ability to follow fluctuating loads,
 high fly-ash carryover, low load smoke
Capable of burning caking coals and a wide
  range of coals (including anthracite), high
  maintenance, low fly-ash carryover,
  suitable for continuous-load operation

Characteristics similar to vibrating-
  grate stokers except these stokers
  experience difficulty in burning strongly
  caking  coals

Low maintenance, low fly-ash carryover,
  capable of burning wide variety of
  weakly caking coals, smokeless
  operation over entire range
                                                      o
                                                      o
                                                      O
CO
i-3
M
O
o
o
(a) pph = Ib steam per hr; 1 pph=*1000 Btu/hr.

-------
 rj       	  -Jr
 Q.      O  W
 o   _-  *r  w
O   o  2  i-
     0*^0



I   e  °  c
 O   iT  C  —
X       <•>  T3
 ^   O  if  «

^   *-  «  o

 o           
CD           C
uu

80

60
40

?n


r\


—
/ '
/ /
//'
+ / 1
it
-/ /
f
/ /
/ /







>---|
E-drm
'30 '50 70
A







v \

^ \ s
t\ \
'90

10.000-16,000 pph












' /
f .
f/t
f / 1
; >

' ,
/ /
f /









'30








—
- « —
—
—
'50











i : ! i
'70
B






N X
^ \ '
„ \ ^
V •
^ \
. \ N
'90











/ ,
' / 1
' t
'//
/ /
/ 1
' / f
f
' /
/ /










'30





_ 	 .
	
---
_ 	 _r
	
	
'50



















* i
i i
i i
i i
i i
70
C
17,000- 100.000 pph
101,000
r ~i


1

1 t







;\\
V \ ^
s \ N

^ \ ^
S \ N
S \







/ /

' / i
'/,
/ / 1
f/'

' t
' 1
f /










:"-":
—
—
—
	 	
—
—


T





'90 '30 '50 70
D
1 1











\ "^
^
\ s
v \ N
x\X
\ \
s N S
^ \











CQ
8
Q

O
H
a
CO
»
M
'90 P
td
O
-250.000 pph 251. 000-500.000 pph y
Figure 1. Estimated Percentage of all Boilers in Four
in 1930, 1950, and 1970 and Forecasted for
Fire




Coal*

















Size Categories Installed
1990






Having a Capability to





N)
O
IO

-------
210                    CLEAN COMBUSTION OF COAL


          • Clean liquid and gaseous fuels from coal conversion
            processes will not be available in large quantities
            by 1990.
          • Firing refuse as a supplementary fuel will increase,
            but will not become significant in terms of percentages.

The coal projection includes the firing of chemically refined solid fuels,
but does not include future synthetic liquid or gaseous fuels derived
from coal.

          The trends in Figure 1 indicate that coal was the dominant fuel
for industrial boilers in the pre-World War II era and that most boilers
had a coal-firing capability.  By 1950, the number of units with a coal-
firing capability dropped dramatically for boilers being installed in the
smaller size categories; in the largest size category, 65 percent of the
boilers had the capability.  By 1970, coal firing nearly disappeared from
the smaller boiler size categories, but 40 percent of the largest boilers
were being installed with a capability to fire coal.

          For the future, a return to the capability for direct firing
of coal was forecast for new installations in the next 15 years, especially
in larger sizes, where about 70 percent of the new boilers are expected
to have coal-fire capability.  In the event of new legislation and/or
fuel priority allocations that prohibit firing gas and oil in large
boilers, all new industrial boilers may be required to have a coal
capability.  Some large industrial companies have already made the
decision to provide for a coal capability in new installations as a
measure of insurance to keep manufacturing plants operating in the face
of an uncertain fuels situation.

                         Trends in Stoker Types

          Figure 2 shows trends in coal-firing methods for industrial
boilers in each of the four size categories as developed in the studies
cited earlier(•*).  Firing methods included in these data are:

          • Stoker types

            Spreader
            Underfeed
            Overfeed (chain grate, traveling grate, or vibrating grate)

          • Other

            Pulverized coal
            Miscellaneous (cyclone, refuse, sawdust, wood, etc.).

The following overall observations can be made from Figure 2:

          • Before the introduction and commercialization of the
            spreader stoker in the 1930's, underfeed stokers
            dominated the smaller size categories and had an
            appreciable share of the market in the larger sizes,
            The overfeed types were significant in all sizes,
            especially larger sizes.  Firing of pulverized coal
            was significant only in the largest sizes.

-------
             Overfeed
  Size
         30  '50   '70  '90
                A
category  10.000 - 16.000 pph
Underfeed
Spreader
Pulverized
coal
 '30   '50  '70  '90
         B
  17,000- 100.000 pph
  '30  '50  '70  '90
         C
  101,000-250,000 pph
Other
    '30   '50  '70  '90
            D
    251,000-500,000 pph
                                                                                                               en
                                                                                                               1-3
                                                                                                               o
                                                                                                               01
                                                                                                               "d
                                                                                                               o
                                                                                                               8
                                                                                                               CO
                                                                                                               S
                                                                                                               H
                  Figure 2.  Estimated Breakdown by Firing Method of Solid Fuel Boilers
                             in Four Size Categories Installed in 1930, 1950, and 1970
                             and Forecasted for 1990 .

-------
212                    CLEM COMBUSTION OF COAL


          • By 1950, the spreader stoker had gained a significant
            share of the market below 250,000 pph, mainly dis-
            placing overfeed stokers below 17,000 pph and underfeed
            stokers from 17 to 100,000 pph.  Pulverized coal made
            some inroads in the size category from 101 to 250 pph
            during this period.

          • In the 1950's and 1960's, the penetration of spreader
            stokers into the share of the underfeed stokers continued
            to the point that it was the dominant type for the range
            of sizes encompassing 17,000 to 250,000 pph.  Pulverized-
            coal firing gained further, nearly dominating the market
            above 250,000 pph by 1970.

          • For the near future, spreader stokers can be expected to
            continue to gain popularity in the smaller sizes, with
            underfeed stokers holding the major share below 17,000 pph.
            Spreader stokers can be expected to decline slightly in
            the market share above 100,000 pph, where firing of pul-
            verized coal is strongest.

It should be noted that introduction and successful commercialization of
the fluidized-bed combustion concept to industrial boilers could displace
the more conventional firing methods by 1990, and, thus, significantly
alter the distributions forecasted in Figure 2, especially above
100,000 pph.

                  EMISSIONS FROM STOKER-FIRED BOILERS

          Table 2 summarizes the emission factors* established by EPA as
averages for fossil-fuel combustion'^).  As noted, the particulate and
sulfur dioxide emissions are significantly higher for combustion of coal
than for fuel oil or natural gas, especially considering that coal can
have an ash content as high as 20 percent and a sulfur content as high as
4 percent.  Also, whereas fuel oil can be desulfurized before firing, there
are currently no commercially viable methods for mechanically treating or
preparing coal to reduce emissions of S02 (other than washing, which
cannot remove organically combined sulfur or fine dispersions of pyritic
sulfur).  Consequently, to reduce emissions from coal-fired combustion
sources necessitates the utilization of downstream emission-control
equipment.

          Because available emission data for stokers are limited, there
are no computational procedures for establishing specific relations among
emission levels, stoker type and design, boiler load, coal characteristics,
and coal size.  The emission factors in Table 2, however, are representative
of a broad grouping of equipment types and can serve as guidelines.

          Generally, stokers that agitate the fuel bed and disturb the
ash increase the fly ash loading in the stack.  Generally, this agitation
*
 The emission factor as used here is a statistical average of the rate
 at which a pollutant is released to the atmosphere per unit of fuel
 consumed^'.

-------
TABLE 2.  EMISSION FACTORS FOR FOSSIL-FUEL COMBUSTION WITHOUT CONTROL EQUIPMENT^
Emission Factor, lb/ 10" Btu of fuel consumed
Sulfur Carbon
Fuel and Type of Combustion Unit Particulates Oxides Monoxide
Coal
Power plant, pulverized
General 0. 64 A(a) 1.25S(b) 0.04
Cyclone burner 0. 08 A 1.52S 0.04
Commercial and industrial stoker
Spreader 0. 52 A 1.52S 0.08
Others 0.20 A 1.52S 0.08
Residual oil
Power plant (also includes 0.053 1.04S 0.02
distillate oil-fired plant
boilers
Commercial and industrial 0.153 1.04S 0.026
Distillate oil: commercial and 0.01 0.95 0.026
industrial
Natural gas
Power plant 0.015 C.0006 0.017
Commercial and industrial 0.18 0.0006 0.017
Nitrogen
Hydrocarbons Oxides


0.012 0.72
0.012 2.2

0.04 0.60
0.04 0.60

0.013 0.7


0. 02 0. 27 to 0. 54
0. 02 0.27 to 0. 54


0.001 0.600
0. 003 0. 120 to 0.230
(a)  The letter "A"  indicates that the weight percentage of ash in the coal should be multiplied by the value given.  Example:  If the factor

    is 0.64 and the ash content is 10 percent, the paniculate emissions before the control equipment would be 10 times 0.64, or 6.4 lb/10

    Btu (about 160 Ib/ton).

(b)  "S" equals the sulfur content [see footnote (a) above] in weight percent.
                                                                                                                                                                              03
                                                                                                                                                                              (-3
                                                                                                                                                                              O
                                                                                                                                                                             CQ
                                                                                                                                                                             §
                                                                                                                                                                             H

                                                                                                                                                                             £

                                                                                                                                                                             trj
                                                                                                                                                                             O
                                                                                                                                                                             H
                                                                                                                                                                             t->


                                                                                                                                                                             O)
to

-------
214                    CLEAN COMBUSTION OF COAL


is intermittent, as is the case for dumping-grate or vibrating-grate
stokers.  The spreader stoker, however, burns up to 50 percent of the
coal in suspension and, thus, has the highest fly-ash loading of all
types of stokers.

          Sulfur oxide emissions from stokers, as well as from all fossil-
fuel combustion equipment, are directly related to the sulfur content of
the fuel.  Presently, there are no combustor designs or equipment opera-
tional modifications that can reduce these emissions except for the still
commercially unproven fluidized-bed combustors.  In addition, unlike the
utility power plant, the economics and the geographical location of stoker-
fired commercial and industrial plants usually are unfavorable for utilizing
tall stacks to reduce ambient S02 to acceptable levels.

          Emissions of carbon monoxide, unburned hydrocarbons and nitrogen
oxides from stoker boilers generally are within air-pollution standards,
provided that the units are operating properly.


               COMPETITIVE SITUATION IN INDUSTRIAL FUELS

          Figure 3 provides a comparison of steam-generation costs for
various fuels on the basis of a load factor of 0.30, a boiler efficiency
of 0.80, and an annualization rate of 16.7 percent.  The estimated costs
include both capital and operating expenses.  It is apparent from the
example shown that the stoker boiler is economically attractive when
the cost differential between stoker coal and other fuels,is greater
than $1/106 Btu.  For example, if stoker coal costs $1/10  Btu (approxi-
mately $26/ton), steam-generation costs would be approximately $3/10  Ib
of steam.  In an oil or gas boiler, the fuel costs would have to be no
higher than $2/10° Btu to generate steam at $3/103 Ib.  In the past, the
cost of all industrial fuels was much less than $1/10° Btu and cost
differentials were about $0.50/10  Btu between fuels.

          Figure 3 also indicates the steam costs from a stoker boiler
equipped with S02 scrubbing equipment.  Because of the relatively high
costs estimated, it is unlikely that S0£ scrubbing will be utilized with
medium-size industrial boilers (unless dictated by Government regulation).


                ECONOMICS OF INDUSTRIAL-STOKER OPERATION

          The economics of industrial utilization of stoker coal are
difficult to assess for reasons that include:

          • Uncertainty in the availability and price of all fuels
          • Wide variation in the cost and quality of stoker-boiler systems
          • Unavailability of extensive operational cost-breakdown
              information
          • Unavailability of proven S02~control technology.

Delineation of the economic situation vith some degree of confidence would
have required an effort that was beyond the scope of this paper.  As a
consequence, the analysis that was developed and is given below should be
regarded only as a guideline and for making relative comparisons between
types of fuel.

-------
              STOKERS FOR INDUSTRIAL BOILERS
                                              215
      7.00
      6.00
      5.00
      4.00
"b
 v^
•ifr
 O
 O

 E
 o
 CD

 V)
      3.00
      2.00
      1.00
      O.OQ.
              Annualizaiion     16.7%

              Load factor     0.30

              Boiler efficiency  0.80
                                l  i
         0.00
l.OO
2.00
3.00
4.00  $/IO  Btu
Equivale
o
o
O)
U_
0
i
0.0
1
0
2 4
1
0.5
i i
10 20
6 8
i
1.0
i
30
10 12
1 1
1.5
1
2.0
i
14
i
40 50 60
Fuel Cost
16 3
_) ,
2.5 $/l
$/ton
I/DDI Of Oil
0 ft of natural gas
of coal
    Figure  3.  Cost Comparison of Various Types of Boiler Systems.

-------
216                    CLEM COMBUSTION OF COAL


                             Capital Costs

          Because coal-fired stoker boilers are more complex, the capital
costs of a stoker-fired boiler are significantly greater than those of
either an oil- or gas-fired unit.  In comparison to an oil- or gas-fired
boiler, the stoker boiler requires:

            Ash-handling facilities
            Fly-carbon reinjection system (for some spreader stokers)
            Very-high-efficiency dust-collection equipment
            Soot blowers
            Larger flow passages and additional heat-transfer surfaces
            More extensive combustion controls
            More extensive fuel-handling and storage facilities
            More extensive field erection work
            More physical space for facilities.

These factors result in the capital cost of a stoker boiler being from two
to four times that of a packaged boiler for oil or gas firing.  Capital-
cost differential will vary, depending on equipment design, manufacturer,
and type of installation.

          As an example of the economic attractiveness of gas- or oil-fired
equipment, one manufacturer estimated that the delivered and erected cost
of a 75,000-pph boiler was $6/pph for an oil/gas packaged unit and $20/pph
for a stoker boiler (with dust control).  The stoker alone can account for
up to 20 percent of this cost with another 20 percent attributed to dust-
collection equipment.

          With the other costs associated with a heating plant, the capital
costs of placing a boiler on line are estimated to be approximately
$10/pph for oil/gas and $25/pph for stoker coal.  (These numbers are based
upon information from a manufacturer and were derived from one set of
assumptions.)  S02~cleanup equipment would add an additional $12/pph to
the cost of the stoker boiler'".

                            Operating Costs

          The average operating costs of boilers are difficult to deter-
mine because of differing accounting procedures.  However, in comparison
with oil- or gas-fired units, stoker boilers clearly have higher operating
costs.  These can be attributed to:

            More frequent and closer attention required by operator
            Higher degree of boiler-operator skill
            Additional mechanical equipment to service
            Ash removal
            Fuel handling.

In addition, the fuel costs must be considered.  These costs vary from
region to region and have "become difficult to predict.  A check on the
current fuel prices in Columbus, Ohio, indicates little difference among
the energy costs of stoker coal and natural gas, while fuel oil costs
about 50 percent more.

-------
                     STOKERS FOR INDUSTRIAL BOILERS                 217
                      CONVERSION TO STOKER FIRING -
                          FACTORS TO CONSIDER
          Stoker fuel can be a viable fuel for meeting industry's demand
for energy.  However, as alluded to earlier, certain factors must be
considered in the design and operation of the overall system that include:

        • Space
          -  boiler and auxiliary equipment
          -  coal handling and ash disposal systems
          -  pollution control equipment

        • Costs
          -  capital
          -  operating

        • Operation
          -  operator skill
          -  high maintenance.

Generally, a boiler that has not been designed for coal firing cannot be
converted.  Additionally, if a new coal-fired boiler is installed, a
substantial amount of space is required for the new and larger boiler
and all the ancillary equipment associated with it.  Also, even with the
higher cost of gas and oil, stoker firing presently in many instances
cannot be  justified on economics alone.

                         BCL Stoker Conversion

          During  1976, Battelle's Columbus Laboratories (BCL) converted
one of their 600-hp  (25,000 pph steam) oil/gas boilers, originally designed
for coal firing,  to stoker  (spreader) firing.  These boilers were originally
installed with oil or gas firing, but the units were designed and assembled
with  conversion to coal firing in mind.  The decision to convert to coal
was not primarily one of economics, but one rather to assure continued
operation  of the  steam plant.  The overall cost of converting one of
the existing boilers was $800,000.  About half the cost was associated
with  the boiler modification and stoker installation while the remaining
half  with  the coal and ash handling equipment, pollution controls, and
engineering design.

           The conversion essentially doubles the space of the steam plant
facility.  Ash and coal silos, above and below ground bunkers, and
bucket elevators  were located adjacent to the steam plant, while the
two-stage  dust collector, forced and induced draft fans, and 18-ton
coal  hopper were  located within the building.

           With the conversion, higher maintenance  is required.   Instead
of the fuel being supplied to the  boiler through a pipe, coal is delivered
through a  system  of bucket  elevators, screw feeds, and conveyors.  As
expected,  several small mechanical difficulties have occurred time  to
time. Like-wise,  operational problems, such as low-load  smoke and exces-
sive  slagging with low ash-fusion  temperature  coals, were experienced.
These and  other problems have been minimized by proper coal selection
and stoker operation.

-------
218                    CLEAN COMBUSTION OF COAL
                              CONCLUSIONS

          It appears that the future market for stoker boilers and the
corresponding demand for stoker coal -will be dependent on factors that
include:

          • National policy decisions that affect the availability of
            fuels
          • Local and national S02 regulations

          * Development of suitable SOg control systems (either in-
            furnace control systems or stack-gas scrubber systems)

          • Availability of clean coal-derived fuels or commercialization
            of fluidized-bed combustion systems with sulfur removal.

The stoker market appears promising for the next 5 to 10 years and per-
haps longer, particularly for the large stokers.   The market for the
small- and medium-sized stoker is less definitive, but sales of medium-
sized industrial stokers will probably increase over the essentially  neg-
ligible levels of recent years.  Although stokers firing high-sulfur coal
emit S02 at levels above those specified by many state source standards,
it is anticipated that some relaxation of these standards may occur,
provided that the local primary ambient standards are being met.   It  is
also possible that economic and practical S02  stack-gas cleanup systems
will be developed, but these systems appear to have many problems.
                              REFERENCES

1.  Steam, Its Generation and Use,  37th and 38th Editions,  The  Babcock
    & Wilcox Co.,  New York (1963, 1973).

2.  Barrett, R.E., Miller, S.E., and Locklin,  D.W.,  "Field  Investigation
    of Emissions  from Combustion Equipment  for Space Heating",  Final
    Report on Contract 68-02-0251 from Battelle's Columbus  Laboratories
    to U.S. Environmental Protection Agency, EPA-R2-73-08Ha (June,1973).

3.  Locklin, D.W., Kropp, E.L.,  "Design Trends and Operating Problems
    Part A - Industrial Boiler Population and  Design Trends", Final
    Report Grant R-802H02 by Battelle's Columbus Laboratories to  U.S.
    Environmental  Protection Agency, EPA-650/2-73-032 (April, 197M •

k.  "Compilation of Air Pollutant Emission  Factors", U.S. EPA,  Office
    of Air Programs, Publication No. AP-1+2  (April, 1973).

5.  "Initial Operating Experiences  with Dual-Alkali  S02  Removal System",
    Presented at the EPA Symposium  on Flue-Gas Desulfurization, Atlanta,
    November k, 197U.

-------
                                                                     219
          INITIAL OPERATION OF THE 30 MWe RIVESVILLE MULTICELL
                  FLUIDIZED BED STEAM GENERATION SYSTEM

                                   by

                  Robert L. Gamble and Newton G. Wattis
                    Foster Wheeler Energy Corporation
ABSTRACT
The world's largest fluidized bed steam generator located at Rivesville,
West Virginia is currently undergoing initial operation.  This unit is
designed to generate  37.8 kg/s  (300,000 Ib/hr) equivalent to 30 MWe of
superheated steam while firing  coal in a fluidized bed of limestone.
Coal was first fired  in this unit in December, 1976 and start-up opera-
tion is progressing.
 INTRODUCTION

 Coal is presently our nation's most abundant energy source.  Current
 environmental regulations require  that coal be mined and consumed in
 an environmentally acceptable manner.  The majority of the energy con-
 sumption in this country is east of the Mississippi river and the ma-
 jority of the coal resources east  of the Mississippi contain quantities
 of sulfur which prohibit direct combustion without means for control-
 ling emissions within current limits.  Direct combustion of coal in
 fluidized beds of limestone has shown to be a near term practical meth-
 od for energy conversion with high sulfur coal.

 The fluidized bed steam generation system located at the Monongahela
 Power Company, Rivesville, West Virginia plant has been sponsored by
 the United States Energy Research  and Development Administration (ERDA).
 The Rivesville system has been designed by Pope, Evans and Robbins, Inc.
 and includes a multicell fluidized bed steam generator designed and
 erected by Foster Wheeler Energy Corporation.  The entire system has
 been retrofit into an existing power plant and steam produced from this
 system will be used to drive an existing turbine-generator and generate
 approximately 30 MWe of electric power.  The system installed at Rives-
 ville is designed to generate 37.8 kg/s (300,000 Ib/hr) of superheated
 steam and is several times larger  than any other fluidized bed steam
 generation system in the world.

 Following the initial start-up operation, de-bugging of the systems at
 the Rivesville plant and installation of test instrumentation, a test
 program of approximately one year  duration will commence.  Successful
 operation of this unit will be a major step toward the commercializa-
 tion of both industrial and utility fluidized bed steam generators
which will be capable of directly  firing low grade and high sulfur
 coals within present environmental emission limits.

-------
 220
                        CLEAN COMBUSTION OF COAL
 FLUIDIZED BED COMBUSTION

A  fluidized bed  is  a bed of granular  particles  supported by a non-
sifting  grid  through which an upward  flow of  air  or, gas is passed with
sufficient velocity to  lift and  float the granular particles.  As the
velocity of fluidizing  gas is increased,  the  bed  will expand and bub-
bles will form similar  to a pot  of boiling water, and in this state
the fluidized bed exhibits the properties of  a  liquid.   Figure 1 sche-
matically represents a  fluidized bed  steam generator.   In the case of
a  fluidized bed  steam generator  firing high sulfur coal, the fluidized
bed particles are normally crushed limestone.   Crushed coal is injected
into the fluidized  bed  of limestone and burns,  converting the fuel
bound  sulfur  to  sulfur  dioxide  (802)•   If the bed is designed and oper-
ated  to  maintain a bed temperature of 815-870C (1500-1600F), the S02
released from the coal  combustion will be chemically absorbed by the
limestone bed material.  Tests have indicated that S02  emissions can
be controlled for virtually all  United States coals by  maintaining ac-
tive limestone bed  material within the fluidized  bed.   This is done by
feeding  raw limestone with the coal and regulating the  bed material in-
ventory  by a  gravity drain system which withdraws spent material and
large  coal ash particles.
                                             FLUE
                   FUEL
                   INJECTION
                      PIPES
                                                 1550*F
                    AIR
                    DISTRIBUTION
                          GRID
                                                fN-211
                 Figure 1.  Fluidized Bed Steam Generator.
Due to the relatively low operating temperature of the fluidized bed
NOX emissions are inherently low and no special systems are required'
for control of NOX emission to meet the current emissions limit.  Par-

-------
                     Fluid!zed Bed Steam Generation
                                                                     221
ticulate emissions can be  controlled conventionally with either elec-
trostatic precipitators  or baghouse filters.

Many thousands of hours  have  been logged in pilot unit operation at the
Pope, Evans and Robbins, Inc.,  Alexandria,  Virginia facility, the
Foster Wheeler Energy Corporation test unit at Livingston,  New Jersey
and other pilot units throughout  the United States and overseas.


RIVESVILLE UNIT DESIGN

The Rivesville systems design can be divided  into four (4)  primary
areas which include  (a)  the air and gas systems,  (b)  the water and
steam systems, (c) the fuel and limestone injection systems and (d)
the bed material recycle system.   A description of each of  these  sys-
tems follows:

a.  Air and Gas Systems  -  The air and gas systems are typical of  that
    found in a utility power  plant steam generating system.   Air  is
    supplied by a forced draft  fan through  a  regenerative air preheat-
    er and the flow  rate is regulated to each section of the boiler by
    individual air regulating dampers.   The air passes through the air
    distribution grid and  the fluidized bed coal  combustion zone,  and
    from this point  flue gas  flows over convection type heat transfer
    surface and to cyclone particle separators,  through a hot (385C,
    730F) electrostatic precipitator,  and on  to  the gas side of the re-
    generative air preheater  and  induced draft  fan at the inlet of an
    exhaust stack.

b.  Water and Steam Circuit - A schematic of  the  steam generator  cir-
    cuit is indicated in Figure 2.   Feedwater from the existing plant
    feedwater system enters the economizer  inlet  and  flows  in series
                        Cell"A"  Cell]*"  CeH"C" Cell"D*
                                  SERIES
                                            PARALLEL
                                            BOILER
                                            CIRCUIT
                          flTENKMTN SHUT
                                           300,000 LB/HR
                                          1350 PSIG  925 °F
                                          STEAM TO TURBINE
            Figure 2.  Demonstration Unit Circuit Schematic.

-------
222
                       CLEAN COMBUSTION OF COAL
    through  four  (4)  economizer  sections  in each of four (4)  independent
    boiler cells.   Feedwater  enters  the drum and flows through the
    downcomers  to  two (2)  forced circulation pumps as seen in Figure 3.
    The  forced  circulation pumps provide  the head required to maintain
    the  proper  mass flow rate in the five (5)  parallel boiler circuits.
    The  boiler  circuits  consist  of four  (4)  horizontal tube boiler
    banks and vertical tubes  which form the boiler enclosure  walls and
    cell partition walls.   A  steam water  mixture exits these  boiler
    circuits to the steam  drum,  and  dried saturated steam flows from
    the  drum through a primary superheater and subsequently through a
    finishing superheater.  An attemporator spray station is  located
    between  the two (2)  superheaters for  final steam temperature trim.
    Steam leaving  the finishing  superheater is routed to the  existing
    steam turbine  where  it  enters at 9 MPa (1300 psig)  and 495C (925F).
   Figure 3.  Multicell Fluidized Bed Demonstration Steam Generator In-
              stalled at Monongahela Power Co.,  Rivesville, WV, Plant.

   The arrangement of  the heat exchange surface indicated  in the circuit
   schematic  (Figure 2) is shown on Figure 4.
                                                  Cell's"
                                                            CellmA"
                     Figure 4.   Demonstration Unit.

-------
                    FLUIDIZED BED  STEAM GENERATION
223
c.  Fuel and Limestone Injection - Crushed limestone is received at the
    plant in pneumatic discharge trucks and blown into storage bunkers
    above the steam generator as indicated in Figure 5.  Coal is either
    received at the plant in a crushed and dried state and loaded into
    the storage bunkers with the plant conveyor systems or is received
    as run of mine and is crushed and dried in the plant systems prior
    to loading in the storage bunker.  A  coal/limestone mix is regulat-
    ed to each fluidized bed cell with rotary flow control feeders at
    the outlet of each of the bunkers.  The coal and limestone mix in-
    to a common pipe as they fall by gravity to a rotary air lock.  The
    rotary air lock prevents back flow of air from the slightly pres-
    surized vibrating distributors, each  of whcih distribute the coal/
    limestone mixture to eight  (8) outlets.  The vibrating distributor
    is pressurized to approximately 14 kPa  (2 psig) with air which
    pneumatically carries the coal/limestone mixture through downward
    sloping injection pipes into the lower section of the fluidized
    bed.  The Rivesville unit contains seven  (7) coal/limestone in-
    jection systems, two  (2) of which are indicated on Figure 5.
                    COAL6TOAAQE
             Figure 5.   Fluidized Bed Fuel Injection System.

     Cells A, B and C (refer to Figure 4)  each receive coal and lime-
     stone from two (2)  sides and Cell D,  which is the smallest cell,
     receives coal and limestone through the single end wall.   The
     Rivesville system uses limestone crushed to -3.2 mm (-1/8  inch).
     and coal crushed to -12.7 mm (-1/2 inch).

     Bed Material Recycle System - The bed material recycle system, is
     used to control the bed material size and inventory in each Qfill*
     Bed material is extracted from each cell through a gravity drain
     and the flow is regulated onto a common vibrating conveyor as in-
     dicated in Figure 6.  Bed material from the common vibrating con-
     veyor may be removed from the system by opening the bypass or may
     flow through a classifier which moves oversize bed material and
     maintains properly sized bed material for recycle back to  each of
     the cells.   Bed material to be removed from the system passes
     through an indirect solids cooler and is discharged to a pneumatic
     system for disposal.  The properly sized bed material is pneumati-
     cally carried up to a bed material storage tank which contains

-------
 224
CLEM COMBUSTION OF  COAL
                                       |STEAM_r-fc~lJ0-§k?CIROSTATIC
                                CYCLONE       VACUUM PRECIPITATOR
                                SEPARATOR      EJECTOR
                                            COOLING WATER OUT
                                           INDIRECT
                                           SOLIDS
                                             ILER
                                              OVERSIZE TO
                                              Dm DISPOSAL
                 Figure 6.  Bed Material  Recycle System.
     multiple outlets so that bed material can be returned to each of the
     four (4) steam generator cells.  This system is  designed to handle
     hot bed material leaving the fluidized beds at up  to  1093C (2000F).
 INITIAL OPERATION

 During  the last three (3) months of 1976, checkout and  initial cold mode
 operation of most of the Rivesville facility systems took place.

 Coal was first ignited in the D cell of the fluidized bed steam genera-
 tor on  December 7,  1976 with the fluidized bed temperature  reaching a
 maximum of approximately 790C (1450F) for a short period of time due to
 insufficient fuel flow into the bed.  Following several other  short
 coal firing periods at the early end of the operations  learning curve,
 coal fires were established and maintained for 33 hours on  December 22
 and 23,  1976.   During this time bed temperature was controlled at an
 average  of 79C (1450F) and all systems performed well.  The bed temp-
 erature  was easily  controlled and responded well to changes in coal
 feed rate.   Coal firing rate was approximately 10% of the steam gene-
 rator full load and a drum pressure of 8.275 MPa (1200 psig) was  main-
 tained while generating saturated steam at approximately 5  kg/s (40 000
 Ib/hr).  During this coal firing operation, firing was maintained with-
 out any  ignition assistance.   Figure 7 indicates the operating condi-
 tions during this thirty three (33) hour run at the end of  which  the
unit was shut  down  for a scheduled outage.

-------
                     FLUIDIZED BED STEAM GENERATION
225
        300
                            D—CELL BED TEMPERATURE
                          First continuous run—Dec. 22-23, 1976
                                         Coal Flow 2800-3000 Ib/hr
                                         Saturated Steam Flow 40,000 Ib/hr
                                         Drum Pressure  1200 psig
                        8   10  12  14  16  18  20  22 24  26  28  30  32  34  36
                                    Time, hrs.
        Figure 7.  D-Cell Bed Temperature First  Continuous Run
                           December 22-23, 1976.
The coal  handling system is designed to handle a maximum of 5% moisture
coal.  During January  1977 a shipment of wet coal  caused operating dif-
ficulties and resulted in only intermittent coal firing operation.
Slide Gate Slide Gate Beds Equalized and
Closed Open C-Cell Ignited

Air 	 	
Distributor

C-Cell
400F
H 	
"


I — //// — '
D-Cell
1800F

Ly///-l

3 Ft.
I

C-Cell

400F-^


D-Cell
1800F
^

it * »




C-Cell

1550F

Jftl : .
D-Cell

1800F

in.




x s
Air Control
Dampers
(1) (2) (3)
                    Figure 8.   Ignition of  the C-Cell.

-------
226                    CLEM  COMBUSTION  OF  COAL
Ignition of the C cell is achieved by allowing hot bed material in the
D cell to flow through an opening in the partition wall between the D
and C cell as indicated on Figure 8.  By starting with a deeper bed in
the D cell flow through the partition wall occurs rapidly and the bed
depth in each cell are equalized in approximately three (3) minutes.
Coal in the C cell is rapidly ignited by the hot material near the
slide gate opening and coal ignition propagates rapidly throughout the
C cell raising the bed temperature to its operating temperature.

The first successful ignition of the C cell occurred on February 10,
1977 and temperatures in the C cell were raised to approximately 870C
(1600F).  Shortly after obtaining the 870C (1600F) bed temperature in
C cell, coal feed was lost, most likely due to the wet coal conditions.
Following this short coal firing operation in the C cell, the unit was
inspected and the boiler found to be in good condition.  Problems with
the forced draft and induced draft fans, however, required repair and
no subsequent coal firing was attempted prior to this writing.
 TEST PLAN

 Following  additional start-up efforts and upon achieving the ability to
 operate  the  remaining boiler cells, test instrumentation will be in-
 stalled  at the Rivesville facility and a one (1) year test program will
 be initiated.  This test program will include determination of heat and
 material balances at various operating loads and a 3,000 hour continuous
 operation  run.   Installation of the test instrumentation should take
 place  during the third quarter of 1977.
 SUMMARY

 Initial  operation has shown that stable combustion of coal can be main-
 tained in  large  fluidized beds of limestone.  As coal feed rate is con-
 trolled, bed  temperatures can be easily maintained and adjusted.

 As  of this writing, only preliminary operation has been achieved at the
 Rivesville facility; however, results of this operation are encouraging
 and indicate  that upon coordinated operation of the systems the Rives-
 ville unit should operate satisfactorily.


 REFERENCES

 Gamble,  R. L. and Warshany, F. R.  "Commercial Development of Atmospheric
 Fluidized  Bed Utility Steam Generators" September 1975, 1975 Joint Power
 Generation Conference, Portland, Oregon

 Mesko, J.  E., "Multicell Fluidized Bed Boiler Design, Construction and
 Test Program", Monthly Status Report #51, December 1976, ERDA Contract
 No. E(49-18)-1237, Report FE-1237-62

-------
                                                                    227
             DEVELOPMENT OF AN EFFICIENT SOLIDS FUEL BURNER

                                   by

                           Norman A. Lyshkow
                 Combustion Equipment Associates, Inc.
INTRODUCTION

      Much interest has been directed to the development of solids
burners for use in various chemical processes.  The compatibility of
the heat source with the particular requirements of the process neces-
sitates detailed study of the chemical system.

      The emissions of the heat source must be analyzed in order to
predict their probable effect on the product.  In certain cases, emission
effect is minimal;  however, in the largest number of cases, solids
burners are sought to replace oil and gas in stringent processes.

      In addition to melt applications, the requirements of these pro-
cesses vary from drying to supplying thermal energy for chemical reac-
tions in rotary kilns.  Coal use has been limited due to product con-
tamination.  The development of a combustor capable of providing a
"Clean Flame" for such applications has required a thorough study in both
the physical and chemical processes of solid fuel combustion.  Removal
of fly ash within the combustor, while providing the necessary tempera-
ture for efficient combustion, directed development efforts to a slagging
type burner.  As fly ash capture efficiency requirements generally vary
with the process, the slagging type burner allows for the most efficient
capture of fly ash as well as a practical method of ash removal.  The
use of a wet wall combustor also simplifies the mathematical analysis
of the system.

      Efficiencies in excess of ninety percent are required in many
processes; theoretical efficiencies greater than 99% are predicted with
pulverized coal.  Extensions of the model combustor with the construc-
tion of new test combustors are planned.  To date, particle efficiencies
have exceeded 95%» while the possibility exists for coal combustion with
particle emission levels below .1 Ib/MBTU.

      The design also permits such efficiencies at combustor pressures
of less than two inches of water.  Computer analyses have determined
dimensions of burners from 1 million to l60 million BTU, with larger
sizes practical.

      Key to the use of solid fuels is an efficient combustor, capable
of utilizing low quality, high ash fuels.  The use of such fuels requires
a method for qualifying them and predicting their combustion characteris-
tics in a real system.

-------
228                    CLEAN COMBUSTION OF COAL
      The need for these predictions and descriptions has promoted the
development of a computer model sufficiently complex to simulate the
combustion requirements and processes of fuel.   The model input must be
the most complete fuel description, thus necessitating the development
of specialized instrumentation and procedures.

      This information development must then be described in a series
of equations.  These equations must compensate  for the variable system
conditions.  Initially, the equations were of the most simple type.
Further expansions were tested in a model combustor to predict their
accuracy.

      Key to the prediction of accuracy was both carbon and particulate
emission.  Emission testing by EPA methods and  the use of an Anderson
impactor provided this correlation.  The computer model devised has
expanded to approximately six hundred lines with many simultaneous iter-
ation loops.  The pitfall of such a complex model is the necessity to
continually maintain relevance.  The complexity of solids combustion, a
dynamic situation, cannot be simply described.   The only constant
criteria of such a model can be its prediction  of accurately sized
burners which meet the requirements of a successful system.
FUEL ANALYSIS
      Elemental and proximate analyses are utilized in the model.   In
addition, a special device was constructed to ignite material under in-
tense radiation and determine ignition time.   The device was  constructed
to eliminate convection and conduction effects during irradiation.   Other
devices were constructed to determine emissivity and envelope specific
gravity.  Particle size analysis was performed and a technique developed
to determine effective particle thickness.  Ash fusion temperatures were
determined to fix the temperature requirements of the combustor.
PROGRAM DEVELOPMENT

      The motion and travel of a mathematically generated system of
particles to predict their interactions was developed,  and includes the
following procedures:

(l)   Continuous analysis of changing characteristics during the ignition-
      combustion cycle is determined for the particles.   The burner is
      also analyzed by sector to determine energy balance of each sector.

(2)   Centrifugal Force—drag force determinations—are utilized to
      predict particle impaction points.

      Since a path of movement for each group of particles is simulated,
a mathematical technique is utilized to determine the radius of curvature
and centrifugal force which predicts the reasonable particle cut of the
combustor.  Tests using an Anderson impactor have found the combustor to
be in close correlation with the mathematical simulation.  The model was
then directed to determine a flow field optimal for a given system of
particles.  The flow analysis on levels of predicted turbulence.  In later
studies of spectral emission, this assumption was reasonably verified.

-------
                          SOLIDS FUEL BURNER                        229
FIELD STUDIES
      A number of combustors designed by the computer model have been
constructed and tested using a vide variety of solid fuels.  Burners have
been constructed from 3 million BTU's to kO million BTU's.   A schematic
of the combustor (Figure l) is presented.  The unit is basically a
cyclone-type burner.  The burner is sized by the appropriate input data
(Figure 2) which generates the dimensions, flow, and volume considera-
tions for a particulate fuel at the desired combustion rate.

      The output data is developed from the fuel characterization
(Figure 3) by the computer model.

      A burner is then constructed to the dimensions provided by the
model and operated at the test facility which consists of a large water-
filled chamber with a steam vent, droplet extractor, circulators and
measuring means to determine water addition.  In addition to the perform-
ance of material and heat balances, the test burner allows gaseous and
particulate testing.

      Thermocouples are located throughout the system and in the chamber
gas effluent.  Particulate is collected in the chamber and at the stack
sampling point.  Data is then analyzed to determine correlation with
mathematical representations (Figure h).

      Chemical data on emissions and slag has been collected (Figure 5).
Further studies in emission are planned.

      To date, low NOX values have been found to exist.  In addition,
concentration of sulfur in the slag has been observed and will be
studied in more detail.
SUMMARY

      Continuous development in mathematical representations, prototype
construction, and materials of construction is underway on this project.
The success of the burner design utilizing fuels as high as 35% ash and
BTU contents as low as 5,500 with resultant particulate emissions as low as
.3^ Ib/mm BTU illustrate the possibilities for utilization of the design
during the present energy situation.

-------
230
CLEAN COMBUSTION OF COAL
                                  PRIMARY AIR VANES

                                    SOLIDS ENTRY
                                               SECONDARY AIR VANES
                                               (IF REQUIRED)
 DIRECTION
    OF
   FLOW

i *
TT
A



                             SUPPORTING FUEL NOZZLES
                             (IF REQUIRED)
                              Figure 1.

-------
             SOLIDS FUEL BURNER                        231
LINKING SYSTEM LIB
ENTER BTU/HR (IN MILLIONS)
73
ENTER BTU/LB
712260
ENTER TYPE OF SUPPORT FUEL
0 -- FOR #2 OIL
G -- FOR GAS
?G
ENTER % CARBON
770.72
ENTER % HYDROGEN
77.36
ENTER % OXYGEN
714.73
ENTER % FIXED CARBON
743.24
ENTER % ASH
78
ENTER ASH SPECIFIC GRAVITY
71.4
ENTER ASH PARTICLE SIZE IN MICRONS
7.2
ENTER ASH FUSION TERMPERATURE FLUID
72350
ENTER % MOISTURE
710
ENTER PARTICLE EMISSIVITY
7.95
ENTER MATERIAL SPECIFIC GRAVITY
71.2
TEST IGNITION TIME MATERIAL AS RECEIVED
7.162
ENTER IGNITION TIME MATERIAL DRY
7.162
ENTER LIFTING VELOCITY of 90% - TILE PARTICLE
765
ENTER 90% - TILE PARTICLE SCREEN SIZE IN MICRONS
7100
ENTER 50% - TILE PARTICLE SCREEN SIZE IN MICRONS
730
ENTER 10% - TILE PARTICLE SCREEN SIZE IN MICRONS
75
ENTER % FOSSIL FUEL
75
ENTER WALL THICKNESS
7.75
ENTER CONE ANGLE
7107
        Figure 2.  Program Input Data.

-------
232                    CLEM COMBUSTION OF COAL
       LBS OF SOLID FUEL/HR                           232.4633
       CUBIC FEET OF GAS/HR                           149.5215
       CFM AIR SOLIDS VANE                            477.2646
       CFM SUPPORT FUEL                               24.1726
       TOTAL CFM COMBUSTION AIR                       501.4373
       AERODYNAMIC                                    .8271E-04
       CYLINDER DIAM. INCHES                          8.6467
       CYLINDER WALL THICKNESS INCHES                 .7500
       MAJOR DIAM. OF BURNER                          18.1491
       LENGTH OF IGNITION SECTION INCHES              NONE
       LENGTH OF COMBUSTION SECTION, INCHES           7.2972
       WIDTH OF SUPPORTING FUEL VANE, INCHES          NOT USED
       ANGLE OF INJESTION IN DEGREES                  69.3950
       INJECTION VELOCITY FEET/MINUTE                 3286.0000
       WIDTH OF SOLIDS FUEL VANE, INCHES              2.5624 COMBINED
       OPTIMAL NUMBER OF INJECTION PORTS              4
       CONE PARTICLE CUT IN MICROS                    1.6099
              Figure 3-  Computer Generated Sizing Data.
          PARTICLE SIZE CORRELATION

          ARITHMETIC MEAN SIZE OF COLLECTED              % OXYGEN
            FLY ASH IN MICRONS

          2.5                                            3.5
          3.6                                            0.0
          2.8                                            1.3
          3.2                                            1.0

          PREDICTED 2.21 MICRONS
                               Figure  U.

-------
                SOLIDS FUEL BURNER                        233
SUB BITUMINOUS COAL
    FLY ASH AS PER CENT OF TOTAL ASH
         4.2

    FLY ASH CARBON
         21.26%

    TOTAL PARTICULATE EMISSIONS IN LB/MBTU
         .40

    S02 EMISSION
         392.5 PPM
         .66 LB/MBTU
         THEORETICAL S02
         1.64 LB/MBTU
         40% OF AVAILABLE SULFUR EMITTED AS S02

    NOX EMISSION
         252 PPM
         .30 LB/MBTU
         THEORETICAL NO  (BY FUEL NITROGEN)
         .86 LB/MBTU
         35% OF AVAILABLE FUEL NITROGEN EMITTED AS NOX
            Figure 5-   Particulate  Emissions.

-------
234                    CLEAN COMBUSTION OF COAL

-------
                                                                    235
                 SESSION IV - POSTCOMBUSTION CLEANUP

   SESSION CHAIRMAN:   Sidney R.  Orem, Industrial Gas Cleaning Institute
     The previous sessions have dealt with preparing coal,  cleaning it
and burning it in a variety of ways.   The papers in this session high-
light the present state of technology on backend cleanup of products of
combustion.  Topics include two methods for high efficiency particulate
control, flue gas desulfurization and flue gas treatment for NOX control.

     Some of you no doubt heard that  the Senate/House Conference
Committee reached a compromise report on the Clean Air Act  Amendments
at 2:15 a.m. yesterday.  Action is expected in both houses  before ad-
journment tomorrow for the August recess.

-------
236                    CLEAN COMBUSTION OF COAL

-------
                                                         237
               "ELECTROSTATIC PRECIPITATION

                     STATE OF THE ART"

                            by

             Dr. R. S. Atkins & D. V. Bubenlck

                     Research-Cottrell

     It is extremely appropriate during  this  symposium  on
the Clean Combustion of Coal to discuss  electrostatic pre-
cipitation.  Electrostatic precipitation has  played  a major
role in controlling coal-fired boiler emissions  and  will
continue to play a significant role  in the future.   In
fact, if it were not for this technology, environmental
pressures would not have permitted the coal-fired boiler
market to have grown to its present  size  (figure 1).
Likewise, many significant improvements  in this  and  other
pollution control technologies developed as a result of
the "1970 Clean Air Act" which necessitated stricter pollu-
tion control requirements and performance standards.  The
precipitator was first applied to fly ash control problems
in 1923 but many significant technological advances  are
occurring now.  With government and  economic  pressures  for
a more coal-dependent economy, we will continue  to see
many more improvements in electrostatic  precipitation to
meet these needs.

     The electrostatic precipitator  is the major high ef-
ficiency particulate control device  for  coal-fired boilers.
Figure 2 illustrates that as late as the 1960's, the
average user required only about 97.5% particulate collec-
tion efficiency.   With today's pollution control laws .
we are designing and supplying systems with an average  re-
moval efficiency of 99.5% and in some instances  a design
requirement of 99.9%.

     The demand for higher efficiencies, the  use of  more
strip-mined low sulfur western fuels, the site-specific
situations and their resulting economics necessitate the
consideration of alternate pollution control  devices.
Baghouses, scrubbers, electrostatic  precipitators and com-
binations thereof are offered by the company  I represent
depending upon customer needs, pollution codes and econom-
ics.   Each situation requires the selection of an appro-
priate control device strategy.  However, today  I will
only discuss the electrostatic precipitator,  where it is
and where it is going.

-------
                                                                                                 to
                                                                                                 CO
      450
                                    900
                                                    r
                                                    UJ
                                                    I
                                                    UJ
                                                    o
                                                    u.
                                                    LJ

                                                    Z
                                                    O


                                                    o
                                                    u
                                                    _J
                                                    _l
                                                    o
                                                    u
99.9



99.8


99.7



99.5


99.3


 99






 98



 97





 95




 93





 9O
                                                                      MAXIMUM
                            AVERAGE _
                                                  o
                                                  o
                                                  §
CQ
1-3
                                                                                                 a
                                                                                                 o
        1920 1930 I94O  1950 I960  1970 1980

                    YEAR
         1920  1930 1940 1950  I960 1970  1980
                      YEAR
FIGURE  1.   COAL CONSUMPTION AND  INSTALLED

            FLY ASH  PRECIPITATOR  CAPACITY

            FOR PUBLIC UTILITIES  IN THE

            UNITED STATES, 1920-1975.
FIGURE  2.   TRENDS IN  THE AVERAGE  AND

            MAXIMUM COLLECTION EFFICIENCIES

            OF FLY ASH PRECIPITATORS IN THE

            UNITED STATES.

-------
                  ELECTROSTATIC PRECIPITATION                239
What is an Electrostatic  Precipitator?

     The conventional  coal-fired  boiler electrostatic pre-
cipitator as we know  it  today  is  a rectangular configura-
tion with the gas  flowing between parallel grounded plates,
called collecting  electrodes,  interdispersed with regularly
spaced discharge electrodes.   Usually the discharge elec-
trodes are flexible weighted wires,  rigid masts or other
electrode geometries  which are of negative polarity to
provide a source of electrons  which  produce ions for
charging the dust  particles.

     The charged dust  is  electrically attracted to the
collecting plates  where  it deposits  and is removed from
the gas stream.  Collected particulate is released from
the collecting  and discharge electrodes during on-stream
operation by rapping  and  vibrating devices.  The collected
dust falls by gravity into hoppers located beneath the
precipitator.   The system is designed for continuous oper-
ation with multiple ducts and  chambers in parallel to
handle the volume  of  gas  and different lengths of treatment
to achieve desired collection  efficiency levels
(Figure 3).

     The precipitator  can be located up or downstream of
the boiler air  heater  and respectively operated at 650 to
850°F (hot precipitator)  and 250  to  350°F (cold precipi-
tator).  The face  velocity through the precipitator typi-
cally ranges from  3 to 6  ft/sec.   Various discharge and
collecting electrode  geometries are  used depending on the
dust properties.   Most units use  8 to 12 inch plate
spacing with various  treatment lengths depending on the
degree of high  tension sectionalization required.  Dis-
charge electrodes  are  equally  spaced on centers between
the plates.  Full  and  half wave D.C. current is used to
negatively charge  the  discharge electrodes with respect to
the grounded plates.   Typical  levels of corona power
density range from 0.5 to 3.5  watts/ft  of collecting sur-
face corresponding to  resistivity levels of 10   to
10  ohm-cm.

     Today's precipitators have a combined charging and
collecting function in the same treatment length.  This
has Resulted in a  compromise between charging and collec-
tion properties to obtain optimized  single-staged perfor-
mance.  Theoretically  better performance may be obtained
by designing units with  independent  charging and collec-
ting sections as illustrated in Figure 4 to optimize their
individual needs.  Two-staged  precipitator research was
conducted many  years  ago  but progressed no further than
the laboratory, possibly  because  the then moderate per-
formance requirements  could not justify the increased
costs.  However, work  in  this  area is again being con-

-------
                                                                                      -P-
                                                                                      O
       H.V. SYST.
     SUPPT. INSUL


     RAP INSUL.

 BUS DUCT

INSUL.
                                                 CHAMBER
                                                  (TYP)/
o
                                                                                       o
                                                                                      CO
                                                                                      1-3
                                                                                      H
                                                                                      O
                                                                                      O
                                                                                      o
            FIGURE 3,  TYPICAL  PRECIPITATOR CONFIGURATION,

-------
               ELECTROSTATIC PRECIPITATION
                                                      241
                  CONFIGURATION
GAS
FLOW
GAS
FLOW
           SINGLE-STAGED PRECIPITATOR
•  •


•  •
             TWO-STAGED PRECIPITATOR
        OPTIMUM  PERFORMANCE REQUIREMENTS
CHARACTERISTICS
CURRENT
FIELD STRENGTH
MAX, FIELD
SPACE CHARGE
DUST
CHARGING
SECTION
HIGH
NON-UNIFORM
AT CATHODE
LOW
NONE ON ANODE
COLLECTING
SECTION
LOW
UNIFORM
AT ANODE
HIGH
COLLECTED
ON ANODE
FIGURE L\,  COMPARISON  OF SINGLE-STAGED AND TWO-STAGED
           PRECIPITATORS,

-------
242                 CLEAN COMBUSTION OF COAL
 ducted  by EPRI,  EPA, research institutes  and  precipitator
 manufacturers to meet the demand for more  efficient,  less
 costly  equipment.  By the end of this decade,  large-scale
 two-staged precipitators with enhanced charging  and  collec-
 ting sections will probably be commercially available.
 These units may  be dry systems or combined dry/wet units
 with additional  capabilities for fine particulate  removal,
 reduced reentrainment and gaseous pollutant control.   The
 two-staged precipitator will be less costly and  less  sensi-
 tive to particle composition.

      In conjunction with High Voltage Engineering  we  are
 applying pulsed  techniques to improve corona  generation
 and  the resulting collection of fly ash.   The  results of
 our  studies continue to be extremely promising.  It has
 shown that this  new electrostatic method greatly increases
 particulate collection at reduced precipitator sizing.

      Several other projects in our laboratories  have  been
 aimed at control of particulates from the  newer  combustion
 methods such as  low and high Btu gasification  units,  atmos-
 pheric  and pressurized fluidized bed boilers,  MHD  genera-
 tors and combined-cycled systems.  Each of these processes
 has  its own distinct collection problems which are some-
 what dissimilar  to conventional boiler applications.   For
 example, in fluidized bed combustion it is desirable  to
 collect the particulate at elevated temperatures of 1000
 to  1500°F and at pressures of 10 to 20 atmospheres.   The
 flue gas stream  contains fly ash, high levels  of car-
 bonaeous materials and reacted and unreacted  limestone.
 Each of these materials will have a different  effect  on
 the  performance  of the precipitator.

      One of the  projects that we are working  on  under an
 EPA  contract is  to develop a precipitator  for  operation
 up  to 2000°F and 500 psia.  Our bench-scale unit has
 successfully operated generating higher than  expected
 corona  current levels.  We are now seeking a  pilot plant
 demonstration.  Today, research is being geared  to provide
 the  needed know-how and technology to serve the  future
 coal conversion  market place.

 What Affects the Design of an Electrostatic Precipitator?

      In the design of an electrostatic precipitator,  the
 range of conditions over which the system  will operate
 must be specified.  As indicated in Table  1,  information
 on gas  flow rate, temperature, and flue gas analysis  as
 well as chemical composition and electrical characteristics
 of the  dust, particle size distribution, and  mass  loading
 should  be obtained.  In many cases this information  is  not
 available,  extremely general or very variable.   In these
 instances,  assumptions based upon fuel analyses, ash

-------
               ELECTROSTATIC PRECIPITATION                 243
   TABLE 1.   DESIGN PARAMETERS AND DESIGN CATEGORIES
              FOR  ELECTROSTATIC PRECIPITATORS.
PERFORMANCE-RELATED PARAMETERS
GAS FLOW
GAS TEMPERATURE
GAS  (TREATMENT)  VELOCITY
SCA
OVERALL MASS  COLLECTION EFFICIENCY
FRACTIONAL MASS  COLLECTION EFFICIENCY
INLET GRAIN LOADING
OUTLET GRAIN  LOADING
GENERATED PLANT  POWER OUTPUT
FUEL BURNING  RATE

FIRING METHOD AND COAL CHARACTERISTICS
FIRING METHOD
% ASH
% SULFUR
% MOISTURE  (AS RECEIVED)
BTU/LB  (WET)
SAMPLE SOURCE
ASTM CLASS
MINE, STATE
MINE, COUNTY
MINE NAME
SEAM NAME

ASH CHEMICAL  ANALYSIS
Si02          Na20
A1203         Li2O
Fe203         P205
TiO2          S03
CaO           SAMPLE SOURCE
MgO           MEAN
K90           DEVIATION
   SAMPLE TYPE

-------
244                 CLEAN COMBUSTION OF COAL
 content, Btu rating, mine  source,  seam  type,  boiler con-
 figuration, firing methods and  efficiency levels must be
 made prior to selecting precipitator  design parameters.
 Specifications which are too general, or  averages of a
 wide variety of proposed fuels  lead  to  very conservative
 equipment sizing.  However, when many sets of data con-
 sisting of ultimate, proximate  and ash  chemical analyses
 are provided for each fuel and  ash source, the supplier
 can more adequately design a unit  to meet the customer's
 reliability and efficiency needs.  Site-specific design
 constraints such as space  limitations,  equipment location,
 minimum velocity given the ductwork  configuration, etc.,
 must also be considered.

      For many applications precipitator design parameters
 are known from prior experience with  the  same coal.  In
 some cases, estimated parameters must be  developed from
 prior experience with similar fuels.  However, problems
 in meeting design efficiency levels do  develop when the
 fuel being fired differs significantly  from that used for
 the initial design.  If better  information is made avail-
 able in the design stages, good performance can be ex-
 pected with the operating unit.

 Ash Resistivity

      The electrical properties  of  the dust often analyzed
 and reported as resistivity are extremely important in
 designing a precipitator.  For  optimum  precipitator per-
 formance the ash resistivity should  be  in the range of 10
 to 10   ohm-cm.  Figure 5  indicates  the typical effect of
 sulfur and flue gas temperature on ash  resistivity.  As
 can be observed, the ash resistivity  increases with de-
 creasing quantities of sulfur and varies  significantly
 with temperature.  Also, factors such as  flue gas moisture
 and alkali, Na~0 and P2^5  contents as well as Fe20~ levels
 in the ash have a major effect  on  particle conductivity
 and collection.

      With low sulfur fuels, the likelihood of back corona
 occurring increases.  Back corona  causes  a reduction in
 the precipitator operating voltage and  current levels
 which in turn decreases its performance.   Also, higher
 resistivity ash requires more intense rapping to remove
 it from the collecting electrodes  and thereby increases
 the possibility for dust reentrainment, structural plate
 failure and equipment maintenance.   Hot precipitators
 limit back corona and ash rapping  problems by operating
 at higher temperatures and under more optimum resistivity
 conditions.

-------
              ELECTROSTATIC PRECIPITATION
                                              245
  10
12
   10"
o
 i
S
X
o
en
   I09
                             /•--0,5-1,0 I S COAL
                                   -1,5-2,0 % S
                                           COAL
                ,-2,5-3,0 I S COAL
     200    250     300    350     400

                   TEMPERATURE, °F
                                          450
  FIGURE 5,   EFFECT OF SULFUR  CONTENT IN COAL ON FLY
             ASH RESISTIVITY,

-------
246                 CLEAN COMBUSTION OF COAL
      Another method to enhance particulate collection  is
 to modify the resistivity of the fly ash.4  This  technique
 has been successfully applied in more than 15,000 MW of
 commercial installations by the addition of chemical ad-
 ditives to the flue gas upstream of the precipitator.  S03
 addition has been used to reduce the resistivity  of low
 sulfur fuels.  NH3 conditioning has been used to  increase
 the resistivity and cohesiveness of fly ash from  high
 sulfur fuels and thus decrease its reentrainment  during
 rapping.  Sodium salts also have been successfully applied
 on a hot precipitator application to reduce the resis-
 tivity of the fly ash and improve electrical performance.
 We have observed that sodium compounds at elevated tem-
 peratures can react with other trace constituents of the
 ash thereby modifying their electrical conduction.  Each
 of these conditioning techniques can increase the relia-
 bility and flexibility of a precipitator to handle a wider
 range of conditions.   It is expected that during  the next
 several years, ash conditioning will be specified as part
 of some new precipitator installations.

 Migration Velocity

      Whereas resistivity is a. rough indicator of precipi-
 tator performance, effective migration velocity is used
 to specify and determine precipitator size.  We have
 developed a good deal of know-how and experience  to re-
 late coal, fly ash and flue gas properties to migration
 velocity.

      The conventional Deutsch-Anderson equation was used
 for many years to relate migration velocity with  collection
 efficiency

                    n  - 1 - e-(A/V)w

 where

      w = Particle migration velocity, (ft/sec)

      V = Gas flow rate, (actual ft /sec)
                                          o
      A = Effective collection surface,(ft )

      n = Overall mass collection efficiency, (fractional).

 The migration velocity w is a function of electrical ener-
 gization and overall  mass collection efficiency.  The
 variation in w within a given precipitator is caused by
 changing particle size distribution as precipitation pro-
 ceeds in the direction of gas flow.  Since the requirement
 of high collection efficiency corresponds to collection
 of submicron particles, it is understandable why w de-
 creases with increasing efficiency requirement.

-------
                  ELECTROSTATIC PRECIPITATION                247
     The modified migration velocity, wk , as presented by
Matts and Ohnfeldt5  can  be treated essentially as a con-
stant for any application.  We use the modified migration
velocity as a more accurate means of predicting precipita
tor performance.

     The following equations used in sizing prec ipitators
relates modified migration velocity with specific collec-
tion area (SCA) and  efficiency.


            SCA = 16'67  In2(l-n)
                       Wk

              n =|~1  -  X  (H.V.) _ 100
                  L     (Ash) (A.C.)J1UU
where
     SCA =  Specific  collection area (ft2/1000 ACFM) =
            16.67  (A/V)

       H =  Overall mass  collection efficiency, (percent)
i
    Mod
               ified  migration velocity, (ft/sec)

       X = Emission  standard, (lb/106 Btu)

    H.V. = Heating value  of  the coal , (Btu/lb )

     Ash = Ash  in  the  coal,  (fraction by weight)

    A.C. = Ash  carryover,  (fraction by weight)

     The required  overall  mass efficiency, therefore, is
a function of  the  coal  heating value and ash content as
well as the fraction of ash  carryover,  which is a function
of boiler type.  The modified migration velocity is a
function of electrical  energization of  the precipitator
and of gas properties.  It is often conveniently linked
with resistivity level, such that for a moderate resis-
tivity of 10^  ohm-cm the value will be between 1.6 and
1.9 ft/sec whereas for  a very resistive dust it may
approach 0.5 ft/sec.

     Resistivity plays  a  significant role in the selection
of wk and power density.   It is generally predicted from
correlation of  ash components and knowledge of environmen-
tal conditions  such  as  moisture content, gas phase com-
position, and  temperature.   Design exponents based on re-
gression analysis are often  used in relating coal pro-
perties to w ,  SCA,  and power density.
            1C

-------
248                 CLEM COMBUSTION OF COAL
 Several Selected Case Illustrations

      To illustrate some of  the  points  made in the prior
 discussion, several western low sulfur cold precipitafcor
 sizing examples will be discussed.   Western low sulfur •-.
 coals are becoming more important  to the  U.S. energy
 picture and have their attendant particulate collection
 problems.  The selected cases represent some of the
 choices available to the  supplier  and  user in developing
 an appropriate control strategy.

      Table 2 illustrates  three  types of western low sulfur
 fuels each exhibiting different precipitator requirements.

      Fuel A fly ash has a low silica content with a moder-
 ately high level of Na20  present mainly in the form of
 sulfates.  This condition in lignite coal  fly ash yields
 a favorable conductivity  effect; hence the ash has low
 resistivity and is easily collected  with  a cold precipita-
 tor.  Both from technical feasibility  and  economic view-
 points no other control device  option  need be considered.

      Fuel B is a Colorado subbituminous coal.  The fly ash
 analysis indicates a moderate Na,,0 level  and a low base
 (< 20%) and low base/Na20 ratio (< 10).   This suggests
 that no special precaution must be taken  to ensure good
 operation.   However, the high  Si02  content tends to sup-
 press the conditioning effect of the sodium present.  As
 a result, the resistivity is higher  than  moderate sug-
 gesting that cold, hot, and cold SO., conditioned electro-
 static precipitators are  all technically  workable options.
 The hot Na»0 conditioned  option would  not  be required be-
 cause of the favorable base/Na.O i-atio which generally
 suggests that hot precipitators will perform well.

      Fly ash analysis of  fuel C, a Wyoming subbituminous
 coal, indicates a base/Na00 ratio  in the  marginal ranged
 suggesting potential for  electrode fouling with hot pre-
 cipitator operation.  Coupled with this is a high P-0
 content which is responsible for corona quenching and
 power-limited hot precipitator  operation.   The high re-
 sistivity shown for fuel  C  in Table  3  means that a
 large cold precipitator would be required.   Improved pre-
 cipitator operation and smaller precipitator size for this
 type of ash can be obtained by  increasing  the sulfate
 content using SO- conditioning.  Sodium salt conditioning
 can be used to condition  the ash to  reduce its base/NaoO
 ratio and bring it into an  improved  hot precipitator
 operating range.  In the  final  analysis,  given the tech-
 nical feasibility of options to be considered, the optimum
 control device must be determined  on the  basis of econom-
 ics and reliability.

-------
                  ELECTROSTATIC PRECIPITATION
                249
   TABLE 2.   TYPICAL SPECIFICATIONS ON SELECTED WESTERN
              FUELS.
                            Fuel A
Fuel B
Fuel C
Fuel Characteristics

  Rank

  Location

As Fired

  % Moisture

  % Ash

  % Sulfur

  Heating Value
     (Btu/lb)



Fly Ash Analysis,  %

  SiO,
   Fe2°3

   CaO

   MgO
   K2O


   Na2°

   SO o
   P2°5
   Base

   Base/Na20
nite
th Dakota
23.9
6.7
.8
8135
20.6
15.6
.6
9.2
31.0
8.8
3.0
3.4
6.6
1.2
55.4
16.3
Subbit.
Colorado
14.8
5.1
.4
10730
45.1
23.2
.8
5.6
8.8
1.7
.6
2.5
10.7
1.0
19.2
7.7
Subbit.
Wyoming
28.0
8.5
.5
8200
33.0
15.1
1.0
6.1
21.6
4.0
.6
1.1
16.0
1.5
33.4
30.4

-------
250
CLEAN COMBUSTION OF COAL
      TABLE 3.  COLD PRECIPITATOR SPECIFICATIONS FOR
                SELECTED WESTERN  LOW SULFUR COALS.
                  FUEL
      COLD

      COLD PPRT.

      SPECIFICATIONS
            LIGNITE

           NO. DAKOTA
    B


 SUBBIT

COLORADO
    SUBBIT

    WYOMING
      SCA @ 99.5%

      wk, (ft/sec)

      p  , (ohm-cm)

      n  @ .1 lb/106 Btu

      n  @ .05 lb/106 Btu
               250

               1.87

           1 x 109

              98.76

              99.39
  440

  1.06

6 x 10

 97.90

 98.95
10
  800

  0.59

3 x 10

 99.04

 99.52
12
      Figures 6 and 7  indicate  the  relative capital in-
 vestment  ($/KW) and annual  cost  (mills/KWH)  for these
 various options.   From  a  strictly economic  standpoint,
 it can be observed that  SO   conditioning  in  conjunction
 with cold precipitation, wnen  feasible,  is the lowest
 capital cost approach.   However, because  of  the relative
 newness of this approach, users  are concerned with its
 long-term reliability.   Hot precipitation can be an at-
 tractive  option and does offer fuel flexibility and relia-
 bility.   Care must be  taken not  to universally apply any
 one solution since we  have  observed that  ash properties
 can affect even the performance  of hot precipitators.
 Before selecting a particulate control strategy, the
 various fuels to be combusted  in the boiler  should be
 reviewed  and their ashes analyzed.  A study  should be made
 to determine particulate equipment sizing and its flex-
 ibility in handling changes in fuel and  in boiler oper-
 ation.  Capital and operating  costs should be calculated
 and only  then should  a particulate control strategy be
 select ed .

-------
                ELECTROSTATIC PRECIPITATION
                                                    251
I-

UJ
en
UJ
Q.
<
O
80

70

60


50


40



30

25


20



 15
   10
    9

    8

    7
        I i  i—I—I—|MI|
1—r
    95.0
                 98.0     99.0.
                      EFFICIENCY, %
             99.8  99.9
    FIGURE 6,   ECONOMIC COMPARISON OF CAPITAL INVESTMENTS
               FOR VARIOUS ELECTROSTATIC PRECIPITATOR
               OPTIONS FOR A 600 flW POWER PLANT,

-------
252
CLEM COMBUSTION OF COAL
       95.0
   98.0     99.0

      EFFICIENCY, %
99B  99.9
      FIGURE 7,   ECONOMIC COMPARISON OF ANNUAL COSTS FOR
                 VARIOUS ELECTROSTATIC PRECIPITATOR OPTIONS
                 FOR A 600 RW POWER PLANT,

-------
                  ELECTROSTATIC PRECIPITATION                253
References

1.  White, H. J.  "Electrostatic Precipitation of Fly Ash:
    Outlook for  Future  Growth" JAPCA Vol. 27, No. 1,
    (January 1977) .

2.  Ibid.

3.  "Technology  for  Electrostatic Freeipitators," Indus-
    trial Gas Cleaning  Institute Inc., Publication .
    No. E-P1.

4.  Atkins, R. S.  and D.  H.  Klipstein, "Improved Precipi-
    tator Performance by  S0~ Gas Conditioning," Proceedings
    of the American  Power Conference, Vol. 37 (1975),
    pp. 693-700.

5.  Matts, S. and  P-0 0*hnfeldt, "Efficient Gas Cleaning
    with  SF Electrostatic Precipitators."

6.  Walker, A. B.  "Operating Experience with Hot Precipi-
    tators on Western Low Sulfur Coals," presented at the
    American Power Conference, Chicago, Illinois
    (April 18-20,  1977) .

7.  Bubenick, D.  V.  "Economic Comparison of Selected
    Scenarios for  Electrostatic Precipitators and Fabric
    Filters," presented at the 70th annual meeting of
    AP,CA, Toronto,  Ontario,  Canada (June 20-24, 1977).

-------
254                    CLEAN COMBUSTION OF COAL

-------
                                                                    255
                  STATE  OF  THE  ART  —  FABRIC  FILTRATION
                            Richard L.  Adams
                 Vice  President,  Systems  and  Technology
                         Wheelabrator-Frye Inc.
                    Air Pollution  Control Division
INTRODUCTION

     If this conference had "been held five years ago, there would not
have been a paper entitled "State of the Art — Fabric Filtration."
Five years ago, electrostatic precipitators were the only particulate
removal device seriously considered for high efficiency removal of fly
ash generated by the burning of coal.  Certainly, a few pilot fabric
filters had been in successful operation by this time but the first
full-scale units were still on the drawing board.  Today, fabric
filters are being installed on both utility and industrial coal-fired
boilers.  Since fabric filters are probably one of the oldest means of
removing solid particulate from the gas stream, what are the reasons
for the long delay in their consideration for use on coal-fired boilers
and their quick acceptance once the barrier was broken?

     The promulgation of the laws pertaining to new source emissions by
the EPA in 1970 brought about requirements for emission control that
were in most cases substantially more stringent than those which had
been in existence up to that time.  These requirements for increased
efficiency of the pollution control devices substantially increased
the size and thereby the cost of an electrostatic precipitator instal-
lation.  In addition, new sulphur regulations have increased the usage
of low-sulphur, Western coal and this has also placed an increased
burden on an electrostatic precipitator.  Finally, because of earlier
difficulties with precipitator performance, a greater degree of con-
servation was included in precipitator sizing and specifications.   The
net result of all of these changes was to cause a marked increase in
the cost of an electrostatic installation and suddenly the fabric fil-
ter was more than competitive in certain areas.  Figure No. 1 shows
the cost of both the fabric filter and an electrostatic precipitator
as a function of plant size and in the case of precipitators, specific
collecting area.  It will be noted that where a cold electrostatic
side electrostatic precipitator has a requirement for an SCA above
approximately 500 the fabric filter will normally result in a less
expensive installation.

     Today, there are several fabric filters in operation, some with
operating histories of over four years.  Within the next year, many
large units (up to 575 megawatts) are scheduled to go into operation.
This paper will briefly trace the history of fabric filtration on
coal-fired boilers over the past five years and present our thoughts
as to the future.

-------
256
CLEAN COMBUSTION OF COAL
                 34-

                 28-

              226-
              S  24-

              i  22-
              UJ  "
              \  20-
              
              S  18-

              2  16-

              ?  14-
              c
-------
                           FABRIC FILTRATION                         257


TYPES OF FABRIC FILTERS

     For those somewhat unfamiliar with fabric filtration, it should be
pointed out that there are two types of fabric filters that may be con-
sidered for use on a coal-fired boiler.  These are the so-called high-
energy fabric filter and the low-energy fabric filter.  In a low-energy
fabric filter, as shown in Figure No. 2, filtration is accomplished by
building a filter cake using a woven filter fabric as a grid or matrix
to support this cake.  The cake is intermittently removed from the fil-
ter fabric by use of mechanical agitation or by backwashing the cloth.

     In a high-energy fabric filter, as shown in Figure No. 3, a felt
is utilized which acts as a depth filter as well as supporting a cake
on the surface of the felt.  In order to remove the collected material
from this type of filter fabric, it is necessary to apply a great deal
more energy, thus the term high-energy fabric filter.  Normally, this
energy is applied by utilizing compressed air which will both agitate
and backwash the filter fabric and remove the filter cake.  Currently,
high-energy filters have been applied only to the smaller industrial
boilers.  The filter fabric utilized has been felted Teflon™ or a
heavy woven fiberglass.  We would expect to see, however, a greater
use of high-energy fabric filters in this service in the future.

     To-date, most of the major installations in the utility industry
in this  country have utilized low-energy filtration.  Fiberglass is
the normal filter fabric employed in low-energy filters but there is
a great  deal of discussions currently with regard to the proper finish
on the fiberglass fabric.  There are also two very different fabric
cleaning systems used today on low-energy fabric filters.  There is  a
system called "deflate and shake" and a system utilizing "reverse
flow."   Both these will be described later.  It is interesting to note
that while fiberglass has been the normal fabric in this country,
synthetics such as acrylics and polyesters have been used with a great
deal of  success 'in Australia.

     To  this day, fabric filtration is still more of an "art" than an
engineering science-  There are approximately ^0 variables that can
 affect the performance of a fabric filter and most of these cannot be
predicted at the time of initial  design.  We are thus in the position
 of having to rely heavily on  actual  field operating experience to make
 judgments as to the proper application of fabric filters.  Just as
there is little or no correlation between resistivity and the precipi-
tator sizing  factor "W" there is  also  little correlation between
 laboratory values of "K" factor or filter drag  and actual field per-
 formance. It might be well, therefore,  for us  to review three  fabric
 filter installations on  coal-fired boilers which have been in  operation
 for  sufficient time to allow  us to make  judgments as  to their perform-
 ance.
 OPERATING EXPERIENCE

      In reviewing these installations, we will examine the operating
 problems encountered with each.   However, these problems have been

-------
258
CLEAN COMBUSTION OF COAL
                      Figure 2
                 Low Energy
                     Filter

-------
  FABRIC FILTRATION
                             259
     Figure 3
High  Energy
    Filter  '

-------
260                    CLEAN COMBUSTION OF COAL


minor and it should be noted that none of these units has caused the
loss of production on a single kilowatt of electrical power.  When
compared to the records of other types of control devices being ap-
plied to utility boilers, we believe this is an enviable record.

     The first full-scale fabric filter installed in the United States
on a coal-fired utility boiler is the installation made at Pennsylvania
Power & Light's Sunbury Station.  There are four units, each handling
222,000 ACFM % 325°F from two 87.5 megawatt boilers.  The first of
these units was started in February of 1973.  The detail design data
for these units is given in Table Wo. 1.

     The performance of the units at Sunbury since their initial start-
up has been excellent.  The problems encountered have been related
primarily to design concepts developed for this job due to peculiar-
ities in space and plant layout, since it was a retrofit.  The fabric
filter installed at Sunbury operates as a pressure unit, i.e., it is
on the discharge side of the I.D. fans, and the gases are forced through
the fabric under pressure and then exhausted to the atmosphere.  Since
reverse gas cleaning is utilized, it is not necessary from a fabric
cleaning standpoint to have completely tight gas shutoff valves on each
compartment.  However, the valves supplied on this installation should
also function to allow maintenance workers to enter the compartments
for inspection and bag replacement and since they are not gastight on
this installation, the only time it is possible to enter the compart-
ments without the use of protective equipment has been during boiler
shutdown.  Adequate attention to this detail during the initial design
phase could have overcome the problem.

     Since the Sunbury baghouses operate under pressure, they were also
designed such that the reverse air fan was located on the dirty gas
side of the unit and the reverse air fans handle fly ash laden gases.
Wear problems have been encountered with these fans and maintenance
in this area has been high.  The fans have not, however, caused a
reduction in fabric filter availability.

     The next units which we would like to examine were installed at
the Nucla Station of the Colorado-Ute Electric Association.  In this
case, there were three units each ventilating a 13 megawatt stoker-
fired boiler.  The detail design data for these units is given in
Table No. 2.  The first of these units was placed into service in
December 1973 and the others followed in early 191^.  These units were
suction-type with I.D. fans located downstream from the fabric filter
and none of the problems associated with the Pennsylvania Power & Light
units at Sunbury were encountered.

     Shortly, after start-up, however, it was noted that there was ex-
cessive wear on the bottom of the fabric tubes.  The  type of tube
attachment used in these collectors is shown in Figure No. U.  A study
of the problem was initiated and it was determined that the density of
the collected material was not 50 Ib. per cubic foot as specified by
the consulting engineer but rather approximately 25 Ib. per cubic
foot.  This, coupled with the hopper emptying procedures, caused an
excessively high level of collected fly ash in the hoppers and thus
caused reentrainment and excessive wear on the bottom of the filter

-------
                  FABRIC  FILTRATION
                                          261
                    Table No. 1
             Pennsylvania Power & Light
                    Sunbury, Pa.
Boiler            Pulverized coal
Size              1*00,000 lb./hr. steam
                  222,000 acfm % 325°F
Coal              Anthracite and coke
Design            2.07 to 1 air/cloth ratio
                  2 gr./acf, 2.5-3.5 inch w.g.
                  99-9$ plus
Boiler

Size

Coal

Design
   Table  No.  2
   Colorado-Ute
 Spreader stoker
 Traveling grate
 132,000  Ib./hr. steam
 86,200 acfm/300°F
 Ave.  0.7% sulfur (bituminous)
'35$ ash
 3.35  to  1 air/cloth  ratio
 Ave.  1.5 grains
 Ij-inch pressure loss, 99-9$  +  design

-------
262
CLEAN COMBUSTION OF COAL
                    Figure 4
             Tube Attachment
                 and Baffles
          air side ^
              baffle depth-8"
                                      t

-------
                           FABRIC FILTRATION                        263


bags.  In order to correct the problem, the hoppers were lowered ap-
proximately two feet to provide added capacity and additional baffling
was installed.  Finally, thimbles for each DUSTUBE which form pro-
tecting extensions were installed on the bottom of the tube sheets.
This arrangement is also shown in Figure No. IK

     It should be noted that a fabric filter almost identical to those
at Colorado-Ute was also installed at this time on a coal-fired boiler
in the Nyssa, Oregon, plant of the Amalgamated Sugar Company.  This
unit was put into service in September of 1973 and had a history very
similar to that described for the units at the Colorado-Ute Electric
Association.  The operating data for the unit at Amalgamated Sugar is
given in Table No. 3.

     The third unit to be examined is the Penna. Power & Light instal-
lation at Holtwood, Pennsylvania.  This unit handles 200,000 CFM or
approximately 50% of the discharge from this 80 megawatt boiler.  The
remaining 50% is handled in an existing wet scrubber.  The unit became
operational in April 1975-  The design and operating data is given in
Table No. k.  Shortly, after start-up of the PP&L installation at
Holtwood, it was noted that there was a gradual increase in the oper-
ating pressure drop across the unit.  This increase did not seem to
be related to any external operating conditions.  An investigation into
the problem revealed that it was most likely related to the finish used
on the fiberglass filter fabric.

     All of the early deflate-and-shake units, including the pilot
plants which had been operated over the prior ten years, had used
fiberglass filter fabric coated with a silicone graphite lubricant.
The filter bags in the PP&L installation at Sunbury, however, were
fiberglass-coated with Teflon B as a lubricant.  The Sunbury bags
were equipped with rings and cleaned by reverse air only.  It was
decided in view of the excellent operating history of Teflon-coated
fiberglass at Sunbury to utilize Teflon-coated fiberglass on the
Holtwood installation.  It wasn't recognized that the finish might
not be compatible with the different cleaning method utilized at
Holtwood.

     Investigation of the increasing pressure drop problem at Holtwood
indicated that it was most likely due to the Teflon B finish used on
the fiberglass bags.  Photomicrographic studies showed that the Teflon
appeared to flow and to fill some of the crevices between individual
fibers.  Figure No. 5 shows this situation.  As a contrast, the
silicone graphite finish, Figure No. 6, appeared to leave the fibers
in a more open condition.  It is unknown why the Teflon B finish
performed satisfactorily in conjunction with the ringed reverse air
cleaning at Sunbury and in a less satisfactory manner in conjunction
with the deflate-and-shake cleaning at Holtwood.

     To confirm the results of the laboratory investigation, a complete
compartment in the Holtwood baghouse was fitted with silicone graphite
fiberglass fabric.  Operating comparisons were made with respect to
volume and pressure drop between the compartment with the silicone
graphite finish and the compartments with Teflon B finish.   The results
indicated that silicone graphite would operate at a substantially  lower

-------
264
CLEAN COMBUSTION OF COAL
                             Table Ho. 3
                      The Amalgamated Sugar Co.
         Boiler            Spreader stoker
                           Traveling grate
         Size              200,000 rb./hr. steam
                           91,800 acfm/300°F
         Coal              0.5% sulfur (bituminous)
                           5$ ash
         Design            3-56 to 1 air/cloth ratio
                           2.0 grains, 3-inch pressure loss
                           99.9% design
                              Table No.  k
                       Pennsylvania Power  & Light
                            Hpltvood, Pa.
          Boiler           Pulverized coal
          Size              1/2  of 700,000 Ib./hr.  steam
                           200,000 acfm 360°F
          Coal              1.8% sulfur  (anthracite)
          Design           2.1+2 to 1 air/cloth ratio
                           J.h  to 8 grains, 2.8-inch pressure  loss
                           99-9$ + design

-------
          FABRIC FILTRATION
265
            Figure 5
      Fibreglass With
Teflon®B  Finish (10,000 X)

-------
266
CLEAN COMBUSTION OF COAL
                    Figure 6
              Fibreglass With
         Silicone Graphite Finish
                 (10,000 X)

-------
                            FABRIC  FILTRATION                       267
pressure drop when handling the same volume of gas.  The Holtwood bag-
house was then changed so that 11 of the compartments, except for the
two control compartments which were left with the original Teflon B
finished bags, were converted to silicone graphite finish.  This con-
version resulted in a substantial lowering of the operating pressure
drop of the unit.

 CURRENT DESIGN PHILOSOPHY
      It may be well to examine current design philosophies  on fabric
 filters to be applied to coal-fired boilers.   It is far too early to
 tell which of these design philosophies will prevail over the long
 run.  Each has advantages and disadvantages and only extended periods
 of operation on many different types of boilers, fired with differing
 coals, will give the ultimate answer as to which represents the  best
 approach to the problem.

      While there is currently a lack of operating data, we  believe
 that pulse-jet type collectors, i.e., high energy equipped  with  heavy
 woven fiberglass filter bags will be an attractive answer for some of
 the industrial segments of this marketplace.   It is possible that this
 type of collector may extend itself into the large utility  installa-
 tions, but additional work history is needed before this type of unit
 can be considered on a major boiler installation.

      Among the low-energy fabric filters, there are two basic design
 philosophies.  One philosophy indicates the use of higher air-to-cloth
 ratios and deflate-shake cleaning to maintain bag pressure  drop.  The
 other philosophy uses lower air-to-cloth ratios and reverse air
 cleaning of ringed bags.  Generally speaking, the design air-to-cloth
 ratio with deflate-shake cleaning would be approximately 3  to 1  and
 the design air-to-cloth ratio for reverse air cleaning will be approxi-
 mately 2.25 to 1 in order to achieve the same overall system pressure
 drop.

      The reason that shake-deflate cleaning can operate at  somewhat
 higher air-to-cloth ratios is because the bags are cleaned  more  uni-
 formly.  Figure No. 7 shows a ringed bag being cleaned by reverse air.
 In the filtering mode, it is normal for the finer particles to end up
 in the top of the bag and the heavier particles to be collected  in the
 lower areas of the bag.  When reverse flow is applied, the  filter cake
 is removed from the lower surface of the bag first.  This opens  up the
 filter fabric in this area and allows short-circuiting of the cleaning
 airflow and thus interferes with the uniform cleaning of the filter
 bag.  No such phenomena exists in the mechanical agitation  of a  filter
 bag where the mechanical energy is distributed evenly throughout the
 length of the bag.

      The Dorsey equation given in Figure No.  8 is the general formu-
 lation for describing the performance of a fabric filter.  In our
 experience, we have found the Dorsey equation to be inaccurate but it
 probably -still represents the best simple mathematical model for per-
 formance of a filter.  Unfortunately, the value "K" is extremely
 difficult to define and in most cases can only be calculated after a

-------
268
CLEAN COMBUSTION OF COAL
                   Figure 7
              Bag Cleaning
                 Methods

-------
            FABRIC FILTRATION                    269
             FIGURE HO, 8
       Dorsey Equation

      operating pressure drop

            AP=KCVXT

K   is the specific dust-fabric filter resistance
    coefficient, usually of order 1 to 10
    (in WG/lb/ft2/ft/min),
C   is the inlet dust concentration (Ib/ft3),
V   is the average filtering velocity
    (ft/min), and
T   is the operating time between cleaning
    of a particular compartment (minutes).

-------
270                     CLEM COMBUSTION OF COAL
unit is in operation.  Use of this equation, however, does allow us to
compare the relative performance of the two cleaning systems as they
are in operation at the Pennsylvania Power & Light Stations at Holtwood
and Sunbury.  Both toilers burn similar coal.  Table No. 5 shows the
comparative data between the two installations.  We do not believe that
this comparison is quantitative but certainly in a qualitative way
indicates that the deflate-shake cleaning system provides substantially
lower "K" or better filter performance than does the reverse-air
cleaning system.

     It has been suggested that bag life may be better with reverse-air
cleaning.  The facts are not available to support this theory at the
present time, but if, indeed, it is true, then the relative merits of
the two methods will have to be judged on an economic basis.  The
deflate-shake method will provide for a smaller and generally less
expensive initial installation.  Should bag life be shorter, this is
partially offset by the increased cost of the bags in the reverse-air
installation due to the fact that there are a greater number of bags
and the bags have sewn-in anticollapse rings.  We strongly urge poten-
tial users to make a complete economic evaluation between the two pro-
posed cleaning methods.  Currently, this would appear to be the most
reasonable form of evaluation.

     It should be pointed out that in light of the successful operation
of fabric filters on smaller boilers, there are currently several units
under construction on boilers in sizes ranging from 350 to 575 megawatts.
The first of these new large installations will go into operation in the
first quarter of 1978 at the Monticello Station of Texas Utilities.   In
this case, there are two fabric filter systems each handling 80$ of the
discharge from a 575 megawatt boiler.  The remaining 20% of the discharge
will be handled by existing electrostatic precipitators.  The design
data on these units is given in Table Wo. 6.   The utility industry is
watching with a great deal of interest these new, large installations
and once they are successfully on-stream, we believe that the fabric
filter will be considered a viable device for all size boilers.

SCL REMOVAL
     No presentation regarding the state of the art  on fabric  filtration
with regard to coal-fired boilers will be complete without  mentioning
that fabric filters are also being considered for the removal  of SOp.
There have been successful pilot operations utilizing a fabric filter
coated with a naturally occurring sodium bicarbonate called Nahcolite.
These filters have been able to remove up to 90$ of  the SOg entering the
system and cost analysis indicates that the total system will  be more
than competitive with the existing wet scrubbing systems.

     Finally, fabric filters are also proposed as particulate  collectors
and chemical contactors following a spray dryer which can remove up to
90% of the S02 entering the system.  We expect to see increased atten-
tion paid to both of these SOg removal processes in  the near future.

-------
                 FABRIC FILTRATION
                                        271
                    Table No.  5
                     Operating
                    Comparisons
                   K = AP/CT^T
    Shake/Deflate
        3.75"           AP
8 gr./cu. ft. * TOGO      C
       2.8 fpm            VX
      UT.O min.           T
         8.9              K
                 Reverse Air
                     3.0"
             2  gr./cu. ft. + TOGO
                    2.0 fpm
                    33.0 min.
                     79-5
                     Table Wo.  6
 Boiler
 Size

 Coal

 Design
 Texas Utilities
Pulverized coal
2 boilers 575 MW each
1,8UO,000 acfmAOO°F each
1.2% sulfur  (lignite)
13$ ash
2.9 to 1 air/cloth ratio
5.6 to 8.8 grains, h.5-inch pressure drop
99.9$ + design

-------
272                     CLEAN COMBUSTION OF COAL
     We believe that the almost-perfect on-stream reliability record of
fabric filters on coal-fired boilers combined with their very economical
first costs and operating costs will result  in an ever-increasing
acceptance on both utility and industrial coal-fired boilers.   In a
way, the use of fabric filters on boilers is in its infancy;  however,
we have learned much over the past five years and we believe  that the
time has now come where the fabric filter can take its  place  alongside
the electrostatic precipitator as a device that can economically meet
the most stringent emission requirements.

-------
                                                                    273
                   STATUS OF FLUE GAS DESULFURIZATION
       THE FEDERAL RESEARCH, DEVELOPMENT AND DEMONSTRATION PROGRAM
                 Julian W. Jones and Michael A. Maxwell
                  U.S. Environmental Protection Agency
              Industrial Environmental Research Laboratory
                 Research Triangle Park, North Carolina
INTRODUCTION
     One of the Nation's major energy-related environmental problems
concerns the need to control sulfur dioxide  (S0_) emissions from
stationary fuel combustion sources.  Flue gas desulfurization (FGD)
is the term used to denote processes for removal of S02 from flue
gas, usually by means of a gas scrubbing operation.  FGD is the
major near-term technological approach to meeting new source per-
formance standards when utilizing high-sulfur coal supplies.  In
addition, U.S. Environmental Protection Agency  (EPA) studies indicate
that FGD is competitive in cost with advanced control methods, such
as chemical coal cleaning and fluidized bed  combustion; therefore,
FGD should play an important role in controlling emissions even in
the 1990's.

     The current Federal energy program calls for a policy of rapid
expansion in the use of coal.  If this policy is to be successfully
implemented, it is essential that FGD technology be fully optimized
for application in the utility and industrial sectors.

     In the Federal energy/environment research and development
program, FGD technology development has been given a high priority.
Several FGD studies, pilot plants, prototypes, and demonstration-
scale facilities have been funded by EPA.  Although prior progress
had been achieved in FGD development, the overall pace of develop-
ment was increased by the initiation, in 1975, of the Federal inter-
agency effort.

     The FGD program is being conducted through EPA's Industrial
Environmental Research Laboratory in Research Triangle Park, North
Carolina (IERL-RTP).  The program has aimed  at demonstrating reliable
and cost-effective FGD processes, including  both nonregenerable
(throwaway) and regenerable  (saleable by-product) systems.  EPA's
key program in the nonregenerable area is the lime/limestone proto-
type test program at the Shawnee Steam Plant of the Tennessee Valley
Authority (TVA), near Paducah, Kentucky.  Other major activities in
the area of nonregenerable technology include full-scale demonstra-
tions of double-alkali scrubbing systems, and a comprehensive program
in scrubber sludge disposal and utilization.

-------
274                    CLEAN COMBUSTION OF COAL
     In the regenerable FGD area, an aggressive program has also
been pursued.  Full-scale demonstrations of the Cat-Ox (producing
80% sulfuric acid), Wellman-Lord (W-L)/Allied Chemical (producing
sulfur), magnesium oxide (producing sulfuric acid), aqueous carbonate
(producing sulfur), and the citrate* (producing sulfur) processes
have been undertaken.

     A number of supporting studies have also been initiated for
both nonregenerable and regenerable processes.  Included are design
and cost evaluations for advanced S02 removal technologies, byproduct
marketing studies, bench-scale research on key processing steps,
investigation of reductants for SO- to sulfur, energy optimization
studies, and a comparison of utility/industry equipment reliabil-
ity/availability.

     To complement these efforts to develop technology, companion
technology transfer efforts are also underway.  Through a series of
briefings, symposia, capsule reports, summary reports, and a survey
of FGD installations, the industry is being aided in its efforts to
stay abreast of the rapidly advancing FGD technology.  The technology
transfer efforts include reporting, to the fullest extent possible,
the status of the many full-scale utility and industrial FGD systems
designed, constructed and operated under private funding.

     The purpose of this paper is to summarize the results and
status of the efforts described above as well as the status of FGD
technology in the electric utility industry.
NONREGENERABLE SYSTEMS

Lime/Limestone Scrubbing

     Lime/limestone processes are the most highly developed FGD
processes, having the greatest amount of operational experience.
Over 90% of the FGD systems, which have been built, are under con-
struction, or are being planned for service by the early 1980's, are
lime/limestone processes.

     The Shawnee Program.  An important part of the lime/limes tone
effort involves the operation of a prototype scrubbing test facility
at Shawnee Steam Plant.  This versatile facility allows comprehensive
testing of up to three 10 MW scrubber types under a variety of oper-
ating conditions.  Bechtel Corporation of San Francisco designed the
test facility and directs the test program, and TVA constructed and
operates the facility.

     The major concerns regarding lime/limestone scrubbing have cen-
tered around potential operating difficulties caused by scaling and
     *The citrate process is based on pilot plant work by the
U.S. Bureau of Mines

-------
                       FLUE GAS DESULFURIZATION                      275


plugging (especially in the mist  eliminator area), the large quanti-
ties of waste sludge generated, and  the high costs (capital and
operating) of scrubbing.   It  is toward these areas of concern that
the Shawnee program has been  directed.

     The testing at Shawnee since 1972 has made major contributions
toward improvement of  lime and limestone  scrubbing technology.  The
most significant results  to date  include:   (1) demonstration, on a
10 Mw scale, that conventional lime/limestone  systems can be operated
reliably (two separate reliability problems have been identified—
scaling and accumulation  of soft  mud-type solids—and methods to
control each have been demonstrated);  (2) evidence that mud-type
solids deposition is a strong function of alkali utilization and at
high utilization  (greater than about 85%) these solids are much more
easily removed;  (3) demonstration of equipment or process variations
which individually improved alkali utilization  (thereby reducing
operating  costs), improved S02 removal efficiency, and favorably
influenced the  system  chemistry;  and (4)  development of useful
industrial tools, such as the design/economic  study computer program
and the computerized Shawnee  data base.

     The EPA Pilot FGD Scrubber Program.  The  FGD pilot plant operated
by IERL-RTP consists of two scrubbers having a flue gas capacity of
about 0.1  MW each.  They  have been in operation since 1972 for the
principal  purpose of providing in-house experimental support for
EPA's larger, prototype scrubber  test facility at Shawnee Steam
Plant.  In addition to supporting Shawnee,  the pilot plant also
provides IERL-RTP with the capability to  independently evaluate new
concepts in lime/limestone scrubbing technology.

     Results from this pilot  unit indicate  that forced oxidation
limestone  scrubbing can be achieved  in a  two-stage scrubbing system
promoting  formation of gypsum, a  more desirable form of calcium
solids, and improving  limestone utilization.   An extension of this
work has recently been undertaken at the  Shawnee test facility.
Further tests at  IERL-RTP have been  directed toward achieving the
oxidation  step  in a single-stage  scrubber,  and information has been
developed  toward application  in larger commercial scrubbing systems.
Thus far,  it has been  demonstrated in the IERL-RTP scrubber that
single-stage forced oxidation can be achieved with no loss of S0»
removal efficiency.

     A study of the formation of  solid solutions in lime/limestone
scrubbers  was completed which verified earlier findings made at
IERL-RTP that sulfate  can be  purged  with  the solids at low oxidation
levels while maintaining  subsaturated liquor.  Larger-scale studies
of this "subsaturated" operation  mode are being made at Shawnee and
Louisville Gas  and Electric's (LG&E) Paddy's Run lime scrubber.

     Bahco Program.  The  AB Bahco Ventilation  (Sweden) lime scrubbing
process has been  installed on about  20 small industrial-size oil-
fired boilers outside  of  the  United  States.  Research-Cottrell is
the licensee in the United States for this  process.  The Bahco

-------
276                    CLEAN COMBUSTION OF COAL
 system appears particularly suited for small  industrial  applications;
 it  is manufactured  in  standard sizes of about  5-50 MW  equivalent.
 The system  is readily  adaptable to a high degree of automation.
 Although  automation results in a somewhat higher capital investment
 cost initially,  labor  costs are low because boiler operating  personnel
 can also  handle  operation of the FGD system.

     The  Air Force  contracted with Research-Cottrell to  install  a
 Bahco system for SOo and particulate control  on up to  seven small
 coal-fired  heating  boilers  (approximately 21 MW equivalent total)  at
 Rickenbacker Air Force Base near Columbus, Ohio.  EPA  is sponsoring
 a  2-year  test program  on this system.  Although numerous mechanical
 problems  have been  encountered since startup  in March  1976, results
 generally have been promising.  Despite the high particulate  loading,
 which at  times has  been predominately very small sooty particles,
 design specifications  for both the particulate and 862 removal
 efficiency  have  been exceeded consistently.

     LG&E Scrubber  Test Program.  In November  1974, results from
 the IERL-RTP pilot-plant testing were reported which showed that
 lime and  limestone  SC^ scrubbers could be operated subsaturated  with
 respect to  dissolved CaSO,'2H?0 (gypsum).  This mode of  operation
 avoids the  problem  of  gypsum scaling on the scrubber internals.
 Subsequent  investigation indicated that at least two commercial
 scrubber  systems were  operating subsaturated with respect to  gypsum,
 one at the  Mitsui Aluminum Plant in Omuta, Japan, and  the other  at
 Paddy's Run Station of LG&E.

     Because of  EPA's  interest in studying the subsaturated mod.e of
 operation on a full-scale system, a program was undertaken at LG&E
 in  the spring of 1976  to evaluate operational and chemical factors
 (identified by scrubber testing at IERL-RTP and Shawnee) which
 appear to have an effect on subsaturated-operation.  The carbide
 lime phase  (baseline tests) of the test program was initiated in
 October 1976 and was concluded in December 1976.  No major scrubber
 operational problems occurred during these tests.  Waste sludge  from
 the system  was collected for sludge treatment/disposal tests which
 are being conducted in conjunction with the scrubber test program.

     The  commercial lime phase of the test program was initiated in
 March 1977; shortly after start-up, scaling occurred in  the scrubber.
 The scaling problem was a result of higher oxidation (than with
 carbide lime) and a lack of gypsum crystals, causing "locally"
 excessive gypsum saturation levels.  The marble bed scrubber  used  in
 the tests is also more prone to difficulties from scaling.  It has
 been concluded that the carbide lime contains an oxidation inhibitor
 in  trace  quantities.   Currently, commercial lime testing has  resumed
 with the  addition of small quantities of magnesium oxide to prevent
 scaling.  High S02  removals, along with no significant operational
 problems, have been reported.

-------
                       FLUE GAS DESULFURIZATION                     277


Double-Alkali Scrubbing

     The double-alkali process provides an alternate wet scrubbing
"throwaway" system to the more prevalent  lime/limestone slurry
scrubbing processes.  Such systems  employ a  clear liquid absorbent
rather than the slurry used in lime/limestone processes.  As a
result, the scrubber in a double-alkali unit is expected to be less
prone to fouling and plugging problems.

     The double-alkali process is now offered commercially by several
companies for control of industrial and utility boilers.  Process
capabilities include 90% or more SCL removal, less than 2% energy
consumption exclusive of reheat energy, close to 100% lime/SO-
stoichiometry, and soda ash consumption in the range of 5% of the
lime on a molecular basis.  However, these processes may in certain
instances be more costly than lime/limestone systems.

     Combustion Equipment Associates  (CEA)/A.D. Little (ADL) Program.
After initial in-house engineering  feasibility studies and laboratory
experiments in 1971 and 1972, EPA contracted with ADL in May 1973 to
conduct a laboratory and pilot-plant study of various double-alkali
modes of operation.  These efforts  included  the study of "dilute"
and "concentrated" systems, lime and limestone regeneration, sulfuric
acid addition for sulfate removal,  and solids characterization.

     In early 1975 the project was  expanded  to include a prototype
test at the 20 MW facility installed at the  Scholz Steam Plant of
Gulf Power Company by The Southern  Company and constructed by CEA.
Testing at the Scholz Steam Plant lasted  from February 1975 to July
1976, with the EPA-sponsored portion of the  testing beginning in May
1975.  As a whole, the prototype system performed very well and
indicated that a double-alkali system could  be a viable FGD system
for coal-burning utilities.  SO^ removal  was generally in the range
of 90 to 99%.

     General Motors  (GM) Industrial Boilers  Demonstration.  GM and
EPA have participated in a cooperative program to demonstrate, test,
characterize, and evaluate GM's "dilute"  mode double-alkali system
for control of SO- emissions from coal-fired industrial boilers.
The program was conducted at GM's industrial boiler complex in
Parma, Ohio.  The system, consisting of four coal-fired boilers
having a steaming capacity of 320,000 Ib/hr  (equivalent to 32 MW
electric generating capacity), was  constructed and operated by GM.
ADL designed and conducted the test program  to evaluate the system
with funding from EPA.  The test program  consisted of three 1-month
intensive test periods and longer term nonintensive testing.  Each
of the intensive tests evaluated a  slightly  different flow scheme.

     LG&E Double-Alkali Demonstration Program.  In September 1976,
EPA contracted with LG&E for a cost-shared,  full-scale coal-fired
utility demonstration of the double-alkali process at the 280 MW Cane
Run No. 6 boiler.  The demonstration project consists of four phases:

-------
278
CLEAN COMBUSTION OF COAL
 (1) design  and  cost  estimation;  (2) engineering design,  construction,
 and mechanical  testing;  (3)  startup and performance  testing;  and (4)
 1  year  of operation  and  long-term testing.  Construction is  expected
 to be complete  by  the  end of 1978, and testing will  begin in early
 1979.   A contract  with Bechtel Corporation has been  initiated to
 design  and  conduct a test program for the LG&E facility  and  evaluate
 the process technically  and  economically.

     FGD Sludge Disposal.  One of the major considerations involved
 in the  selection,  design, construction and operation of  nonregenerable
 FGD systems is  the disposition of large quantities of sludge.   Since
 1972, efforts have been  underway in the EPA FGD research and  develop-
 ment program to evaluate and demonstrate sludge disposal techniques.
 Currently twelve projects are being conducted in this area;  these
 are listed  in Table  1.   These projects address the two major  environ-
 mental  concerns associated with FGD sludge disposal:  (1)  the water
 pollution potential  of soluble materials, and (2) the land degrada-
 tion potential  of  physically unstable wastes.

              Table  1:   EPA  Projects in FGD Waste Disposal
           Project  Title

 FGD Waste  Characterization,
 Disposal Evaluation, and Transfer
 of  FGD  Waste  Disposal Technology

 Lab and Field Evaluation of 1st
 and 2nd Generation FGD Waste
 Treatment  Processes

 Ash Characterization and Disposal

 Studies of Attenuation of FGD
 Waste Leachate by  Soils

 Establishment of Data Base for
 FGC Waste  Disposal Standards
 Development

 Shawnee FGD Waste  Disposal Field
 Evaluation

 Louisville Gas and Electric
 Evaluation of FGD  Waste Disposal
 Options

 FGD Waste  Leachate/Liner Compat-
 ibility Studies

 Lime/Limestone Wet Scrubbing
 Waste Characterization

 Dewatering Principles and
 Equipment  Design Studies

 Conceptual Design/Cost Study
 of  Alternative Methods for Lime/
 Limestone  Scrubbing Waste Disposal

 Evaluation of Alternative FGD
 Waste Disposal Sites
                  Contractor/Agency
             The Aerospace  Corporation
             U.S. Army  Corps  of  Engineers
             Waterways  Experiment  Station

             Tennessee  Valley Authority
             U.S. Army  Materiel  Command
             Dugway  Proving Ground
             Stearns, Conrad  and Schmidt
             Consulting Engineers,  Inc.
             (SCS Engineers)
             Tennessee  Valley Authority
             The Aerospace Corporation
             Louisville Gas & Electric
             Company (Subcontractor:
             Combustion Engineering)

             U.S. Army  Corps  of  Engineers
             Waterways  Experiment  Station
             Tennessee  Valley Authority

             Auburn  University
             Tennessee Valley  Authority
            Arthur  D.  Little,  Inc.

-------
                       FLUE GAS DESULFURIZATION                     279
     Results in this program area have been substantial.  FGD sludge
chemical characteristics, to a large degree, have been quantified.
Sludge liquors exceed drinking water standards for total dissolved
solids (IDS), with high concentrations of calcium, sulfate, and
chloride (and in some cases, magnesium and sodium).  In addition,
excessive concentrations of several trace metals have been noted.
The chemical composition of FGD sludge solids consists of calcium
sulfite hemihydrate, calcium sulfate dihydrate (gypsum) and/or
hemihydrate, and calcium carbonate, plus any fly ash collected in
the scrubber.  The percentage of each solid constituent is primarily
a function of the alkaline additive (e.g., lime, limestone), the
percent sulfur in the coal, and the manner in which the scrubber
system is operated (e.g., whether forced oxidation is applied,
whether fly ash is collected separately).  Although the fly ash has
been shown to be a major contributor of  trace elements to the sludge
solids and liquor, separate collection of fly ash does not necessarily
mean that concentrations of all these elements in the sludge liquor
will be insignificant.  In summary, chemical characterization of FGD
sludge has shown the need for protection of drinking water supplies
from intrusion by sludge leachates.

     The physical properties of FGD sludge vary considerably from
system to system; chemical composition is related to, but does not
adequately define, the sizes and types of the sludge solid crystals.
Many FGD sludges tend to liquefy easily, even after substantial
dewatering.  Several approaches to improving physical stability con-
tinue to be studied, including stabilization using underdrainage and
compaction, production of gypsum, and chemical treatment ("fixation")
for landfill.  Chemical treatment of FGD sludge has been shown to
result in significant structural improvement, a 50-75% reduction in
major solubles (e.g., chloride) in the leachate and an order of
magnitude (or more) reduction in permeability.  Further testing of
these disposal methods, including revegetation (reclamation) of
disposal sites, is planned.

     The costs of FGD sludge disposal vary considerably, depending
on the disposal system design, and site-specific factors such as
labor costs or the cost of a pond liner  (if one is installed).  Pre-
liminary cost estimates for a typical high-sulfur-coal-burning plant
are about $4-$9 per metric ton (dry basis, including ash) for
ponding, and about $8-$12.50 per metric  ton  (same basis) for chemical
treatment and landfill.  The ponding costs do not include reclamation
costs.  More detailed economics for these disposal methods have been
defined by TVA, and will be reported soon.  The next phase of the
EPA-sponsored study at TVA will include  gypsum disposal.

     Costs of FGD sludge disposal represent a major part (up to 20%)
of the capital and operating costs of an FGD system.  These costs
can be drastically reduced by improved absorbent  (e.g., limestone)
utilization, controlled solids quality,  and by improved sludge
dewatering equipment.  One approach to control solids quality which
is currently under study is an attempt to develop a procedure to
obtain consistent, easily dewatered sulfite solids.  An alternative

-------
280                    CLEM COMBUSTION OF COAL
approach would be to use forced oxidation to produce only gypsum
crystals, which are normally much larger than calcium sulfite crystals,
A complementary approach is to improve the performance of dewatering
equipment.  Separation of the clarification and thickening steps can
result in improved performance of gravity settlers, with a substantial
reduction in the equipment size.  With the exception of controlling
sulfite solids quality, all of these improvements have been shown to
be feasible and are currently making their way into the process
supply market.  However, further development/refinement of these
techniques is continuing; their full commercial use is expected in
the next 2-3 years.

     Coal-mine disposal of FGD sludge has greatly interested engineers
in the flue gas desulfurization industry for many years, because of
established means of transportation between the coal mine and the
power plant, and the need for material to fill the void left by
mining of the coal.  In addition, many plants may not have sufficient
land area for on-site disposal.  Recent technical/economic assessments
indicate that active Midwestern surface mines and Eastern/Midwestern
room-and-pillar underground mines are the most promising candidates
for this disposal approach.  One utility plans to begin disposal of
FGD sludge and ash in a surface mine this summer.  Plans under the
EPA program are to conduct a 2-year monitoring/assessment effort at
this utility site.  Successful demonstration of this disposal approach
could make conversion to coal quite feasible even in areas where
land for disposal is limited.

     Ocean disposal of FGD sludge is also being assessed because
many plants in the Northeast may have difficulty switching to coal
for lack of disposal sites; however, many of these plants do have
access to the ocean.  It was also recognized that the major soluble
chemical constituents in FGD sludge are found in relatively high
concentrations in seawater.  Studies of ocean disposal of FGD sludge
by Arthur D. Little for EPA have identified several potential environ-
mental problems.  It appears that these problems could be alleviated
by either chemical treatment to a "brick-like" form (possibly creating
an artificial reef) or oxidation to gypsum (followed by a widely
dispersed disposal operation).   The costs of these approaches are
being defined and are expected to be somewhat higher than for chemical
treatment/landfill near the plant.  Pilot disposal simulation studies
are underway to define the environmental effects of both untreated
and treated FGD sludge disposal in the ocean.

     Currently there are no Federal regulations which specifically
address the disposal of FGD. sludge.  However, the Resource Conserva-
tion and Recovery Act (RCRA) of 1976 calls for the eventual Federal
regulation of disposal of hazardous solid wastes and the issuance of
guidelines (to be used by the states) for disposal of non-hazardous
solid wastes.  The RCRA specifically identifies solid wastes and
sludges, including those generated by air pollution control devices,
as being covered by the Act.  Although no official designation'
(hazardous or non-hazardous) has been placed on FGD sludges, it is
currently assumed that they will be considered non-hazardous, with

-------
                       FLUE GAS DESULFURIZATTON
                            281
disposal guidelines to be issued in the next 2-3 years.  An effort
has been underway since mid-1975 to prepare a preliminary technical
support document which could be potentially useful in setting FGD
waste disposal guidelines.  A draft of the document is currently
under review.

     Studies of the characteristics of coal ash and the effects of
coal ash disposal have been underway and are continuing.  These
efforts are not as extensive as those for FGD sludge.  However, they
are no less significant because of the increasing generation of coal
ash, and because many FGD sludges contain significant quantities of
fly ash, either collected in the scrubber or added to the FGD sludge
prior to disposal (e.g., in a chemical treatment process).  A report
has been issued which summarizes and evaluates existing data on the
characteristics of coal ash from studies made by TVA and others.

     FGD Sludge Utilization.  This part of the FGD program has only
become active in the past 2 years; it currently consists of five
projects, listed in Table 2.

            Table 2:  EPA Projects in FGD Waste Utilization

     Project Title                        Contractor/Agency
Gypsum By-product Marketing  Studies

Lime/Limestone  Scrubbing Waste
Conversion Pilot Studies

Fertilizer Production Using
Lime/Limestone  Scrubbing Wastes

Use  of FGD Gypsum in Portland
Cement Manufacture
 FGD Waste/Fly Ash  Beneficiation
 Studies
Tennesssee Valley Authority

Pullman-Kellogg


Tennessee Valley Authority
Babcock & Wilcox
Portland Cement Association
C.E. Lovewell

TRW, Inc.
      Since FGD  sludge  is  a  relatively  new by-product, utilization in
 the United States  is not  yet  a  commercial reality.  However, conversion
 of FGD  sludge to gypsum (or direct  production of  gypsum) for use in
 wallboard and portland cement manufacture is practiced  extensively
 in Japan.  Although the Japanese experience has primarily been with
 oil,  gypsum-producing  FGD experience with coal is increasing in
 Japan and the United States.  However,  thus far there have been no
 full-scale commitments by American  utilities to produce FGD gypsum
 for sale.  This situation may change as the utilization of coal for
 electric power  expands and  an energy and resource conservation ethic
 begins  to take  shape.

      Tools for  the development  of FGD  gypsum market strategies have
 been  developed.  Studies  currently  underway at TVA include a thorough
 economic evaluation of several  gypsum-producing FGD processes—e.g.,

-------
282                    CLEM COMBUSTION OF COAL


limestone/gypsum, Chiyoda (H-SO./gypsum), and Dowa (aluminum-based
double alkali/gypsum)—and a detailed U.S. marketing study of FGD
gypsum for wallboard.  A report on this study is expected later in
1977; indications are that processes less complex than those used in
Japan will be necessary for a profitable situation to occur.  Wallboard
production using FGD gypsum from a Southeastern utility has been
successfully demonstrated.  Feasibility demonstration of FGD gypsum
used in portland cement in cooperation with trade associations is
planned.

     Development of FGD sludge utilization in fertilizer is continuing
at the pilot level at TVA.  Spreading the material over a relatively
large land area in this manner would not only alleviate the disposal
problem, but would also minimize the potential localized environmental
impact of a highly concentrated waste; i.e., the leachate's chemical
constituents would be highly diluted by rainfall and interaction
with the soil.  Further development of the fertilizer production
process is needed to establish its viability.

     Conversion of FGD sludge to elemental sulfur with recovery of
calcium carbonate (using coal as the reductant) for recycle to the
scrubber has been studied on a pilot level by Pullman-Kellogg and
Ontario Hydro.  Further studies are planned under a contract currently
being negotiated with Pullman-Kellogg.  This effort should be under-
way later in 1977.

REGENERABLE SYSTEMS

     Since its inception in 1970 EPA has assisted in the development
of several recovery processes capable of producing sulfuric acid,
elemental sulfur, or liquefied SO-.  These processes have been
pursued in hopes of conserving a valuable natural resource and
reducing overall S02 control costs.  Most of the EPA efforts have
been directed towara full-scale demonstrations of a number of leading
processes; however, support has also been given to bench-scale and
pilot plant efforts.

     Magnesium Oxide Scrubbing Program.   The Mag-Ox scrubbing
process—developed by Chemical Construction Company (Chemico) and
Basic Chemicals, and currently offered commercially by Chemico—is
one of the more promising regenerable FGD approaches.  The process,
which produces sulfuric acid, is widely applicable to both existing
and new power plants.  It is also amenable to the centralized pro-
cessing concept; i.e., spent sorbent can be regenerated at a central
plant capable of servicing a number of power or industrial plants.

     In 1974, EPA and Boston Edison completed a co-funded demonstration
program of a 155 MW capacity scrubbing/regeneration system.  Results
obtained during 2 years of operation indicated:  (1) SO- removal
efficiencies in excess of 90% were obtained consistently, and (2) more
than 5,000 tons of saleable sulfuric acid of high quality was recovered
from the stack gas and sold commercially.  A number of problems were
encountered that were primarily equipment, rather than process,
related; however, continuous, long-term, reliable operation was not
achieved.

-------
                       FLUE GAS DESULFUBIZATION                     283
     In 1973, Potomac Electric Power Company installed a 100 MW Mag-
Ox scrubbing system at its coal-fired Dickerson Station.  EPA pro-
vided the Mag-Ox regeneration system for Potomac Electric's use in
processing spent scrubber sorbent.  Results indicate S0~ removal
efficiencies greater than 90% are possible and particulate removal
of 99.6% was attained.  Over 2000 tons of sulfuric acid was produced
and marketed.  Unfortunately due to a shortage of funds this demon-
stration did not run long enough to completely answer all questions
regarding absorbent recycle, absorbent losses, and process reliability.

     Philadelphia Electric will soon begin operating a 120 MW MgO
scrubbing facility at the Eddystone Unit 1.  After initial startup
in September 1975, this unit was shut down when the regeneration-
sulfuric acid system at Olin Chemicals, Paulsboro, New Jersey, plant
was permanently closed.  Regeneration will now take place at Essex
Chemical's acid plant in Newark, New Jersey.  EPA plans to supply
consulting for startup, operation, and test program formulation.

     Wellman-Lord/Allied Chemical Demonstration Program.  EPA and
Northern Indiana Public Service Company (NIPSCO) have jointly funded
the design and construction of a flue gas cleaning demonstration
plant utilizing the Wellman-Lord S0~ recovery process and the Allied
Chemical SO  reduction process to convert recovered S0_ to elemental
sulfur.  The operational costs for the system will be paid by NIPSCO,
and a comprehensive test and evaluation program will be funded by
EPA.  The demonstration system has been retrofitted to the 115 MW,
coal-fired Unit 11 at the D.H. Mitchell Station in Gary, Indiana.
Construction of the facility was completed in August 1976.

     Startup activities and acceptance testing were delayed by
boiler problems which developed when Unit 11 was shut down for
annual maintenance.  The boiler and the FGD plant were restarted
in June, and integrated operation and acceptance testing will follow
shortly thereafter.  Long-term duration testing will begin after
acceptance.

     Bureau of Mines (BOM) Citrate Demonstration Program.  EPA and
BOM have entered into a cooperative agreement to pool funds and
technical talents to demonstrate the citrate process developed by
BOM.  A concurrent development program, carried out by an industrial
consortium headed by Pfizer Chemical Company, also led to a pilot
operation of the process.  Based on the results of these two pilot
programs, EPA and BOM have.initiated the demonstration of this tech-
nology on a 53 MW coal-fired boiler at St. Joe Minerals Corporation
in Monaca, Pennsylvania.  Construction of the facility is expected
to be completed by mid-1978.  After acceptance, a 1-year test and
evaluation program will be initiated.

     Aqueous Carbonate Demonstration Program.  EPA and Empire State
Electric Energy Research Corporation (ESEERCO), a research organiza-
tion sponsored by New York's eight major power suppliers, have
recently contracted to fund jointly the design and construction of a
demonstration of Atomics International's sulfur-producing aqueous
carbonate process.  The demonstration system is being retrofitted to

-------
284                    CLEM COMBUSTION OF COAL


Niagara Mohawk Power Company's 100 MW coal-fired Huntley Station  in
Tonawanda, New York.  Construction of the facility is expected  to be
completed by mid-1979.  After acceptance testing, a 1-year test and
e*\ra 1 imlH cm m-ncri-am is nlarmpd.
evaluation program is planned.
     Catalytic Oxidation  (Cat-Ox) Demonstration Program.  The Cat-Ox
process is Monsanto Enviro-Chem Systems' adaptation of the contact
sulfuric acid process.  EPA and Illinois Power Company attempted
to demonstrate the process on a 103 MW coal-fired boiler at Illinois
Power's Wood River Station between 1970 and 1976.

     Detailed design, construction, shakedown testing, and performance
guarantee testing were completed in July 1973.  The unit met all
guarantees and was subsequently accepted.  Because of the shortage
of natural gas, the burners were modified to allow either oil- or
gas-firing.  However, design and startup problems precluded suc-
cessful operation and initiation of the comprehensive 1-year test
program.

     In view of the problems and long delays encountered, a thorough
technical and economic study was made of the costs and benefits of
continuing the Cat-Ox demonstration at the Wood River Station.
Results of this study led to the decision to end the project in
December 1976.

     Ammonia Scrubbing with Bisulfate Regeneration Pilot-Plant Program.
In 1970, EPA and TVA jointly undertook the development of a completely
cyclic ammonia scrubbing/ammonium bisulfate regeneration process
which has as its major product a concentrated stream of SO- which
can then be used to produce sulfuric acid or elemental sulfur.  This
process was evaluated at a 3,000 ft-Vmin (5,000 m^/hr) pilot unit
located at Colbert Steam Plant.  While initial developmental efforts
at the pilot unit were concentrated on the absorber, later work
included investigation of all subunits of the system except the
electrical decomposer.  It became apparent that the process had two
major problems:  (1) the formation of a persistent fume which could
not be consistently controlled or eliminated by reasonable control
efforts, and (2) unfavorable economic projections due primarily to
energy consumption by the decomposer.  As a result of these problems,
the development project was terminated during the summer of 1976.
FGD SUPPORT STUDIES

     Key supporting studies in several problem areas of FGD technology
have been sponsored by EPA to further the advancement and application
of commercial systems.  In many cases, the studies undertaken are
broad general assessments which are directed toward a wide variety
of potential users.  Examples of results from these studies are
described below.

Comparative Economics of SO,, Control Processes

     The purpose of this continuing EPA/TVA project is to study the
most promising S02 removal processes advancing toward commercialization.

-------
                       FLUE GAS DESULFURIZATION                     285


It includes selection of those processes which have the greatest
degree of development and which are potentially attractive both
technically and economically.  These evaluations include preparation
of flowsheets, material balances, and layouts; definition of process
equipment; preparation of capital investments and operating costs;
and analysis of design and economic variables for cost sensitivity
analysis.  Currently, this is being done for the citrate and double-
alkali processes using the limestone system for comparison; results
should be available by the fall of 1977.

By-product Marketing Studies

     TVA is conducting a program for EPA (1) to determine the quantities
of FGD by-product acid sulfur and gypsum that would be produced at
power plant sources, and (2) to analyze the markets for these by-
products.  The computer model which has been developed considers
compliance of actual utility plants with SO- emission standards and
potential by-product market demand for each plant (based on current
markets and freight rates).  Reports on the marketing of sulfur,
sulfuric acid, and gypsum (for wallboard) are expected in the fall
of 1977.

Reductant Gases

     Conversion of SO  to elemental sulfur requires the use of a
reductant; in the pas? major emphasis was placed on the use of
natural gas.  Since this is now an impractical approach, other
sources of reductant gas must be sought, such as coal gasification.
A study conducted for EPA by Battelle-Columbus concludes that a
gasifier-based reduction system would significantly increase the
complexity of the overall FGD system.  However, the gasifier could
be used to supply part of the energy for the FGD process, including
stack gas reheating.

Magnesium Oxide Scrubbing Support Studies

     Two studies in support of magnesium oxide scrubbing have been
conducted for EPA by Radian Corporation.  In the first study, Radian
evaluated the feasibility of producing elemental sulfur directly
from magnesium sulfite.  This would expand the applicability of
current magnesium oxide processes which only produce sulfuric acid.
The second study is concerned with the mechanism of formation of
trihydrate and hexahydrate forms of magnesium sulfite (MgSO^-SH-O,
MgSO -6H-0).  The hexahydrate crystals separate and handle easily;
the trihydrate crystals require less drying energy but are more
difficult to separate and handle.  This study has attempted to
generate information on formation mechanisms and operating conditions
that can be used to control the type of crystal formed.

Comparison of Availability and Reliability of Equipment Utilized in
the Electric Utility Industry

     For the past several years one of the major objections of  the
utility industry to installing FGD systems has been that reliabili-

-------
286                    CLEAN COMBUSTION OF COAL
ty/availability of FGD systems is much lower  than for other major
utility equipment items, such as boilers, turbines, generators,
electrostatic precipitators, and gas turbines.   It is desirable  that
good  information and data on this problem be  gathered so  that  a
valid comparison of performance can be made.

      To meet this need and to provide information as input for a
current National Academy of Sciences study of SO^ control technology,
a  study by EPA was initiated (with Radian Corporation) as a jointly
sponsored project of EPA and the Council on Environmental Quality.
The study concluded that a statistically meaningful comparison of
reliability/availability of these components  cannot now be made,
primarily because of the small number and short  service time of  FGD
system data  (a meaningful comparison can probably be made in 1979) .
However, the study also concluded that the mechanical reliability  of
some  types of conventional equipment now being used by the electric
utility industry is not much different from that of similar items
used  in FGD systems.
 TECHNOLOGY  TRANSFER

      For  several years, EPA has disseminated FGD technology data and
 information through  the traditional outlets of FGD symposia, industry
 briefings,  capsule reports, and project summary reports.  Six  symposia
 have  now  been held,  the last one in March 1976.  The next symposium
 is  scheduled for November 8-11, 1977, in Hollywood, Florida.   Progress
 on  lime/limestone technology has been reported through industry
 briefings,  the  last  one in October 1976.  Thus far, three capsule
 reports have been issued on the EPA/TVA/Bechtel Shawnee program.
 Numerous  project reports have also been issued.

      In order to improve the effectiveness of the FGD technology
 transfer  effort, the survey of utility FGD installations and,  more
 recently, the Engineering Applications/Information Transfer (EA/IT)
 program were initiated.  These are described below.

 Survey of Utility FGD Installations

      To meet the continuing demand for technical and economic  data
 on  operational, under construction, and planned future utility FGD
 units, EPA  has  employed PEDCo-Environmental to monitor this field of
 technology  and  prepare periodic reports for use by utilities,  system
 vendors and designers, and regulatory authorities.  In addition to
 detailed  technical reportst PEDCo is providing bimonthly status
 reports indicating the number of each type of S02 control system in
 operation,  under construction, or planned in the United States, and
 the megawatt capacity controlled or to be controlled.  The bimonthly
 status report gives  technical and economic information on all  known
 U.S.  utility FGD systems categorized in 15 tables and 4 appendices
 to  promote  ease of use.  Some of the information from the latest
 PEDCo report April-May 1977) is summarized in the final section of
 this  paper.

-------
                       FLUE GAS DESULFURIZATION                     287
Engineering Applications/Information Transfer  (EA/IT)

     The EA/IT program is a comprehensive effort designed to augment
traditional EPA technology transfer efforts  so that dissemination of
information will be more efficient and  effective.  The program
includes several new and innovative reporting activities, including:

   - A series of quarterly reports on FGD research, development and
     demonstration efforts sponsored by EPA.

   - A series of Cost/Reliability Handbooks  to assist potential
     users in choosing the specific SOX control strategy, FGD system,
     and FGD system design which best fit their needs.

   - Lime and Limestone Scrubbing Data  Books (cooperative EPA/Electric
     Power Research Institute  efforts).

   - Non-utility  (industrial)  combustion source survey report assess-
     ing applicability of various SO  control  strategies.
                                    X


FULL-SCALE UTILITY FGD APPLICATIONS

Overview

     According  to the latest PEDCo survey, 119 utility boilers
representing over 50,000 MW of electric generating capacity will be
controlled by FGD by  the mid-1980's.  The current status of these
systems is given  in Table  3.

                  Table 3:  Status of Utility FGD Systems

                                           No. of
          Status                             Units       MW

     Operational                              27        7319
     Under Construction                      .29       12648
                               Subtotal        56       19,967

     Planning
          Contract Awarded                    20        9797
          Letter  of Intent                     5        1892
          Requesting/Evaluating Bids           5        3565
          Considering only FGD Systems       ^3_       14856

                               Total          119       50,077

     Of the units which are operational or under construction,  91%
of the generating capacity or  roughly 18,000 MW will  be  controlled
by lime/limestone scrubbing systems.  Of the units for which  con-
tracts have been  awarded,  88%  of the generating capacity or roughly
8000 MW will also be  controlled by lime/limestone systems.  However,
of those units  in the remaining stages  of planning shown above
(e.g., letter of  intent) for which a process type has been  selected,
the trend appears to  be away from lime/limestone, with only 55% of  the

-------
288                    CLEAN COMBUSTION OF COAL
 generating  capacity  being planned  for  control  by  these  systems.
 However,  of the  20,467 MW in  these remaining stages,  over  13,500 MW,
 or  66%, represent  units  for which  the  specific FGD  system  has  not
 yet been  selected.   Therefore,  it  is perhaps too  soon to tell  if
 there will  be  a  substantial move toward  the regenerable FGD processes.
 Nevertheless,  as operating experience  with regenerable  systems
 increases,  particularly  with  the Wellman-Lord  and magnesium oxide
 systems,  they  will no doubt occupy a significant  portion of the  FGD
 market in the  1980's.

      Table  4 summarizes  the FGD systems  which  were  operational as of
 May 1977.
 CONCLUSIONS

      The experience  level  of  FGD  technology  is rapidly developing.
 The development  of lime/limestone scrubbing  is far ahead  of  other
 FGD systems  primarily  because of  information gathered from operational
 systems  and  through  development sponsored by EPA.  If future power
 plant scrubber systems are designed  to  take  advantage of  data  already
 available, improvement in  overall reliability, cost, and  effective-
 ness should  transpire.  Additional work is needed in this area of
 technology,  however, to improve performance, reduce costs, and
 increase reliability.

      Double-alkali technology is  rapidly developing as a  viable
 means of SO™ control.  The potential advantages of such systems  to
 improve  SO-  removal, sludge characteristics, and operating reliability
 first need to be confirmed through the  full-scale demonstrations,
 then applied as  an alternate  to lime/limestone technology.

      The area in which the most work remains to be done is in  regen-
 erable process technology.  The presently planned demonstrations
 need to  be completed as soon  as possible.  In addition, further  work
 is  needed to confirm the long-term effectiveness and economics of
 magnesium oxide  scrubbing.

-------
                    Table 4:    Summary of  Operational  Utility  FGD  Systems*


1.
2.

3.

4.

5.

6.

7.
8.
9.
10.
11.
12.
13.
14.

15.

16.

17.

18.

19.
20.

21.

22.

23.

24.

25.
26.
27.

Company
Arizona Public Service
Columbus and Southern
Ohio Electric
Commonwealth Edison

Duquesne Light

Duquesne Light

Gulf Power Co.

Kansas City Power & Light
Kansas City Power 4 Light
Kansas City Power & Light
Kansas Power & Light
Kentucky Utilities
Louisville Gas & Electric
Louisville Gas & Electric
Montana Power Co.
•
Montana Power Co.

Nevada Power

Nevada Power

Nevada Power

Northern States Power Co.
Pennsylvania Power Co.

Philadelphia Electric Co.

Tennessee Valley Authority

Tennessee Valley Authority

Northern Indiana Public
Service
Northern States Power
Springfield City Utilities
Tennessee Valley Authority

Unit
Coal Percent
Unit Name Capacity, Hw FGD Process Type
Cholla No. 1

Conesville No. 5
Will County No. 1

Elrama Power Station

Phillips Power Station

Scholz Nos. IB & 2B

Hawthorn No. 3
Hawthorn No. 4
LaCygne No. 1
Lawrence No . 4
Green River Nos. 1 & 2
Cane Run No. 4
Paddy's Run No. 6
Colstrip No. 1

Colstrip No. 2

Reid Gardner No. 1

Reid Gardner No. 2

Reid Gardner No. 3

Sherburne County No. 1
Bruce Mansfield No. 1

Eddys tone No. 1A

Shaunee No. 10A

Shawnee No. 10B

D. H. Mitchell No. 11

Sherburne County No. 2
Southwest No. 1
Widows Creek No. 8
115

400
167

510

410

23

140
100
820
125
64
178
65
360

360

125

125

125

710
835

120

10

10

115

680
200
550
Limestone Scrubbing

Lime Scrubbing
Limestone Scrubbing

Lime Scrubbing

Lime Scrubbing

Chiyoda 101

Lime Scrubbing
Lime Scrubbing
Limestone Scrubbing
Limestone Scrubbing
Lime Scrubbing
Lime Scrubbing**
Lime Scrubbing**
Lime/Alkaline Ash
Scrubbing
Lime/ Alkaline Ash
Scrubbing
Sodium Carbonate
Scrubbing
Sodium Carbonate
Scrubbing
Sodium Carbonate
Scrubbing
Limestone Scrubbing
Lime Scrubbing

Magnesium Oxide
Scrubbing
Lime/Limestone
Scrubbing
Lime/ Limes tone
Scrubbing
Wellraan-Lord

Limestone
Limestone
Limestone
Sulfur
0.

4.
4.

1.

1.

5.

0.
0.
5.
0.
3.
3.
3.
0.

0.

0.

0.

0.

0.
4.

2.

2.

2.

3.

0.
3.
3.
44-1.0

5-4.9
0

0 -2.8

0 -2.8

0 (max)

5 -3.5
5 -3.5
0
5
8
5 -4.0
5 -4.0
8

8

5-1.0

5 -1.0

5 -1.0

8
7

5

9

9

5

8
5
7
Waste Disposal

Start-Up Date Method
10/73

2/77
2/72

10/75

7/73

3/75

11/72
8/72
2/73
12/68
9/75
8/76
4/73
10/75

7/76

4/74

4/74

7/76

3/76
4/76

9/75

4/72

4/72

6/77

4/77
4/77
4/77
(Evaporation) Ponding
Chemical Treatment/
Landfill
Chemical Treatment/
Landfill
Chemical Treatment/
Landfill
Chemical Treatment/
Landfill
(Lined) Ponding
(Gypsum)
Ponding
Ponding
Ponding
Ponding
Stabilization/Ponding
Ponding
Stabilization/Landfill
Ponding

Ponding

(Evaporation) Ponding

(Evaporation) Ponding

(Evaporation) Ponding

(Lined) Ponding
Chemical Treatment/
Ponding
Not Applicable —
By-product Acid Produced
Ponding

Ponding

Hot Applicable -
By-product sulfur produced
(Lined) Ponding
Stabilization/Ponding
Ponding









^
f
s

p
>
rr\
\JJ
a
M
Ul
t-i
^
S
H
G
y>
f_g
H
0
&














 *Source:  PEDCo Summary Report, April-May 1977, Contract 68-02-1321, Task No. 28
**Carbide Lime
NJ
00
VO

-------
290                    CLEAR COMBUSTION OF  COAL
BIBLIOGRAPHY

1.   Ando, J. "Status of Flue Gas Besulfurization and Simultaneous
     Removal of SO,, and NO  in Japan."  In Proceedings;   Symposium
     on Flue Gas, Desulfurization, New Orleans, March 1976, Vol.  I,
     EPA-600/2-76-136a (NTIS PB 255 317), May 1976, pp. 53-78.

2.   Borgwardt, R. H.  "EPA/RTP Pilot Studies Related to  Unsaturated
     Operation of Lime and Limestone Scrubbers."  In Proceedings;
     Symposium on Flue Gas Desulfurization, Atlanta, November  1974,
     Vol. I.  EPA-650/2-74-126a (NTIS PB 242-572), December 1974.

3.   Borgwardt, R. H.  "IERL-RTP Scrubber Studies Related to Forced
     Oxidation."  In Proceedings: Symposium on Flue Gas Desulfurization,
     New Orleans. March 1976, Vol. I.  EPA-600/2~76-136a  (NTIS PB
     255-317), May 1976, pp. 117-143.

4.   Borgwardt, R. H.  "Improving Limestone Utilization in FGD
     Scrubbers."  AIChE Symposium Series, "Air-1976," in  press.

5.   Borgwardt, R. H.  Sludge Oxidation in Limestone FGD  Scrubbers.
     EPA-600/7-77-061 (NTIS PB 268-525), June 1977.

6.   Bucy, J. I., J. L. Nevins, P. A. Corrigan, and A. G. Melicks.
     "Potential Utilization of Controlled SO  Emissions from Power
     Plants in Eastern United States."  In Proceedings: Symposium
     on Flue Gas Desulfurization, New Orleans, March 1976, Vol. II.
     EPA-600/2-76-136b (NTIS PB 262-722), May 1976, pp. 647-700.

7.   Crowe, J. L., and H. W. Elder.  "Status and Plans for Waste
     Disposal from Utility Applications of Flue Gas Desulfurization
     Systems."  In Proceedings;  Symposium on Flue Gas Desulfurization,
     New Orleans, March 1976, Vol. II.  EPA-600/2-76-136b (NTIS PB
     262-722), May 1976", pp. 565-577.

8.   Devitt, T. W., G. A. Isaacs, and B. A. Laseke.  "Status of Flue
     Gas Desulfurization Systems in £he United States."   In Proceedings;
     Symposium on Flue Gas Desulfurization, New Orleans,  March 1976,
     Vol. I.  EPA-600/2-76-136a (NTIS P-B 255-317), May 1976, pp.  13-
     51.

9.   Epstein, M.  EPA Alkali Scrubbing Test Facility:  Advanced
     Program - First Progress Report.  EPA-600/2-75-050 (NTIS  PB
     245-279), September 1975.

10.  Epstein, M.  EPA Alkali Scrubbing Test Facility:  Summary of
     Testing Through October 1974.  EPA-650/2-75-047 (NTIS PB
     244-901), June 1975.

11.  Epstein, M., H. N. Head, S. C. Wang, and D. A. Burbank.
     "Results of Mist Elimination and Alkali Utilization  Testing  at
     the EPA Alkali Scrubbing Test Facility."  In Proceedings:
     Symposium on Flue Gas Desulfurization, New Orleans,  March 1976,
     Vol. I.  EPA-600/2-76-136a (NTIS PB 255-317), May 1976  pp
     145-204.

-------
                       FLUE GAS DESULFURIZATION                     291


12.  Lowell, P. S., W. E.  Corbett, G. D. Brown, and K. A. Wilde.
     Feasibility of Producing  Elemental Sulfur from Magnesium
     Sulfite.  EPA-600/7-76-030  (NTIS PB 262-857), October 1976.

13.  Head, H. N.  EPA  Alkali Scrubbing Test Facility:  Advanced
     Program - Second  Progress Report.  EPA-600/7-76-008  (NTIS PB
     258-783), September 1976.

14.  Hissong, D. W., K. S. Murthy, and A. W. Lemmon, Jr.  Reductant
     Gases for Flue Gas Desulfurization Systems.  EPA-600/2-76-130
     (PB 254-168), May 1976.

15.  Hollinden, G. A., R.  F. Robards, N. D. Moore, T. M.  Kelso, and
     R. M. Cole.  TVA's 1-MW Pilot Plant:  Final Report on High
     Velocity Scrubbing and Vertical Duct Mist Elimination.  EPA-
     600/7-77-019  (TVA PRS-19), March 1977.

16.  Interess, E.  Evaluation  of  the General Motors' Double Alkali
     S0? Control System.   EPA-600/7-77-005  (NTIS PB 263-469), January
     1977.

17.  Jones, B. F., P.  S. Lowell,  and F. B. Messerole.  Experimental
     and Theoretical Studies of  Solid Solution Formation  in Lime
     and Limestone  SO.-, Scrubbers, Vol. I—Final report, and Vol. II—
     Appendices.  EPA-600/2-76-273a and -273b  (NTIS PB 264-953 and
     264-954), October 1976.

18.  Jones, J. W.,  "Disposal of  Flue Gas Cleaning Wastes," CHEMICAL
     ENGINEERING, Vol. 84, No. 4, pp. 79-85, February 14, 1977.

19.  Jones, J. W.,  Brna, T. G.,  Crowe, J. L., Flora, H. B. and Ray,  S.  S.
     "Environmental Management of Effluents and Solid Wastes from Steam
     Electric Generating Plants." In Proceedings;  Second National
     Conference on  the Interagency Energy/Environment R&D Program,
     June  1977, in  press.

20.  Kaplan, N. "Introduction  to  Double Alkali Flue Gas Desulfuriza-
     tion  Technology."  In Proceedings:  Symposium on Flue Gas
     Desulfurization,  New  Orleans. March 1976, Vol. I.  EPA-600/2-
     76-136a  (NTIS  PB  255-317), May 1976, pp.  387-422.

21.  Koehler, G., and  J. A. Burns.  Magnesia Scrubbing Process as
     Applied to an  Oil-Fired Power Plant.   EPA-600/2-75-057  (NTIS
     PB 247-201), October  1975.

22.  Koehler, G.  Magnesia Scrubbing Applied to a Coal-Fired Power Plant.
     EPA-600/7-77-018  (NTIS PB 266-228), March 1977.

23.  LaMantia, C. R.,  R. R. Lunt, R. E. Rush,  T. M. Frank, and N.
     Kaplan.  "Operating Experience—CEA/ADL Dual Alkali  Prototype
     System at Gulf Power/Southern Services, Inc.  In Proceedings:
     Symposium on Flue Gas Desulfurization, New Orleans,  March 1976,
     Vol.  I.  EPA-600/2-76-136a (NTIS PB 255-317), May 1976, pp.  423-
     471.

-------
292                    CLEAN COMBUSTION OF COAL
 24.   Leo,  P.  P.  and  J.  Rossoff,  Control of Waste and Water Pollution
      from  Power  Plant Flue  Gas  Cleaning Systems:  Second Annual R
      and D Report.   (To be  published for EPA).

 25.   Lowell,  P.  S.,  F.  B. Messerole,  T.  B. Parsons.   Precipitation
      Chemistry of Magnesium Sulfite Hydrate in  Magnesium Oxide
      Scrubbing.  EPA report in  press.

 26.   Lunt, R. R., C. B.  Cooper,  S.  L.  Johnson,  J.  E.  Oberholtzer,
      G. R. Schimke,  and W.  I. Watson.   An Evaluation of the Disposal
      of Flue  Gas Desulfurization Wastes in Mines and the Ocean:
      Initial  Assessment,  EPA-600/7-77-051, May  1977.

 27.   McGlamery,  G. G.,  H. L.  Faucett,  R.  L.  Torstrick,  and L.  J.
      Henson,  "Flue Gas  Desulfurization Economics."  In  Proceedings:
      Symposium on Flue  Gas  Desulfurization,  New Orleans, March 1976,
      Vol.  I.  EPA-600/2-76-136a (NTIS  PB 255-317), May  1976,  pp.  79-99.
                                          *
 28.   McGlamery,  G. G.,  Stern, R.  D.  and Maxwell, M.  A.   "The  Federal
      Interagency Flue Gas Desulfurization Program."   In Proceedings;
      Second National Conference on  the Interagency Energy/Environment
      R&D Program, June  1977,  in press.

 29.   PEDCo-Environmental, Inc.   "Summary Report -  Flue  Gas Desulfuriza-
      tion  Systems."  April-May  1977.   EPA Contract 68-02-1231,  Task
      No. 28.

 30.   Rossoff, J., R. C.  Rossi,  R. B.  Fling,  W.  M.  Graven,  and  P.  P.  Leo,
      Disposal of By-Products  from Nonregenerable Flue Gas Desulfuriza-
      tion  Systems:   Second  Progress Report,  EPA-600/7-77-052,  May
      1977. (In Print).

 31.   Tennessee Valley Authority.  Pilot-Plant Study  of  an Ammonia
      Absorption  - Ammonium  Bisulfate Regeneration  Process, Topical
      Report Phases I and II.  EPA-650/2-74-049a (NTIS PB 237-171),
      June  1974.

-------
                                                                    293
      STATUS OF FLUE GAS TREATMENT TECHNOLOGY FOR CONTROL OF NO
                AND SIMULTANEOUS CONTROL OF SOY AND NO         X
                                              A       X
                 J. David Mobley and Richard D. Stern
             Industrial Environmental Research Laboratory
                 U.S. Environmental Protection Agency
             Research Triangle Park, North Carolina  27711
                               ABSTRACT
     The status of flue gas treatment technology for control of
NOX and simultaneous control of NOX and SOX applicable to stationary
combustion sources is presented.  Dry processes and wet processes
are described and applications discussed with respect to performance,
operating experience, and economics.

     As a result of a very stringent NOX ambient standard in
Japan, the Japanese NOX flue gas treatment technology appears to
be the most advanced in the world.  For this reason, Japanese
technology is emphasized in the paper.  EPA's past, current, and
planned flue gas treatment program is also discussed.
                             INTRODUCTION
     Nitrogen oxides (NOX) in the atmosphere have been determined
to have adverse effects on human health and welfare.  To aid in
preventing these adverse effects, the Industrial Environmental
Research Laboratory at Research Triangle Park, N. C. (IERL-RTP) is
leading the U.S. Environmental Protection Agency's  (EPA) efforts
to develop and demonstrate NOX control technologies for stationary
combustion sources.  There are two main technologies being developed:
combustion modification and flue gas treatment.

          Combustion modification (CM) technology attempts
     to minimize the formation of NOX during the combustion
     process.  CM techniques include staged combustion,
     low excess air operation, flue gas recirculation,
     water injection, and burner redesign.  CM technology
     should be able to reduce NOX emissions from stationary
     combustion sources by 50% or more in a relatively
     cost effective manner.  CM technology will not be
     discussed in this paper; however, additional information
     is available from other sources.1

-------
294                   CLEM .COMBUSTION  OF  COAL
           Flue gas  treatment  (FGT)  technology attempts to
      remove NOX from the gaseous  products of combustion.
      FGT techniques include dry selective catalytic reduction
      processes and  wet scrubbing  processes.   FGT technology
      should be able to reduce NOX emissions  by 90% and has
      the potential  for 90% control of both NOX and SOX
      emissions.

      NO  FGT research and development programs have received a
 relatively low level of funding by EPA since it has not been determined
 conclusively that high NOX removal efficiencies will be required to
 achieve and maintain the current  National Ambient Air Quality
 Standards (NAAQS).   However,  there are significant uncertainties
 which may affect the required level of NOX control.  Due  to these
 uncertainties, EPA  is proceeding  with small  scale NOX FGT experimental
 projects in parallel with control strategy and technology assessment
 studies.  One phase of the assessment is  an  evaluation of Japanese
 FGT technology which has progressed to the point of being commer-
 cially applied to gas- and oil-fired sources.   In addition, the
 Japanese are developing processes for application to flue gas from
 coal-fired sources.  EPA is investigating the Japanese and other
 worldwide technologies for potential application to the U.S. coal-
 fired situation to  save both  development  time and money.   Through
 these actions, the  basic foundation will  be  established if the
 technology is required in the United States  and acceleration of
 the development program becomes necessary.
                      FLUE GAS TREATMENT PROCESSES
      There are two main categories  of flue  gas  treatment (FGT)
 processes for the control of NOX and the simultaneous  control of
 NOX and SOX emissions from stationary combustion sources:   dry
 processes and wet processes.   A description of  the most  promising
 processes in each category and their developmental status  in Japan
 is discussed below.

 DRY PROCESSES

      The following dry process types are being  developed:

                    Selective catalytic reduction
                    Selective noncatalytic reduction
                    Adsorption
                    Nonselective catalytic reduction
                    Catalytic decomposition
                    Electron beam radiation

      Of these, only selective catalytic reduction (SCR)  has achieved
 notable success in treating combustion flue gas for removal of NO
 and has progressed to the point of  being commercially  applied.  The
 other process types are much less attractive at this time.  Selective
 noncatalytic reduction processes do not achieve high NO   removal
                                                       X

-------
                      NO /SO  FLUE GAS TREATMENT                    295
efficiences and adsorption processes are not applicable to combustion
sources.  Nonselective catalytic reduction, catalytic decomposition,
and electron beam radiation processes are at a very low level of
development.  These process types will not be discussed in this
paper, but additional information is available from other sources.2»3»4,5

Dry NOx Processes

     Selective catalytic reduction processes are based on the
preference of ammonia (N^) for NOX over other flue gas constituents.
Since the oxygen enhances the reduction, the reactions can best be
expressed as:

                                   catalyst
               4NH3 + 4NO + 02 	*-  4N2 + 6H20         (1)
                                    catalyst
               4NH3 +  2N02 + 02  	*-  3N2 + 6H20         (2)
     Reaction  (1) predominates  since approximately 90-95% of the NO  in
combustion flue  gas  is  in  the form of NO.   Since 1 mole of NH3 is
required per mole of NO, most processes operate with an NH^/NO mole
ratio of 0.9 to  1.1.  The  reaction temperature is usually in the range
of 300 to 450°C, but space velocities vary  considerably depending on the
process.  Under  these operating conditions, NOX removal efficiencies of
90% or greater are typical.^

     The catalysts used in SCR  processes vary with process developer.
However, there are some general traits known about the catalysts.  The
catalyst carrier or  substrate is usually alumina, silica, or titanium
dioxide.  Alumina is satisfactory for application to flue gases without
SOX such as from natural gas firing.  However, alumina tends to react
with SOX, particularly  803, to  form aluminum sulfate.  This "poisons"
the catalyst by  decreasing the  available surface area and the catalyst
activity.  Titanium  dioxide and silica are  less susceptible to attack by
803 and are applicable  to  flue  gas from heavy oil or coal firing.  The
active metal on  the  substrate may include Co, Cr, Cu, Fe, Mn, Ni, Pt,
and V or combinations thereof,  but the exact composition of the catalyst
is usually proprietary.  These  metals or their oxides can also react
with the SOX to  form sulfates.  Many of these sulfates are also cataly-
tic in the reduction of NO with NH3, and therefore, can be tolerated.
The catalysts are normally designed to have a life of at least 1
year.4,5

     The formation of ammonium  sulfate and bisulfate is a major concern
with SCR processes.  Ammonium bisulfate will form downstream of the
reactor if NH3 and 803  are present in sufficient quantities and if the
gas temperature  drops sufficiently.  It is very difficult to avoid the
conditions for formation;  for example, ammonium bisulfate will form if
the gas contains 10  ppm of NH3  and 10 ppm of 803 at a temperature of
210°C.  Ammonium sulfate will form at lower temperatures.  Fine partic-
ulate  emissions of  these  compounds are a concern, but the major problem

-------
296                    CLEAN COMBUSTION OP COAL


with ammonium sulfate and bisulfate  is deposition  on  heat  exchanger
surfaces.  Since these compounds are very  corrosive and  interfere with
heat transfer,  the heat exchanger must be  made of  corrosion  resistant
material and must be cleaned periodically  by  soot  blowing  or water
washing.  Approaches to preventing this formation  entail use of  an
ammonia decomposition catalyst or operation at lower  NH3/NO  ratios.4,5

     Another concern with SCR processes is catalyst plugging.  Signi-
ficant progress has been made in avoiding  plugging problems  through
reactor and catalyst design.  Fixed  bed reactors,  such as  parallel
passage, tube,  and honeycomb, are being designed which can tolerate
particle loadings typical of coal firing.  (Unless otherwise specified,
the fixed bed reactors referred to in this paper will be this  open
passage type.)  Moving bed reactors  are also  being developed which can
tolerate and remove moderate amounts of par tides. ^»^ However,  the
particle concentrations acceptable to a moving bed reactor are approxi-
mately an order of magnitude less than those  tolerated by  a  fixed bed
system.  The space velocity through  a moving  bed reactor is  expected  to
be about double that of a fixed bed  reactor,  but the  pressure  drop
across the moving bed reactor should be less.

     Published  information on the cost of  SCR processes  is limited and
estimates available are based on different design  premises.  The
reported estimates of the required capital investment range  from  $10  to
$80/kW with an  average of about $30/kW.  The  revenue  requirements range
from 0.2 to 3.3 mills/kWh with an average  of  about 1.7 mills/kWh.2»5

     A list of  Japanese SCR process  developers is  given  in Table  I along
with information on their developmental status.  There are 16  commercial
scale plants in operation treating 100,000 to 750,000 Nm3/hr of flue  gas
(33 to 250 MW).  In addition, there  are 11 prototype  plants  treating
from 15,000 to  99,999 Nm3/hr of flue gas and  numerous pilot  and bench
scale plants in operation.6  The prototype and commercial  scale plants
are achieving 90% control of the NOX from  flue gas derived primarily
from gas- and oil-fired sources.  The operating experience in  Japan
qualifies NOX control by SCR as a viable control technique in  the U.S.
when high NOX removal efficiencies are required from  gas-  and  oil-fired
sources. SCR technology has not been demonstrated  in  Japan on  coal-fired
sources although several of the pilot plants  are currently evaluating
such an application and larger scale demonstrations are  expected  in the
near future.  In fact, a northern utility  in  Japan in considering
installation of a 90 MW SCR system on a coal-fired boiler  which will
begin operation in 1980.6

Dry Simultaneous NOX/SOV Processes

     There are  two noteworthy variations of SCR processes  which
have the capability to simultaneously remove  NOX and  SOX:  the
activated carbon process and the Shell copper oxide process.

     The activated carbon process requires a  special  carbon  bed
which acts as an adsorbent for SOX and as  a catalyst  in  the
reduction of NOX with NH3.  When the bed is saturated with SOX,
flue gas is switched to a fresh bed, the carbon is regenerated'

-------
                  TABLE I.
MAJOR NO  SELECTIVE CATALYTIC REDUCTION PLANTS IN JAPAN
        x
Process Developer
Asahi Glass
Hitachi Ltd.
Hitachi Ltd. -Mitsubishi
Hitachi Zosen
Hitachi Zosen
Hitachi Zosen
Hitachi Zosen
Japan Gasoline
Japan Gasoline
Kurabo
Mitsubishi H.I.
Mitsui S.B. -Mitsui P.C.
Mitsui S.B. -Mitsui P.C.
Mitsui Toatsu
Mitsui Toatsu
Mitsui Toatsu
MKK-Santetsu
Osaka Gas
Seitetsu Kagaku
Sumitomo Chemical
Sumitomo Chemical
Sumitomo Chemical
Sumitomo Chemical
Sumitomo Chemical
Sumitomo Chemical
Sumitomo Chemical
Sumitomo Chemical
Plant Owner
Asahi Glass
Kawasaki Steel
PC Mitsubishi PC
Toshin Steel
Idemitsu Kosan
Shindaikyowa P.C.
Kawasaki Steel
Kashima Oil
Fuji Oil
Kurabo
Fuji Oil
Mitsui Pet. Chem.
Ukishima Pet. Chem.
Mitsui Toatsu
Mitsui Toatsu
Osaka Pet. Chem.
Nippon Yakin
Osaka Gas
Seitetsu Kagaku
Sumitomo Chem.
Sumitomo Chem.
Higashi Nihon Met.
Sumitomo Chem.
Sumitomo Chem.
Nihon Ammonia
Sumitomo Chem.
Sumitomo Chem.
Plant Site
Keihin
Chiba
Yokkaichi
Hime j i
Chiba
Yokkaichi
Chiba
Kashima
Sodegaura
Hirakata
Chiba
Chiba
Chiba
Takaishi
Takaishi
Takaishi
Kawasaki
Sakai
Kakogawa
Sodegaura
Anegasaki
Sodegaura
Anegasaki
Niihama
Sodegaura
Sodegaura
Sodegaura
(Nm
70
350
150
70
350
440
750
50
70
30
200
200
240
170
87
90
15
63
15
30
100
200
200
200
250
250
300
„ Capacity
/hr) (-VMW)
,000
,000
,000
,900*
,000
,000
,000
,000
,000
,000
,000
,000
,000
,000*
,000*
,900*
,000
,800*
,000
,000
,000*
,000*
,000*
,000*
,000*
,000
,000
23.
116.
50.
23.
116.
146.
250.
16.
23.
10.
66.
66.
80.
56.
29.
30.
5.
21.
5.
10.
33.
66.
66.
66.
83.
83.
100.
3
6
0
6
6
6
0
6
3
0
6
6
0
6
0
3
0
3
0
0
3
6
6
6
3
3
0
Source of
Gas
Glass Furnace
Coke Oven
Oil-fired Boiler
Heating Furnace
CO-fired Boiler
Oil-fired Boiler
Sintering Machine
Heating Furnace
CO Boiler
Oil-fired Boiler
Oil-fired Boiler
Oil-fired Boiler
Oil-fired Boiler
Ammonia Plant
Ammonia Plant
Heating Furnace
Oil-fired Boiler
Gas Generator
Oil-fired Boiler
Oil-fired Boiler
Gas-fired Boiler
Heating Furnace
Gas-fired Boiler
Heating Furnace
Heating Furnace
Oil-fired Boiler
Oil-fired Boiler
Start-up
Apr.
Oct.
Dec.
June
Nov.
Dec.
Nov.
Nov.
Mar.
Aug.
Late
Sep.
Aug.
Oct.
Mar.
Sep.
June
July
June
July
Feb.
May
Feb.
Mar.
Mar.
Mar.
Oct.
1976
1976
1975
1976
1975
1975
1976
1975
1976
1975
1977
1975
1976
1976
1976
1976
1976
1976
1975
1973
1975
1974
1975
1975
1975
1976
1976





<^,
0
03
O

f
M
Q
Cfl
Kj
W
Kj
§
i-3








Flue gas contains minimum amounts of SO  and particles
                                       X
                                                                                     NJ
                                                                                     \O

-------
298                    CLEM COMBUSTION OF COAL
and  a  concentrated S02 stream  is produced which  can be  used to
generate  a  salable byproduct.  The  process has the  potential
for  removing  90% of both pollutants, but its  application may be
limited to  flue gas containing relatively equal  concentrations
of NOX and  SOX.^  The economic projections for the  process  are
about  $65/kW  for the capital investment and 6.3  mills/kWh in
revenue requirements.  The  activated carbon process is  being
evaluated on  a pilot plant  scale in Japan by  Takeda,  Unitika,
and  Sumitomo  Heavy Industries.-*

      In the Shell process,  copper oxide reacts with 862 to
form copper sulfate.  The copper sulfate and, to a  lesser
extent, the copper oxide act as catalysts in  the reduction  of  NOX
with NH3-   As in the activated carbon process, a multiple bed
system is required so that  a bed is available for acceptance
while  regeneration takes place.  In the regeneration  cycle,
hydrogen  is used to reduce  the copper sulfate and a concentrated
S02  stream  is produced which can be used to generate  a  salable
byproduct. 4  The economic projections for the process are $131/kW
for  the capital investment  and 5 mills /kWh for the  revenue  requirements.
The  process has been installed at the Showa Yokkaichi Sekiyu
Company  (SYS) plant in Yokkaichi, Japan on a  commercial scale
 (120,000  Nm3/hr) and has removed 90% of the SOX  and 70% of  the
 WET PROCESSES

      The  following wet  process  types are being developed:

                     Oxidation-Absorption Processes
                     Absorption-Oxidation Processes
                     Oxidation- Absorption-Reduction Processes
                     Absorption-Reduction Processes

 Wet N0_ Processes
 ..... -     A.

      The  first  two process  types are generally for NOX  control
 only.  In oxidation-absorption  processes,  the relatively  insoluble
 NO  is  oxidized  in the gas phase to N02 which is  absorbed  into
 the liquid phase.  The  typical  oxidizing agents  used  are  ozone
 and chlorine dioxide.   The  absorbents vary with  process developer.
 The process seems more  feasible for flue gas containing equimolar
 mixtures  of NO  and N02  which  is not typical of combustion flue
 gas.   Much of the absorbed  N02  remains in  the liquid  phase in
 the form  of nitrate  salts which are water  pollutants.1^

      In absorption-oxidation  processes, NO is absorbed  directly
 into  the  liquid phase and then  oxidized.   Liquid oxidizing
 agents such as  sodium hypochlorate or hydrogen peroxide are
 used  to convert the  NO  to a nitrate salt.  Due to the insolubility
 of  NO, relatively large absorbers are required.   In addition,
 the process is  not applicable to flue gas  containing  S02  since
 the more  soluble SO^ would  consume the liquid oxidizing agent
 by  converting the absorbed  sulfite ion into sulfate. 4

-------
                      N(VS°x FLUE GAS TREATMENT                    299


     Due to their complexity, limited applicability, and water
pollution problems, wet processes  cannot  compete economically
with dry selective catalytic  reduction processes for control of
NOX in combustion flue gas. Therefore, these process types will
not be addressed further  in this paper, but additional information
is available from other sources.2,3,4,5

Wet Simultaneous NO^/SO^.  Processes
 11     "     * •'••"• i •• inJv   lA,      i,,

     The attractiveness of wet processes  is their potential for
simultaneous removal  of NOX and SOX.  Oxidation-absorption-
reduction processes and absorption-reduction processes are
designed for this type of control.

     The oxidation-absorption-reduction processes basically
evolved from flue gas desulfurization (FGD) systems.  A gas
phase oxidant is injected before the scrubber to convert NO to the
more soluble N02-  The N02 is then absorbed into an aqueous solution
with S02-  The absorbent  varies with the  type of FGD system being
modified.  The absorbed S02 forms  a sulfite ion which reduces a
portion of the absorbed nitrogen oxides to molecular nitrogen.  The
remaining nitrogen oxides are removed from the waste water as nitrate
salts.  The remaining sulfite ions are oxidized into sulfate by air
and removed as gypsum.  The percentage of nitrogen oxides going
either to the preferred molecular  nitrogen or to the troublesome
nitrate salts is uncertain; however, it is estimated to be about 50-
50, but this can vary considerably.^

     The oxidation-absorption-reduction processes have the potential
to remove 90% of both SOX and NOX  from combustion flue gas.^
However, there are several drawbacks remaining to be overcome before
the processes can be  widely applied.  The process chemistry is
complex and use of a  gas  phase oxidant, such as ozone or chlorine
dioxide, is expensive.  Chlorine dioxide, although cheaper than
ozone, adds to the waste  water problems created by the nitrate
salts.  Chlorine dioxide  also causes concern due to the possibility
for chlorination of organics  in the waste water to produce carcinogenic
compounds.

     Despite these drawbacks, the  potential of a simultaneous
removal process warrants  further research and development.  Table II
lists the process developers  evaluating oxidation-absorption-reduction
technology.  One small commercial  scale plant (33 MW), four prototype,
and three pilot plants are currently being operated in Japan.   As
with SCR processes, published economic data are limited, but the
average reported capital  investment is about $110/kW and the average
revenue requirement is about  7.5 mills/kWh.

     The absorption-reduction processes were seemingly developed to
avoid the use of a gas phase  oxidant.  A  chelating compound, such as
ferrous-EDTA (ethylenediamine tetraacetic acid) which has an affinity
for the relatively insoluble  NO, is added to the scrubbing solution.
The NO is absorbed into a complex with the ferrous ion and the S02
is absorbed as the sulfite ion.  The NO complex is reduced to molecular

-------
TABLE II.
MAJOR WET NO /SO  CONTROL PLANTS IN JAPAN
            x   x
                                                                                         o
                                                                                         o

Process Developer
Plant Owner
Plant Site
(Nm3
Capacity
(MW)
/hr) Approx.
Source of
Gas
Start-up
OXIDATION - ABSORPTION - REDUCTION PROCESSES
Chiyoda
Ishikawa j ima H.I.
Mitsubishi H.I.
Osaka Soda
Shirogane
Sumitomo Metal-Fujikasui
Sumitomo Metal-Fujikasui
Sumitomo Metal-Fuj ikasui
Chiyoda
Ishikawa j ima H.I.
Mitsubishi H.I.
Osaka Soda
Mitsui Sugar
Sumitomo Metal
Toshin Steel
Sumitomo Metal
Kawasaki
Yokohama
Hiroshima
Amagasaki
Kawasaki
Amagasaki
Fuji
Osaka
ABSORPTION - REDUCTION
Asahi Chemical
Chisso Corp.
Kureha Chemical
Mitsui S.B.
Asahi Chemical
Chisso P.C.
Kureha Chemical
Mitsui P.C.
Mizushima
Goi
Nishiki
Chiba
1
5
2
60
48
62
100
39
,000
,000
,000
,000
,000
,000
,000
,000
0
1
0
20
16
20
33
13
.3
.6
.6
.0
.0
.6
.3
.0
Oil-fired
Oil-fired
Oil-fired
Oil-fired
Oil-fired
Oil-fired
Boiler
Boiler
Boiler
Boiler
Boiler
Boiler
Heating Furnace
Oil-fired
Boiler
Aug.
Sep.
Dec.
Mar.
Aug.
Dec.
Dec.
Dec.
1973
1975
1974
1976
1974
1973
1974
1974
PROCESSES


5

600
300
,000
150
0
0
1
0
.2
.1
.6
.05
Oil-fired
Oil-fired
Oil-fired
Oil-fired
Boiler
Boiler
Boiler
Boiler
Apr.
Apr.
Apr.
Apr.
1974
1974
1975
1974
                                                                                         o

                                                                                         i
                                                                                          o
                                                                                          Q
                                                                                          M

                                                                                          O
                                                                                          o
                                                                                          o

-------
                              FLUE GAS TREATMENT                    301
nitrogen by reaction with the sulfite  ion.  A regeneration  step
recovers the ferrous chelating  compound  and oxidizes  the  sulfite  ion
into sulfate which is removed as  gypsum.-*

     The absorption-reduction processes  also  have  the potential to
remove 90% of both the NOX and  SOX  in  combustion flue gas. 4  Although
the processes seem to have advantages  over the oxidation-absorption-
reduction processes, there are  obstacles to be overcome before the
processes can be widely applied.  Even with the addition  of the
chelating compounds, a large absorber  is required  to  absorb the NO.
The replacement, recovery, and  regeneration costs  of  the  chelating
compounds, although potentially less than the gas  phase oxidants,
are still significant.  The process chemistry is complex  and is
sensitive to the flue gas composition  of SC>2,  NOX, and oxygen.  The
molar ratio of SC>2 to NOX must  remain  above approximately 2.5 and
the oxygen concentration must remain low.^

     Table II also lists the process developers evaluating absorption
reduction technology in Japan.  There  are four pilot  or bench scale
plants currently being operated. 6  The average reported capital
investment is about $96/kW and  the  revenue requirements are about
6.3 mills/kWh.5

     Table III summarizes the status,  cost, and performance of the
dry and wet processes for control of NOX and  simultaneous control of
NOX and SOX.
CONTROL OF NOx, SOX, AND PARTICLES


     Over 90% control of NOX,  SOX, and particles may eventually be
required in the U.S. for stationary  combustion sources, especially
for new, large, coal-fired  sources.   Schematics of some of the
alternatives for such an overall  control system are shown in Figure
1.

     Perhaps the ideal situatiori  would be represented by one control
device that simultaneously  removes NOX, SOX, and particles.  Such a
control device is not yet available,  but it is conceivable that a
wet scrubber system could remove  all three pollutants.  Such a
system is represented in Schematic A of Figure 1.  Since the recovery
of heat from the flue gas is important, a heat exchanger is shown in
the schematic.  A reheater  is  also shown since most wet scrubber
systems require reheat of the  flue gas prior to its discharge from
the stack.

     The most developed overall control system is NOX control by
selective catalytic reduction  (SCR),  SOX control by flue gas desulfur
ization (FGD) , and particle control  by an electrostatic precipitator
(ESP) .  The sequence of these  control devices is variable as illus-
trated in Schematics B, C,  and D.  Schematic B shows particle control
first, followed by NOX control, and  then S02 control.  In this
configuration, the SCR system  could  be a moving bed type which would
require most of the particles  to  be  removed first.  To maintain high

-------
                     TABLE III.  STATUS OF FLUE GAS TREATMENT PROCESSES IN JAPAN

Process Type
Dry Selective Catalytic Reduction
(with ammonia)
Dry Simultaneous NO /SO
} XX
Activated carbon
Shell Copper Oxide

Developmental
Plants
16 commercial
11 prototype

3 pilot
1 commercial

Removal Efficiency (%)
NO SO
x x
_> 90 N.A.

^ 90 ^90
^70 ^90

Approx.
Capital
Cost
($/kW)
30

65
110

Approx.
Revenue
Requirements
(mills /kWh)
1.7

6.3
5.0
0
f
o
o
1
H
O
!2|
Wet Simultaneous NO /SO
                   X   X


     Oxidation-absorption-reduction
1 commercial

4 prototype

3 pilot
< 90
> 90
131
7.5
                                         o
                                         o
     Absorption-reduction
4 pilot or bench    < 90
          > 90
              96
             6.3

-------
                 NCL,/SOX FLUE GAS  TREATMENT
303
                         Wet  Scrubber
                         for  KOX,  SOX,  &
                         Participates
                     SCR
                 (Moving  Bed)
          SCR
       (Fixed Bed)
ESP


Dry MOX/SOX
(Activated
carbon)
ESP


Wet
NO /SO
x' x
   ESP - Electrostatic Precipitator for Particle Control
   FGD - Flue Gas Desulfurization for S02 Control
   SCR - Selective Catalytic Reduction for NOX Control
FIGURE 1 - POTENTIAL EQUIPMENT CONFIGURATION FOR NO , SO ,
                        AND PARTICLE CONTROL
                                                   x'   x'

-------
304                    CLEAN COMBUSTION OF COAL


temperatures needed by the SCR system, a hot-ESP would be necessary.
The requirement for a hot-ESP could be avoided by use of a fixed bed
SCR system which could tolerate high particle concentrations.  This
configuration is illustrated in Schematic C.

     Since the sulfur compounds in the flue gas lower the resistivity
of the fly ash and improve the collection efficiency of the ESP, it
is not deemed advantageous to have the ESP follow the FGD system.
However, it is uncertain if ammonia, ammonium sulfate, or ammonium
bisulfate will leave the SCR system and adversely affect the performance
of the FGD system.  To avoid this possibility, the SCR system could
follow the FGD system as shown in Schematic D.  However, the dis-
advantage of this configuration is the extensive reheat that would
be required to raise the flue gas temperature to a level suitable
for the SCR system.

     The simultaneous NOX/SOX systems present an apparent simplification
of the control systems as shown in Schematics E, F, and G of Figure 1.
Schematic E depicts the activated carbon process and Schematic F
illustrates the Shell process for dry simultaneous NOX/SOX control.
Two configurations are presented due to the different tolerance of
particle concentrations and to the different operating temperatures
of the processes.  The wet simultaneous NOX/SO  control systems are
shown in Schematic G and can be represented by either the oxidation-
absorption-reduction processes or by the absorption-reduction processes.

     At this time it is mere speculation as to which configuration
will emerge as the optimum overall control system from a technical
and economic standpoint.  In all probability, the optimum system
will be dependent on specific flue gas and site considerations as
well as user preference.
                 EPA'S NO  FLUE GAS TREATMENT PROGRAM
                         A

     High NOX removal efficiencies may not be required to achieve
and maintain the current National Ambient Air Quality Standards
(NAAQS).  Further, the current New Source Performance Standards
(NSPS) for NOX can be achieved by implementation of combustion
modification techniques which are more economical than FGT processes.
However, there are significant factors which may cause the NAAQS and
the NSPS to become more stringent.  These include:  the alarming
increase in NOX emissions from stationary combustion sources pro-
jected for the next decade, the possibility of a short-term NO?
NAAQS, the relationship of NOX emissions to levels of photochemical
oxidants, the impact of increased use of coal resources, the impact
of relaxing mobile source emission standards, the role of NO
emissions as precursors to other pollutants of concern such as PAN
(peroxyacyl nitrates), other nitrates, nitrosamines, and nitric
acid, and the health effects of NOX and related pollutants.

     If more stringent standards are promulgated, then NO  FGT
technology may be required to meet the standards to protect human
health and welfare.  Therefore, the NOX FGT program is proceeding

-------
                      N0x/S0x FLUE GAS TREATMENT                    305


with experimental projects progressing toward full scale demonstration
of highly efficient NOX and simultaneous NOX/SO  control technology
in parallel with control  strategy and technology assessment studies.
The results of these  studies will assist in determining the appropriate
scale of the experimental projects.  EPA's past, current, and planned
activities in these areas are summarized below.

CONTROL STRATEGY AND  TECHNOLOGY ASSESSMENT STUDIES

     The control strategy and technology assessment studies are
mainly research projects  to examine various aspects of NO  control
technology and to determine if and when NO  FGT technology will be
needed in the U.S.                        x

Assessment of Japanese Technology

     Since Japanese technology in this field is more advanced than
any other country's,  EPA  has sponsored the publication of periodic
reports and papers to facilitate the transfer of information on NOX
and NOX/SOX abatement technology from Japan.  These documents have
been mainly prepared  by Dr. Jumpei Ando of Chuo University in Tokyo,
Japan.2,3,4,7,8  Dr.  Ando is also assisting EPA in activities
associated with the Stationary Source Pollution Control Project of
the US/Japan Environmental Agreement which includes a subproject on
NOX and NOX/SOX FGT technology.

     In addition to monitoring published information from Japanese
sources, EPA personnel have made periodic trips to Japan to observe
testing facilities and to discuss the technology with process
developers and operating  personnel.  The most recent trip was in
March 1977.

Ozone Oxidation of NO9

     Gas phase oxidation  of NO to N0£ is essential for wet oxidation-
absorption-reduction  processes.  Therefore, a task order was issued
by IERL-RTP to the Research Triangle Institute to analyze the supply,
demand, production economics, and energy consumption of this key FGT
process step.  The results of the study indicate that only a stoichio-
metric amount of ozone is required to achieve essentially complete
conversion of NO to N02 which may be subsequently scrubbed from the
gas stream.

     The energy requirements and the capital and operating costs
were examined for ozone generation with both air and oxygen as input
to the ozone generator.   Approximately 13% more energy is required
for ozone generation  from oxygen than from air.  The capital invest-
ment for ozone generation from oxygen is about 3 times as large as
that required from air, and operating costs are about twice as
large.  For a 500 MW  plant with air as input to the ozone generator,
the estimates for oxidizing 200 ppm of NO were:  energy requirement,
1.1 X 108 kWh/yr or 3.1%  of station capacity; capital investment,
$17.60/kW; and operating  costs, 2.0 mills/kWh.  The 200 ppm concentra-
tion is representative of a coal-fired source with combustion

-------
306                    CLEAN COMBUSTION OF COAL
modification techniques applied or an oil-fired  source without
supplementary NOX control applied.  The estimates  are for  oxidation
of NO  only; the  energy requirements and cost of  control  for  N0£
would  be additive.

     The report  indicates that unless there is a significant improve-
ment in ozone generation technology, wet processes using ozone for
oxidation  of NO  to N02 will be very expensive.   However, since these
processes  have potential for simultaneous NOX/SOX  control, the
energy and cost  impacts may be more acceptable.

N0y Control Strategy Assessment

     IERL-RTP contracted with Radian Corporation to determine the
potential  effectiveness of applying NOX controls to large  stationary
combustion sources.   The Chicago Air Quality Control Region (AQCR)
was selected for a modeling study of emissions from point, area, and
mobile sources to determine the relative impact  of each  category on
ambient NOX concentrations.  The calibrated dispersion model pre-
dictions of annual average concentrations indicate that  the  major
point  sources, which contributed nearly 40% of the total NOX
emissions  in Chicago, accounted for less than 10% of the ambient N02
levels in  1974.  Preliminary investigation of expected short-term
concentrations of total NOX shows that major point sources may
contribute as much as 80% of measured NOX levels.  Therefore, it
appears that stringent NOX control for large point sources may be
required to meet a potential short-term N02 standard, but  cannot be
justified  currently on the basis of the existing annual  average N02
standard.10  However, NOX emissions from stationary combustion
sources are expected to increase significantly in the next decade.
As a result of these findings, the Chicago AQCR  modeling study was
expanded to determine more accurately the short-term ambient N02
levels, to project the annual and short-term N02 concentrations to
1985,  and  to assess the use of NOX emission control on stationary
combustion sources to attain or maintain compliance with possible
N02 ambient short-term and annual average standards.  The  results of
this study should be available by early 1978.

     Another Radian Corporation study is seeking to determine the
key factors relating to "if" and "when" NOX FGT  technology will be
needed in  the U.S.  Since research and development of a  technology
should lead its  application by several years, it is necessary to
monitor factors  which could require implementation of NOX  FGT
technology in the near future.  By these efforts, the decision to
emphasize, maintain, or terminate the research,  development,  and
demonstration of NOX FGT technology can be made  on the best  available
information.  This study should also be available in early 1978.

Economic Assessments of NOy FGT Processes

     The Tennessee Valley Authority (TVA), through an interagency
agreement  with EPA, is developing comparative economics  of NOX and
NOX/SOX FGT emission control processes.  This state-of-the-art
review will be conducted in two phases.  Phase I will evaluate and
summarize  the technical feasibility of all candidate NOX control

-------
                      NOX/SOX FLUE GAS TREATMENT                    307


processes being offered in the U.S. and Japan.  The Phase I report,
which includes descriptions of about 45 processes, should be published
in the summer of 1977.5  Phase II will concentrate on the most
promising processes identified in Phase I and will perform a preliminary
economic assessment of each, including development of material and
energy balances.  In addition, a direct comparison of the economic
and technical feasibility of the dry and wet processes will be made
to determine the most effective method to remove NOX and SOX from
combustion flue gas.  The project is cofunded by IERL-RTP and the
Electric Power Research Institute.

     EPA is planning a third phase of the project to prepare detailed
economic projections of as many as four of the most promising process-
es.  This activity should be complete in late 1978.  Further, it is
contemplated that a study will be conducted during this phase to
determine the impact of ammonia utilization by SCR processes.  The
cost and energy requirements of ammonia generation for a typical
utility application will be examined.  In addition, the impact on
the supply, demand, and cost of ammonia worldwide will be analyzed.
This study may be available by the end of 1977.

EXPERIMENTAL PROJECTS

     EPA's experimental projects have been directed toward enhancing
the evolution of FGT technology from bench scale research to full
scale demonstration on coal-fired sources by the mid-19801s.  The
technology must be applicable to utility and large industrial combustion
sources and must achieve highly efficient NOX and simultaneous
NOX/SOX control in a relatively energy efficient and economical
manner.

Bench Scale Catalyst Research^-^-

     In 1975, a research grant was awarded to the University of
California at Los Angeles (UCLA), School of Engineering and Applied
Sciences to further the development of promising catalysts.  The
study extended the catalyst screening work performed earlier by UCLA
under an IERL-RTP contract with TRW, Inc.12  The objectives of the
grant were to optimize the compositions of vanadium and iron-chromium
catalysts for selective reduction of NOX with ammonia and to perform
long-term durability studies of the optimum catalyst compositions in
flue gas containing sulfur dioxide.  The results of the study,
completed in mid-1976, indicated that a 15% loading of vanadium
oxide on alumina support material and a 10% loading of iron oxide-
chromium oxide on alumina support material with an iron/chronium
ratio of 9:1 were the optimum catalyst formulations.

     Parametric tests showed that both catalysts were selective, in
that only the NOX was reduced.  The tests also showed strong enchancement
of NOX conversion rates due to the presence of Q£ under typical
operating conditions.  C02, H20, and S02 did not affect the NOX
reduction in the concentration ranges applicable to power plant
exhaust.  Both catalysts were most active between 400°C and 425°C
and required excess NH3 for maximum activity.  Long-term durability
tests of both catalysts in the presence of SOX indicated no degradation

-------
308                    CLEAN COMBUSTION OF COAL
 in  catalyst performance.  Typical conversion levels  for  the  vanadium
 and iron-chromium catalysts operating at 400°C  in  simulated  flue  gas
 were about 90% and  80%, respectively, at 20,000 hr~l space velocity.
 In  addition, preliminary tests of iron-vanadium and  iron-chromium-
 vanadium  catalysts  indicated 99% removal of NOX from the simulated
 flue gas.

 Pilot Plant Evaluation of Gas and Oil Firing13

      In 1973, a  contract was awarded to Environics,  Inc. to  evaluate
 the performance, reliability, and practicality  of  a  SCR  system with
 ammonia and a platinum catalyst on alumina support material.  A
 pilot plant, treating a slipstream from an operating utility boiler,
 was designed, installed, and tested on gas and  oil firing.  Laboratory
 testing was conducted to supplement the pilot plant  testing.  Satisfactory
 results were found  on gas-fired operation with  85-90% NOX removal
 achieved  for over 4000 hrs at a space velocity  of  50,000 hr~l.
 Results on oil-fired operation indicate that the catalyst system was
 not suitable for flue gas containing SOX.  The  maximum NOX removal
 efficiency achieved was 65% with the average only  50%.   Fluctuation
 in  flue gas temperature and catalyst plugging with soot  and ammonium
 sulfate caused problems on oil-fired operation.

 Pilot Plant Evaluation on Coal Firing

      The  next phase of the experimental program is evaluation of FGT
 processes on a coal-fired application.  A request  for proposal was
 issued in September 1976, and best and final offers  are  currently
 being evaluated.  Contract award is anticipated by the end of
 September 1977.  It is contemplated that two contracts will result
 from this procurement process.  One will be for a pilot plant to
 evaluate  removal of NOX emissions, and the other will be to evaluate
 simultaneous removal of NOX and SOX.  However, budgetary constraints
 and technical considerations may impact on the  final decision in
 this  regard.

      The  pilot plants must treat a flue gas volume equivalent to 0.5
MW  and achieve a NOX removal efficiency of 90%.  For the simultaneous
 control of NOX and  SOX, 90% removal of both pollutants must be
 achieved.   The pilot plant projects will each consist of a 24 month
program which will be conducted in four phases.  Phase I includes
 the preparation of a detailed process design and an estimation of
 capital and operating costs for the pilot plant.  Following erection
of  the pilot plant and mechanical acceptance testing in Phase II,
 the contractor will perform system start-up and debugging, parametric
 testing,  and optimization testing over a wide range of flue gas
 conditions during Phase III.  Phase IV provides for testing and
evaluation of the plant during 90 days of continuous operation.  It
is  currently anticipated that final reports will be published on the
results of the pilot plant operations in early 1980.  A project
manual conveying the total concept of the proposed plant is planned
for early 1978.   In addition to these projects, there is the possibility
of  a  pilot plant project being initiated in 1977 with the Tennessee
Valley Authority.

-------
                      NOX/SOX FLUE GAS TREATMENT                    309
     The pilot plant projects will enable an assessment of the
technical, environmental, energy, and economic aspects of applying
N0x and NOX/SOX FGT technology to the U.S. coal-fired situation.
This information, in conjunction with the control strategy and
technology assessment studies, will provide technical and budgetary
direction and emphasis for EPA's NOX and NOX/SOX FGT program.
                              REFERENCES
1.   Bowen, J. S., G. B. Martin, R. D. Stern, and J. D. Mobley.
     "Stationary Source Control Technology for NOX."  The Second
     National Conference on the Interagency Energy/Environment R&D
     Program, Washington, D.C., June 6 and 7, 1977.

2.   Ando, J., R. D. Stern, and J. D. Mobley.  "Status of Flue Gas
     Treatment Technology for Control of NOX and Simultaneous Control
     of SOX and NOX in the United States and Japan."  American
     Institute of Chemical Engineers, 69th Annual Meeting, Chicago,
     Illinois, November 28 - December 2, 1976.

3.   Ando, J., H. Tohata, and G. A. Isaacs.  NO  Abatement for
     Stationary Sources in Japan.  PEDCo-Environmental Specialists,
     Inc.  EPA-600/2-76-013b (NTIS No. PB 250 586/AS), January 1976.
     U.S. Environmental Protection Agency, Research Triangle Park,
     N.C.

4.   Ando, J., H. Tohata, and K. Nagata.  "NOX Abatement for Stationary
     Sources  in Japan - August 1976."  PEDCo-Environmental Specialists,
     Inc., (Draft Report, to be published Summer 1977 by EPA).

5.   Faucett, H. L., J. D. Maxwell, and T. A. Burnett. "State-of-
     the-Art Review of Processes for Removal of Nitrogen Oxides from
     Power Plant Stack Gas."  Tennessee Valley Authority (Draft
     Report, to be published Summer 1977 by EPA).

6.   Ando, J., personal communication with J. D. Mobley, 06/25/77.

7.   Ando, J., and H. Tohata.  Nitrogen Oxide Abatement Technology
     Japan - 1973.  Processes Research Inc., EPA-R2-73-284
     (NTIS No. PB 222 143), June 1973.  U. S. Environmental Protection
     Agency, Research Triangle Park, N. C.

8.   Ando, J.  "Status of Flue Gas Desulfurization and Simultaneous
     Removal of S02 and NOX in Japan."  In Proceedings: Symposium on
     Flue Gas Desulfurization, New Orleans, March, 1976, Vol. I,
     EPA-600/2-76-136a (NTIS No. PB 255 317), May 1976, pp 53-78.
     U.S. Environmental Protection Agency, Research Triangle Park,
     N. C.

-------
310                    CLEAN COMBUSTION OF  COAL
  9.   Harrison, J. W. Technology and Economics of Flue Gas NOV
       Oxidation by Ozone, Research Triangle Institute, EPA-600/7-76-
       033  (NTIS No. 261 917/AS), December 1976.  U.S. Environmental
       Protection Agency, Research Triangle Park, N.C.

  10.  Eppright, B.R., personal communication with J. D. Mobley, 09/27/76.

  11.  Nobe, K., G. L. Bauerle, and S. C. Wu.  Parametric Studies of
       Catalysts for NOV Control from Stationary Power Plants, University
       of California, Los Angeles, EPA-600/7-76-026  (NTIS No. PB 261
       289/AS), October 1976.  U.S. Environmental Protection Agency,
       Research Triangle Park, N.C.

  12.  Koutsoukos, E. P., J. L. Blumenthal, M.  Ghassemi, and G. L.
       Bauerle.  Assessment of Catalysts for Control of NOX from
       Stationary Power Plants, Phase I, Volume I, Final Report,
       TRW, Inc., EPA-650/2-75-001-a  (NTIS No. PB 239 745/AS), January
       1975.  U.S. Environmental Protection Agency, Research Triangle
       Park, N.C.

  13.  Kline, J. M., P. H. Owen, and Y.  C. Lee.  Catalytic Reduction
       of Nitrogen Oxides with Ammonia:   Utility Pilot Plant Operation,
       Environics, Inc., EPA-600/7-76-031 (NTIS No. PB 261 265/AS),
       October 1976.  U.S. Environmental Protection Agency,  Research
       Triangle Park, N.C.

-------
                                                                    311
                 SESSION V - WHERE DO WE GO FROM HERE?

   SESSION CHAIRMAN:  VICTOR S. ENGLEMAN, SCIENCE APPLICATIONS, INC.
     The papers in previous sessions have presented the foundation on
which we stand and have supplied the building blocks to achieve clean
combustion of coal.  This closing session takes a look into the future
in terms of policy, supply, and technology for direct coal utilization.
It is clear from the conference thus far that precombustion, combustion,
and postcombustion technologies will all "be important in achieving "Best
Available Control Technology" for coal utilization.  The National Energy
Plan will have a major impact on the overall energy picture and the
incentives and requirements for coal combustion.  Increased coal utili-
zation will require increased supplies which will require new mines and
increased capacity in the delivery system.  New developments and improve-
ments in technology will be required to achieve the likely environmental
requirements in a technically and economically sound manner.  The papers
in this session focus on these three key areas in the future of clean
combustion of coal.

-------
312                     CLEAN COMBUSTION OF COAL

-------
                                                                    313
                          Transcript of Talk on

                        THE NATIONAL ENERGY PLAN
                          )
                                   by
                           C. William Fischer
                          Department of Energy
                            Washington, D.C.
August 5, 1977

     I spoke to Jim Schlesinger a week ago today in connection with the
job of putting together a planning mechanism for the new Department of
Energy and he gave me some very succinct and terse guidance which I
believe is particularly relevant to this conference and which is why I
am more eager to come here and listen to you than to talk.  He said,
"You know I have been around Washington a little bit."  I should say
that Jim Schlesinger, now Secretary Schlesinger, is quite a remarkable
guy.  His nomination and confirmation for the job of first Secretary of
Energy in less than one day must have set a speed record.  Hearings on
the nomination were held yesterday morning from 8:00 to 11:30, and in
the afternoon the Senate voted to confirm.  But there is another record
here:  This is the fourth cabinet-level agency Jim Schlesinger has
headed in the last seven years.  Prior to becoming the President's
chief energy advisor, Schlesinger was Chairman of the Atomic Energy
Commission, Director of the Central Intelligence Agency, and Secretary
of Defense.  Prior to that he was Assistant Director of the Bureau of
the Budget which was his forst job in Washington.  His directions to me
the other day were very clear.  He said that in the defense business and
in some of the other places where he has worked, you make a decision on
a project, you make some estimates about what it ought to cost to imple-
ment, and then you summon the four-star generals and you say this is
what I want you to do, this is how much money you have to do it with,
and you may also call in the president of North American Rockwell and
say that if you do it for that amount within that time schedule this is
what the fee will be, and if you don't make the time schedule or you
don't make the performance standard, then the fee will be reduced accord-
ingly, and away you go from there.  That is exactly the opposite of what
the Department of Energy must do.  The Department, if it is going to do
its job for the Nation, should set a broad policy framework of overall
objectives and then determine, with the help of the people who are going
to have to work with these energy problems, how to get from here to
there.  In setting the broad policy goals we should ask, "What is feas-
ible?"  The output of this department is not going to be a set of things
or a set of widgets; it is going to be a set of decisions that have to
be made by millions of consumers and thousands of leaders in the corpo-
rate structures and the Federal, State and local public institutions of
this country.   The Federal Government will not be able to design,
develop, direct ajid implement this program as though it is in control

-------
314                    CLEAN COMBUSTION OF COAL


of all, or even the majority, of the tools necessary to solve our energy
problem.  We are going to have to do an awful lot of listening, and we
are going to have to do a lot of cooperative work with the people who
have to make the programs work.  I thought I would share these thoughts
with you.  It is a perspective that is far from messianic.  Just the
opposite; it is something that is, I think, a realistic recognition of
what it will take to move this country toward more plentiful energy
resources and away from the profligacy in the use of energy that we have
come to know, in part because of our pricing policy.  This basic philos-
ophy of listening to the public, is particularly applicable to me.  I am
not an engineer; I am not an expert; I have no specialized knowledge of
the technical parameters within which we must operate.   That is why I
said last night that after this conversation I would like to hear your
recommendations on legislation to facilitate siting, both for nuclear
and nonnuclear power.

     As we begin our discussion this morning, it might  be worth it to
review the gravity of the situation that we face.  The  world community
is now using about 60 million barrels of oil a day.   We are increasing
our demands for that scarce commodity each year by about 5 percent and
we are just now loading the first tanker out of Alaska.  The entire
yearly Alaskan North Slope production, when at full yield, is going to
be roughly equivalent to the world's increase in consumption over a
period of nine months.  The whole of Texas oil production would equal
the world's increase in oil consumption for one year; total Saudi Arabian
production is roughly equivalent to the projected net increase in world
consumption over the next three years.  World production can probably
keep increasing for six to eight years.  Sometime in the late 80's or
early 90's, however, it won't go up much more.  If we could hold the
growth rate in world demand to five percent yearly,  we  would use all of
today's proven reserves—notice I said proven reserve,  not inferred, not
potential—by the end of the next decade.  There are lots of arguments
about how much resource there is in the ground.  Oil companies are making
all kinds of noises that if you just gave them a little more incentive
they could turn all inferred and potential reserves into proven reserves.

     In that regard, I would like to talk about the general structure of
the National Energy Plan, and, specifically, about this incentive ques-
tion that has to do with basic pricing policy.  Later,  I would like to
concentrate on the coal part of the program and give you an update on
what the specific provisions were in the President's Plan, what the Ways
and Means and the Commerce Committees have done to the  regulatory and
tax provisions, and what are the major issues pending for House floor
resolution today, and I mean literally today, that could alter the Plan
significantly.  Then we will open it up to questions and discussion.

     Alaskan flow, at maximum production, will equal about two years'
growth in U.S. energy demands.  If we don't act now, by 1985 we will be
using about 33 percent more energy than we are using today and because
we probably can't substantially increase domestic oil production, we
will have to. import twice as much oil.  To give you a frame of reference,
in 1970 we were paying $2.7 billion a year for imported fuels.  By 1976
the cost of imports was $36 billion.  In 1977 we expect to be paying
$1*5 billion for oil imports by 1985-  What are we doing about this?
Before the National Energy Plan we were continuing merrily on our way.

-------
                          NATIONAL ENERGY PLAN                      315
We Charted a twelve-month moving average of gasoline consumption in the
United States.  The chart shows that in August of 1973 we were consuming
6.6 million barrels of motor gasoline per day.  By August of 1975 we
were still using about 6.6 MMB/D but this statistic reflected the period
of the 1973-7U embargo when there were physical restrictions on supply,
long lines for gasoline and so on.  By June of 1977 this country was
consuming 7.1 million barrels of gasoline per day.  Since June of 1975
we have been increasing consumption at the rate of 3.8 percent a year.
Production rates, on the other hand, have decreased.  Another twelve-month
moving average for production shows that in August of 1973 this country
was producing 9.3 million barrels of crude per day.  By June of 1977
production was down to 8 million barrels per day—the equivalent to a
decreasing rate of 3 percent per year over the past two years.  In
January of 1973 we were importing U.8 million barrels a day; by January
of 1977, 7-5 million barrels a day—a rate of increase of 17.2 percent
per year since June of 1975-  The Administration's program consists of
more than 150 specific provisions which are now moving through the United
States Congress faster than I have ever seen any bill of that size, on
any subject, move.  Earlier this week I was engaged in a public television
discussion with Representative Jim Jones from Oklahoma and the Moderator
commented that in April when the President gave his message there was
much concern expressed, but that now things have quieted down and there
seems to be an energy gap in the Congress.  Well, the first thing I did
was to deny that.  I had never seen such a sustained level of intense
Congressional inquiry, debate, and activity.  It may be that the coinci-
dence of the timing of the National Energy Plan hitting the Congress and
the Speaker's dedication to getting control of the House of Representa-
tives may have facilitated fast action on the bill.  I don't agree with
this interpretation of the reasons for this speedy consideration of the
bill by the House.  I think Congress is exhibiting genuine interest in
the energy problem.  But the road is not smooth.

     We have some of our greatest problems in the coal area—which perhaps
reflects the technical difficulties we encounter in this area and the
difficulties that have been discussed here in balancing our enhanced
concern for the environment with our consciousness that we have to move
at least in the intermediate term to more plentiful sources of energy.
In some cases these difficulties can be resolved together—their solutions
are not mutually exclusive nor inconsistent.  I am thinking specifically
of automobile emissions.  It now looks as though going to cleaner tail-
pipes may also get us more efficiency.  Cutting down the size of automo-
biles, which our European friends have seen fit to do for quite some
time, actually accomplishes' both goals simultaneously.  But there are
plenty of areas where we have very real and difficult technical problems
to solve.  You are in a position to help find some of the solutions.
Beyond finding solutions to technical problems, some necessary compromises
will have to be made to get us where we need to go in terms of balancing
our need to burn coal, for instance, and our desire to protect the envi-
ronment .

     The Administration's program is basically four-pronged.  First, we
must have adequate price incentives to produce additional resources
without undue enrichment.  That is the guts of the new pricing policy
which says that we are going to try to get the price of scarce fossil
fuels up to their replacement value without causing undeserved windfall

-------
316                    CLEM COMBUSTION OF COAL
gains to producers.  The mechanism for avoiding windfall gains to pro-
ducers will be a tax coupled with a rebate to consumers to help compen-
sate for higher fuel prices.  There is room for disagreement on whether
or not feeding these taxes that are imposed on the higher prices at the
pump back to the consumer will in fact put a demand restraint in the
system.  It remains to be seen whether prices together with the rebate
will have the full effect of reducing consumption.   I believe that it
will, because people react to their immediate consumption patterns and
if they have a rebate in the future, they will not necessarily spend it
on fuel.  The second prong of the National Energy Act is the strong
incentive to conserve without an unfair burden on the consumer.  Adequate
price incentives for production and correspondingly higher prices are the
front part of this element of the plan.  Prices will go up for both
natural gas and oil—especially for new production, at the margin.  We
define "new" production in terms of where the well is drilled with regard
to an already-existing producing well—at which depth and at what distance
in relation to that last producing well.   In case of gas, the Administra-
tion has agreed to a new reservoir concept, the details of which are yet
to be worked out.  And prices for those "new" commodities will be essen-
tially uncontrolled.  Theoretically, they would go to the world price.
Additionally, residential, commercial and industrial consumers would be
encouraged to conserve through adopting energy efficiency measures, such
as retrofitting homes and businesses.  Tax credits are provided for this
purpose.  Third are the proposals for strong incentives to use more
abundant fuels, that is, the conversion program and the development of
safe nuclear energy.  Fourth is the vigorous research and development
program on new and renewable energy resources.

     With regard to the incentives for oil production, one of the propos-
als , known as the Jones-Schroeder Amendment, would allow producers a tax
break against their well-head tax due.  The credit would equal one-half
a percent a month for each month for which exploration expenditures
exceeded 25 percent of gross revenues.  At the end of the first year the
credit would equal 6 percent of the well-head tax due.  This amendment
which the Administration opposed was defeated yesterday on the floor of
the House.  The reason I am discussing the amendment is because it goes
right to the heart of the major producer's case—that there are inadequate
incentives for exploration and that more incentives are needed to develop
new fields.  First the plow back would be in addition to proposals that
are in the President's bill to put secondary and tertiary enhanced
recovery at the world price and all newly discovered oil at the world
price by 1980, plus adjustments to account for the domestic rate of in-
flation.  It is also in addition to the Administration's provisions that
would allow domestic producers to receive a greater-per-barrel margin
than producers any place else in the world.  We surveyed the different
world markets and found that with the world price being set on new oil
production, domestic producers would receive a greater-per-barrel margin
than they would any place else in the world including the North Sea and
other fields.  The Ways and Means Committee estimated that the Jones-
Schroeder proposal would increase oil industry revenues by $89 million
in 1978; we estimate that the amendment would result in an oil industry
revenues increase of up to $100 million.   If the tax break is made to be
cumulative, then, by the end of 1980, the plow back would increase the
oil industry's revenues by about $2 billion and $5.2 billion by the end
of 1985.

-------
                          NATIONAL ENERGY PLAN                      317


     Another problem with the Jones-Schroeder proposal is that the
25 percent threshold is so low the producers would be rewarded for doing
something that they are doing already.  According to the Bureau of the
Census 1975 Annual Survey of Oil and Gas, independent producers were
already reinvesting 63 percent of their revenues, apart from royalty
revenues which account for 12 percent of their total revenues, for
exploration and development.  If you make the adjustment for royalty oil
you are still above the 25 percent threshold.  In 1975 the majors—the
top eighteen oil producers—were reinvesting 35 percent of their revenues.

     Additionally, the plow back would cause considerable administrative
difficulties since it would be based on taxes and consequently the crude
oil equalization tax would have to be levied on producers rather than
first purchasers as was proposed in the Administration bill.  Taxes
would have to be collected from over 1600 producers rather than refiners
and other first purchasers of which there are about 300.  Furthermore,
the cash balances that the major producers will have without the plow back
after their investment in exploration and development is adequate.  These
funds are, of course, not excess profits.  They are funds for use in
other business purposes; they are essential to the business enterprise.
They can be resources for alternative investments, or can be paid out to
shareholders as dividends.  But these are funds that could be used for
additional investment for exploration and development which are not being
so used.  These producer cash balances amounted to $2 billion in 1972 and
went up to $U.9 billion in 1975.

     Still another problem is that the credit is available only to pro-
ducers of flowing oil.  A producer has to be established to have the
25 percent threshold ascertained.  New entrants to the field would not
be able to earn the credit.  Major oil companies—the top 18—own two-
thirds of all old domestic oil for which the plow back would be available,
and, therefore, the majors would benefit the most.  The reported domestic
profits of the largest 18 oil companies increased from $3.5 billion in
1972 to $5-7 billion in 1975—a 60 percent increase in reported profits.
Profits for 1975 were second only to 197^ profits in the history of the
industry.  Almost all of this increase in income comes from domestic
rather than foreign production.

     The government is currently quite limited in its ability to examine
the industry profit structure.  The Administration has put forward a
proposal, authorized under existing law and mandated under the Department
of Energy Organization Act, to improve the financial reporting by major
producers.  The major producers are not averse to the proposed Petroleum
Company Financial Reporting System—they recognize that they have credi-
bility problems.

     Capital investment in development and exploration has increased in
recent years.  The average annual level in the early 1970Ts was about
$9 billion a year; the actual spending in 1975 was $21.7 billion.  In
February of this year the Oil and Gas Journal projected that 1977 expen-
ditures for development and exploration would be up to $26.7 billion.
The sum total of this evidence leads us to believe that the profits are
up, the investments are up, and the physical activity is up with regard
to exploration and development without price decontrol.  That gets us
back to the question of whether the degree of enrichment that would

-------
 318                    CLEAN COMBUSTION OF COAL
result for producers from decontrol and the rise in prices to consumers,
which the producers themselves are advocating, is necessary to stimulate
exploration and development activity.  ¥e think there is reason to ques-
tion the producers' arguments.  With regard to physical activity, the
number of rotary rigs in operation within the U.S., including Alaska, off-
shore, and the outer Continental Shelf, has increased significantly.  In
1968, out of 2100 total rigs, 1500 were active and 600 were inactive.  In
1972 the numbers dipped to about 1^00 active, and the total number of
rigs dropped to about 1760.  By 1976 the total number of rigs was back
up to 2200.  There are now almost 2000 active rigs.   The number of exist-
ing but inactive rigs is down to 225-  About 90 percent of the total gas
and oil rigs were active as of 1976.  I've seen these figures in the
annual rotary rig census of the Reid Tool Company.   Hughes Tool does a
similar census and they have some more recent numbers.   Their figures
indicate that the number of rigs active in 1977 is  also around 1900 to
2000.  So exploration and development activity is up and the number of
inactive rigs is significantly down.  And this is all prior to the
President's policy under which new oil goes to the  world price and new
reservoirs go to the world price all of which makes for additional incen-
tives.  By the way, 2% miles from a currently producing well is not a
great distance in determining what is so-called new gas and oil.

     In spite of all this activity, however, additions  to reserves have
declined, and proven reserves have dropped by 12 percent since 1972, to
a proven domestic reserve level of 30.0 billion barrels at the end of
1976.  What has happened is that as the prices have risen, and they have
risen substantially even under the controlled situation, we have made
wells in fields that were earlier passed over.   Instead of stepping out
into new reserves and new pools, producers are going back and opening up
previously uneconomic reserves that don't expand our reserve.  As it
turns out there is also no significant effect on production—domestic
oil production is continuing to fall.  So this raises the question whether
higher prices do yield significant additional production.  For gas there
have been increases in intrastate prices from 1970  averaging about 25<£
per Tcf to $1.30, to $1.97 and in some intrastate cases to above $2.00
(in some new contracts in parts of Texas and Louisiana) and interstate
increases from 35<£ to an average of $1.21 per Tcf in 1976.  Gas well
drilling is today more than twice the level it was  in 1971, but again
proven reserves have shrunk by over 25$ from the 1967 high of 293 tril-
lion cu ft.

     A few words about the oil and gas replacement  program or the so-
called coal conversion program.  The proposal has two pieces and affects
two major sectors.  The pieces are the regulatory and the tax provisions.
They affect major fuel-burning industries as well as utilities.  I will
discuss the regulatory proposals and what has happened to them in the
Congress.  Our regulatory proposals would cover industries and utility
units capable of consuming fuel at a fuel heat input rate of 100 million
Btu per hour or greater or a combination of several such units which have
a total capability of consuming fuel at a fuel input rate of 250 million
Btu's per hour or greater.  In industry it includes boilers, combustors,
combined cycle units and internal combustion engines.  I am going to go
through these provisions and tell you what the principal committees in
the House, where the action is right now, have done to those provisions.
But before I begin, to give you some perspective, the total amount of oil

-------
                          NATIONAL ENERGY PLAN                      319


imports that the President's program would save is kit million barrels a
day.  Total U.S. consumption of oil in 1976 vas 17.1* million barrels a
day.  Of that 17-U MMBD consumed in 1976, our imports were 7.3 million
barrels a day.  If we don't do anything, by 1985 our oil imports will go
to 11.5 million barrels a day.  Under the President's program, imports
would level off at 7-1 to 7-2 MMBD which means about a IK 5-million-
barrels-a-day savings.  Of that, 2.2 million barrels a day or about
50 percent is to be achieved through the oil and gas replacement program.

     The House Commerce Committee, which has jurisdiction over the
regulatory provisions has made a fundamental change in our proposal with
regard to industry.  The Committee bill excludes combustors from the
proposed regulatory activity.  If this provision is finally approved it
will result in an oil savings loss of somewhere between 250 thousand
barrels and half a million barrels a day for that provision alone.  Under
the President's bill no oil and gas will be burned by new utilities.
Exceptions are made for environmental site limitations.  Existing utili-
ties will not be allowed to burn gas after 1989.  If a utility is coal-
capable it will be ordered to switch to coal.  If it is not coal-capable
it can be ordered to shift from gas to oil.  The Commerce Committee
version deletes this last provision.  The Administration proposal makes
exceptions for existing utility units based on environmental grounds or
site limitations or the use of synthetic fuels.  With regard to new
industrial units no oil or gas may be used for boiler fuel.  Nonboiler
installations may also be prohibited by regulation or order from using
oil or gas.  Once again, exceptions will be made based on site limita-
tions, environmental grounds, or technical impossibility.  The House
Committee has made no change in these provisions and they are up for a
vote on the floor of the House today.  Existing industrial units which
are coal-capable may be prohibited by regulation or order from using oil
or gas.  If they are capable of using mixtures or combinations of fuels,
they may be ordered to use the minimum possible amounts of oil or gas.
Exceptions here also are based on site limitations, environmental grounds
or technical impossibility.

     I should also discuss briefly the tax provisions.  On the industry
side there would be a single tax rate effective in 1979 at a level of 900
a barrel for oil used starting in 1979 and would phase up to $3 a barrel
for oil used in 1985, adjusted for inflation.  Effective in 1979 gas
used would be taxed at a rate equal to the difference in Btu equivalent
price between gas and distillate fuel minus $1.05 per million Btu of gas
used during that year.  The $1.05 deduction phases out, so that by 1985
the tax will equal the difference in price between gas and Btu equivalent
distillate.  The Ways and Means Committee went from a one-tier tax
provision and made a distinction between the tax rate for boilers,
turbines and other internal combustion engines, and other nonexempt
industrial users, mostly coal-capable nonboilers.  The Committee also
basically reduced the initial tax from 900 (in the case of industrial
oil) to 300 a barrel, but phasing up to the same $3 a barrel by 1985-
These reductions in taxes are estimated to lose, in terms of oil saved,
another half a billion barrels of oil.  Industries and firms using more
than 500 billion Btu's or 85,000 barrels of oil equivalent a year—account-
ing for less than 2 percent of the total number of industrial firms or
about lilOO out of 100,000 firms— will be taxed under these provisions.

-------
320                    CLEM COMBUSTION OF COAL


     The Ways and Means Committee included an exception from this tax
for process use of oil or natural gas if other fuels are not usable for
technical, economic or environmental reasons.  The Administration's
proposal, as well as in the Ways and Means proposal, made exceptions for
transportation including aircraft, rail and water, farming, ammonia feed-
stocks, refining, automotive or special fuels, and for gas that is
reinjected for the extraction of additional gas.   The Ways and Means
Committee added exemptions for residential facilities, commercial
facilities, and any facilities not an integral part of manufacturing,
processing or mining.  Additionally, Congressman  Steiger introduced an
amendment in the Ways and Means Committee that would extend exemptions
from the regulatory provisions for environmental  and technical feasibility
reasons to the application of the tax.   The Administration's bill provided
that, even though you might be exempt from the prohibition against using
oil and gas, you would still have the tax imposed.  The Steiger amendment
would say that if you were exempt from the regulation you are also exempt
from the tax.  The Steiger amendment adds another loss in savings of
barrels of oil a day.  The Gorman amendment that  is up on the floor today
would eliminate that additional exemption.  Finally, the Ways and Means
Committee has added two other exemptions to the bill.  Industries that
can show that they would be placed at a competitive disadvantage would
be exempt.  That adds up to another 100,000 barrels of oil a day.
Another amendment provides a procedure whereby the Secretary would have
discretionary authority to reclassify industrial  uses of oil and gas to
a lower tier tax rate or altogether exempt such uses from tax.  That
would add another loss in savings of a quarter of a million barrels a
day.

     With regard to the utility oil tax, the Administration proposed a
250 per million Btu or $1.50 a barrel flat tax beginning in 1983 but
with inflation adjustments beginning on the date  of passage.  The Ways
and Means Committee put a 250 a million Btu or $1.50 a barrel flat tax
beginning in 1983, but with inflation adjustments beginning in 198l.
The utilities gas use tax begins in 1983 and is keyed to the regional
distillate price as a target.  In 1983 the tax would be 500 a million Btu
less than the target price.  By 1988 it would be  equal to the target
price, that is, the regional distillate price.  Inflation adjustments are
made from 1975-   Once again, the Committee would  start the tax in 1983,
initially at 550 per million Btu and going to 750 per million Btu in
1985.  The tax can't exceed the regional residential gas prices and it
is inflation-adjusted from a later date.  This loses the additional one-
half a billion barrels that I mentioned previously.  All those additional
amendments and exemptions reduce the 2.2 million  barrels a day oil saving
that is supposed to result from the coal conversion program to somewhere
between .8 and 1.0 million barrels a day.  The Administration's bill also
contains a rebate program which is basically a tax exemption from this
oil and gas use tax for investment in alternative energy properties,
including boilers not using oil or gas, coal utilization equipment, low
Btu gas control equipment, required pollution control equipment, but
excluding buildings and structural components. The Ways and Means
Committee added to that high-Btu-gas-from-coal equipment, up to the
turbine stage, and for geothermal and hydroelectric power.

     For utilities the rebate is the same as the  industrial rebate in
the Administration's bill except that in the case of the industry there

-------
                          NATIONAL ENERGY PLAN                       321


is an alternative.  They can either take the rebate or they can take an
additional 10 percent investment tax credit on top of their existing
lO^percent investment tax credit.  This does not apply to utilities.  The
utility rebate is available only for investments that replace or convert
existing oil- or gas-fired facilities.  Industrial firms are not limited
in the Administration's bill to conversion or replacement investments.
The Ways and Means Committee provides that the rebate for utilities is
also available for phasing down instead of only being available for the
conversion and puts the incentive on phasing down and converting from
base load to peak load use of oil- or gas-fired facilities.  That also
would mean a substantial loss in energy savings.

     I would say that the major issue on the House side that is up for
action today is the Corman-Steiger debate on whether the regulatory
exemption should also apply to the taxes.  The rationale here, of course,
for allowing the tax to be imposed even when the regulations are not, is
that you need a variable incentive to move people away from the use of
oil and gas or to allow them, in the case of large firms—and these
taxes only apply to large firms—to take advantage of being able to
convert where they can convert, and pay the taxes where they choose not
to convert or where they can't convert.  Those taxes would then be recap-
tured for them in terms of their conversion investments on their units,
because they can trade off inside the firm the taxes they pay on one
unit with the rebates they get for the conversion of another unit.

     I hope I have given you some idea of the importance of these pro-
visions.  I appreciated very much getting from you your suggestions on
what the Federal government ought to do to facilitate this conversion
to more plentiful sources of energy both in terms of its regulatory
activities and the siting question and in terms of the technical activity
in research and development.

-------
322                     CLEAN COMBUSTION OF COAL

-------
                                                                    323
                  THE OUTLOOK FOR COAL THROUGH 1985

                           Robert L. Major
                      Manager - Market Research

                          AMAX Coal Company
                         Indianapolis, Indiana


     Although the Carter Administration has indicated that it is
counting on coal to be the "swing fuel" which will allow us to reduce
our dependence on imported oil, it has refused to come to grips with
the real and potential barriers to the increased mining, transporta-
tion, and utilization of coal.

     The future role of coal must be evaluated within the context of
the total energy picture.  While it is obvious that the United States
needs a "comprehensive national energy policy," there has been little
agreement as to what specific policies should be included in such a
national energy policy.  In general, industry had in mind a program
which relied on the market forces of supply, demand, and price to
allocate scarce energy resources among the various competing uses and
to achieve desired fuel switching.  However, as Walter Mead stated
recently in an article in Science:

     ". . .to political Washington, the cry for a national energy
     policy is interpreted as a demand for more government decision-
     making and less reliance on the market."*

Mead further states that most professional energy economists have
argued for less government and more reliance on the market to solve
the "energy problem."  This preference does not stem from a political
conservatism as much as from their awareness of the "poor record of
government interference  in the energy market.  That record is one of
massive and repeated resource misallocation" (emphasis addedT-*  The
result of such an approach is that great emphasis is being placed on
the "stick" and little on the "carrot."  A "stick" policy which seeks
to force certain actions which run counter to the internal economics
of the energy user is likely to be met with strong opposition, while
policies which seek to create economic incentives ("carrots") to
switch fuels or to use energy more efficiently are much more likely to
succeed.  With these brief remarks as an introduction, I would now
like to discuss the general factors underlying the markets for coal.
     *Mead, W. J., 22 July 1977, "An Economic Appraisal of President
Carter's Energy Program," Science, vol. 197, p. 340.

-------
324                    CLEM COMBUSTION OF COAL


 General  Market  Factors

      Coal  markets  tend  to  be segmented by coal  types.   The primary
 segmentation  of the  market is  between  metallurgical  and steam coals.
 Among the  steam coals  there is  a  further  segmentation  by coal  types--
 bituminous, subbituminous, lignite,  and anthracite.   Each of these
 coals tends to  serve special  niches  within the  overall  coal  market.
 Some of  the markets  are much more constrained than others.   Metal-
 lurgical coking coals  are  specialty  coals which command premium
 prices.  The  demand  for these  coals  is determined by the output of pig
 iron and steel, both here  and  abroad.   The bulk of the  coals exported
 overseas from the  United States (excluding Canada) is of the metal-
 lurgical variety.  Although there has  been some movement of  lower
 grade (high-volatile) metallurgical  coals into  the "compliance" steam
 coal* market  in recent  years,  the markets for metallurgical  and steam
 coals are  essentially  separate.   Because  the main growth sector over
 the next decade will be in the  steam coal  market, little further
 discussion of metallurgical  coal  markets  will be included in this
 paper.

      Steam coal  is the  largest  segment of the coal industry, and
 electric power  plants are  the  largest  consumers  of steam coal,  account-
 ing for  some  68 percent of total  coal  consumption in  1976.   Lesser
 amounts  of steam coal are  used  by industry for  process  steam,  space
 heating, and  self-generation of electricity.  Other  uses  are minor and
 are expected  to essentially disappear  in  the future.  Potentially, the
 largest  user  of steam coal  in  the future  will be coal-based  synthetic
 fuel  plants.  However,  because  of technical, economic,  and financial
 problems,  it  now appears that coal gasification  and  coal  liquefaction
 will  not be a significant  factor  in  the market  until 1990 or later.

      Traditionally,  the utility and  industrial  coal markets  have been
 strongly regionalized because of  the limitations of  transportation
 costs, and consumers have  tended  to  buy coal from the nearest  coal
 supply areas.   The "shape"  of the market  was determined  by the  de-
 livered  cost of various coals vis-a-vis the delivered cost of  other
 coals and  other  fuels.   This regionalization of  the markets  has  been
 broken down to  some  extent  by the  new  air  pollution control  regula-
 tions for  power  plants, which have forced  utilities to  use more  dis-
 tant  "compliance coals" rather  than  local   high-sulfur coals.  As a
 result,  large tonnages  of  Western  compliance coals have  been moving
 east  into  the Upper Midwest markets at the expense of previously used
 Illinois Basin  coals.   In  an attempt to slow down this movement  and to
 restore the use of "Eastern" coals,  the Carter Administration  has
 proposed a policy in which  scrubbers would be mandated on all new
 power plants regardless of  the  sulfur  content of the coal burned
     *"Compliance coal" is used here to designate any coal whose
sulfur content is sufficiently low so that it can be burned without  a
scrubber and still be in compliance with the 1.2-pound of SOo/MM
Btu standard.                                               *

-------
                          OUTLOOK FOR COAL                         325
     Anthracite coal production is mostly limited to the northeastern
part of the State of Pennsylvania.  At one time, it was mined in
significant quantities and supplied a large share of the energy market
in the Northeast.  However, high mining costs combined with competi-
tion from cheaper oil and gas have resulted in this industry's shrink-
ing to a mere shadow of its former self.  At present, anthracite
accounts for less than 1.0 percent of the total U.S. coal production.
However, this coal does possess several advantages.  It is well
situated to serve the populous Northeast energy markets and it is a
high-Btu, low-sulfur fuel.  As a result, FEA has recently established
a task force to assess the feasibility of increasing the use of anthra-
cite in the future.

     Among the steam coals, lignites play a special, more restricted
role.  Although the United States has vast reserves of lignite which
can generally be mined at low cost, the low heat value (6,000-7,000
Btu per pound) and high moisture content of the lignites tend to
restrict their use to markets near the lignite deposits.  With few
exceptions, lignite is utilized at mine-mouth locations for electric
power generation.  At present, the bulk of the lignite produced in the
United States is mined in the states of North Dakota and Texas.  While
the use of such local lignites reduces the demand for coals in more
distant supply areas, the converse is not true, because lignite cannot
compete in distant markets.

     In the case of bituminous and subbituminous coals, the product is
sufficiently high in quality that long distance movements of such coal
are competitive under certain conditions.  However, in most cases the
markets for these coals are strongly regionalized.  It is useful to
discuss the markets for these coals in terms of the main coal supply
regions—Appalachia, the  Illinois Basin, and the "West."  The West, in
turn, contains a number of subregions, each of which serves its own
markets.

    - The great bulk of the new expansion in coal production in the
United States through 1985 will occur in the West, especially in the
Powder River Basin of Wyoming.  It is important to discuss this expan-
sion, because there is a  great deal of confusion about this develop-
ment.  The demand for this Western coal arises from three distinct
markets, each of which is controlled by different economic variables.
The first market demand for Western coal is from utilities and in-
dustries located in the Western coal supply areas to meet local
demands for energy.

     The second market demand is for Western coal to serve utilities
and industries located in the West but outside of the coal-producing
areas.  This market will  be served both by coal used in mine-mouth
power plants whose electrical output is "wheeled" to distant market
centers and by coal which is transported to distant markets by train,
barge, and/or slurry pipelines.  This new demand for Western coal is
primarily stimulated by the shortfall in natural gas supplies, which
is causing utilities and  large industrial energy users in the West
(especially in the West South Central Region)  to build new coal-fired
plants to replace their old gas-fired plants.

-------
326                    CLEAN COMBUSTION OF COAL
      The  third  market  demand  is  the  so-called "compliance coal" market.
 This market is  stimulated  by  the desire of utilities (mainly in the
 Upper Midwest)  to avoid  the installation of expensive flue gas desul-
 furization units ("scrubbers"),  which would be required if they con-
 tinued to use high-  and  medium-sulfur coals from the Midwest.   These
 utilities and industries,  in  some cases, are paying a considerable
 premium over the cost  of local coals in order to purchase low-sulfur
 (compliance) coals from  the West or  from Appalachia.  The impact of
 this penetration has been  most severe on the market for Illinois Basin
 coals.

 Future Coal Demand

      Table 1 indicates our forecast  of the total demand for coal in
 1985 by major consuming  sectors.   You will note that we are more
 optimistic than the Carter Administration regarding the demand for
 utility coal, but much less sanguine in regard to industrial  demand.
 For the sake of comparison, we have  also included the forecasts pre-
 pared by Joel Price of Dean Witter and by Island Creek Coal  Company.
 The key point to note  is the  consensus that the Carter Administration
 is too optimistic, and that it is unlikely that U.S. coal  production
 will reach 1.27 billion  tons  by  1985.

      Table 2 indicates the trends in the demand for coal  by electric
 utilities through 1985.   These data  indicate estimated coal  demands by
 NERC region.  The boundaries  of  the  NERC regions are shown in  Fig-
 ure 1.  Although the overall  growth  rate of 7.3 percent per year may
 appear to be high, we  feel that  it is reasonable because the phase-out
 of natural gas  as boiler fuel, and the delays in bringing new  nuclear
 plants on stream will  force utilities to rely heavily on coal  for
 their future fuel supplies.  The very high growth in the Southwest
 regions of ERGOT and SPP is mostly a response to the phasing out of
 extensive gas-fired capacity.

      Despite the rapid growth in utility coal use in the West, from
 89 million tons in 1976  to 337 million tons in 1985, some 59 percent
 (488 million tons) of  the  total  coal used by utilities in 1985 will
 still be consumed east of  the Mississippi River.  Secondly,  although
 there has been  much talk about Western coal being used in the  East,
 only about 56 million  tons, or 11.5  percent, of the 1985 coal  supply
 for Eastern utilities  will be Western coal.  Therefore, more than
 85 percent of the coal expansion in  the West is expected to serve
 Western markets.

      The future demand for coal  by industry is very difficult  to
 predict because of the great  uncertainty which exists regarding the
 Federal Government's policies on coal  use and environmental  regula-
 tions.  However, if we assume that the long-term switch away from coal
 can be reversed, and that  part of the new energy demands by industry
 can be captured by coal, then we estimate that the potential  regional
 demand for industrial  coal in 1985 could be as shown in Table  3.  The
 boundaries of the regions  used in Table 3 are indicated in Figure 2
 Under either scenario, the bulk  of the future industrial coal  usage*is
 expected  to occur in only  two regions: the Midwest and the Southwest

-------
                                            Table 1
                    COMPARISON OF UNITED STATES COAL DEMAND PROJECTIONS FOR 1985
                                     BY MAJOR CONSUMING SECTORS
CONSUMING SECTORS


Electric Utilities

AMAX
COAL (1)
824
no
107-154
92
-

(Million Tons)
ISLAND
CREEK (2)
744
TOO
90
90
-


DEAN
WITTER (3)
778
no
104-134
87
-


CARTER
ADMINISTRATION (4)
779
105
278
90
12
1
Metallurgical                  110            100                 110                105                                     §
                                                                                                                             1-3

Industrial                 107-154             90             104-134                278                                     o
                                                                                                                             Q

Export


Synthetics

                                                                                                                             o
Other


TOTAL                     1133-1180          1024           1080*-1110*             1265
 (1)  AMAX Coal Company, Market Research Department, July 1977.

 (2)  Barker, Stom'e, Jr., February 8, 1977, "Coal - Current Status and Future Prospects", speech
       before the Washington Society of Investment Analysts, Washington, D.C.

 (3)  Price, Joel, June 15, 1977, "The Supply and Demand for Steam Coal: Implications and Challenges",
       an address before the Richmond Society of Security Analysts, Richmond, Virginia.

 (4)  Executive Office of the President, Energy Policy and Planning, June 2, 1977, Replacing Oil and
       ^ With Coal ajid^ Other Fuels In Tne Industrial and Utility Sectors, page III-2.


   * Does not add due to independent rounding.

-------
328
CLEAN COMBUSTION OF COAL
                                  Table 2
                   UTILITY COAL DEMAND, BY NERC REGIONS*
                               (Million Tons)
                                                     Annual Percent
NERC Region
ECAR
ERCOT
MAAC
MAIN
MARCA
NPCC
SERC
SPP
WSCC
TOTAL
1976
142.2
12.0
32.8
57.3
28.4
7.8
104.3
9.1
39.9
433.8
1980
166.6
39.9
38.6
73.4
45.9
10.1
117.5
48.3
70.7
611.0
1985 Growth, 1976-85
207.4
69.3
40.3
86.1
66.9
13.8
139.9
101.4
98.9
824.0
+ 4.3
+21.5
+ 2.3
+ 4.6
+10.0
+ 6.5
+ 3.3
+30.7
+10.6
+ 7.3
   * See Figure 1  for regional  boundaries.
     SOURCE:  National Electric Reliability Council, 1977
              Preliminary data.

-------
UNITED  STATES
 OF   AMERICA
    FIG.* I
   SCALE OF MILES
 0 100 200 300 400
                                                                                                                  o
                                                                                                                  o
                                                                                                                  3
                                                                                                                  o
                                                                                                                  o
                                                                                                                  U)
                                                                                                                  NO
                    FIG. * I - BOUNDARIES OF NATIONAL ELECTRIC RELIABILITY COUNCIL REGIONS

-------
330
                       CLEAN COMBUSTION OF COAL
                              Table  3
                    INDUSTRIAL COAL  DEMAND  IN  1985
                            BY  REGIONS*
                                           (Million Tons)
                               LOW GROWTH**          HIGH GROWTH***
   New England
   New York/New Jersey
   Mid-Atlantic
   South  Atlantic
   Midwest
   Central
   North  Central
   Southwest
   West
   Northwest
          Total
  0.7
  3.6
 15.9
 13.4
 31.6
  4.7
  4.4
 25.6
  4.1
  2.6
106.6
  1.3
  4.8
 18.5
 18.1
 38.7
  5.9
  6.2
 49.0
  7.7
  3.9
154.1
     *See Figure 2 for FEA industrial energy regional boundaries.
    **"Low Growth" scenario assumed that industry within each region
         will continue to use coal at the same rate that it did in
         1975 plus 10 percent of the incremental increase (1975-1985)
         in energy demands will be met by coal.
   ***"High Growth" scenario is similar to low growth, but assumes
         that 20 percent of the incremental demand (1975-1985) will
         be met by coal.

-------
                                                                                         (NEW ENGLAND
                            NORTH CENTRAL
UNITED  STATES
  OF   AMERICA
    FIG. #2
   SCALE OF MILES
  0 100 200 300 400
  T  I   I    I   I
                                                                                                                Q
                                                                                                                O
                                                                                                                o
                                                                                                               LO
                                                                                                               CJ
                    FIG. #2 - BOUNDARIES OF FEA INDUSTRIAL ENERGY  DEMAND  REGIONS

-------
332                    CLEAN COMBUSTION OF COAL
 However,  it should be noted  that depending on the" regional  availability
 of natural  gas and future pricing policies of the Federal  Government,
 the regional  shares could be altered  significantly.   Despite this, it
 is our opinion that the total  industrial  coal demand in 1985 will  be
 in the forecast range.

 Regional  Demand

      Now that we have estimated the future demand for coal  through
 1985, let us examine the question as  to which regions will  supply the
 coal to meet the forecast demand.  Table 4 indicates our forecast of
 the 1985 coal demand by market sector and by supply  region.   Despite
 the large increase in Western coal  production between now and 1985,
 Eastern and Midwestern coals are expected to continue to play a
 significant role in meeting  our future demands for coal.

 Supply/Demand Balances

      Recent research by AMAX Coal has indicated that while  the overall
 coal industry is expected to have surplus capacity through  the early
 1980's, potential  shortfalls* in steam coal  capacity could  develop in
 the Illinois Basin by 1979 and in Appalachia by 1981.   While part  of
 the potential shortfalls in  the East  could be filled by surplus
 Western coal, this option is not likely to offer a total  solution  to
 the supply problem.  Lastly, because  there are a number of  environ-
 mental, legal, and technical problems which might reduce the effective
 supply of Western  coal  by as much as  50 million tons or more below the
 currently forecast levels, the "surplus"  capacity projected for the
 West may never materialize.

 Summary

      If public policy conflicts regarding the mining,  transportation,
 and utilization of coal are  not quickly resolved, coal's future role
 will be diminished and our dependence on  imported oil  will  continue  to
 increase.   Many of the problems and uncertainties which face the coal
 industry are the result of our failure to develop a  consistent set of
 policies for coal.   Melvin Laird, in  an excellent little pamphlet
 entitled Energy -  A Crisis In Public  Policy makes the following
 observation:
 *Note:     These  potential  shortfalls  were  derived  by the  difference
           between  the estimated  demand  and the  projected  supply
           (which is  equal  to  current  capacity minus  depletions  plus
           announced  capacity).   Because the data on  new mine capacity
           in  the 1981-1985 period  are less complete  for the East,
           there  is a tendency to underestimate  actual  supply.   The
           projected  shortfalls should be taken  as  indicators of the
           regions  in which there is a potential demand for additional
           mine capacity.

-------
                                          Table 4



                      COAL DEMAND IN 1985 BY MARKET AND SUPPLY REGION
                                         (Million Tons)
                                               SUPPLY    REGION
SECTOR/MARKET
Electric Utilities
Industrial*
Metallurgical Coal
Exports**
— Metallurgical Coal
— Steam Coal
Total
APPALACHIA
285
52-65
97

76
16
526-539
ILLINOIS
BASIN
152
18-23
5

-
-
175-180
TEXAS
LIGNITE
56
13-26
-

-
-
69-82
NORTH
DAKOTA
LIGNITE
23
1-2
-

-
-
24-25
WESTERN
COAL
308
23-38
8

-
-
339-354
TOTAL
824
107-154
no

76
16
1133-1180
                                                                                                                              s
                                                                                                                              o
                                                                                                                              o
                                                                                                                              p
 * Includes minor usage by commercial and retail customers.


** All exports allocated to Appalachia as exports from other regions have been small
   and irregular.
                                                                                                                              U)
                                                                                                                              u>
                                                                                                                              to

-------
334                    CLEAN COMBUSTION OF COAL
      "The  failure  of  the  Federal  Government  to  come  to  grips with the
      crucial  environmental/energy trade-offs is partially attributable
      to  the  fact that,  in the  Congress  as well  as  in the  executive
      branch,  policy direction  for energy and environment  is  treated
      separately.   With  missionary zeal, those in government  dealing
      with  each  set of issues push their own  programs toward  conclu-
      sions that are often contradictory and  sometimes actually  impos-
      sible of resolution  by the decisionmaker in the private sector,
      at  the  end of the  regulatory chain."*

 We,  in the coal industry,  are  ready and willing to do our part  in
 solving  the  energy problem.  However, unless many of the  policy road-
 blocks are resolved quickly, it will be very difficult, if not  impos-
 sible, for coal to play as large  a role in our  future energy supply as
 is  currently projected.   Even  under the most optimistic assumptions,
 it  is questionable whether we  will reach the Carter  Administration's
 goal  of  almost  1.3 billion tons by 1985.  Coal  can be a part of the
 solution to  the energy  problem of the United States,  but  it  is  not a
 panacea.
     *Laird, M. R., 1977, Energy - A Crisis In Public Policyi
American Enterprise Institute for Public Policy Research, Washington,
u * i>>»9 p. y •

-------
                                                                    335
                    WHERE DO WE GO FROM HERE IN R&D?

                                   by
                            S. William Gouse
                     Deputy Assistant Administrator
                           Fossil  Energy,  ERDA
     ERDA is going into the Department of Energy on October 1, 1977.
Figure 1 shovs the projection of the major Fossil Energy activities (not
including mining and coal preparation) for four years "beyond FY '78.
Once in the Department of Energy, we will have added to the program coal
mining research and development and coal preparation work.  The numbers
for FY '78 are pretty firm, totaling a little over $600 million of new
contract authority.  The numbers for FY  '79 and "beyond are what the
program managers are projecting as their needs for carrying out the
program that has been initiated in the past and for the next fiscal year.
That also includes some new starts that they believe they will be able
to make as a result of research now underway.  As you can see, it goes
over a billion dollars per year in FY '80.

     The chart shows a general breakdown of activities.  Coal conversion
includes gasification and liquefaction.  Utilization means coal-oil
slurry combustion, atmospheric and pressurized fluid-bed combustion,
and direct combustion; as well as advanced high temperature turbines
for use of coal-derived fuels.  Advanced research and supporting tech-
nology means, what it says with respect to research, but includes mate-
rials and component development work in support of the total program.

     MHD is self-explanatory.  Demonstration plants cover the large
commercial scale projects we have under contract or are about to enter
into contract.  Petroleum, natural gas, and oil shale are not a subject
of this Conference, but they are in the Fossil Energy Program of ERDA
and they are likely to be in the program in the Department of Energy.
It is a large program.

     At this point in time, there are about 800 work efforts under con-
tract in place around the world.  I do not know how many subcontracts
there are.  There are the order of 200 contracts in universities alone
on fairly basic studies.  I believe our feeling, at the present time,
is that the level of support in new or exploratory R&D has been limited
by the number of ideas rather than by funds or other policy judgments.
We are probably funding more than we ought, rather than the other way
around.

     The growth of the program has been very rapid.  In FY '73, Fossil
Energy's budget was the order of $50 million.  Such rapid growth cannot
take place without errors in judgment.  In addition, some of our programs
are in place by Congressional direction.  This rate of growth of an R&D

-------
                       FIGURE  1
U)
u>
              FOSSIL ENERGY OVERVIEW
         BUDGET AUTHORITY DISTRIBUTION
             BY MAJOR PROGRAM AREA

                     FY 1977-1982
                            MODIFICATIONS AT ENERGY RESEARCH CENTERS
4001—
    1977
             1978
                       1979
                                1980
                                          1981
                                                   1982

-------
                               COAL R&D                             337


activity is faster than is prudent when one thinks in terms of making
wise investment decisions.  However, the nature of the problem facing
the country caused Congress to launch an accelerated activity with
supplemental appropriations in FY '7^, with the understanding that such
an acceleration would bring with it more risk than had been the practice.
When such an increase in obligational authority takes place, with pres-
sure to obligate rapidly, you are forced to put under contract the best
of what is available at that particular time.  Once something substan-
tial is under contract, jobs are at stake and it is difficult to stop.

     Efforts are made to redirect and make projects more effective if
it is determined that the original thrust is not as useful as it might
have been.  However, even though one finds that certain undertakings
could be judged to have been mistakes from a Federal point of view, it
may not be so.  In the process, we train people; we improve the general
knowledge base in fossil fuels; and, of course, we learn considerable
about the behavior of components, materials, and chemistry of various
unit operations involved in coal conversion and utilization.

     So far as Fossil Energy's research, development and demonstration
program in ERDA is concerned, we believe all of it supports the National
Energy Plan with respect to improving the prospects and potential for
increased use of coal, and making available as reserves, large resources
of oil and gas in tar sands, bitumens and various tight formations.  We
do not see any breakthroughs coming and we do not see any results that
indicate that getting coal, shale, tight formations, etc., is going to
be inexpensive or simple.  People fail to observe that we gave up exten-
sive use of coal in many applications for good reasons.  The world used
considerable amounts of coal oil between whale oil and natural petroleum
and stopped because natural petroleum is much easier.  In addition, many
parts of the world used coal-derived gas to operate cities—some until
the late 1950's.  Again, that was slowly phased out for good reasons.
Natural gas is cleaner, easier, and less expensive.  Now, we have a
great deal of coal and we may go back to it.  But, I think we should
understand that the reason we go back is not because it is easy or inex-
pensive, but because it is the only choice we can see in front of us.

     There are many people entering the research and development arena
today who make the mistake of net looking into the history of coal use
around the world.  This, of course, has to be done carefully.  If you
look too hard at what has been done, then you fall into the trap of
beginning to believe that there are no new ideas left to be found.  That
is not true.  But, if one approaches R&D without looking at what has
been done, then there is a great deal of time and effort wasted in re-
inventing the wheel.  Coal is a very difficult material—every piece is
different, even inches apart in the same seam.  It has many constituents,
including potential toxic substances which have not yet been regulated.
Some products of coal are carcinogenic.  The manufacture of coal-derived
products has to be done with appropriate regard for the health and safety
aspects of employees.  While one can refine coal-derived liquids into
specification-grade products that we are accustomed to dealing with,
there is movement now to examine the impact of petroleum-derived liquids
on health and safety.  Regulations that would come from there would, of
course, apply to coal liquids.

-------
338                    CLEAN COMBUSTION OF COAL
     Today, one can purchase a license to operate coal liquefaction and
gasification plants.  They are very expensive and not competitive with
alternative fuels.  Atmospheric fluid-bed combustion appears to be
coming along well.  You probably will be able to buy one with commercial
guaranties in the not-too-distant future, especially at the small indus-
trial scale.  Advanced power systems are also under development.  Higher
fuel costs will justify more complex technical systems to increase
efficiency.  The ideas involved really are not new, but the technology
involved will require materials which do not now exist.

     MHD is the Fossil Energy Program's long shot.  It is an interesting
approach for several reasons.  One is that it offers the highest ultimate
conversion efficiency when examined in terms of energy in coal to elec-
tricity off the busbar.  The other reason has nothing to do with genera-
tion of electricity.  To make MHD concepts work on coal, you must have
a. combustor up front that delivers products of combustion at the order
of U500°F.  The present system seems to be favoring a one- or two-stage
slagging combustor.  If such combustors are actually made to function
reliably, it will have a great deal of fallout in the development of
coal gasifiers.  MHD actually requires more severe service than coal
gasification schemes under development.  The MHD people are really out
at the frontier of all the technology they are concerned with.  They
have had to think in terms of revolution in their concepts, rather than
evolution.  Even if only partial success is achieved in the MHD objective,
the fallout will be substantial and beneficial to many applications.  In
addition, one has to inject and recover seed material.  The recovery must
take place at high temperature and, thus, extremely difficult hot gas
cleanup conditions have to be dealt with.  Again, if this sort of problem
is solved in MHD, then it undoubtedly will have application to the other
combined cycle operations involving cleanup of coal-derived gases.  The
ultimate configuration for coal-fired MHD that provides the maximum
potential efficiency would be a three-cycle type of plant; that is, an
MHD converter, some kind of gas turbine and a steam plant.  This is a
complex engineering system involving extraordinary operating conditions
and being able to keep all parts operating at the same time for reasonable
periods of time.  This probably is one of the greatest engineering chal-
lenges facing fossil energy R&D.

     We are just beginning to look at three liability and operating
problems of such systems as advanced combined cycles.  The Petroleum and
Natural Gas Program has several thrusts.  In the Enhanced Oil Recovery
Program the objective is to be able to economically extract very large
amounts of oil, in place, left behind after primary recovery has been
completed.  In the gas area, the principal thrust in either the Eastern
or Western formations is to develop technology to economically exploit
the large amounts of gas available in tight formations.  What we mean
here is a conventional well will not produce at a rate sufficient to
make very useful recovery or investment possible.  Fracturing or other
techniques may raise production rates to where exploitation becomes
attractive.  In the gas area, we are essentially trying to convert a
resource to a reserve.  In the Eastern Devonian fields, there are con-
siderable numbers of producing wells.  These are usually in areas with
extensive natural fracture patterns and, thus, have sufficiently high
production rates.  Another interesting characteristic of wells in
Devonian fields is that they produce for very long periods of time,

-------
                               COAL R&D                             339
albeit at a lower rate.  Many holes have been drilled in Devonian fields--
in tens of thousands.  Most of them have been declared dry holes.  I think
you are aware that, generally, if a hole is not a producer, it is written
off as drilling expense.  If one produces from a hole, then the hole
must be capitalized.  Many times it is not useful to produce from low-
production rate holes because the investment recovery is too slow.  It
is conceivable that one could increase production in Devonian shales in
a significant way by just not requiring capitalization of low producers.

     Within Fossil Energy, we are estimating that gas from tight forma-
tions may be less expensive than gas from coal.  However, until enough
work is done, this will not be well understood.

     Synthetic pipeline quality gas from coal will be the order of $3 or
more per million Btu.  Our lowest estimates are for processes that are
in the development stage rather than for processes that are in the stage
of commercial availability.  Some of this low-potential cost of the more
advanced processes may be real.  On the other hand, as is often true,
development of economics are attractive early in the development process
because one has not yet fully developed their technology and come to
grips with all the problems.  ¥e are estimating low-Btu gas to be the
order of $2.30 a million Btu or thereabouts with technology that is
currently available.

     The major uncertainty is in the environmental area, but I do not
think that this is a serious problem.  It is mainly uncertainty because
such units have not been operated under modern standards.  There does
not appear to be any unusual process requirements for cleanup of low-Btu
gas processes.  ¥e have eight industrial or commercial low-Btu demonstra-
tions under contract at the present time for various applications over a
range of sizes using a variety of coals.  It will almost all be operating
in 1980.  Assuming reasonable success, these ought to demonstrate the
capability of this technology for meeting certain potential demands.  If
natural gas prices go in the direction people seem to think they are
going in, these industrial low-Btu applications ought to be commercially
attractive.  Also, one can operate a low-Btu gasifier system to meet very
tight environmental standards—tighter than is possible in a direct
combustion application.

     Liquefaction of coal seems to indicate a product at the order of
$30 a barrel; shale liquids, the order of $20 a barrel.  While these
numbers sound high compared to world price of oil, they are not high in
a functional way.  Gasoline at 75<£ a gallon without taxes is really
pretty interesting compared to walking.  Of course, the economy is not
adjusted to that sort of cost and people are not accustomed to thinking
about it.  People often ask why we have both a shale liquids program and
a coal liquids program.  They are different raw materials and produce
products of different characteristics.  Coal liquids are very aromatic.
Shale liquids are not nearly so.  The route to gasoline is probably
through coal.  The route to jet fuels and other distillates is probably
through shale.  One can produce either product from either source, but
the economics are more attractive with a particular feedstock/end-product
combination such as shale oil to distillate rather than shale oil to
gasoline.

-------
340                    CLEAN COMBUSTION OF COAL
     In addition, there is the Fischer-Tropsch process currently in
operation in South Africa.  If operated with high temperature entrained
gasifiers, it is potentially a very clean process from an environmental
point of view and, with development of more selective catalysts, capable
of producing a wide range of hydrocarbon liquids.

     At the present time, this looks to be the most expensive process.
Again, perhaps we know most about it and, therefore, we have the most
realistic number.

QUESTION:   WHAT ABOUT THE ENERGY EFFICIENCY OF THE FISCHER-TROPSCH
            PROCESS?

ANSWER:     It all depends on how you look at the process.   Even in
Fischer-Tropsch, if you try to make all one product (such as gasoline),
the efficiency is probably hd percent.  However, coal does not really
want to behave that way.  Neither does shale.  Most processes involving
coal or shale are somewhat similar to processes for finding oil:  You
put something in, you heat it up and a range of products come out.   If
one can arrange a process that produces a useful product mix, even the
Fischer-Tropsch process may be as much as 60 percent efficient.

     Other liquefaction processes may be 70 percent efficient.  While
it may be easier in the early stages of building synthetic fuels industry
to have a single-product plant, it is probably likely that large-scale
commercialization would have multi-product plants.

QUESTION:   WHAT ABOUT FLASH HYDROPYROLYSIS?

ANSWER:     We are moving ahead rapidly with very rapid hydropyrolysis.
The work on the West Coast with Rocketdyne is very encouraging—maybe
because we have so little data.  At this time, this process route seems
to offer very high grade hydrocarbon products with potentially very
large capacity per unit of investment in steel.

     In general, as we have learned more about any process, it is sur-
prising how the economics of all converge on something fairly high and
not too far from existing technology.  Look at a coal liquefaction or
gasification plant of nearly any kind—the order of 30 percent to UO per-
cent of the product price is the coal feedstock.  About 50 percent or so
is the cost of money.  The rest is other costs.  R&D can make its biggest
impact on decreasing the amount of capital required.  Actually,  if we
could figure out a way to use the interest rate with our R&D funds, we
would have a very high benefit/cost ratio.  The capital charges become
very small if one could go to a forty-year write-off and 3 percent money.

     However, we have been doing many process evaluations trying to
understand where the capital and operating costs are.  The principal
objective is to identify targets of opportunity for Federal research
dollars.

CONCLUDING REMARKS:

     I have tended to indicate the problems associated with coal conver-
sion are very difficult.  They are.  It is not going to be inexpensive,

-------
                                 COAL  R&D
                                                                    341
tov ^straightforward.  However, research and development will help
to make the plants more reliable, more easily operable, more environmen-
tally acceptable and safer from an occupational health and safety point

-------
342                     CLEAN COMBUSTION OF COAL

-------
                                                                       343
                                TECHNICAL REPORT DATA
                         (Please read Instructions on the reverse before completing)
1. REPORT NO
 EPA-600/7-78-073
                           2.
                                                      3. RECIPIENT'S ACCESSION NO.
, TITLE AND SUBT.TLE
                proceedings rf ^ Engineering

Foundation Conference on Clean Combustion of Coal
                                                     5. REPORT DATE
                                                     April 1978
                                                      6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Victor S. Engleman, Conference Chairman
                                                     8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Science Applications, Inc.
L200 Prospect Street
La Jolla, California 92038
                                                     10. PROGRAM ELEMENT NO.
                                                     EHE624A
                                                      11. CONTRACT/GRANT NO.
                                                      EPA Purchase Order
                                                       DA-7-03662B	
 12. SPONSORING AGENCY NAME AND ADDRESS
 EPA, Office of Research and Development*
 Industrial Environmental Research Laboratory
 Research Triangle Park, NC 27711
                                                      13. TYPE OF REPORT AND PERIOD CC
                                                      Proceedings; 5/77-2/78
                                                                           COVERED
                                                     14. SPONSORING AGENCY CODE
                                                       EPA/600/13
 15.SUPPLEMENTARY NOTES IERL-RTP project officer is G. B. Martin (MD-65, 919/541-2235).
 (*)Cosponsors! project officers are A. Macek of ERDA and A. W. Deurbrouck of USBM
 (both now part of DOE).                                            	
 16. ABSTRACT
 The proceedings document the 27 presentations made during the Engineering Founda-
 tion Conference on Clean Combustion of Coal, at Rindge, NH  August 1-5, 1977.
 Sponsored by the Environmental Protection Agency,  the Energy Research and Devel-
 opment Administration, and the Bureau of Mines (the last two now part of the Depart-
 ment of Energy), the Conference  dealt with the technical, economic, environmental,
 and policy aspects  of clean combustion of coal. The five Conference sessions dealt
 with problem definition, precombustion processes, combustion processes, postcom-
 bustion processes, and future prospects. The Conference was intended to provide an
 assessment of the status and trends of clean combustion of coal.
17.
                             KEY WORDS AND DOCUMENT ANALYSIS
                DESCRIPTORS
                                          b.IDENTIFIERS/OPEN ENDED TERMS
                                                                  c.  COSATI Field/Group
 Pollution
 Combustion
 Coal
 Coal Preparation
 Coal Gasification
 Flue Gases
                     Combustion Control
                     Combustion Effi-
                       ciency
Pollution Control
Stationary Sources
Flue Gas Cleaning
Combustion Modification
13 B
2 IB
2 ID
081
13 H
 3. DISTRIBUTION STATEMENT

 Unlimited
                                          19. SECURITY CLASS (ThisReport)
                                          Unclassified
                        21. NO. OF PAGES
                             347
                                          20. SECURITY CLASS (Thispage)
                                          Unclassified
                                                                  22. PRICE
EPA Form 2220-1 (9-73)

-------