xvEPA
         United States
         Environmental Protection
         Agency
          Industrial Environmental Research
          Laboratory
          Research Triangle Park NC 27711
EPA-600/7-79-170
July 1979
Control Technologies for
Particulate and Tar
Emissions from Coal
Converters

Interagency
Energy/Environment
R&D Program Report

-------
                  RESEARCH REPORTING SERIES


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

    1. Environmental Health Effects Research

    2. Environmental Protection Technology

    3. Ecological Research

    4. Environmental Monitoring

    5. Socioeconomic Environmental Studies

    6. Scientific and Technical Assessment Reports  (STAR)

    7. Interagency Energy-Environment Research and Development

    8. "Special" Reports

    9. Miscellaneous Reports

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

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

-------
                                            EPA-600/7-79-170

                                                     July 1979
   Control  Technologies  for Particulate
and  Tar  Emissions  from  Coal  Converters
                              by

                   C. Chen, C. Koralek, and L Breitstein

                       Dynalectron Corporation
                       Applied Research Division
                        6410 Rockledge Drive
                       Bethesda, Maryland 20034
                       Contract No. 68-02-2601
                     Program Element No. EHE623A
                  EPA Project Officer: Robert A. McAllister

                 Industrial Environmental Research Laboratory
                  Office of Energy, Minerals, and Industry
                    Research Triangle Park, NC 27711
                           Prepared for

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

-------
                                  ABSTRACT


          Raw product gases from coal converters generally contain participates
and tars which must be controlled to a level compatible with environmental
regulations.  Control technologies for removing particulates and tars from
product gases were identified and evaluated.

          Particulate and tar emissions in raw product gases from several
types of coal gasifiers were characterized in terms of their total quantities,
chemical composition, and size distribution.  The emissions data were organized
according to generic gasifier type, with fixed, fluid, and entrained-bed gasi-
fiers being considered.  The design and operating features of each identified
control technology were described, with emphasis on characterizing collection
efficiencies as a function of particle size and other parameters.  These data
were also organized into generic categories such as cyclones, wet scrubbers,
electrostatic precipitators, fabric filters and granular bed filters.

          The applicability of each of the identified control technologies
was assessed with respect to the generic gasifier types for combined cycles
and gas-fired boilers.  These assessments were based on existing and pro-
posed environmental regulations and process requirements for product gas
purity.  The fate of the particulate and tar emissions was assessed in the
purified product gases, liquid effluents, and solid wastes or sludges.  Gaps
in the data base required for these assessments were identified.
                                    ii

-------
                                  Contents
Abstract	  ii
Figures  	   v
Tables	.	vii
Acknowledgement  	   x

1.    INTRODUCTION	   1
      1  .1  Background	   1
      1.2  Methodology 	   2

2.    CONCLUSIONS AND RECOMMENDATIONS  	   3
      2.1  Conclusions 	   3
      2.2  Data Gaps and R&D Requirements  	   k
          2.2.1  Characterization of Particulate and
                 Tar Emissions	   5
          2.2.2  Performance Characteristics of Control
                 Devices 	   5
          2.2.3  End-Use Requirements  	   7
          2.2.k  Fate of Pollutants	   7

3.    CHARACTERISTICS OF PARTICULATE AND TAR EMISSIONS  	   8
      3.1  Fixed-Bed Gasifiers 	   8
          3.1.1  The Lurgi Process	11
          3.1.2  MERC Process	13
      3.2  Fluidized-Bed Gasifiers 	  13
          3.2.1  Winkler Process	16
          3.2.2  C0 Acceptor Process	16
      3.3  Entrained Bed Gasifiers	23
          3.3-1  Koppers Totzek Process  	  23
          3.3.2  Bi-Gas Process	27
      3.A  Summary of Data	27

A.    ALTERNATE CONTROL TECHNOLOGIES  	  32
      k.]  Cyclones	32
          A.I.I  Conventional Cyclones 	  32
          A.1.2  Multiclones	33
          4.1.3  Rotary Flow  Cyclones	33
      k.2  Wet Scrubbers	36
          k.2.]  Spray Towers	38
          k.2.2  Cyclonic Spray Scrubbers  	  38
          k,2.3  Mechanical Scrubbers	38

-------
                                 Contents (cont'd)


          4.2.4  Packed Bed Scrubbers	   39
          4.2.5  Orifice Scrubbers   	   39
          4.2.6  Impingement Scrubbers   	   39
          4.2.7  Venturi Scrubbers   	   39
          4.2.8  Other Wet Scrubbers	   !tO
      4.3  Electrostatic Preci pi tators   	   /iO
      4.4  Surface Filters  	   k^
          4.4.1  Fabric Medium Filters	   /!
            4.4.1.1  Baghouses	   45
            4.4.1.2  Ceramic Fabric Fi1ters  	   k$
          4.4.2  Porous Medium Filters   	   4?
            4.4.2.1  Porous Ceramic Filters  	   kj
            4.4.2.2  Porous Metal Filters 	   48
      4.5  Granular Bed Filters	   48
          4.5.1  Fixed Bed Filter	   48
            4.5.1.1  Ducon Filter	   i9
            4.5.1.2  Rexnord Filter  	   49
          4.5.2  Intermittently Moving Bed Filter 	   53
            4.5.2.1  Panel Bed Filter  	   53
          4.5.3  Moving Bed Filter	   53
            4.5.3.1  Combustion Power  Company Filter  	   53
      4.6  Other Collection Devices 	   56
          4.6.1  A.P.T. Dry Scrubber	   56
          4.6.2  Molten Salt Scrubber	   56
          4.6.3  Electrofluidized Bed  Collector 	   59
          4.6.4  Charged Filter	   59

5.    APPLICABILITY OF CONTROL TECHNOLOGIES  	   62
      5.1  End Uses for Product Gases	   62
      5.2  Environmental Regulations and Process Requirements 	   62
          5.2.1  Boiler Fuel  	   62
          5.2.2  Gas Turbines	   63
      5-3  Performance Evaluation 	   63

6.    FATE OF POLLUTANTS	   78
      6.1  Fixed Bed Gasifiers	   81
      6.2  Fluid Bed Gasifiers	   85
      6.3  Entrained Bed Gasifiers	   85

References	   95
                                      iv

-------
                                FIGURES


Number                                                                      Page

  1    Particle Size Distributions  for Fixed Bed Gasifiers  .........   15

  2   Particle Size Distributions  for Fluid Bed Gasifiers  .........   22

  3   Particle Size Distributions  for Entrained Bed  Gasifiers  .......   28

  k   Effect of Temperature on Conventional Cyclone
        Efficiency  ............................   3k

  S   The Effects  of High  Temperature and Pressure on the
        Collection of a High Efficiency  Conventional  Cyclone  .......   3*
  6   Fractional  Removal  Efficiency Curves for Multiclones  ........  35

  7   Fractional  Removal  Efficiency Curves for Aerodyne
        Series "SV" Dust  Collector  ....................  35

  8   Fractional  Removal  Efficiency Curve for Donaldson
        Tan jet Cyclone   ..........................  37

  9   Typical  Fractional  Removal  Efficiency of Venturi
        Scrubbers  .............................  ^'

 10   Removal  Efficiency  As A Function of Particle Size for
        a Typical Electrostatic Precipitator Installation  .........  ^3
 11    Typical  Fractional  Removal  Efficiency of Fabric Filters  .......  k6

 12   Schematic Illustration of Ducon Filter ................  50

 13   Rexnord  Gravel  Bed  Filter  ......................  51

 H   Typical  Collection  Efficiencies for Rexnord Filter   .........  52

 15   Cross-Section Through a Panel  Bed Filter ...............  5^

 16   Combustion Power Company's  Moving Bed Filter .............  55
                                     v

-------
                              FIGURES (cont'd)
Number                                                                       Page

 17   Collection Efficiency  of Combustion Power Company's
         Moving Bed Filter	    57

 18   Comparison of Experimental  and Theoretical  Particle
         Collection Characteristics  of the A.P.T.  Dry
         Scrubber	    58

 19   Fractional Removal  Efficiency  of Apitron Charged Filter  	    61

 20   Graphical  Procedure for  Estimating Overall  Collection
         Efficiency for Particles  up to 6 Micrometers
         in  Diameter	    68

 21    Absorption of Napthalene By  Particulates  	    9*
                                     vi

-------
                                   TABLES
Number                                                                     Page

  1    Status of U.S.  and Foreign Low and Medium-Btu
        Fixed-Bed Gasification Systems  	    9

  2   Operating and Raw Gas Stream Characteristics  -
        Fixed Bed Gasifiers  	10

  3   Composition of Lurgi  Raw Product  Gas	12

  4   Composition of MERC Raw Product Gas	]k

  5   Status of U.S.  and Foreign Low and Medium-Btu
         Fluid Bed Gasification Systems	17

  6   Operating and Raw Gas Stream Characteristics  -
         Fluid Bed Gasifiers	13

  7   Composition of Winkler Raw Product Gas	19

  8   Percent Composition of C02 Acceptor Raw Product
         and Lockhopper Vent Gases	21

  9   Status of U.S.  and Foreign Low and Medium-Btu
         Entrained Bed Gasification Systems  	   24

 10   Operating and Raw Product Gas Stream Characteristics -
         Entrained Bed Gasifiers   	   25

 11    Composition of Koppers Totzek Raw Product Gas   	   26

 12   Composition of Bi-Gas Product Gas	29

 13   Summarized Particulate and Tar Loadings and Particle
         Size Distributions	30

 14   Particulate Removal Efficiencies Required for
         Different End Uses	65
                                      vii

-------
                              TABLES (cont'd)
Number                                                                    Page

 15   Collection Efficiency of High Efficiency Cyclone
         For Particulates  from MERC Fixed Bed Gasifier
         Employing Illinois #6 Coal  ..................    67

 16   Overall  Participate  Removal  Efficiencies of Generic
         Control  Technologies  for  Typical  Gasifier Outputs  .......    69

 17   The Applicability  of a Conventional  Cyclone for
         Gasification  Processes  ....................    70

 18   The Applicability  of a Rotary Flow Cyclone  for
         Gasification  Processes  ....................    71

 19   The Applicability  of a Venturi  Scrubber for
         Gasification  Processes  ....................    72

 20   The Applicability  of an  Electrostatic  Precipitator
         for Gasification  Processes  ..................    73

 21    The Applicability  of a Fabric Filter for
         Gasification  Processes  ....................    7^
 22   The Applicability of a Granular Bed  Filter  for
        Gasification  Processes   ....................   75

 23   Summary of Applicability Assessments ...............   77

 2k   Trace  Elements - Estimated  Volatility   ..............   79

 25   Organic Composition of Quench  Liquor -  Fixed Bed
        Gasifier   ...........................   82

 26   Trace  Elements in Grab Samples by  SSMS  ..............   83

 27   Levels of Trace  Elements in  Liquids  from  the
        Quench Liquor and By-Product Tar  Versus  Water
        Quality Standards .......................   34

 23   Organic Composition of Cyclone Dust  Extract - Fixed
        Bed Gaslfiers .........................   86

 29   Characteristics  of Particulates -  Fixed 3ed Gasifier  .......   86
                                     viii

-------
                                TABLES  (cont'd)


Number                                                                   Page

 30   Laboratory Synthane Gasifier Trace  Element
         Analysis of All  Streams	8?

 31   Mass Spectrometric  Analyses  of Benzene-Soluble
         Tar from Synthane Gasifier		89

 32   Scrubber Water Effluent  Inorganic Species  -
         Entrained Bed Test Gasifier	-	91

 33   Scrubber Water Effluent  Elemental Composition -
         Entrained Bed Test Gasifier   	92

 3*t   Scrubber Water Flash Gas Composition -  Entrained
         Bed Test Gasifier	93
                                       ix

-------
                                ACKNOWLEDGEMENT
          The authors wish to express their gratitude to Or. Clarence A.  Johnson
and Mr. Daniel M. Kennedy for their many helpful suggestions throughout the course
of this work.  Thanks are also due Mssrs. Chester A. Vogel  and William J. Rhodes
of the EPA's Industrial  Environmental Research Laboratory (lERL-RTP)  for  their
continuing interest.

-------
                                  SECTION I

                                 INTRODUCTION
1.1       BACKGROUND

          The energy supply problems of the United States and most of the major
industrialized nations are well-known and documented.  Current projections indicate
that the world demand for petroleum and natural gas will  exceed supply sometimes
during the 1980's.  One obvious approach to increasing domestic fuel  supplies, and
consequently to reducing demand for imported gas and oil, is to utilize the vast
coal resources of the United States to produce synthetic oil and gas.

          In recent years, the electric utility and industrial sectors of the
economy together accounted for about 55 percent of the energy consumption in the
United States.  Natural gas and petroleum supplied about 80 percent of the indus-
trial energy consumption and 30 percent of the utility consumption.  The use of
coal-derived fuels to replace natural gas and petroleum  in these areas could have
important economic benefits to the United States, in addition to reducing the
nation's dependence on foreign, unreliable sources of energy.  Such coal-derived
products might be employed in a wide variety of end uses, such as industrial process
heat, industrial and utility boilers, gas turbines, and  reducing or synthesis gas
for various industries.

          In the case of coal-derived product gases from coal gasifiers, each par-
ticular end use of the gases would have different environmental regulations and/or
process requirements governing the allowable particulate and tar levels  in the
product gases.  Thus, the use of coal-derived product gases to replace natural gas
and oil on a large scale will require adequate control technology to  remove tars
and particulates from the product gases to levels compatible with the various
possible end uses.  The overall objective of this study was, therefore,  to assess
the applicability of alternate control technologies both commercially available and
under development for the removal of particulates and tars  from coal-converter
product gases.

          The study described herein was performed under EPA Contract Number 68-02-
2601 for the Fuels Process Branch of the Environmental Assessment and Control
Division of the  Industrial Environmental Research Laboratory at Research Triangle
Park (1ERL-RTP).

-------
1.2       METHODOLOGY

          The first step in carrying out these control technology evaluations in-
volved the identification and collection of pertinent sources of information.  Com-
puterized literature searches covering the Chemical Abstracts, Engineering Index,
Pollution Abstracts, U.S. and foreign patents, government publications, and
numerous journals were made to identify sources of information.  These computerized
searches were complemented by thorough library and patent searches.  In addition,
other EPA contractors, process developers, and equipment vendors were contacted
for relevant data.

          After reviewing the identified sources of information, appropriate data
therefrom  were employed to carry out several sub-tasks, as summarized briefly
be1ow.

          The initial sub-task involved the characterization of the raw product
gases from fixed, fluid, and entrained-bed gasifiers, with emphasis on the particu-
late and tar contents of the gases.  Gasifier emission data were characterized in
terms Of total particulate and tar  loads, the chemical composition of both the
particulates and tars, particle size distribution, and other pertinent exhaust gas
parameters.   Representative emission factors for the above generic gasifier classes
were thereby developed.  A second sub-task involved the identification and descrip-
tion of the performance of alternate control technologies for particulates and tars,
with emphasis on characterizing the particulate removal efficiency of each control
device.

          The results of the above two sub-tasks were then employed to assess the
applicability of each control technology to the various gasifiers under considera-
tion.  Applicability assessments were made for two end uses representing low-to-
moderate and high degrees of required particulate removal.  Both applicable environ-
mental regulations and process requirements were considered in these assessments.

          Another sub-task involved assessments of the ultimate fates of the par-
ticulates and tars with respect to their presence in gaseous, liquid, and solid
discharge streams.  All available data from experimental and commercial installa-
tions were employed as the basis for these assessments.

          Finally, the data available to carry out the above sub-tasks were reviewed
and examined to identify gaps and deficiencies in the available data base and the
resulting R  D requirements.

-------
                                  SECTION 2

                       CONCLUSIONS AND RECOMMENDATIONS


2.1       CONCLUSIONS

          Applicability assessments were made for various combinations of particu-
late control  devices and gasifier-end use pairs.   These assessments were made for
the three major generic classes of coal gasifiers, including fixed-bed, fluid-bed,
and entrained-bed, and for two end uses for the product gases.   The first end
use consists of combined-cycle power generation with the fuel  gas combusted in a
turbine, which represents a relatively high product gas cleanup requirement.   The
second end use consists of boiler fuel, which represents a low-to-moderate cleanup
requi rement.

          Detailed applicability assessments were made for the following six
generic classes of control devices:  (1) conventional cyclones, (2) rotary flow
cyclones, (3) venturi (wet) scrubbers, (4) fabric filters, (5)  electrostatic
precipitators (ESP's), and (6) granular bed filters.

          The results of these assessments are summarized in Table 23 of Section
5.  Due to the uncertainties in the emission characteristics from the different
types of gasifiers, these results are presented in terms of best-case, worst-case,
and average (or typical)- case analyses.  The worst-case condition represents the
estimated upper limit of particulate load, with a relatively high percentage of
small  particles, which are difficult to remove.  The best-case condition represents
the estimated lower limit of particulate load with a relatively low percentage of
small  particles.  All available data on the characteristics of gasifier emissions
were considered in estimating these upper and lower bounds.

          In the case of the combined-cycle end use, fabric filters, venturi
scrubbers, and rotary flow cyclones were found to be applicable for at least some
gasifiers.  The other control devices listed above would not be applicable for com-
bined cycles.  Fabric filters can probably be used for all entrained-bed and fluid-
bed gasifiers that do not produce tars.  Fabric filters would probably not be
applicable to fixed-bed gasifiers due to the need for a quenching operation to
condense and remove tars, and the likely presence of droplets or "sticky" particles
in the cooled gases.  Venturi scrubbers are applicable to all gasification types
except the worst-case fluid-bed.  If coupled with an upstream cyclone, the venturi
would also be applicable to this case.

          Since the particulate removal requirements for boiler fuel are not as
restrictive as for combined cycles, the number of control devices applicable to

-------
 boiler  fuel  gases  is  increased  considerably  as  compared  to combined cycles.
 Venturi  scrubbers  are  applicable  to  all  gasifier  types.   Electrostatic precipitators
 are applicable  to  the  best  and  average-case  fluid-bed  and entrained-bed gasifiers.
 If coupled with  an upstream cyclone,  ESP's would  be applicable  to all fluid and
 entrained-bed gasifier cases.   Due to the high  carbon  content and low resistivity
 of particulates  from fixed-bed  gasifiers, ESP's are not  applicable to the case of
 fixed-bed gasifiers.   Results for the other  types of control devices under con-
 sideration are also presented in Table 23-

          Sufficient data to serve as a  basis for applicability assessments were
 not available for  several relatively  advanced control  devices still in the develop-
 mental  stage.  The need for additional data  for these  control devices is discussed
 in Section 2.2.

          Data on  the  fate  of the particulates  and tars  emitted in the product
 gases,  in terms  of their ultimate presence and  concentrations in solid, liquid, and
 gaseous  discharge  streams,  are  very  preliminary and limited  for all gasifier types.
 The conclusions  presented below should,  therefore, be  considered tentative until
 confirmed by additional data.   The distribution of these  particulates and tars  in
 the various discharge streams is dictated by both the  removal technology and the
 physical and chemical characteristics  of the contaminants.

          In the case of fixed-bed gasifiers, the quench  liquor employed to condense
 and remove the tars contains high concentrations of phenolic compounds.  These com-
 pounds,  together with ammonia and dissolved acid gases, must be removed from the
 quench  liquor.  As much water as possible should be recycled to minimize the dis-
 charge of liquid effluents.   Mercury  tends to concentrate in the tar, while most
 other volatile elements tend to become concentrated on the particulates.  Selenium
 concentrations in  the quench liquor  are very high.

          In the case of fluid-bed gasifiers, most of  the available data on the
 fates of the various contaminants were obtained with the  Synthane unit, which also
 produces tars.  Since most  other fluid bed gasifiers do not  produce tars, these
 data may not be  representative of this generic gasifier type.  The available data
 indicate that many of the trace elements tend to concentrate in the particulates
 and char.  Some of the more volatile  elements such as  As, Pb, and Hg are also
 found in potentially harmful concentrations  in the tar.

          In the case of entrained-bed gasifiers, organics tend to concentrate on
 the particulate matter, as  opposed to  scrubber water.   Volatile elements such as
 Hg, Se, and As are not absorbed in the scrubber water.   Tars are not produced by
 entrained-bed gasifiers and, therefore, do not present  a disposal problem.

 2.2       DATA GAPS AND R & D REQUIREMENTS

          As discussed below, the data required for these analyses were, in many
cases, quite limited.  Thus, the results and conclusions  summarized in Section 2.1
and presented in detail in  Section 5  should generally  be  considered as preliminary
and tentative until additional,  more complete data become available.   The deficien-
cies  in the  available data base and the associated R & D  requirements are discussed
below.
                                       k

-------
2.2.1     Characterization of Particulate and Tar Emissions

          The greatest deficiency in the data base employed for the applicability
assessments presented in Section 5 was generally found to involve the characteriza-
tion of the particulate and tar emissions from the various types of gasifiers under
consideration.  As discussed in Section 2.1, the emission characteristics were,
therefore, presented in terms of best-case, worst-case and average (or typical)-
case conditions.  The conditions under which most of the reported data were
obtained are usually unspecified.  In addition^ the precise influence of many
operating conditions (e.g., coal feed rate, steam/coal ratio,  steam/air (0_)  ratio,
pressure, temperature, etc.) on the emission characteristics is unknown,  thus,
more complete data are required for all gasifier types to develop "typical"
emission characteristics with a greater degree of accuracy.

          The particulate removal efficiency of most control devices is especially
sensitive to the particle size distribution.  Such data were generally found  to
be very scarce and incomplete.  More complete data over a broad, specified range of
gasifier operating conditions are needed.   In the case of fluid and entrained-bed
gasifiers, particle size data were not available below approximately 35 and 20 micro-
meters, respectively.  Extrapolation of the existing data for large size particles
down to the small size particle range was, therefore, required for these two types
of gasifiers.  Since the concentrations of the smaller size particles in the product
gases are of particular interest, due to the fact that the larger particles are
easily removed while the particles below approximately 5 micrometers are much more
difficult to remove, future R S D programs should concentrate on the collection of
particle size data down to the sub-micron size range.

          Particle size distribution measurements are usually based on either aero-
dynamic or optical properties of the particles.  Measurements in the same gas
stream by these two different techniques often yield  inconsistent results.  Parti-
cle sizes are especially difficult to measure at high temperature  and pressure
(HTHP) conditions.  The collection of  reliable particle size distribution data for
coal-gasifier product gases will require the development of improved methods and
instrumentation suitable for HTHP conditions.

          Additional data are also needed to accurately estimate particulate and
tar loadings from the various types of gasifiers, particle and tar compositions, and
other pertinent properties such as particle  resistivity.   It should be noted that
complete data sets were not available for any of the  gasifier types.  For example,
the particle size distribution might be available for a specific type of gasifier
at a given or unspecified set of conditions, whereas  particulate loadings and
compositions might be available for another  type of gasifier within the same
generic class, but at a different set of conditions.  There is, then, a need for
R & D programs to provide complete data for  all of the above parameters at the
same, specified gasifier operating conditions.

2.2.2     Performance Characteristics of  Control Devices
          The performance characteristics of the commercial types of control devices
are generally well known and documented.  There are deficiencies  in the data base
for several of the more advanced and recently developed types of  control devices.

-------
  which  are  described  in  Section k.   The adequacy of  the available data for the
 various types of  control  devices  under consideration  in this  study are discussed
 below.

          Adequate data  are available to describe  the particle collection efficiency
 as  a  function of  size  (i.e.,  particle diameter) for  conventional cyclones, venturi
 and other commercial wet  scrubbers,  electrostatic  precipitators  (ESP's) operating
 at  low  to moderate temperatures,  and conventional  baghouse  filters.

          The vendor-supplied collection efficiency  data presented in Section A.1.3
 for the Aerodyne  rotary  flow  cyclone are very encouraging.  Additional data are
 required to confirm these  results, however, since  tests by  Westinghouse Corporation
 showed  significantly  lower collection efficiencies.  Since  data obtained at high
 temperatures are  very  limited, additional data are especially needed at these
 cond i t ions.

          The availability of data for several advanced types of wet scrubbers
 under development to  improve  the  collection of fine, sub-micron particles is very
 limited.  These newer  types of scrubbers include foam, steam-assisted, and electri-
 cally augmented devices, which are discussed  in Section A.2.8.   In addition to1 the
 need  for more collection efficiency  data, more performance  testing is needed to
 assess  their operational  reliability.

          Electrostatic  precipitators have undergone only limited testing at high
 temperatures (up  to nearly 1100C) and nressures Cup to 52  MPa).  While these
 limited data are  encouraning, more data are required to determine the collection
 efficiencies and  operational  reliability under these conditions.  Since ESP's have
 been  thoroughly tested at  less severe conditions and established as a commercial
 technology for decades, additional testing at HTHP conditions should be a high
 priori ty R & D i tern.

          Conventional baghouse filters are limited  to operation at temperatures
 below 300C.  Ceramic  fabrics are being developed  to extend the range of possible
 operating temperatures.  The  availability of data  on ceramic fabric filters is
 very  limited, with most of the data  being collected at ambient conditions.  More
 data are, therefore, needed to determine the collection efficiency and operability
 of  ceramic fabric filters.

          Porous  ceramic filters appear to be very promising for highly efficient
 collection of particles down  to the  sub-micron size  range at high temperatures.
 While preliminary data at high temperatures are encouraging, additional testing
 with  larger-scale control devices are required for confirmation.  This type of
 filter  should,  then,  be given serious consideration as a high priority R & D item.

          Granular bed filters (GBF)  generally appear to be promising for HTHP
operation.   Most GBF systems are still  in the development stage.  The GBF
 developed by Combustion Power Company is the most advanced of this generic class of
control  device.    It  is commercially available for operation below 430C and at
ambient  pressures.  No data are available at HTHP conditions.   More data are needed
 for the  Combustion Power Company and other GBF systems to define their collection

                                       6

-------
efficiencies and operational  reliability.
          In addition to the control  devices discussed above,  several  novel  devices
are in the early stages of development,  with only very limited preliminary data
being available.  Such devices,  discussed in Section k.6,  include the  A.P.T.  dry
scrubber, molten salt scrubber,  electrof1uidized bed, and  the  Apitron  charged
filter.  The latter appears to have especially high collection efficiencies down
to sub-micron size particles, but operation   is restricted to  the same temperature
limits as a conventional baghouse filter.

2.2.3     End-Use Requirements

          The applicability assessments  involving the combustion of product gases
in gas turbines are very sensitive to the assumed limitation on the particulate con-
centration in the combusted gases.  More data are required to  accurately determine
the tolerance of gas turbines to particulates.

2.2.k     Fate of Pollutants

          In all cases, data on the fates of the particulates  and tars, with respect
to their presence and concentrations in  solid, liquid, and gaseous discharge
streams are very preliminary and limited.  Data for entrained-bed gasifiers are
especially sparse.  As previously discussed, most of the data for fluid-bed gasifi-
ers were obtained on the Synthane unit,  which produces tars and  is, therefore,
not representative of most other fluid-bed gasifiers.  Thus, additional data are
especially needed for more representative fluid-bed gasifiers.

          Additional sampling is required to  identify and determine the concentra-
tions of contaminants  in quench water, solid wastes, tars and scrubber water under
better defined conditions.  Laboratory analyses should include trace metals and
identification of the chemical forms in  which they appear, as well as other  inor-
ganic and organic compounds.  Studies to determine the leachability of trace
elements from captured particulates and  tars  into quench and scrubber water, as
well as to ground water after ultimate disposal, would be very useful.  Volatility
studies are also needed to determine gaseous emissions that would result  from
liquid and solid discharge streams.

-------
                                    SECTION  3

                CHARACTERISTICS  OF  PARTICULATE AND  TAR  EMISSIONS


          As  an  initial  step  in the evaluation of  technologies for  the control of
 participates  and  tars  in gaseous streams  originating from  coal gasifiers, emissions
 and process data  were  obtained  for a wide variety  of gasifiers.  These data,
 gathered  from an  array of  sources,  both published  and  unpublished,  were used to
 develop "typical" product-gas characteristics for  each of  the three principal
 generic classifications  of gasifiers (i.e., fixed-bed, fluid-bed, and entrained-
 bed).  Although the availability of pertinent data was found to be  quite  limited,
 sufficient data were obtained for  this study to  permit reasonable characteriza-
 tions of  the  particulate and  tar emissions  from  these  generic types of gasifiers.

 3.1       FIXED-BED GASIFIERS

          In  fixed (gravitating) bed gasifiers,  sized  coal  is typically fed from
 above into the gasifier.   The fuel  bed, which rests on a grate, is  maintained at
 a constant depth  (typically 50  cm).  Steam and air or  oxygen travel  countercur-
 rently upward and through  the bed.   The fuel bed consists of several distinct
 regions or zones.  In  the  upper or  preheat zone, the rising hot gases both dry and
 devolatize the coal.   Below,  in the  adjacent endothermic gasification or  reduction
 zone, the water-gas reaction occurs.  The oxidation or combustion zone at the
 bottom provides the necessary heat  for gasification.   In the oxidation zone,the
 residual material  is discharged as  ash or slag.  Fixed bed-gasifiers have several
 advantages over the other  generic  types.  The heat economy of the fixed-bed gasi-
 fier Is generally excellent, mainly  because of the countercurrent flow of the gas
 and coal.  Due to the  lengthy duration of the coal's residence time, the combus-
 tible solids  content in the ash residue is minimized.  Other advantages include
 simplicity of operation and the advanced  state-of-the-art of fixed-bed gasifiers.

          A major drawback to fixed-bed gasifiers  is their difficulty in processing
 caking coals.   In addition, tars and oils tend to  form in the upper preheat zone
of the fuel  bed,which  can cause plugging of the beds.  The tars and oils also
 result in increased gas cleanup requirements, and  associated wastewater treatment
and tar/oil  separation operations.    Another disadvantage is that the coal  must be
 sized to maximize output, and to minimize fines that can plug the bed.

          Table   1    lists a number of fixed-bed  gasifiers.  Typical  process para-
meters and raw gas characteristics  for each of these processes are presented In
Table 2,     Data were gathered from a variety of  references,as noted in the table.
Characteristics include particle  loading,  particulate compos! t ion, and tar load

-------
     TABLE  1   STATUS OF U.S. AND FOREIGN LOW-AMD MEDIUM-BID FIXED-BED GASIFICATION SYSTEMSO)
                                                           Number of Gasifiers Built
Gasifer


Lurgi

We 11 man Galusha


Woodal1-Duckham/
Gas Integrals

Chapman (Wilputte)

Riley Morgan

Pressurized Wellman
Galusha (MERC)

GFERC Slagging
   Li censor/developer
Lurgi MineralotechnIk GmbH

McDowell Wellman
Engineering Co.

Woodal1-Duckham (USA) Ltd.
Willputte Corporation

Riley Stoker Corporation

ERDA


ERDA
LBGa
5
150
72
12
1
'"
-
MBGb SNGC Location
39 22 Foreign
US/Foreign
8 Foreign
US
US
US
ld - US
Scale
Commercial
Comme re i a 1
Commerci al
Commercial
Comme re i a 1
Demonstra-
tion
Demonstra-
tion
           Low-Btu Gas
          Medium-Btu Gas
          'Synthesis Gas
          Under Construction

-------
                                                                 TJWL  2
                                                            RW Mi* S1KMK OMMCTENHTItS
                                                             f IXCD BED SASfFlEB
                  CM!
                                                  Pressure
                                                                 Uwdl
                                                                fnm>(q
                                                    ng      FrtltuUt*
                                                                                           Tar
                                                                                     Tar
                                                                                                                                           J/i"*
MtttitcU*    *IQ-8Z8
tottwrinou*
coke
                                                   US)
                                                                                           10-50
                                                                                                                          1.1 *I0
                                                                                                                                         10
                    Kl typei"
                                                               0,5-S.O        C-75-*)*      t*r-lj
                                                                                     t*r i-
                                                                                     0.77t
                                                            Mh*IQ-ast    t*r oil     t*r oil
                                                                          75         5-0.19t
                                                                                                                          cwn
                                                                          (195)
                  bftualnout    (250-UOO)


                                (1000-I
                                e.rt
                                05)
                                o.to
                                (15)
                                                                                           o)1 antfttr
HI ley
 nthnclt.    $70-630
 frltunlnout
 klnq fcft.
                                                  D.IO
I-"l
D.8-1.7
tar fD-ltt

Ur all
                                                                                                                          161,
  5,4)
                   types
                                0.10-1.10    0.5-6.0
                                CT5-300)     (0.2-J.5)
                                                                              t-75-0*      30
                                        t-Sl.l
                                        H-7.
                   bit.  chtr
                                                  0.66-1.86
                                                                                            tr olI-I?
                                                                                              (Wl
                                                                                                                           $53

-------
and composition.   In cases  where data differs from one reference to another,  a
range is provided.

          The following operating parameters are expected to influence the composi-
tion and/or concentration of impurities in the raw product gas from fixed-bed
gasi fiers:

          Coal  Analysis
          Ash Analysis
          Coal  Particle Size Distribution
          Coal  Feed Rate
          Steam/Coal Ratio
          Steam/Air (02) Ratio
          Recycle Composition and Rate
          Method  of Coal Feed
          Number, Design,and Location of Injectors
          Bed Pressure
          Bed Temperature and Temperature Distribution
          Rotational Speed and/or Vertical Travel of Stirrer

          Unfortunately, the conditions under which the reported data were collec-
ted are usually not specified.  In addition, the precise influence of the above
parameters  on the gasifier emission characteristics is unknown.  For the purposes
of this study,however, the data obtained for the various types of gasifiers and
presented herein  can be assumed to be "typical" or "representative" of the parti-
cular technologies under consideration.  Lurgi and MERC have the most complete
data available.  These two processes are discussed below.

3.1.1     The Lurgi Process

          The Lurgi process, developed by Lurgi Mineralotecnik GmbH of West
Germany, is a proven high pressure coal gasification process.  The gasifier  is a
vertical, cylindrical steel pressure vessel that operates at  2.16-3-3 M Pa
(300-A65 psig).  Boiler feed water that is circulated through a water jacket sur-
rounding the gasifier shell recovers heat,  improving the overall efficiency of
the process.  A coal lock hopper above the gasifier feeds crushed and screened
coal to the coal  bed.

          Steam and air, or oxygen, are distributed into the coal bed through
the rotating grate.  The continuously  rotating grate supports the coal bed,
assuring a constant and even withdrawal of ash.  The ash drops  into an ash lock
hopper, an  integral part of the gasifier.

          A typical composition of raw product  gases  from  a Lurgi gasifier  is
shown  in Table 3-     Particulate size distribution data  is  limited.   It  has
been estimated that 1*0-80% of the particulate matter  is  smaller than  100  microns.
In addition to the  particulates and  tars  in  the  raw product  gases, other  poten-
tial sources of particulates  include   the coal  and ash  lock  gases.  The composi-
tion of the coal  lock gas  is  determined by  the  method of pressurizing  the coal
lock.   In addition  to the  components  in  the  raw  gas and  the  lock  filling  gas


                                      11

-------
            TASLE 3  COMPOSITION OF LURGI  RAW PRODUCT GAS (2)



Component
CO
H2
CH/t
C2H|,
C2H6
C02
N2 + Ar
2
H2S
COS + CS2
(kg/kg coal)
Mercaptans
Thiophenes
S02
H20
Naphthas
(kg/kg coal)
Tar (kg/kg coal)
Tar Oil (kg/kg coal)
Crude Phenols
NH3 (kg/kg coal)
HCN (kg/kg coal)
Particulates (coal
fines, ash)
(kg/kg coal)
Trace elements
HHV (dry basis):
Gasification media:

Subb i tuminous A


15.1
41.1
11 .2
0.5
30.4
1.2
ND
0.5
(9.2 x lO"*1)
PR
PR
PR
PR
(8.6 x 1C'3)
(3.0 x 10";)
(3.2 x 10"Z)
PR ,
(2.0 x 10~?)
(6.0 x 10"b)

0
(3.7 x 10"^)
PR
1.14 x 107 J/Nm3
(307 Btu/scf)
Steam/oxygen
Coal Type
High Volatile
C Bituminous

17.3
39.1
9.4
0.7
3K2
1.2
ND
1.1
(5.4 x 10"1*)
PR
PR
PR
PR
(1.0 x 10"2)
(3.8 x 10",)
(3.5 x 10 *)
PR ,
(4.0 x ]Q~l)
(6.2 x 10~5)


(5.6 x 10"3)
PR
1.11 'x 107 J/Nm3
(298 Btu/scf)
Steam/oxygen

Subb i turn i nous C


17.4
23-3
5.1
0.63
14.8
38.5
ND
0.23
PR
PR
PR
PR
PR
(1.6 x 10~2)
Pr?
?*.
0^
PR
PR


PR
PR
7.28 x I06j/Nm3
(195 Btu/scf)
S team/a i r
ND = presence of component not determined

PR *= component is probaby present, amount not determined

All components are presented as % volume, unless otherwise indicated.  Component
volume % is given on a relative basis to all other components that have a value
for volume % 1 is ted.

                                     12

-------
(either cooled raw gas or vent gas  from acid gas  removal), the coal  lock gas  may
contain entrained coal fines.   Particulate emissions have been estimated at
52 g/hr (0.116 Ib/hr) .  The ash lock gas composition is  also determined by the method
of pressurizing the lock.  Steam may be added to  the ash lock to pressurize  the  sys-
tem.   The ash in the hopper may be  cooled with a  water spray or quenched.   This
contacting of water and the hot ash produces steam and ash dust.  Uncleaned  char
can react with the steam and any organic compounds in the quench water can be
thermally quenched.  These reactions contribute to the composition of the gas stream
emitted when the ash lock is depressurized.  In any case, gases from the gasifier
can flow into the lock as it fills  with ash.  Particulate emissions from the ash
lock exhaust  fans with cyclone were estimated to be 9'  g/hr (0.2 Ib/hr).

3.1.2     MERC Process

          The MERC stirred-bed reactor has been demonstrated in a 20 TPD pilot plant
operated by the Department of Energy at the Morgantown Energy Research Center.
Under development since 1958,  this  gasifier is basically a WeiIman-Galusha unit
modified for pressurized operation.

          Sized coal (50 percent less than 1.25 cm) is transported to the coal feed
hopper.  Inert gas is used to pressurize the hopper to a pressure slightly greater
than the gasifier operating pressure.  The coal collects on the grate, forming a  bed.
The gasifier is a cylindrical, vertical steel pressure vessel with a stirrer  in  the
center and a rotating grate on the bottom.  Air mixed with superheated steam  is
fed into the gasifier beneath the grate.  The gases flow upward through the
descending coal.  The rotating stirrer, which also moves up and down, prevents the
coal  from caking.  The raw product gas exits through the side outlet near the top
of the gasifier.

          A typical composition of the raw product gas from the MERC gasifier is
shown  in Table A.  Data on the size distribution of particulates in the raw product
gas from the MERC gasifier are presented  in Figure 1.  Due to the  large differences
in the size distributions for the different data sets, the available data were
classified as worst-case and best-case for the purposes of this study.  Worst-case
data indicate a relatively high concentration of small particles, which are
difficult to remove from the gas stream, whereas best-case indicates a  relatively
low concentration of small particles.   In both cases, the size distribution
approximates a log normal function at the smaller  particle sizes.  The Lurgi  data
in Table 2 are also presented for comparative purposes.

3.2       FLUIDIZED-3ED GASIFIERS

          In a fluid!zed-bed gasifier, crushed coal is fluidized by the gas,  passing
up through the bed.  Because of the  intimate solids-gas mixing  in  the gasifier, the
temperature  is uniform throughout the bed,  resulting  in an exit  temperature  roughly
equal  to that of the bed.

          Since fluid-bed gasifiers have  higher particulate  loadings  in the  raw
product gas  than  is typical of fixed-bed  processes, more extensive solids removal
is required.   In contrast, however,  tar  production is minimal,  and the  heat  content
of the product gas  is  lower, due to  the  smaller yield of hydrocarbons.  Fluid-bed
gasifiers have the advantage of being able  to utilize a  variety  of coal types.  The

                                       13

-------
           TABLE k  COMPOSITION OF MERC RAW PRODUCT GAS (2)
                                         Coal  Type
                           Subbiluminous A
 Component
 CO
CH
Ar
 CS2
C0
N2
02
H2S
COS
Mercaptans
Thiophenes
S02
H20  (kg/kg coal)
Naphthas
Tar  (kg/kg coal)
Tar ON
Crude Phenols
NH3
HCN
Participates (coal
   fines, ash) (kg/kg coal)
Trace elements

HHV (dry basis):
   Joule/Nm3
   (Btu/scf)

Gasification media:
                                16.0
                                19.0
                                 3.5
                                12.6
                                48.4
                                 ND
                                 0.2
                                 PR
                                 ND
                                 ND
                                 ND
                              (0.64)
                                 PR
                              (.03*0
                                 PR
                                 PR
                                 PR
                                 ND
                              (0.17)

                                 PR

                                 5.6 x  10*
                                (150)
                            Steam air
                                                            High Volatile
                                                              b i turninous
  21.6
  18.7
   2.9
   0.2

   7.3
  48.9
   ND
   0.4
   PR
   ND
   ND
   ND
(0.37)
   PR
(0.20)
   PR
   PR
   PR
   ND
                                                         (3.5 X
                                                            PR
       10"3)
                                                            6.1  x
                                                           064)
                                                        Steam/a!r
ND = presence of component not determined.

PR = component is probably present, amount not determined.
All components are presented as % by volume, unless otherwise indicated.
volume % is given on a relative basis to all other components that have a
volume % 1isted.
                                                                          Component
                                                                          value for

-------
   soc-
   JOT
    JO
01

-------
gasifiers can  be operated at a range of output rates with little loss in efficiency.

          Typically, there are two general categories of ash handling in fluidized
beds.   In the dry-ash mode, the bed temperature is maintained below the ash soften-
ing  temperature.  Part of the ash, mixed with some coal, is "drained off" the bed
while the bulk is discharged with the off-gas and is then collected.  The ash-aglom-
merating bed, on the other hand, operates at higher temperature.  The ash forms low
carbon-content agglomerates, eventually settling to the bed bottom.   The ash is
then cont inuously wi thdrawn from the bottom of the bed.

          In Table 5,the status of selected fluid bed gasifiers is summarized.   It
can be seen that Winkler is the only commercially available fluid-bed gasifier.
Typical parameters and raw gas characteristics for each of these processes are shown
in Table 6.

          Of the flgid bed gasifiers listed in Table 6,Winkler and C02 Acceptor
were found to have the most complete data available.

3.2.1     Winkler Process

          Coal is crushed to 0-1 cm and is sent to the gasifier feed hopper.  Screw
feeders are used to transfer the coal into the gasifier.  Steam and oxygen are
added near the bottom of the reactor and pass up through the coal, fluidizing the
particles.  The coal reacts with the oxygen and steam in the bed to form H2, CO, C02
and CH.^.  At the bed temperatures, approximately 760 C (1400 F), heavy hydrocarbons
and tars are not produced, and the ash remains solid.  The heavier ash particulates
drop through the fluidized bed and are discharged, while the lighter particles are
carried upward through the bed with the product gas.  Approximately 70% of the ash
in the feed coal leaves the reactor in the form of particulate carryover.

          A typical composition of raw product gases from a Winkler gasifier is
shown \n Table 7.  Particle size distribution data on the Winkler Process are not
available.

          Besides particulate removal from the raw product gas stream, other gas
streams that require particulate control  include the coal bin nitrogen vent gases
the dry ash bin nitrogen vent, and the ash slurry settler vent.  In order to
minimize the potential  for explosion, nitrogen is used to blanket  the coal dust feed
bins.  Coal  particles can then be entrained in the vent gases.  Nitrogen is also
used to blanket the dry ash to prevent the char in the ash from reacting or com-
busting.  Ash particulates can then be entrained in the nitrogen vent gases.  The
ash slurry settler unit may contact any of the raw gas components  that dissolve or
condense in the direct/scrubber/cooler.   The ash washed from the raw gas stream is
separated from the quench liquor in a settler.  Components of the  gases that are
dissolved or condensed  in the quenching liquor can evaporate.  These gases are then
vented.   Entrained droplets of gas quenching liquor or ash slurry  may potentially be
present in the vent gases.

3.2.2     CO- Acceptor  Process

          The C02 Acceptor  Process developed by CONOCO Coal  Development Company,

                                      16

-------
TABLE  5   STATUS  OF IKS,  AND  FOREIGN  LOW-AND  MED1UM-BTU FLUID BE'O  GASIFICATION  SYSTEMS 0)
                                                            Number of Gasifiers  Built
Gasi fier

Winkler

SCR Low-Btu


Hygas



Synthane
CO- Acceptor
U-Gas


Battelle/Carbide

COGAS

Westinghouse
   Li censor/developer

Davy Powergas

Bituminous Coal
  Research, Inc.

Institute of Gas
 Technology


ERDA


ERDA


Institute of Gas Technology
Phillips Petroleum Corp,

Battelle Memorial Institute

COGAS Development Company

Westinghouse Electric Co.
a h
I D P UQ e*
LBU Bu
23
1
1
1
1
1
I
1
1
SNGC Location
H Foreign
US
US
us
us
us
us
us
us
   Scale
Demonstrat ion
Demonstration
(High-Btu)

Demonstration
{High-Btu)

Demonstrat ion
(High-Btu)

Pilot
PDU

PDU

PDU
                Low Btu Gas
                Medium-Stu Gas
                Synthesis Gas

-------
                                                                                       TABLE 6

                                                                    OPERATING AND  RAW GAS STREAK CHARACTERISTICS
                                                                               FLUID BED GASIFIERS
oo

Wtnkler(2.3.4)

Synthane(3,4,5)

CO; Acceptor
(3.''. 5)

HygasO.4)


CoGas(3)

Hydrane(3)

Unloo Car-
bide(3)
West!nghouse(3)

U-Gas(3)


BCR (3)

lgnlfluld(5)

Coal Type
several coal
types
all types

lignite
sub -bitu-
minous
all coals


all types

all types



"variety
of coals"
non-caking
caking req.
pretreatnt.




Temperature
590-730
(1100-1450)
760
(1400)
815
(1550)

1100
(2000)

870
(1600)
540-815
(1000-1500)
870-980
(lfrOO-1800)


840-1040
(1550-1900)



590-715
(1100-1320)
Pressure
HPa(psla)
0.10
05)
6.90
(1000)
1.03-2.06
(150-300)

6.90-10.3
(1000-1500)

0.21-0.41
(30-60)
6.90
(1000)
0.69
(100)
0.90-1.38
(130-200)
0.34-2.41
(50-350)

41.62
(*235)
O.tO-0.50
(15-75)
Part Icul ate
Loading Participate
g/nm3(9r/scf) Composition
C-30*
ash-70*
4.8-12 C-BOt
(2-5) ash-20*
26 C-8*
(10) a$h-88t

120 C-55*
(50) ash-4ot
Oz-5*




0.0-1.2
(0.05-0.5)
8
(3-3)





84
(35)
Tar Loading HHV-OjBlown

-------
           TABLE 7   COMPOSITION OF WINKIER RAW PRODUCT GAS (2)
                                           Coal  Type
   Component
CO
H2
C2Hi,
C2H6
C02
N
Ar
H2S
COS + CS2
Mercaptans
Thiophenes
S02
H20
Naphthas
Tar
Tar Oil
Crude Phenols
NH3
HCN
Parti culates
  (coal fines, ash)
Trace Elements

HHV  (Dry Basis):
J/Nm3 (Btu/scf)
                     Subbiturn!nous A
   22.0
   14.0
    1.0
    ND
    ND
    7.0
   56.0
    ND
    PR
    ND
    ND
    ND
    ND
    PR
    NP
    NP
    NP
    ND
    ND
    ND
    PR

    PR

4.66 x 10*
   (125)
Gasification media:    Steam/air
                                         Lignite
   35.5
   40.0
    2.8
    ND
    ND
   19.9
    1.8
    ND
    PR
    ND
    ND
    ND
    ND
    PR
    NP
    NP
    NP
    ND
    ND
    ND
0.46 kg/kg
 coal DAF
    PR

1.01 x 107
   (272)

Steam/02
                                        Subb i turn!nous
   37.0
   37.0
    3.0
    ND
    ND
   20.0
    3.0
    ND
    PR
    ND
    ND
    ND
    ND
    PR
    NP
    NP
    NP
    ND
    ND
    ND
    PR

    PR

1.0 x I0y
   (270)

Steam/02
ND = presence of component not determined

PR = component  is probably present, amount not determined

NP = component  is probably not present

All components  are presented as % by volume,  unless  otherwise  indicated.
Component  volume % is  given on a relative basis  to all  other  components that have
a value  for  volume %  listed.
                                      19

-------
 has  undergone  testing  at  27230  kg/day  (30  TPD)  at  a  pilot  plant  located  in Rapid
 City,  S.D.   This  process  consists  of two  fluidized bed  reactors,  including the coal
 gasifier  itself and  a  regenerator  for  spent  limestone or dolomite.  Control of parti-
 culates and  tars  will  be  discussed in  this  report  only  for the gasifier, since the
 regenerator  is unique  to  this process  and,  therefore, is not  representative of other
 gasification processes.

          Lignite or subbi turn! nous coal is  crushed to 0.1 to  1 cm  and  is  fed into
 the  bottom of  the gasifier.  Steam enters  through  a  side nozzle at  roughly the same
 bed  height as  the coal feed.  Above the fluidized  bed,  hot recirculated  acceptor,
 CaO  is fed into the  gasifier and falls through  the bed.  The  gasifier fluid bed is
 a mixture of acceptor  and  char, with the  lighter char concentrated  in the upper zones
 of the gasifier.  The  following exothermic  reactions occur in the main gasifier:
                CaO + C02

                CO + H0
The heat produced by these reactions and the sensible heat of the acceptor supply
the heat necessary to drive the endothermic gasification reactions:

                C + H20  - *- CO + H2
                2C + H20 + H2 - - CH^ + CO

The raw product gas then leaves the gasifier through an internal cyclone.

          The carbonated acceptor flows out of the reactor and  is conveyed to the
bottom of the regenerator by air or recycled gas.  Residual char is withdrawn
through a standleg near the top of the fluidized bed and is transferred to the regen
erator.  The char is burned at 1010 C (1850 F) with air in the regenerator
fluidized bed.  The carbonated acceptor is calcined at this temperature to CaO and
C02, with the CaO then being returned to the gasifier.  Particulate matter is
removed from the regenerator flue gas by means of a cyclone.

          It should be noted that the particulate matter released from the gasifier
consists primarily of char and ash, with a negligible amount of acceptor being
present.  Thus, the emissions characteristics for the gasifier  should be typical of
other types of fluid-bed gasifiers.

          Reported compositions of the raw product gases and the lockhopper vent
gases are presented in Table 8.  The particle size distribution is shown on Figure 2
The C02 Acceptor data in Figure 2 were obtained  in an air-blown system using coal
with 7-5% ash.  The particulates tested consist of the char that escaped the
internal cyclone of the gasifier.

          In addition to the C02 Acceptor data in Figure 2, particle size
distribution data are also presented for the Ignifluid process.  The Ignifluid
gasification system employs a fast fluid bed and temperatures of 1095 C (2000 F)
or more, which results in a "sticky" ash that tends to agglomerate.  The particle
sizes are, therefore, relatively large, as can be seen from Figure 2.

                                      20

-------
                         TABLE 8

               COMPOSITION OF C02 ACCEPTOR RAW
           PRODUCT AND LOCKHOPPER VENT GASES (11 )
                        Raw Product Gases    Lockhopper Vent Gas

CH^                             8.9

CO                             12.2

C02                             4.8                33.1

H2                             50.8

N2                              0.1                65.1

NH                             <0 . 1
H20                            21.0                 1.3

Inert                           1.3

so2                                                
-------


5M






U
 r/
Particle Size, micromet
9- - ~1



















































/
f
///
9.1

















/

y
/ /
/ /
' /
/ /
/ /
/
t
1















y
/
*
T

/
w
f

















/
/


/
J
'







10


\












t
f
/



f
/









1

/
/




f
//
/








\









/



/
^
-
^
/

















/

X
f
*


/
O
^
















/

/



o
f

}?
















^.
f




_/"
f


















^
c.
*'


S^








^^BM






x*"













O
VH 





f*'




x*











^<
















 . " -






^ 	









COj-Acceptor Data
Best-case based o
Ignlfluid data (5
 ~xtrapo1ated bey
range of data
Average of Worst
Best Cases





















(11)
n
).
ond
and





































30 39 tO SO (0 10 tO 10 S 99 . 
FIGURE 2.
        Cumulative Probability
Particle Size Distribution for Fluid Bed Gasiflers

-------
3.3       ENTRAINED BED GASIFIERS

          In the entrained-bed gasifier,  the gases that are fed to the reactor and
gases formed during the gasification reactions carry or "entrain" pulverized coal
through the gasifier.   Because the flow is concurrent,  the reaction rate decreases
as the coal particles  travel  through the  reactor.   Since high temperatures are
maintained to minimize the reactor size,no tar or  heavy oils are produced.

          An advantage of entrained-bed gasifiers  is their capability to process all
coal types. This is due to relatively large separations between the particles, which
minimizes agglomeration.  Other advantages of entrained-bed gasifiers include a
smaller bed reactor volume per unit of energy production, and the ease of disposal
of fused ash in comparison to loose flyash.  Disadvantages of entrained-bed gasifi-
ers are primarily due to the low concentration of  fuel  in the gasifying medium and
the concurrent flow.  Only 85-90% of the  carbon can be  economically gasified in a
single pass.  Due to the high temperatures of the  gases leaving the reactor, heat
recovery is necessary to reduce energy losses and  improve the thermal efficiency.
This is complicated by the presence of solids in the stream and the possible
deposition of molten ash on heat transfer surfaces.

          The status of some of the more  promising entrained-bed gasifiers  is shown
in Table 9-  It can be seen that Koppers-Totzek is the only commercially available
entrained-bed gasifier.  Typical process  parameters and raw gas stream characteris-
tics are presented in Table 10.

          Of the entrained bed gasifiers  listed in Table 10,Koppers Totzek and Bi-Gas
were found to have the most complete data available.   In addition to the  list of
parameters in Section 3-1 for fixed-bed gasifiers, the gas velocity and residence
time are expected to affect the product gas characteristics from entrained-bed
gasifiers.  As  in the case of both fixed and fluid-bed gasifiers, the many  para-
meters that can affect the product gases  are usually unspecified, and the effects
of these parameters on the product gas characteristics can generally not be quanti-
fied.  For the purposes of this study, however, the product gas characteristics
presented herein can be assumed to be "typical" or "representative" of other gasi-
fiers within the same generic classification.

3.3.1     Koppers Totzek Process

          The Koppers Totzek  (K-T) process  is a commercially proven  low pressure
coal gasification process.  The gasifier  is a horizontal eliptical double walled
steel vessel with refractory  lining and has either two or  four  truncated  cone-
shaped heads mounted on either end of the ellipsoid.  The  coal,  steam and  oxygen
are  injected  through burners.  A waste heat boiler at  the  top of the gasifier
recovers heat from the hot effluent gases.  The carbon and volatile matter  are
gasified, and between 50-70?  (4) of the liquefied  coal ash  falls  into a water
quench tank in  the form of molten slag.  The  remainder of  the coal ash  is present
as  particulate  carry-over  in  the  raw  product gas.  Table  II  indicates the projected
composition of  the  raw  product gas.

          Typically,  in  the  K-T process,  gas departing  the gasifier  is  quenched by


                                       23

-------
NJ
                                       TABLE 9

     STATUS OF U.S. AND FOREIGN LOW-AND MEDIUM-BTU  ENTRAI NED-BED  GASIFICATION SYSTEMS(l)
Gasif ier
Koppers-Totzek
Bi-Gas
Texaco
Combust i on
Eng i nee ring
Babcock 6 Wilcox
Co a 1 e x
Licensor/developer LBG
Koppers Company,
Inc.
B i turn i nous Coal
Research , Inc.
Texaco Development
Corpo ra t i on
Combustion Engineer- 1
ing Co rporat i on
Babcock 6 Wi 1 cox 1
Corpo rat i on
Inex Resources, 1
Number of Gasifier
MBGb SNGC Locat
39 Forei
1 - US
1 US
US
US
US
s Bui It
i on Sea 1 e
gn Comme re i a 1
Demons t ra -
t i on
Demon s t ra -
t i on
Demons t ra -
t i on
Demons t ra -
t i on
Pi lot
    Foster Wheeler
 Inc.

Foster Wheeler
 Energy Corp.
US
Pi lot
     Low-Btu Gas

     Medium -Btu Gas

    cSynthes i s Gas

-------
                                                                                     TABLE  10
                                                                  OPERATING AND RAW PRODUCT GAS STREAM CHARACTERISTICS
                                                                                   ENTRAINED BED GASIPIERS
N)

(Coppers Totzek
(2.3.4)
BI-Gas(2.3,)



Texaco (2,3)

Combustion
Englneering(J)
8 S W(3)

Coa1ex(2)

Poster
Wheeler(2,3)




Coal
Type
al 1 types

lignite
sub-bit.
bitumin.

lignite
bl tumln.
all types

all types

all types

non-caking





Temperature
O (Op)
H80
(2700)
7*5-1180
(1375-2160)


200-260
CtOD-SOO)
870
(1600)
980
(1800)
925-950
(1700-17*0)
upper stage
980-1150'
(1800-2100)
lower stage
1370-15W
(2500-2800)
Pressure
HPa(psie)
0.10
(15)
1.62-10.3
(235-1500)


2.10-S.27
(300-1200)
0.10
(15)
0.10-2.10
(15-300)
0.10
(15)
2.41
(350)




Participate
Loading Paniculate
g/nn3(gr/scf) Composition
30-60 C-10*
(12-25) ash -90*
230 char-96-89*
(J6) ash-12-IO*
volatiles-
2-1*














Tar Loading
g/nn3
(gr/sef)
None

None




None



None


None





HHV-02Blown
Tar Composition (Btu/scf)
l.lxlO7
(290)
1 -3xl07
(356)


9.*x106
(253)


l.lxlO7
(30)








HHV-AIr Blown
(Btu/sef)


5.3x10*



6.5xl06
(175)
4.7xl06
027)
3.3xl06
(102)
I(.9xl06
(133)
6.6x10*
(177)





-------
            TABLE  11   COMPOSITION OF KOPPERS-TOTZEK RAW PRODUCT GAS  (2)
                                          Coal  Type
 Component

 CO
 H2
 CH/,
 C2Hi
Ar
 C02
 N2
 02
 H2S
 COS  +  CS2
 Mercaptans
 Thiophenes
 S02
 H20
 Naphthas
 Tar
 Tar  Oil
 Crude  Phenols
 NH3
 HCN
 Particulates  (coal
   fines, ash)
   (kg/kg OAF coal)
 Trace  elements

 HHV  (dry basis):
   Joule/Nm3
  (Btu/scf)
Lignite A
56.87
31-30
PR
ND
ND
10.0
1.18
ND
0.60
0.05
ND
ND
PR
PR
ND
ND
ND
ND
<0.2
PR
(0.08)
PR
1.1 x 107
(290)
B bituminous
52.35
35.66
PR
ND
ND
10.0
1.12
ND
0.82
0.05
ND
ND
PR
PR
ND
ND
ND
ND
<0.2
PR
(0.06)
PR
1.1 x 107
(290)
C bituminous
52.51
35.96
PR
ND
ND
10.0
1.15
ND
0.36
0.02
ND
ND
PR
PR
ND
ND
ND
ND
<0.2
PR
(0.08)
PR
1.1 x 107
(290)
Gasification Media:
                    Steam/02
Steam/02
Steam/02
ND * presence of component not determined

PR  component is probably present, amount not determined

All components are presented as % volume, unless otherwise  indicated.  Component
volume % is given on a relative basis to all other components that have a value
for volume % 1 is ted.
                                     26

-------
water sprays.  This quenching is done to solidify the entrained slag particles to
prevent deposition on the downstream boiler tubes.   The gas then passes through  a
waste heat boiler where high pressure  steam is  produced.   After leaving the waste
heat boiler, the gas is cleaned by a venturi scrubbing system.   The particle size
distribution data presented in Figure 3 was collected at a sampling point between
the waste heat boiler and the water quenching operation.

3.3.2     Bi-Gas Process
          This two-stage entrained flow gasifier developed by Bituminous Coal
Research, Inc. (BCR) is currently in the pilot plant stage.  A 120 ton/day pilot
plant has been constructed in Homer City, Pennsylvania.

          Run-of-the-mine coal is crushed, dried, and pulverized to .05 cm (-200 mesh)
The coal is combined with hot gas recycled from the gas purification section,  and
is fed  into the upper stage of the gasffier.   The coal  reacts with gas from the
lower stage  and steam,and yields the product gas.

          Residual char is removed from the gas by cyclones and is recycled with
steam and air.  The hot gas flows  to the upper stage for reaction with the coal.
The molten slag collects and drains from the bottom of the lower stage into the slag
pot where it  is water quenched.

          Table 12 presents data on the composition of the raw product gas.  Avail-
able particle size distribution data are shown in Figure 3.

          In addition to the  raw product gas, another emission source is the slag
lock gas.  When the slag lock is depressurized to discharge accumulated slag,  a
gaseous discharge stream will be emitted that may contain slag particles, any  of
the  raw gas components, plus volatiles found  in the slag quench water.  Controls,
dependant upon the makeup of the stream, may  include a cyclone  or combustion  in
a flare or boiler.

3.^       SUMMARY OF DATA

          The particulate and tar loading data and the particle size distribution
data presented in Sections 3*1 through 3-3 are summarized  in Table 13.  The best,
worst,  and average-case data  for the particulate and tar  loadings from fixed-bed,
fluid-bed, and entrained-bed  gasifiers were estimated from the detailed data for
the  individual gasifier types in Tables 2, 6, and 10.  It can be seen that fixed-bed
gasifiers produce the smallest particulate loadings, while the entrained-bed gasifiets
produce the greatest particulate  loadings.  Only the fixed-bed gasifiers produce a
significant amount of tars.   The particle size distribution data  in Table  13 were
obtained from Figures  1, 2 and 3.

           It  should be noted  that the data in Table 13 should not be  interpreted
as being indicative of any particular type of gasifier, but  instead portray charac-
teristics expected of processes  in general for the generic categories under considera-
tion.   The best and worst-case data for particulate and tar  loadings  should reason-
ably reflect  the  extremes to  be  expected  from each generic class  of gasifier.   Due

                                      27

-------
-oo
wcv

50C


100
in
i.
V
+j
 '"
1
a"
/>
01
"o
4-
i.
10
0.
K
























X
/'/ /
/



















y
f
jfr
f^
Sf
7


















/
//
f S _^r
xr_-X^
<"'






0.1 '















/
/
/
/






















/
/*

X
X









>0







!






/
t



X



I


/





/
s
X











/^




X
/

^



! 1






1


















/
9


yl
y


s

















/


/
J
V


^x















	 <
/
/


/
i*
^


^X
''A














/









LJ-X'
r













,70 JO <0 50 0 70 J











^















s
r















x
vXO*
















x^
















^










o BI-Gas Data (15)
A Koppers Totzek Data
(for Coke, Char,
Anthracite) (H)
	 Extrapolated beyond
range of data
	 Average of Worst and
Best Cases
































0 30 95 39 #. JSi
                                                                  Cumulative Probability
                                          f I CURE 3.  Particulate Size Distribution for Entrained Bed Gasifiers

-------
                TABLE 12  COMPOSITION OF BI-GAS  PRODUCT GAS (2)
                                                  Coal  Type
Component

CO
H2
CHi,
C2H1,
CzH6
C02
N2 + Ar
02
H2S
COS + CS2
Mercaptans
Thiophenes
S02
H20
Naphthas
Tar
Tar Oil
Crude Phenols
NH3
HCN
Particulates (coal
   fines, ash)
Trace elements

HHV (Dry basis):
   J/Nm3
   (Btu/scf)

Gasification Media:
Western Kentucky
  #11 Bituminous
     40.6
     22.5
     14.3
      ND
      ND
     12.3
      0.6
      ND
      1.3
      PR
      ND
      ND
      ND
      7.7
      NP
      NP
      NP
      ND
      PR
      ND

      PR
      PR
    1.30 x  10
      (350)

    Steam/02
                                                             Illinois To
                                                             Bituminous
                                                                18.1
                                                                13.1
                                                                 3-6
                                                                 ND
                                                                 ND
                                                                 8.3
                                                                1*5.8
                                                                 ND
                                                                 0.5
                                                                 0.1
                                                                 ND
                                                                 ND
                                                                 ND
                                                                10.2
                                                                 NP
                                                                 NP
                                                                 NP
                                                                 ND
                                                                 0.4
                                                                 ND

                                                                 PR
                                                                 PR
                                                             5.29 x 10
                                                                (142)

                                                             Steam/Air
ND = presence of component not determined

PR = component is probably present, amount not determined

NP = component is probably not present

All components are presented as % volume, unless otherwise indicated.  Component
volume % is given on a relative basis to all other components that have a value
for volume % listed.

                                      29

-------
                              TABLE 13  $UHJWFf.?2EO PWtTICtltftTt  ftKU TAR
HAl
O


~iXtt4 3srf
StCwe.
Worst Case
Average
test Case
*jrst Case

Best Cas,a
Average
Participate
Lo-adi rig
Is/1*^
0.5
6,0

1.2
I2Q.O

33.0
ua.G
Tar /*ercertt Particles
Load^fg Sp-ecifEad C" i-s-be
^nr-^/ T t
ffl .0 <0 . 1 Q  J
50 . 0 * 0 . 1 h . 0
(S,0 
-------
to the limited amount of particle size data available for this study, the data in
Table 13 were obtained by considering all  the available data for each generic class
of gasifier, and by employing engineering  judgment to estimate values representa-
tive of each class.

          The data in Table 13 will  be utilized as a basis for estimating the parti-
culate and tar loadings on the control technologies described in the next section,
and for assessing the adequacy of each control  technology for reducing the particu-
late and tar concentrations to acceptable  levels.

-------
                                     SECTION  4

                         ALTERNATE  CONTROL TECHNOLOGIES


           Alternate  control  technologies  for removing  the particulate and tar
 emissions  from the product  gases  of the various  types  of gasifiers discussed in
 Section  3    are described  in  this section.   Control  technologies both commercially
 available  and  under  development are considered.   Detailed descriptions of most of
 these  control   devices  are  readily  available in  a wide variety of books, reports,
 and  papers  in  the open  literature.   Brief summaries  of important design and oper-
 ating  features  are presented  herein, with emphasis on  characterizing the particulate
 removal  efficiency of each  control  device.   These performance characteristics are
 then employed  in Section 5    in assessing the  applicability of each control device
 to each  type of generic gasifier  previously  discussed.

 k.]        CYCLONES

           Cyclones utilize  the centrifugal force  created by a spinning gas stream
 to separate  particulates from the carrier gas.   Spinning motion is imparted to the
 carrier  gas  by  tangential gas inlet, vanes or  a  fan.  Particulates, which have a
 greater  applied centrifugal force than that  of the gas molecules, move outward to
 the  wall and are carried down a receiver.  The major classes of cyclones include
 conventional cyclones,  multiclones, and rotary flow  cyclones.

          Conventional  cyclones have the  advantage of being a proven technology
 and  a simple device with no moving  parts.  However,  they suffer from the dis-
 advantage of having  low removal efficiencies for particulate sizes less than
 5 micrometers.   Because of  their  relatively  low capital and operating costs,
 cyclones are commonly used  as pre-cleaners to  remove most of the large particles
 in a gas stream upstream of a more  expensive control device (e.g., venturi  scrubber
or electrostatic precipitator) required to remove the smaller size particles.   To
 improve  the collection efficiency,  multiclones and rotary flow cyclones were
developed.   Each class of cyclones  is discussed below.

A.1.1     Conventional Cyclones

          Design procedures for conventional  cyclones have been highly developed
from theoretical considerations,  supplemented to some extent by empirical  tech-
niques gained from operating performance data.  The particulate removal  efficiency
has been found  to be very dependent on particle size.  It is, therefore, necessary
to define the particulate size distribution  in order to successfully predict the
removal efficiency that will be obtained with a particular design for a given set


                                     32

-------
of operating conditions.   High efficiency cyclones are generally considered to be
those with a body diameter less than 0.23 meters, since the smaller the body
diameter, the larger the separation force created.

          Participate collection efficiency increases with an increase in partic-
ulate diameter, particulate density, inlet velocity, cyclone body length, ratio of
cyclone body diameter to outlet diameter, and the smoothness of the inner wall.
Efficiency decreases as the gas viscosity, gas density, body diameter and gas
outlet diameter increase.   A typical set of data on the efficiency of collection
as a function of temperature are shown in Figure k.  Since the gas viscosity is
proportional to temperature, an increase in temperature results in a decrease in
the efficiency.

          A further illustration of the effect of high temperature and pressure
on particulate collection is presented in Figure 5.  This is done by determining
the collection efficiency of a cyclone operating at various temperatures and
pressures for the same inlet velocity.  Curve 1 in this figure shows a typical
efficiency curve for a high efficiency cyclone operating at ambient conditions.
The collection efficiency drops significantly as the temperature increases to
1100C, as shown in Curve 2.  Curve 3 shows the collection efficiency decrease,
again for small particulates,as the pressure increases from 0.1 M Pa to 1.5 M Pa.

4.1.2     Multiclones

          Multiclones are designed to increase the conventional cyclone perform-
ance by reducing the diameter of the cyclone while maintaining a constant inlet
velocity since the centrifugal force applied to the particulates varies inversely
with the diameter of the cyclones.  These devices are fabricated by manifolding
together banks of smaller cyclones, usually with a common inlet plenum chamber,
dust storage bin and outlet plenum chamber.  In order to achieve the same level
of collection efficiency as a single tube with the same diameter and inlet vel-
ocity, it is necessary to equalize the gas loading between the cyclones to prevent
backflow, plugging or re-entrainment from the dust bin.

          Combustion Power Company has tested two stages of multiclones in their
CPU-400 Pilot Plant to remove fly ash and fine bed material particulates  (46).
Results show that the cyclone tubes of both stages are prone to plugging  in the
lower cone body due primarily to suspected cross  flow between the tubes.  The
performance of these units is similar to that of small conventional cyclones  (47)

          Multiclones have been commercialized, as well as being demonstrated at
high temperature (788C) and high pressure (5-0 M Pa).  Grade efficiency  data
reported by Environmental Elements Corporation multiclones are shown in Figure 6.

4.1.3-    Rotary Flow Cyclones

          Rotary flow cyclones are designed to augment the normal tangential  swirl
of the inlet gas by the addition of a secondary airflow.  By doing so,  the possi-
bility of short-circuiting  of particulates from  inlet to outlet  is greatly
reduced.  Two  rotary-flow type cyclones have been  developed  to  improve  cyclone
collection efficiency,as discussed below.

                                     33

-------
*<>
 X
 u
 c
 0>
     80
                                       7.6 cm A p
                           Temperature,1 C


        FIGURE *.  Effect of Temperature on Conventional
                           Cyclone Efficiency
u
c
u
                                                Constant Inlet Velocity
                                                      Condit ions

                                                      20C, 1x105 Pa

                                                      1,IOOC, IxlO5 Pa
                                                      1,100C, 1.5x10 Pa
                                                         _L
JL
JL
                   46     8    10  12   14   16   13   20   22   2k   26   23


                       Particle Diameter,  micrometers
        FIGURE  5.  The  Effects  of High  Temperature and Pressure on the
                   Collection of  A High Efficiency Conventional CycloneC5)

-------
                  99.9
   100  r-
*   80  -
                              5       10      15      20

                            Particle  Size,  micrometers
25
                       FIGURE  6.   Fractional  Removal  Efficiency Curves
                                  of Multiclones by Environmental
                                  Element  Corp.  (18)
                                Particle Size, micrometers
        FIGURE  7.   Fractional  Removal  Efficiency Curves for Aerodyne Series
                   "SV" Dust Collector
                                         35

-------
           In  the  Aerodyne  rotary  flow cyclone,  particulate-laden  gas enters  the
 collection  chamber and  passes  a stationary  vane which  imparts  a  rotary motion  to
 the  flow.   Particulate  matter  is  thrown  toward  the  outer wall  by  centrifugal  force
 where  it  is swept downward  to  the collection  hopper by  the  secondary flow.   The
 vendor curves  for the  Aerodyne Series "SV" rotary  flow cyclone  are presented
 in  Figure  7-  Westinghouse  has  tested an Aerodyne Tornado Cyclone with a primary
 air  flow of  1.42  m3/min, and a  secondary flow of 0.84 m3/min.  The grade effi-
 ciency data obtained  in these  tests  show a  discrepancy  with  respect to the claimed
 performance by  the manufacturer.   This may  be due to the difficulty of holding
 design removal  specifications when testing  a  small  unit with  less than 11.3  m'/tni n
 capacity  (50).  Thus, the  fractional  collection efficiency  curves presented  in
 Figure 7 need to  be further verified.

          The Donaldson Tanjet  Cyclone is designed  to incorporate a low-volume,
 high velocity tangential secondary gas flow to  induce a strong vortex flow on a
 primary axial flow stream.  The secondary flow  should be at  the same temperature
 as the primary  flow to  prevent  thermal stressing of the assembly.  The tangential
 jets are designed to provide a  radially  outward flow component in the separation
 zone,which aids in particulate separation.  Westinghouse has operated a Tanjet
 unit at 3-7 mVmin with a secondary  flow of clean air at a  rate of 0.4 mVmin at
 a pressure drop of 3-3  M Pa.  These  tests confirmed that the unit maintains  its
 performance at high temperatures  ( 5).   The grade efficiency curve is presented
 in Figure 8.  The Tanjet program  was  discontinued by Donaldson in 1976 to await
 the development of an attractive  market.  Further work  to improve the collection
 efficiency and to refine the collection  hopper  are  needed.   In addition, the use
 of clean secondary air, at  pressure,  must be  considered as a significant process
 cost and should be included in any overall  design.

 4.2       WET SCRUBBERS

          Wet scrubbers are available in  a wide  variety of designs.   All  wet
 scrubbers, though, operate  on a common principle of contacting a polIutant-laden
 gas with a liquid  (usually water)   that captures  the pollutants.  Wet scrubbers can
 be utilized to remove both  particulates  and/or  tars.  The objectives  of good
 scrubber design are to  provide good  liquid-gas  contact, minimize energy consumption
 and equipment size, and minimize water requirement.   All wet scrubbers produce a
 liquid slurry for  disposal  or further treatment.  Most  modern applications attempt
 to concentrate the solids to simplify their ultimate disposal, and to recirculate
 as much of the scrubbing liquid as possible.

          The collection efficiency of wet scrubbers is strongly dependent on
 particle size.  In order to achieve high collection efficiencies with  small
 particles, a high energy input is   required.    For particles above approximately
 10 micrometers,  simple wet scrubber designs are  usually adequate, with pressure
 drop of 0.25 kPa  being  typical.   Fine particulates with diameters of  1  micrometer
or less require more complex scrubbers with pressure drops usually well  above
 1.25 k Pa.   In exceptional  circumstances, pressure drops up  to 25 k Pa have
been employed.

          Wet scrubbers  have been   found  to be very effective in removing  tars
 from raw product gases.   Commercially available gasification systems  generally

                                    36

-------
         90
         30
         70
         60
     r
     
-------
 have employed various types of wet scrubbers  to quench  and cool  the  gases  and
 knock out the tars, along with a portion of the particulates.

           The various types of wet scrubbers  in common  use are  described briefly
 below.

 it.2.1     Spray Towers

           Gas flows upward through a mist of  falling liquid droplets  produced  by
 spray nozzles or atomizers.  The settling velocity  of the  drops  must  exceed  the
 upward gas flow to avoid droplet entrainment  and carryover.   Spray towers  are
 often used as precoolers when large quantities  of gas are  involved.   Operating
 characteristics include low pressure drop,  ability  to handle scrubbing  liquids
 with a high solids content, and moderate liquid requirements.   Spray  towers  are
 usually limited to the collection of particulates with  diameters of  10  micro-
 meters  or larger.

 4.2.2     Cyclonic Spray Scrubbers

           These devices, combining the  collection principles of  scrubbers with the
 cyclonic removal  of liquid droplets, are of two types.   In  the  first  type, a
 spinning motion is imparted to the gas  stream by tangential  entry.   In  the second
 type, the rotating motion is  given to the gas stream by  fixed vanes and impellers.
 The  sprayed liquid droplets are carried to  the  wall  by  centrifugal forces, thereby
 reducing droplet  entrainment  and carryover  from the  scrubber.   Because of  the
 centrifugal  action and reduced tendency for entrainment, smaller droplets  and
 higher  gas  velocities can be  employed,  as compared  to a  conventional  spray chamber
 resulting in increased removal  efficiencies.   A  wide  variety  of configurations  can
 be employed in  arranging the  locations  of the water  sprays.  The pressure  drop
 typically ranges  from 3-75 to 7-5 h  Pa,  and efficiencies of  90% or more can be
 obtained for particles  with diameters above 5 micrometers.

 4.2.3     Mechanical  Scrubbers

           Water is  sprayed into a fan or other  rotating  element to mechanically
 break the liquid  up  into small  droplets.  Spray  nozzles  at the  inlet  offer an
 opportunity  for impact ion.   Centrifugal  force and impingement on the  blades are
 utilized for further  collection  and  water separation.  The power requirement is
 typically  in  the area of 1.5  k  Pa pressure  drop when  fans are employed on the
 rotating element.  The  collection efficiency  is  generally comparable  to other
 collectors with a  similar pressure drop.

          An  especially  high  performance  type of  mechanical  scrubber  is known
 as a  disintegrator, which  consists of a  casing  that  houses a series of rotating
 and stationary  bars.  Water is  injected  axially  through  the  rotor shaft and Is
 separated  into  fine droplets  by  the  high  relative velocity of rotor and stator
 bars.  The particulate-laden  gas  also enters axially and passes through the dense
 spray zone where the  particulates  are subjected  to intense bombardment by the  fine
water droplets.  Advantages of  this  scrubber are high collection efficiency for
 small particulates and  low  space  requirements.  The principle disadvantage is  its
 very  large power requirement.

                                    38

-------
          Disintegrators have been employed on  blast and electric  furnaces
as well as Koppers-Totzek coal gasifiers.   Due  to close clearances,  disintegrators
are sometimes subject to particulate buildup problems if the inlet dust  loading
is high.  Then, they are usually preceded  by a  cyclone or wet scrubber to reduce
the particulate loading.

A.2.k     Packed Bed Scrubbers

          A tower packed with irregularly  shaped objects which resist corrosion
may be used for dust and mist collection as well as for gas  absorption.   The
packed bed may be held in place (fixed), free to move (fluid), or covered with
water  (flooded).  The irrigating liquid serves  to wet, dissolve,  and/or wash the
entrained particulates from the bed.  This design generally  works  best for liquid
particulates, since solid particles sometimes tend to plug up the bed.  Collection
efficiencies are typically about 90 percent for particulates of 2 micrometers and
larger.

4.2.5     Orifice Scrubbers

          This collector atomizes the liquid and forms aerosols by generating high
velocities through restricted openings.  The gas stream comes into contact with
scrubbing liquid at the entrance to a constriction.  Liquid is picked up and
carried into the restriction where greater 1iquid-particulate interaction occurs,
resulting in high frequency impact ion.  Upon leaving the restriction, most water
droplets disengage from the gas stream by gravity since the gas velocity is greatly
reduced downstream from the restriction.  Small droplets are subsequently removed
by centrifugal force and impingement.  The pressure drop for orifice type scrubbers
typically ranges from 0.75 to 1.5 k Pa.  Collection efficiencies generally approach
90% for particles 2 micrometers or larger.

k.2.6     Impingement Scrubbers

          This type of unit produces droplets as the gas passes through wetted
perforated plates and impinges on baffles.  Here the intention is to expand the
surface area of the liquid through use of the gas stream's kinetic energy.  One
or more impingement baffle stages can be employed, depending on the degree of
particulate removal required.  A water spray zone  is often employed below the
bottom plate to cool and humidify the gases and to remove the larger particulate
matter.  The efficiency and pressure drop of impingement scrubbers are generally
comparable to  the orifice type.

A.2.7     Venturi Scrubbers

          This type of scrubber employs a venturi-shaped  constriction and throat
velocities considerably higher  than  those experienced with  the orifice type.
The high gas velocities at the  throat atomize  the  scrubbing  liquid and the  tur-
bulence created  leads to increasingly high collection efficiencies at  increasing
energy inputs.  Liquid  can be supplied  at or ahead of  the throat  through  piping
or jets.  The  collection mechanism  of the venturi  is primarily impaction.   As  with
wet collectors  in general, the  collection efficiency  increases with  higher  pressure
drops.  Different pressure drops  are  achieved  by designing  for varied gas

                                     39

-------
velocities in the throat.  Some venturi scrubbers are manufactured with adjustable
throats allowing a range of pressure drops for a given air volume.  The collection
efficiency of the venturi scrubber can generally be considered highest of the
wet collectors.

          High energy venturi scrubbers are capable of fine particulate control
with an efficiency close to that of fabric filters.  As discussed previously,
these high energy scrubbers utilize inert!al impaction as the principal collection
mechanism.  The effectiveness of inertial impaction decreases markedly at  partic-
ulate sizes below 1 micrometer unless high velocity differentials are maintained
between the collecting body  (water drops) and the particulates.  The contribution
of inertial impaction could also be enhanced by decreasing the size of the
collecting bodies.  In both cases, higher energy consumption is required for
improving fine particulate collection.  Figure 9 shows a typical fractional
collection efficiency vs. particulate size for venturi scrubbers (51).  It can be
seen that the venturi  scrubber approaches 100% removal for all particulates larger
than 1.5 to 2 micrometers.

          In view of its high removal efficiencies, the venturi scrubber has
been applied to many difficult collection jobs, including the removal of sub-
micron size iron oxide fumes in the iron and steel industry.

4.2.8     Other Wet Scrubbers

          In addition to the commonly used wet scrubbers discussed previously,
there are several new scrubber types, including foam scrubbers, steam-assisted
scrubbers, and electrostatically augmented scrubbers, that are available for fine
particulate removal.   The foam scrubber, which combines gaseous absorption along
with particulate collection, should be capable of removal of even the very finest
particulates with less than twenty inches pressure drop.  On commercial instal-
lations, electrostatic augmentation has been proved to have achieved both low
energy consumption and operating cost.  Steam-assisted scrubbers use a considerable
amount of energy, but waste heat can be used when available to reduce the net
energy consumption and total operating costs.  Each of these new types achieves
high efficiency on fine particulates, at least under some conditions.  At present,
the obvious drawback of these designs is their relatively high initial cost com-
pared to the venturi  and other types of wet scrubbers, although this initial cost
differential may be offset by their lower energy consumption and operating cost.
In addition, these newer type scrubbers generally have only a limited amount of
operating and performance experience.

4.3       Electrostatic Precipi tatoj^s

          Electrostatic precipi tators for cleaning particulates from gases have
been used by industry  for over seventy years.  They have also been found to be an
efficient means to detar the gases.  Electrostatic precipitators operate by using
a high voltage, direct current to create gas ions which impart an electrical
charge to particulates by bombardment.  The charged particles are collected by
exposing them to an electric field, which causes them to migrate and deposit on
electrodes of opposite polarity.   The electrode cleaning system is dependent upon
the type of precipitator.  The conventional  dry-type precipitator collects

                                    40

-------
   99.99

    99.9
    99.8

    99.0

    95.0
    90.0
u
c
A)
    50.0
.-  20.0
u-

^  10.0

5   5.o
     1.0
     0.5
    0.2
    0.1
    0.05
    0.01
     I   I  I  I  I I  I
              T    I   I
I     I
L LlJJ
J    111 I  III
         0.01
                  0.1
                                                            1.0
                       Particle Diameter, micrometers
          FIGURE 9. Typical  Fractional Removal Efficiency of Venturi Scrubbers (51)

-------
 particulates  on  a  dry  electrode  and  removes  them periodically by mechanical
 shaking or rapping.  Dry  precipitators  are  installed  in  industries with widely
 varying gas  conditions, temperature,  and  pressure.  The  newer wet-type precipitator
 collects and  removes particulates with  a  thin,  continuous  flowing fiIm of water.
 The  operating temperatures  are generally  less than 65C.

           The resistivity of  particulates  is  a  critical  factor  in the design and
 operation of  a dry precipitator.  Resistivity refers  to  the  resistance of partic-
 ulates  to the flow of  an  electrical  current.  It is affected by the  physical
 characteristics  of the particulates,  gas  temperature, particulate composition,
 and  the concentrations of certain flue  gas  components such as water  and sulfur
 trioxide.  Particulates with  low resistivity  (below 50 Q.  m) are difficult to
 collect efficiently since they tend  to  be  loosely adhered  to the collector and are,
 therefore, easily  reentrained in the  gas  stream.  On  the other hand, if the partic-
 ulate resistivity  is too  high (above  0.2  GQ. m), the  voltage drop across the
 deposited particulate  layer becomes  so  large that the discharge electrode electron
 emission rate is caused to  drop, which  leads to a decline  in the overall collection
 performance.   Particulate resistivity is  usually a strong  function of temperature.
 The  resistivity  is generally  found to reach a maximum value around 150C, and to
 decrease continuously  as  the  temperature  is either decreased or increased.  Hot-
 side precipitators which  operate at  temperatures up to 5*0C were, therefore,
 developed for certain  applications involving high-resistivity particulates.  The
 moisture content of the gas also affects particulate  collection by altering the
 resistivity,  which decreases  with increasing concentrations of moisture.

           Particle collection efficiency  is strongly  dependent on the migration
 or drift velocity, which  is the  rate  at which particles become charged and move
 to the  collecting  electrode.   In addition to being a  strong function of the
 particle resistivity,  the migration velocity is also  directly proportional to the
 particulate size.  Figure 10  presents the collection  efficiency as a function of
 particulate size for a high-efficiency precipitator operating on a coal-fired
 power plant for  a  field test  carried  out by Southern  Research Institute (52).
 Collection efficiencies in excess of  3Q% on a mass basis were noted  for partic-
 ulates  in  the  range of 0.1 micrometers to 1 micrometer.  The overall efficiency
 of the  test was  99.6$.

          Among  the advantages of electrostatic precipitators are generally
 efficient particulate  removal, even at sub-micron sizes and low energy consumption.
 Disadvantages  include  high capital and maintenance costs,  relatively large space
 requirements,  and  greater difficulty  in maintaining design collection efficiencies
 than with many other types of control equipment.  Several parameters can signifi-
 cantly  lower  collection efficiency if they depart from design limitations,
 including gas  volume flow, dust concentration and particle size distribution,
 resistivity and  power  levels.

           In  addition  to the  more common dry precipitators, wet precipitators have
 been developed in  recent years to solve some of the problems experienced with the
 dry units.  The wet collectors utilize water or some other liquid to wash the dust
 from the collecting electrodes.  The basic principle of precipitator operation is
essentially the same for the  dry and wet systems,  with the major difference being
 in the method of removing the collected dust.

-------

-------
           Wet electrostatic  precipitators  have  an  important advantage over dry
 units  in  that the  efficient  collection  of  a wide  range of particulates  is
 enhanced.   Since the  particulates  are  generally thoroughly wetted  in this type
 precipitator, they are  preconditioned  and  are effectively charged  and precipi-
 tated,  regardless  of  the  degree  of resistivity  of  the material.  Since  the resis-
 tivity  of the water film  is  low, the particle resistivity does not affect collection
 efficiency.   In addition, the  continuous removal of collected particles by the
 flowing water fi1m eliminates  the  reentrainment problems of dry systems.  The prin-
 cipal  problems of  wet precipi tators involve scaling and corrosion and the need to
 handle  a  liquid slurry, as compared to  the dry  solid waste that results from dry
 electrostatic precipitation.   There are presently  more than 30 wet precipitators
 in  commercial operation.

           Electrostatic precipitators,  in  general, exhibit high collection
 efficiency in the  fine  particulate range,  and low  pressure drop under normal
 operating conditions.   Precipitators can handle relatively large gas flows and
 operate in a wide  range of temperatures and pressures.  However, they also have
 major  disadvantages such  as  high Initial cost and  little adaptability to changing
 process conditions.

           Research-Cottrel1, under an EPA  contract, has demonstrated the ability
 of  an electrostatic precipitator to generate stable corona at temperatures up to
 1093C  and pressures  up to 51.5 MPa.  Furthermore, their results indicate that con-
 ditions for electrostatic precipitation may actually become more favorable as
 temperature and pressure  increase  (53)-  Collection efficiencies and operational
 reliability need to be  determined  under these  conditions, however, before pre-
 cfpftators can be  considered for commercial use under these conditions.

 4.4       SURFACE  FILTERS

           Surface  filters are  generally classified as fabric or porous medium
 filters.   These control devices use either fabric or porous material, including
 natural or synthetic fibers, ceramic fabrics, and porous metals or ceramics as the
 filter  medium.  In  general, particulates are initially captured and retained on the
 filter  medium by direct Interception, inertial  impact I on, diffusion, electrostatic
 attraction, and gravitational  settling.  Once a filter cake is accumulated on the
 upstream  side, further collection  is accomplished by cake sieving as well as the
 above mechanisms.    In large part,  the filter cake  is a principal mode of partfc-
 ulate separation,  and Its growth is allowed to  continue until the pressure drop
 increases  to a specified  value.  Cleaning methods  include fabric flexing, reverse
 air flow  through the medium, and solvent washing.

 A.A.I      Fabric Medium Filters

           One of the oldest and most widely used techniques for removing partic-
 ulates  from a gas  stream  is the use of  fabric filters.  The baghouse design is
 very commonly used, and is Inherently highly efficient even for small particulates.
 However, commercially available baghouses are not suitable for use at high temp-
eratures.   A number of high temperature resistant ceramic fabrics  have become
 commercially available.   Due to the lack of a suitable high temperature, inorganic
 fiber lubricant  needed for the fiber-to-fiber abrasions, many of these developed

                                    kk

-------
ceramic fabrics are presently unsuitable for filtration purposes.   Still,  ceramic
fabric filters offer a potentially promising solution to the problem of controlling
particulates in the high-temperature,  high-pressure environment.    Typical frac-
tional collection efficiencies as a function of particulate size  for fabric filters
are shown in Figure 11.

4.4.1.1   Baghouses   There are two major types of baghouses.   Envelope-type bags
provide maximum surface area per unit  volume, but suffer from dust  bridging pro-
blems.  Tubular bags are open at one end and closed at the other, with the direction
of filtering being either inside-out or outside-in.  Tubular filter bags are often
sewn together to form multibag systems.

          Different gas characteristics require different filter  media for proper
operation.  Two basic types of material are commonly used in baghouses, con-
sisting of woven and felted fabrics.  Woven fabric acts as a support on which a
layer of dust is deposited to form a microporous layer capable of removing addi-
tional particulates by   sieving and other basic filtration mechanisms.  Felted
fabrics are complex, labyrinthine masses of randomly oriented fine  fibers.  The
relative thickness provides the advantages of maximum particulate impingement and
changes of direction of flow to entrap fine particulates.  Felted fabric filters
usually can be operated with higher air to cloth ratio than woven fabric filters.

          Baghouses have been applied to many industries.  In the case of the
electric power utility industry, increasing attention is being paid to baghouses,
though only a few plants presently employ them.  With increasing emphasis on
burning low sulfur coal^which often results in reduced  efficiencies of
electrostatic  orecipitators, baghouses have become a possible alternate for partic-
ulate control.  The most recent installations of baghouse filters at the Nucla
Station of Colorado-Ute and the Sundbury Station of Pennsylvania Power and Light
have been proven to be successful (54)  (55).  Extensive tests performed at the
Nucla Station show overall collection efficiencies greater than 99-9% with exit
grain loadings of  less than  1.1 cg/m^.  The cleaning part of the operational cycle
contributes most to the emissions.  The baghouse was found to be very efficient  in
collecting submicron particulates.

          The advantages of baghouse filters  include very high collection effi-
ciencies, even for sub-micron particles, relatively  low energy use and pressure
drop  (typically less than 7-5 k Pa), and particles collected in dry form, which
simplifies ultimate waste disposal.  Disadvantages include large space  require-
ments, high initial costs, and proven  temperatures limited to about 290C.

4.4.1.2   Ceramic  Fabric Filters    Filtration by conventional fabrics has proven
to be highly efficient in controlling  particulates at  low  temperature and pressure.
To achieve the same level of collection efficiency at high temperature and pressure,
ceramic fabric filters appear to be promising.  Advantages of ceramic  fibers over
conventional fibers are as follows:   (1) a number  of commercially available cera-
mic fibers have been found to be functional  in high-temperature environments; and
(2) ceramic fibers have finer diameters than  conventional  fibers, usually about
3 micrometers for  ceramics vs.  10 to 20 micrometers  for conventional  fibers.  By
examining those particulate  removal mechanisms which apply to fabric  filtration, the
increase  in removal efficiency by the  use of  smaller diameter fibers  should

-------
5
     99.99


     9:1

     99.0


     95.0

     90.0
"o    50.0
20.0

10.0

 5.0


 1.0
 0.5
 0.2
 O.i
 0.05  -

 0.01
     0.0}
                   I    I
I I I
                       A   I  I I  I  I I I
           I     I   I  I  I f I I
                                     0.1
                                  Particle  Diameter, micrometers
I    f    I
                                                                                 5.0
             FIGURE  II.   Typical  Fractional  Removal  Efficiency of Fabric  Filters  (51)

-------
compensate for the reduction of efficiency caused by high temperature and
pressure.  Nevertheless, difficulties with the filter house and cleaning mecha-
nisms have to be solved before ceramic filters can be commercialized.   Currently,
Acurex Aerotherm, under EPA contract, is  investigating the suitability of commer-
cially-available ceramic fabrics for high temperature filtration.   Preliminary
results from their ambient tests confirm that some of the available ceramic mat-
erials appear to have good filtration properties.  It was also concluded that
innovative cleaning and media support techniques  can be designed to be com-
patible with the special properties of ceramic media (5$).

4.4.2     Porous Medium Filters

          Porous materials suitable as filter media are commercially available
for filtering particulates from gases at high temperatures.  Two categories of
porous medium filters have been under development, including porous ceramic and
metal filters.

4.4.2.1   Porous Ceramic Filters   Recently, efforts have been devoted to the
development of porous ceramics for high temperature and high pressure filtration.
These materials offer the advantages of high melting temperature and mechanical
strength, low thermal expansion, and compatabi 1 ity with corrosive environments.

          In a two dimensional representation, ceramic membrane filters may be
thought of as a uniform solid phase interspersed with a fluid phase (holes), and
are generally characterized by low porosity as compared to fiber filters.  In a
simplified view, membrane filters work because the membrane pore diameter  is
smaller than that of the particulates to be collected; the particulates are thus
prevented from passing through the membrane.

          Westinghouse Research Laboratory, under funding by EPA,  is currently
performing a large scale effort directed towards applying ceramic  filters  to partic-
culate removal from gas streams at high temperatures  (57).  This experimental work
has been carried out in two phases.  Phase 1  investigated the development  and use
of membrane type ceramic materials as hot gas filters.  The first  part of  the effort
was limited in success due to the fragile nature of the ceramic membranes  formed
and because it proved to be difficult to control the pore size distribution of  .
the filters.  It was concluded that ceramic membrane  filters will  not be suitable
for commercial use. in large electric power generating applications.

          Phase  II examined several commercially available types of ceramic mat-
erials,  including porous thick-walled filters and thin-walled  (0.2 mm) monolithic
honeycomb structures.  Test results applied to both categories of  materials are:
(1) filters of this nature are suitable for operation at  temperatures exceeding any
current  coal conversion process requirements;  (2) the filters generally exhibit
virtually 100% effective  removal of a submicron  test  particulate;  and  (3)  cleaning
methods  can be devised  to allow continuous operation.  One of the  most promising
materials tested was a  ceramic  crossflow monolith produced by 3M Company  under the
trademark of ThermaComb.  Test  results under  both ambient and high temperatures
(730C)  for ThermaComb  are as  follows:   (1) effective cleaning  for continuous
operation can be achieved relatively easily with back washing pulses;  (2)  the

-------
 filters  have relatively  low pressure  drops  at  moderate  face  velocities;  (3)  the
 filters  have very  high surface  area to volume  ratios, which  makes efficient  use of
 pressure vessel  containments; and   (k)  a  system based on  the use of  this type of
 filter material  appears  to be viable  as a hot  gas  cleaning system.

 k. k. 2.2    Porous Metal FiIter    Porous  metal media,  in  general, have the advantage
 of durability,  in  addition to high  filtration  efficiency.  However,  porous metal
 filters  require  high  pressure drop  and have low face velocities.

           Brunswick Corporation had under development in  recent years a porous
 metal  filter for use  in  high temperature, high pressure application.  This device
 relies on  depth  filtration, which occurs  by particulate capture on individual
 metal  fibers.  This approach reduces  the  pressure  drop  and enhances  throughput.
 As  a  result  of  laboratory  tests at  ambient  conditions,  99% removal efficiency is
 claimed  for  0.3  to 0.5 micrometer   particulates at air  velocities of 36.6 m/min,
 or about 25  times  normal baghouse air-to-cloth ratios.  While  the filter could be
 adequately cleaned over  a  short time  period, the pressure drop across the filter
 was  found  to gradually increase to  unacceptable levels  as the  number of filtration-
 cleaning cycles  increased.   A suitable  cleaning method  to enable operations  over
 long  periods of  time  was not identified in  the development program.   Due to  the
 lack  of  a  viable cleaning  method, Brunswick has discontinued this development work
 on  their Brunsmet  filter  (58).

 A.5        GRANULAR BED FILTERS

           A  granular  bed filter employs a stationary or moving bed of granules,
 sand,  gravel, coke or sintered  material,  as  the filter medium.  In order to  main-
 tain  a steady operating performance,  a  granular bed filter needs to  remove the
 collected  particulates from the collecting  surface.  Several different designs are
 reported in  the technical  literature.   In general, they may  be classified as con-
 tinuously  moving,  intermittently moving,  or  fixed  bed filters with respect to the
 cleaning methods.

           Granular bed filters  are  a  promising technique for high temperature and
 pressure operation.   They have  the  advantages  of being able  to use either inert or
 sulfur-absorbent material,  and  being  able to accomodate high face velocities while
 incurring  a  moderate  pressure drop.   In addition,  they have  the potential to achieve
 the same high collection efficiency as  fiber filters while being somewhat easier
 to clean.  The collection mechanism is similar to  that of fiber filters, with
 impaction  predominating and  particulates  being collected in  the interstices of the
 filter.  After the initial  collection at  the filter surface  produces a filter cake,
 further  collection is essentially accomplished  by  cake sieving.  Most recently,
 granular bed filters  have  received  increased attention,  and  a number of research
 projects are underway to further develop  these  systems.   Each class of granular
 bed filters  is discussed below.

 k.5.1      Fixed Bed Filter

           The fixed bed filter  removes collected particulates periodically by
either a back flow of clean  gas  or mechanical  shaking.   The bed material itself is

-------
not moved or replaced.   The Ducon and Rexnord Filters are examples of this type,
and are discussed below.

/j.5.1.1   Ducon Fi 1 ter   The Ducon filter employs screens to retain the granular
bed particles while permitting removal of the collected particulates by a blow
back technique.  The arrangement and operation of the unit is shown schematically
in Figure 12.

          Performance correlations developed from data on cat-cracker regenerator
off-gases  show that the Ducon filter achieved overall collection efficiencies
ranging from 85% to 98% at temperatures around 480C (60)-  Other tests on the
Ducon filter have been conducted by Westinghouse, Bureau of Mines, and IGT, with
efficiency of 33% reported (47).  However, no specific grade efficiency data are
available.  Exxon Research and Engineering Company is currently evaluating the
Ducon filter for reducing emissions of particulates in the flue gas from pres-
surized fluidized bed coal -.ombustion at high temperature and pressure (530C,
IMPa)  (61).   Stable operation for up to 2k hours was demonstrated with a collection
efficiency of about 90%.   Principal problem areas involve filter cleaning at
system conditions, filter plugging and loss of filter media.

4.5.1.2   Rexnord FiIter   The Rexnord gravel bed filter consists of granular
filter beds in conjunction with a mechanical collector,as shown schematically in
Figure 13-  After the raw gas enters the unit, it initially passes through a
cyclone collector which separates the large coarse particles.  The precleaned gas
rises from the cyclone through a vortex tube and then flows downward through
annular gravel beds so that the remaining fine particulates are deposited on the
quartz grains and in the interstices of the bed.  To clean the filter, the unit
uses a raked-shaped double-arm stirring device to loosen the filter cake, and a
backwash of clean air to blow the particlesout of the bed.

          The results of tests performed on a full scale  Rexnord  filter system used
for controlling particulate emissions from a Portland cement plant clinker cooler
were reported by J.D. McCain  (62)-  The system consisted of eight  filter units.
During normal operation, seven of these eight units were on-line  in the forward
flow direction, with one being cleaned and  renewed by backflushing with heated
ambient air.  These tests were performed at temperatures  from 74C to  130C and at
atmospheric pressure.  The collection efficiency of the Rexnord filter determined
by standard mass train techniques on a source producing particulates having a
mass median diameter of about 200 micrometers ranged from 35% to 98% during three
days of testing, throughout which the collector was not operating  in an optimum
mode.  Overall efficiencies determined from cascade  impactor data during a second
two-day test series under an  improved operating mode were found to be 99-**% and
33.7%.  Fractional collection efficiencies as determined  using the cascade impactors
are shown in Figure 14.  The poor collection efficiency of particulates less than
1 micrometer in size is due to the attrition of particulates which occurs  during
backwash  and raking cycles.

          Gravel bed filters, designed to treat  127  m^/min,  are commercially
available.  The unit utilizes mild steel  construction  and can operate up  to  370C.
Rexnord is currently seeking  funding  to develop a filter  for  high  temperature  and
pressure  operation,(i.e.   870C,  1 M  Pa).

                                    49

-------
                        Collection Cycle
         Dust Laden

         Gas
vn
o
 Cleaning Cycle
Collected1*

Dust
                                                                -.
                                                                Y':1
                                                                fjw*j;
                                                                       O
                                                                       o<
                                                                       ov
                                                                     Purge  Gas
                   t
                  '
                                                                                '
                                                                                * 1*
                        Fluidized

                        Granules
                      FIGURE  12.   Schematic Illustration of Ducon  Filter (59)

-------

'
*
BACKFLUSH r|_
DUCT H
^
ft f ft M f A 














	 1
1
^
















 *^,
^^i












t



X n


x^.









J=^
j
nit

 , ,


1L
Jun
-^









J
3

II




LL,
n





\
1


^
'ft



1










V

:^3
UK
HUM



^
alUJ
iUilll


i i -i i ii

J

"
^/

	 	 IV MIX I. UMIVC
^RAKE MEGHAN
r.o AVFI RFn


	 VOKTEX TUBE
CM -rr p THAMF

 	 SCREEN
CLEAN AIR
PU A MRFR


\frto-rcv TIIQC
   V\JM't.A lUDt
PRIMARY
COLLECTOR

rvn TtMF
                                        SEPARATOR
                                         COUNTERWEIGHTED
                                         VALVE
FIGURE  13.  Rexnord  Gravel Bed Filter  (59)

-------
     100
                  f   I   Tj
o
c

0)
o
u
0)
o
o
    -100
    -200
I	 (    I   I
         0.5
                                   Aerodynamic Diameter, micrometers
                                                                               I    f   r
                                                                                        10
         FIGURE 14.   Typical  Collection Efficiencies for Rexnord Filter (62)

-------
4.5.2     Intermittently Moving Bed Filter

          The intermittently moving bed filter uses  periodic cleaning to remove
filter cake along with a portion of the granular medium in the bed.   The Squires
panel bed filter is representative of this type and is discussed below.

4.5.2.1   Panel  Bed Fi 1 ter   The panel bed filter was developed at City  College of
New York under the direction of Prof. A.M. Squires.   This filter uses two grades  of
sand which form two vertical walls contained in an open, louvered framework,  as
illustrated in Figure 15.  Fine sand exposed to a dirty gas provides the surface
for filter cake  formation.  Coarse sand, separated from the fine sand by a series
of closely spaced horizontal plates, acts as a baffle to contain the fine sand.
Cleaning is done by a puffback technique,  which consists of interrupting the flow
of dirty gas to be filtered, and then passing a clean purge gas in the reverse
direction across the panel.  This reverse surge creates a mass movement  of the sand
bed toward the gas-entry face  causing sand from each gas-entry surface  to spill.
This spilled sand carries with it the deposited particles accumulated during the
filtration step prior to cleaning.  The expelled sand is immediately replaced by
downward movement of fresh sand from the overhead hopper.

          In laboratory tests with redispersed power station fly-ash, the panel
bed was found to afford high filtration efficiencies  (beyond 99-95% removal of
fly ash) at 150C, with practicable gas throughput and pressure drop.  Efficiency in
these tests for removal  of sub-micron particulates is estimated at beyond 99%.
This estimate was supported by separate experiments employing a monodisperse aerosol
of 1.1 micrometer  particulates (63).   In a high temperature application (5^0C)  at
the Morgantown Energy Research Center, overall collection efficiencies in the range
of 95-96% have been recorded.  No specific grade efficiency data are available (5).
The panel bed filter has gone through the bench-scale stage of development.  There
is a need to further test the system in more rigorous conditions to ensure the
reliability of the system.  However, no future work  is scheduled, as further
support is being sought.

4.5.3     Moving Bed Fi Her

          The moving bed filter removes collected particulates by circulating part
of its bed media continuously.  The Combustion Power Company has developed a cross
flow moving pebble bed filter which  is  discussed in  the following section.

4.5.3.1   Combustion Power Company Filter   The Combustion Power Company's moving
bed filter and its media circulation loop are  illustrated schematically  in Figure 16.
The unit houses an annular filter element which retains the granular filter media
between two vertical, louvered screens.  As the raw  gas passes through the filtering
media, internal  impaction  removes entrained particulates from  the gas which then
emerges into an interior collection plenum.  The collection of particulates is
accomplished by depth filtration, which occurs by particulate capture on individual
granules in the bed.  To avoid particulate saturation, the media  is continuously
recirculated and cleaned,using a  gravity-bed recirculation system.

          Preliminary data from an ongoing experiment on moving bed  filtration have
been presented by G.L. Wade  (64).  The  results of a  linear  regression analysis of
                                     53

-------
DIRTY
GAS
               SPACE SPACE
                FOR   FOR
               FINE  COARSE
               SOLID SOLID
CLEAN
GAS
         FIGURE 15.  CROSS SECTION OF PANEL BED FILTER (63)".

                   54

-------
      PARTICLE-LADEN AIR
         (TO BAGHOUSE)
                                     DISENGAGEMENT VESSEL
                                      FLUIDIZED BED
                                      FLUIDIZING AIR
                                       MEDIA RETURN PIPE
DIRTY AIR
                                       FRONT PANEL
                                       FILTER BED

                                       OUTLET PANEL
                                       MEDIA LIFT PIPE

                                       MEDIA OUTLET PIPE
                                           TRANSPORT AIR
                      EJECTOR AIR
FIGURE  16.   Combustion Power Company's  Moving  Bed  Filter  (6*0
                           55

-------
 collection efficiencies  vs.  mean  diameter of  inlet participates  for 45 cold flow
 tests  is  shown  in  Figure  17-  This  regression  analysis  indicates that reasonable
 operating conditions  may  be  specified  for which  total efficiency in excess of 35%
 is  readily attainable.

          Commercial  devices, restricted to temperatures below about 430C and to
 pressures near  atmospheric,  have  been  available  for  a few years.  Combustion Power
 Company has  fabricated a  unit to  operate at 650  C with  a pressure of a 1 MPa and
 treat  a 11.8 rWmin gas  stream  at a  grain loading of 9.2 g/mj or less.  Further
 testing is needed  to  determine  the  collection  efficiencies of this unit at high
 temperature  and pressure.

 4.6       OTHER COLLECTION DEVICES

          In addition to  the relatively advanced control technologies previously
 discussed, several novel  devices  at  a  relatively early  stage of development have
 undergone some  preliminary experimental testing.  Several of these novel devices
 are  discussed briefly below.

 4.6.1     A.P.T. Dry  Scrubber

          Air Pollution Technology  (APT), Inc., has  developed a dry scrubber system
 which  can be used  at  high temperature  and pressure for  the collection of fine
 particulates  on  larger particulates  (65).  The relatively large particulates used
 as collection centers for fine particulates can then be cleaned and recycled.

          Particulate collection  is  primarily by inertial impaction, and to a
 lesser extent,by diffusion for smaller particulates.  APT, supported by EPA, has
 done a preliminary experimental  evaluation of  the dry scrubber system.  Some
 preliminary  conclusions from this work are as  follows:

          o   The experimental data on  the primary collection efficiency
              of the system agree well with predictions  based on a math-
              ematical model which was  first developed for wet scrubbers.
              The data are shown in Figure 18.

          o   The system has  the same primary collection efficiency/power
              relationship as a venturi  type wet scrubber.

          o   The overall  efficiency of the system depends on the re-entrain-
              rnent characteristics of the specific system, in addition to the
              primary efficiency.

4.6.2     Molten Salt Scrubber

          Battelle Memorial  Institute  is developing a molten salt scrubbing
technique for removal of both particulates and sulfur compounds  from producer  gas.
This technique  is based on the use of a molten alkali salt as the scrubbing liquor
in a conventional venturi scrubbing apparatus.  Particulate removal  in the venturi
scrubber  is  effected by particulate  inertial   impaction onto molten  salt droplets.


                                     56

-------
    100
     90
o
c
30
c
o
o

-------
     1.0
 c
 o
 O
 IB
o

-------
As is true with any venturi  scrubber,  this  molten salt scrubber has the advantage
of a potentially high collection efficiency.   However, there are some unique
problems associated with this design:   (1)  the corrosiveness of the scrubbing
alkali medium presents a difficult material  handling problem; (2)  molten salt
decomposes at 900C and reacts with other gaseous pollutants; (3)  carryover of
molten liquid droplets; and (k) high energy requirements.

          Under DOE funding, Battelle has designed a process demonstration unit
for operation in a fully continuous mode (66).  Preliminary data show that the
overall collection efficiency of the molten salt scrubber is not as great as ex-
pected.  This reduced efficiency is largely attributable to the re-entrainment of
droplets of the scrubbing liquor.  This project,however, has been discontinued
recently.  A final report is being prepared by Battelle.  There are no plans for
further research.

1.6.3     Electrofluidlzed Bed Collector

          An electrofluidized bed collector was developed by K. Zahedi and
H.R. Melcher (6?) for the removal of submicron partlculates.  In this device, an
electric field is imposed on a fluidized bed of particles by means of external
electrodes, thus causing a polarization of the bed particles, which in turn acts
as the collecting electrode.  This device is designed to collect fine particulates
with high efficiency, and is reported to have potential applicability in the high
temperature and pressure domain.

          There are three general stages to the gas cleaning process  in this
device, as in the case of a conventional electrostatic precipitator.  First,
the particulates are transferred from the gas to the surface of the bed particles.
Second, agglomerates of the captured particulates are formed as a  result of adhesion
on the bed particles.  Finally, the agglomerates are  removed by means of the out-
flow of pollutant in a fluidized state.

          The electrofluidized bed is still largely in the  research stage.  Pre-
liminary results show promise of its being extremely effective  in  removing fine
particulates at temperatures up to at least 200C and at atmospheric  pressure.

A.6.4     Charged Filter

          This device, which combines the principles of electrostatic pre-
cipitation and fabric  filtration  has been developed by American  Precision
Industries,  Inc. and  is called the "Apitron".   It provides  high efficiency per-
formance, and may be applicable to high  temperature flue gas systems  with the
development of fabrics suited  to high temperature operation.

          The Apitron device  is essentially a wire-tube precipitator  with a  con-
centric bag  filter and charged screen.   The first stage of  this device  performs
as a  high velocity precipitator which charges all particulates, while removing
about 90% of the  total.  The  fine  particulate matter  is then  collected  by the
fabric  filter  in  a manner that permits velocities four  to  five  times  those  used
in conventional  baghouses.
                                     59

-------
          Recently the EPA has conducted an extensive evaluation of the Apitron
system.  Efficiencies of 99.99% for 0.1 micrometer  particulates have been ob-
tained in the tests.  Figure 19 shows a typical sub-micron collection efficiency
curve  (68).

          In addition to its efficient fine particulate collection, the Apitron
has the advantage that its electrostatic charging promotes the development of a
more loosely packed, porous dust layer on the fabric, permitting air cloth ratios
to be increased by a factor of 2 to 5.  This results in an overall  equipment size
reduction of similar proportions.
                                    60

-------
    90.0
    99.0
    99.9
c
(U
w   99.99

>
    99.999
    99.9999
           10
             -2
                                                           I    I   I  I
                 	L
<  i  i 11 11	i	i  i  t  i 111 r
                                                10
                       Particle  Diameter,  Micrometers
            FIGURE  19-   Fractional  Removal  Efficiency  of Apitron  Charged
                        Filter  (68)
                                  61

-------
                                 SECTION 5

                   APPLICABILITY OF CONTROL TECHNOLOGIES


5.1       END USES FOR PRODUCT GASES

          Product gases from coal gasifiers might be employed for a wide variety
of end uses, including industrial and utility boilers, fuel for industrial process
heat, gas turbines, and reducing or synthesis gases for a variety of industries.
Each particular end use for the product gas has different environmental regulations
and process requirements governing the allowable particulate levels in the product
gases.  For the purposes of this study, two particular end uses were selected for
consideration.  These end uses were selected so as to cover a wide range of parti-
culate removal requirements for the control devices under consideration.  The use
of product gases as a boiler fuel was selected to represent those end uses with
low to moderate particulate cleanup requirements.  On the other hand, the use of
product gases as a fuel for gas turbines was selected to represent end uses with
relatively restrictive particulate removal requirements.


5.2       ENVIRONMENTAL REGULATIONS AND PROCESS REQUIREMENTS

          Environmental regulations and process limitations on the allowable par-
ticulate contents of the product gases were identified for each of the end uses
discussed above.


5.2.1     Boiler Fuel

          For the purposes of these assessments, existing environmental regulations
for direct coal-fired boilers are assumed to apply to boilers firing coal-derived
fuel  gases.   The New Source Performance Standard established by EPA to limit
partfculate emissions  from fossil-fuel fired steam generators is ^3 nanograms per
Joule heat input (0.10 lb/10 Btu).  This is equivalent to 0.24 grams of particu-
lates per cubic meter for low-Btu fuel gas with an average heating value of
                                     62

-------
0.56 MJ/nr (150 Btu/scf),  or 0.48 grams  of participates  per  cubic meter for medium-
Btu fuel gas with a heating value of 1.12 MJ/m3  (300  Btu/scf).   Thus,  depending  on
the heat content of the gas to be used in the boiler  fuel  and  assuming that combus-
tion of the fuel gas adds  only a negligible amount  of particulates,  the particulate
removal system that is selected might have to remove  particulates down to a level
as low as 0.2A g/m3.

          Adverse health effects and haze problems  are most  acute with sub-micron
particulates, which generally account for only a very small  portion  of the total
particulate emissions.  While present EPA regulations apply  only to  the total  par-
ticulate load, future regulations may specifically  limit the emissions of the  sub-
micron size particulates.

5.2.2     Gas Turbines

          A particularly promising end use for coal-derived product  gases is as  a
fuel for gas turbines employed in combined-cycle power stations.  In a combined
cycle, electricity  is produced both by expansion of pressurized gases in a gas tur-
bine and by expansion of steam in a steam turbine.   The fuel gas would be burned
and then expanded in the gas turbine, which generates electricity and also drives
the air compressor.  The exhaust gases from the turbine can then be used to deliver
heat to a steam generator, which drives a steam turbine to produce additional  power.
Combined cycle plants are generally more energy efficient than comparable convention-
al power plants.

          The tolerance of a gas turbine to particulates  is not known with a high
degree of certainty.   In general, the turbine tolerance for particulates depends upon
size and physical and chemical properties.  Stringent specifications  for fuels to be
burned  in gas turbines have been established by various turbine manufacturers.   The
most stringent turbine tolerance established to date  is 0.003^ g/m3 for  low- Btu fuel
gas, based on an air/fuel ratio  of 8/1 with dust-free air and turbine inlet temper-
ature of 980C  (69). However,  results of theoretical  calculations made by OOE's  High
Temperature Turbine Technology Program (5) suggest a max!mum tolerance of 0.00^6 g/m
of expansion gas, or 0.0^1 g/m3 of fuel gas, with no particulates larger than 6
micrometers.  Until more data  are developed, these recommendations probably provide
the most reliable  indication of turbine particulate  tolerance.  Thus,  in these assess-
ments on control technology applicability, the particulate tolerance of  the gas  tur-
bine will be taken  as 0.0^1 grams of  particulates per cubic meter of  fuel gas, with
no particulates  larger than 6  micrometers being allowed.  It should be noted that
there are presently no environmental  regulations governing the emission  of parti-
culates from gas turbines.

5.3       PERFORMANCE  EVALUATION

          The applicability of a  particulate control  technology  for any  given end
use-gasifier combination  is determined by many  factors  such as  the  particulate  char-
acteristics, cost of  control,  and system  reliability.   However,  it  is  important to
first  determine whether a  particulate control technology  is capable of meeting  the
basic  requirements  of  particulate removal before proceeding with  more detailed
evaluations.  Thus,  in  this section,  the  applicability  of each control  technology for

                                      63

-------
the end uses under consideration are evaluated strictly on the basis of the capa-
bility of that control technology to achieve the  required degree of particulate
removal.

          A preliminary analysis was first conducted to assess the range of removal
efficiencies required for each end use, based on  the best, average, and worst-case
effluent grain loadings for each gasifier, as discussed in Section 3 and presented
in Table 13.  This was done by computing the required removal efficiency as a func-
tion of particulate loadings for each generic type of gasifier.  The results show
a quite wide range of removal efficiency is required,as shown in Table 1k .

          The removal efficiency of any particulate control technology is a strong
function of particulate size.  Thus, a meaningful applicability assessment of con-
trol technologies requires knowledge of the particulate size distribution in the
gases to be treated, along with removal efficiencies of the control technologies
as a function of particle diameter.  Typical particle size distribution curves for
each generic gasifier type are presented in Figures 1,2and 3  of Section 3-  Typical
fractional removal efficiency curves for the generic control technologies under
consideration are presented  in Section 4.  However, in most cases, these removal
efficiency curves are limited only to the range of small particulates.  This is due
to the fact that the removal efficiency is much more dependent on particle size for
the smaller particles, with most removal efficiency curves dropping off rapidly at
small particle diameters from a nearly constant value at the larger particle sizes.
For example, both fabric filters and high efficiency venturi scrubbers are capable
of complete removal of particles greater than 6 micrometers under certain operating
conditions.  However, their removal  efficiencies  degrade considerably as the parti-
cle size decreases to the sub-micron range.

          For particulates less than 6 micrometers, a graphical method was used to
calculate the mean removal efficiency as a function of particulate size distribution
for each gasifier effluent.  For the particulates greater than 6 micrometers a
representative value of removal efficiency was chosen for each generic control  de-
vice,with the exception of the conventional cyclone.  This is due to the fact that
the removal efficiency of a conventional cyclone  usually reaches a maximum at a
much larger particle size than 6 micrometers, while for most other control  devices
the removal efficiencies are nearly constant for  particles greater than 6 micrometers,
Thus, the same general  graphical  method was used  to calculate the mean removal
efficiencies of a cyclone for each gasifier effluent over the particulate size ranges
below and above 6 micrometers.

          The graphical  method employed for calculating the mean particulate removal
efficiency for particles between 0 and 6 micrometers in diameter is derived from the
following equation:
          E " ] ~ Jo  f(Dp) d(Dp)
Where E = total removal efficiency for particulates less than 6 micrometer,
      f = removal efficiency as function of Dp,
     Dp = particulate diameter, micrometers.

                                     6k

-------
                         TABLE  14   PARTICIPATE  REMOVAL EFFICIENCIES REQUIRED
                                          FOR DIFFERENT END USES

Gas i f i e r
Fixed Bed
Best Case
Average Case
Worst Case
Fluid Bed
Best Case
Average Case
Worst Case
Entrained Bed
Best Case
Average Case
Worst Case
Effluent Loading
(g/m3)
0.5
3.0
6.0
1.2
26
120
30
110
230
Requi red Removal
End Use 1
Turbine Tolerance
0.04l(g/m3)*
(before combustion)
91.80
93.63
99.29
96.58
99.84
99-97
99.86
99-96
99.98
Efficiencies (2)
End Use 2
EPA Emission Standard
0.24(g/m3)
(Low-Btu Gas)
52.0
92.0
95.9
80.0
99.1
99.8
99.2
99.8
99.9
* No particles greater  than 6 micrometers in diameter are permitted.

-------
          The graphical method of integrating this equation is illustrated by the
following example for calculating the collection efficiency of a conventional cyclone
operating on an effluent with the characteristics of the raw product gases from the
MERC fixed bed gasifier burning Illinois #6 coal.  Particulate size distribution
data for the MERC fixed bed gasifier burning Illinois #6 coal  are presented in
Figure 1  .  Also, data for the fractional removal efficiency  of a conventional high
efficiency cyclone are presented in Figure  5 .   These data were then tabulated, as
shown in Table  15 .   Figure 20 was then obtained by plotting  the last two columns
of Table  15.  The area under the curve was graphically integrated by determining
the point at which the areas above and below the line become equal.  An overall re-
moval efficiency of 36 percent for the particulates less than  6 micrometers was
thereby obtained.

          Table 16 presents a compilation of calculated removal efficiencies for
particulates less than 6 micrometers for each combination of generic control device
and gasifier effluent.  These results were obtained by the graphical method described
above.  Results are presented for the best, average, and worst-cases of particle
size distribution as  discussed in Section 3 and presented in Figures  ' through 3.
It should be noted that sufficient data were not available for several of the more
advanced and recently developed control technologies discussed in Section k to
estimate their overall particulate collection efficiencies.  Results for these devices
are, therefore, not presented in Table 16 .

          With the overall removal  efficiency of each generic  control device thus
far developed, the applicability assessments were then carried out on the basis of
the estimated particulate loadings from each gasifier, as presented in Table  13  .
The amount of particulates not removed, and consequently that  remain in the product
gases, were then calculated.  The results for each generic control device under
consideration are presented In Tables  17 through  22  .  The applicability was deter-
mined by comparing the amount of particulates not removed to the maximum allowable
amount of particulates for each end use.   These results are shown in the last two
columns of Tables  17 through  22  .

          Conclusions drawn from these results are discussed below separately for
End Use 1 (combined cycle fuel gas) and End Use 2 (conventional boiler fuel gas).
As for End Use 1, the very restrictive requirement of removing all particles larger
than 6 micrometers has limited the potential control devices to fabric filters, a
high efficiency venturi scrubber, and the Aerodyne rotary flow cyclone.  Among these
three control devices, the fabric filter was found to be the only device capable of
achieving the required product gas purity for End Use 1 for all gasifier effluents.
This requirement is to reduce the concentration of particulates with size less than
6 micrometers to below 0.0^1 g/m^ for all cases, as shown in Table   '** .  The high
efficiency venturi scrubber is applicable for End Use 1 for all gasifier effluents,
except for the worst-case fluid bed gasifier, as shown in Table  19  .  However, with
a high efficiency cyclone upstream as a scalping device, the venturi scrubber is
capable of achieving  this requirement for the worst-case fluid bed, based on the
assumption that the particulate size distribution for particulates less than 6
micrometers remains unchanged after passing through the cyclone.  The Aerodyne rotary
flow cyclone Is found to be inapplicable for the average and worst-case fluid bed


                                    66

-------
                                   TABLE 15
                    COLLECTION  EFFICIENCY OF HIGH EFFICIENCY CYCLONE
                      FOR  PARTICULATES FROM MERC FIXED BED GASIFIER
                             EMPLOYING ILLINOIS #6 COAL
Particulate Size(Dp)
  micrometers

       16

       15
       14
       13
       12
       11
       10
        9
        8
        7
        6
        5
        4
        3
        2
        1
 Amount < Dp,*
\ by weight
     3
     2.5
     2.0
     1.5
     1.3
     1
     0.7
     o.4
     0.3
     0.25
     0.11
     0.07
     0.02
     0.01
     0.001
  Cyclone
Efficiency,**
  >99
    99
    99
    98.5
    98
    97
    96
    95
    94
    92
    90
    87
    83
    77
    68
    53
       *   Cumulative size distribution data for MERC gasifier.
       **  Collection efficiency for particles with  diameter of Dp.
                                      67

-------
c
O
u
V
O
o
50
    30
                                 36% Removal
        Dp=2  micrometers

           Dp=3
                                   Op=5
                                                   I
       0         0.05         0.1                       0.2

             Amount less than stated size,  % by weight

      FIGURE 20.   Graphical  Procedure for Estimating Overall  Collection
                  Efficiency For Particulates up to 6 Micrometers  in
                  D i ame te r
                               69

-------
                    TA3LE 16

OVERALL PARTICULATE REMOVAL EFFICIENCIES OF GENERIC
 CONTROL TECHNOLOGIES FOR TYPICAL GASIFIER OUTPUTS

o
6
<6
>6
<6
>6
*6
>6
<6
>6
<6
>6
<6
>6
<6
>6
^6
?6
% Participate Removal Vs. Particulate Size
Conventional
Cyclone
86
99
86
93
86
97
82
98.8
79
98.7
76
98.5
3i
98.6
83
98.8
82
98.9
Rotary
Cyclone
98.5
100
98.8
100
99
100
90
100
91.5
100
93
100
96.5
100
97.3
100
98
100
1
Venturi
Scrubber
99.93
100
99. 9*
100
93.95
100
97
100
97.8
100
98.5
100
99.7
100
99.83
100
99.95
100
Fabric
Filter
99.99
100
93.99
100
99.99
100
99.2
100
99. A
100
99.6
100
99. 9*
100
99.97
100
99.99
100
E.S.P.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
93
99.8
98.4
99.8
98.8
99.8
99
99.3
99.2
99.8
99.4
99.8
Granular .
Bed Filter
94.6
95
94.7
95
94.7
95
94.4
95
94.5
95
94.5
95
94.6
95
94.7
95
94.7
95

-------
                TABLE 1?  THE APPLICABILITY OF CONVENTIONAL  CYCLONE FOR GASIFICATION PROCESSES

o
0)
ra
o
6
<6
>6
<6
>6
*6
>6
<6
>6
<6
>6
<6
>6
<6
>6
<6
>s
?
0.3
99.7
5.4
94.6
10.5
89.5
i
99
3
37
5
95
0.5
99.5
0.7
99.3
0.8
99.2
Percent
Partlculate
Removed
%
86
99
86
98
86
97
q?
95.3
?q
q3.7
76
98.5
84
98.6
81
98.3
82
93.9
Partlculate
Not Removed
g/m3
0.0003
0.0046
0.0021
0.0557
O.o86q
0.1603
n r>Q23
O.OH7
o.i 6fcq
0.127^
1.44
1.65
0.023
0.419
o.m
1.312
0.339
2.512
Applicability
for Combined
Cycle End Use a
N.A.
N.A.
N.A.
N.A.
N A
N.A.
N.A.
N.A.
N.A.
Applicability
for Boiler
Fuel End Useb
Appl i cab le
ADD! i cable
H.A.
Annl 5/*aKlo
N.A.
N.A.
N.A.
N.A.
N.A.
3 Based on maximum allowable
b Based on maximum allowable
particulate load
participate load
of 0.041 g/m3.
of 0.24 g/m3.

-------
                  TABLE 18  THE APPLICABILITY OF A  ROTARY CYCLONE FOR GASIFICATION PROCESSES

o
V
eo
T3
s
Lu
t3
(U
CO
tJ
3
U.
o

c
UJ
Best
Average
Worst
Best
Average
Worst
Best
Average
Worst
Participate
Loading
g/m3
(gr/scf)
0.5
(0.22)
3.0
(1.3D
6.0
(2.62)
1 .2
(0.52)
26.0
(11.36)
120.00
(52. M)
30.0
(13.11)
no.o
(43.07)
230.0
(100.51)
Percent
Part iculate
Distribution
Size
(>*m)
<6
>6
<6
>6
<6
>6
*6
>6
<6
>6
<6
>6
<6
>6
<6
>6
<6
?6
?
0.3
99.7
5.4
94.6
10.5
89.5
l
99
3
97
5
95
0.5
99.5
0 7
99.3
0.8
99.2
Percent
Particulate
Removed
%
93.5
100
qft 3
100
99
100
qn
100
91. 5
100
93
100
qfc.q
100
97.1
100
98
100
Particulate
Not Removed
g/m3
0.00002
0
n rn93
0
0.0069
n
0.0012
q
n.nfiW*
0
0.421
0
n nnt;i
0
p.n?i
0
0.036
0
Applicability
for Combined
Cycle End Use a
Appl i cable
Appl i cable
Appl i cabl e
Appl \ cab le
N.A.
N.A.
Appl icable
Appl icable
Appl icable
Applicability
for Boiler
Fuel End Useb
Applicable
ADP! i cable
Ann] 1 r.ah 1 p
Appl i cab le
Annl i cah IP
N.A.
Appl icable
Appl i cable
*\ppl icable
a Based on maximum  allowable  partlculate  load of 0.041
b Based on maximum  allowable  particulate  load of 0.24 g/m.

-------
                  TABLE 19  THE APPLICABILITY OF A VENTURI  SCRUBBER FOR  GASIFICATION  PROCESSES

o
V
CO
o
0)
X
IZ
o
6
<6
>6
<6
>6
<6
>6
<6
>6
<6
>6
<6
>6
<6
>6
<6
>6
X
0.3
99.7
5,4
94.6
10.5
8q.q
l
39
3
97
5
95
0.5
99.5
0.7
99.3
0.8
39.2
Percent
Particulate
Removed
%
99.93
100
qq.q4
100
99.95
inn
q7
100
97.8
100
98. 5
100
99.7
100
99.83
100
99.95
100
Particulate
Not Removed
g/m3
0
0
o.onoi
0
0.0002
Q
n nnf)1^
0
0.017
0
0 Qto
0
o.oooq
0
o.oou
0
0.0009
0
Applicability
for Combined
Cycle End Use a
Appl icable
Appl icable
Appl 1 cab 1 e
Appl icable
Appl i cabl e

N.A.
Appl icable
Appl icable
ADD! icable
Appl icabi lity
for Boiler
Fuel End Useb
Appl icable
Appl icab 1e
Appl ! rah 1 p
Appl i cabl e
App 1 i r.ah 1 p
Appl icable
Appl icabl e
APD! icable
Appl icable
tsJ
   a Based on maximum allowable
   b Based on maximum allowable
particulate load of 0.041  g/m^.
particulate load of 0.24 g/m3.

-------
TABLE 20  THE APPLICABILITY OF AN ELECTROSTATIC PRECIPITATOR FOR GASIFICATION  PROCESSES

TD
0)
CO
TJ
(1)
X
IZ
o
0)
OQ
T3
3
LL.
"O
0>
co
T3

4->
C
ID
Best
Average
Worst
Best
Average
Worst
Best
Average
Worst
Particulate
Loading
g/m3
(gr/scf)
0.5
(0.22)
3.0
(1.30
6.0
(2.62)
1 .2
(0.52)
26.0
(11.36)
120.00
(52.44)
30.0
(13-11)
110.0
(43.07)
230.0
(100.51)
Percent
Particulate
Distribution
Size
(Xm)
<6
>6
<6
>6
<6
>6
^6
>6
<6
>6
<6
?6
<6
^6
<6
>6
<6
>6
?
0.3
99.7
5. *
9^.6
10.5
39.5
1
qq
3
97
5
95
0.5
99.5
0.7
99.3
0.8
99.2
Percent
Particulate
Removed
%
__
--

__
--
--
q8
97 8
98.4
99. 8
9S.8
99.3
09
99.8
99.?
99.3
99.4
99.3
Particulate
Not Removed
g/m3
	

_ _
w_

--
n nno2
n ono?
o.on
o.o^n
0.071
0.227
0.002
0.060
o onA
0.210
0.011
Q.Vtf
Applicability
for Combined
Cycle End Use a
N,A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
Appl icability
for Boiler
Fuel End Useb
N.A.
N.A.
N.A.
Appl i rah 1 <=
Appl i cab le
N.A.
Appl icab le
ADD! i cab le
N.A.
a Based on maximum allowable
b Based on maximum allowable
                 partIculate load of 0.041
                 particulate load of 0.24

-------
             TABLE 21  THE APPLICABILITY OF A FABRIC FILTER FOR GASIFICATION PROCESSES

-o
0)
CD
n
*
IZ
O
V
CO
T3
3
U.
TJ
0)
CD
1
tm)
<6
>6
<6
>6
<6
>6
*6
>6
<6
>6
<6
>6
<6
>6
<6
>6
<6
?6
%
0.3
99.7
5,H
9^.6
10.5
89.5
i
99
^
q?
5
95
0.5
99.5
0.7
99.3
0.8
99.2
Percent
Participate
Removed
%
99.99
100
99. 99
100
99.99
100
99 ?
100
qq.4
mo
99.6
100
99. 94
100
QCLQ7
100
99.99
100
Parttculate
Not Removed
g/m^
0
0
0.00002
0
0.00006
0
p.ooooq
0
0.0042
n
0.0024
0
o.ooooq
0
n nnn?^
0
0.00018
0
Applicability
for Combined
Cycle End Use a
Appl i cable
Appl icable
Appl i cab 1 e
Applicable
Appl Icable
Appl i cable
Appl icable
Appl icable
Appl icable
Applicability
for Bo Her
Fuel End Useb
Appl icable
Appl icable
Appl icah 1*>
Appl i cable
Appl I cab le
Appl icable
Appl Icable
Appl icable
Appl icable
3 Based on maximum allowable
b Based on maximum allowable
partIculate load of 0.041
particulate load of 0.24 g/m

-------
               TABLE 2.2  THE APPLICABILITY  OF  A GRANULAR BED  FILTER  FOR GASIFICATION PROCESSES
1 	
TJ
O
00
TJ
0>
X
iZ
o
^m)
<6
>6
<6
>B
<6
>6
*6
>6
<6
>(,
<6
>6
<6
^6
<6
?6
<6
^6
X
0.3
99-7
5. ^
9A.6
10.5
89.5
1
99
3
97
5
95
0.5
99.5
0.7
99.3
0.8
99.2
Percent
Participate
Removed
%
3k,6
95
9*.7
95
9'4.7
95
94.4
95
94. q
q?
94. <5
95
94.6
95
94 7
95
94.7
95
Participate
Not Removed
g/m3
0.00008
0.0229
0.0035
0,142
0.033
0 . 269
0.00067
0.025
n_p4i9
1 .918
0 . * 1 8
5.422
0.0077
1.972
n n3Q7
5.231
0.095
10.95
Appl icabi 1 ity
for Combined
Cycle End Use a
N.A.
N.A.
^J.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
Applicability
for Boiler
Fuel End Useb
Appl i cable
Appl i cable
N A
Apol i cable
N.A.
N.A.
N.A.
N.A.
N.A.
a Based on maximum  allowable  parti
b Based on maximum  allowable  parti
culate load of 0.041  g/m3.
culate load of 0.24 g/m3.

-------
gasifier, as shown in Table  13 .  By considering a cyclone upstream as a scalping
device, the Aerodyne rotary flow cyclone was found to be capable of meeting the
requirement of End Use 1  for the average case,but not the worst case of the fluid
bed gasifier.  It should be noted that the results presented herein for the rotary
cyclone should be considered tentative until the vendor-supplied dat* employed in
these assessments are confirmed.  Uncertainties in the fractional efficiency data
for the rotary cyclone are discussed in Section k.

          Since the particulate removal requirement of End Use 2 is not as restric-
tive as End Use 1, the number of control devices applicable to End Use 2 is in-
creased considerably as compared to End Use 1.   The results in Tables 21  and  19
show that both the fabric filter and the venturi scrubber are capable of achieving
the requirements of End Use 2 for all gasifier effluents.  The Aerodyne rotary flow
cyclone was found to be applicable to all the gasifier effluents except for the
worst-case fluid bed gasifier.  However, with a conventional cyclone upstream as  a
scalping device,  it would be applicable to this worst case as well.  A conventional
high efficiency cyclone by itself is found applicable to the best and average cases
of the fixed bed gasifier, and the best case of the fluid bed gasifier.  The CPC
granular bed filter is found to have the same applicability as the high efficiency
cyclone mentioned above.   Two cyclones in series are capable of achieving the same
efficiency as one Aerodyne rotary flow cyclone.  A cyclone followed by a CPC granular
bed filter would be applicable to two more cases than the CPC filter by itself,
namely the average case of the fluid bed gasifier and the best case of the entrained
bed gasifier.  As for an electrostatic precipitator, it was found that a dry-type
device is not applicable to fixed bed gasifier effluents.  This is due to the fact
that the particles in these effluents have very high carbon contents (55 percent
to 80 percent), which results in low resistivity of the particles and inefficient
collection.  The electrostatic precipitator was found to be applicable to the best
and average cases of the fluid and entrained bed gasifiers for End Use 2.  With a
cyclone upstream as a scalping device, the electrostatic precipitator would also  be
able to achieve the required removal efficiency for the worst cases of the fluid
and entrained bed gasifiers.

          As discussed previously, the applicability assessments presented herein
are based only on the ability of the various control devices to meet specified par-
ticulate collection efficiencies.  Other possible limitations such as economic
feasibility, energy requirements, or operational reliability are not considered in
these analyses.  For example, a fabric filter could not be employed for gasifier
effluents containing high levels of liquid or "sticky" particles.  Thus, fixed-
bed gasifiers, in particular, may not be compatible with fabric filters, due to the
quenching operation commonly used to condense and remove tars and oils.  As discussed
in Section 2, the characteristics of the particulates in the gasifier effluents are
generally not sufficiently well knowi to make such an assessment.

          The results of the applicability assessments discussed above are sum-
marized in Table 23.

-------
                     TABLE  23   Summary  of  Applicability Assessments
Appl icabi
End Use/Control Device
COMBINED-CYCLE
conventional cyclone
rotary cyclone
venturi scrubber
fabric filter
E.S.P.
granular bed filter
rotary cyclone*
venturi scrubber*
fabric filter*
E.S.P.*
granular bed filter*
BOILER FUEL
conventional cyclone
rotary cyclone
venturi scrubber
fabric filter
E.S.P.#
granular bed f i 1 ter
rotary cyclone*
venturi scrubber*
fabric filter*
E.S.P.*
granular bed filter*
lity
of
Fixed Bed
B
X
X
P
X
X
P
X
X
X
P
X
X
X
P
X
W
X
X
P
X
X
P
X
X
P
X
X
P
A
X
X
P
X
X
P
X
X
X
P
X
X
X
P
X
Control
Flui
B
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Devices
d
W
X
X
X
X
X
X
X
X
X
Bed
A_
X
X
X
X
X
X
X
X
X
X
X
X
X
X
for Gas
ifier
Entrained
B
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
W
X
X
X
X
X
X
X
X
X
X
X
X
X
Types**
Bed
A
X
X
X
X
X
X
X
X
X
X
X
X
X
X
*   A conventional cyclone is assumed to be employed as a scalping device
    upstream of the indicated primary control device.
**  B - Best Case
    W - Worst Case
    A - Average Case
    P - Designates probable inapplicability due to operating problems, although
        particulate removal is adequate.
    X - Designates control device is applicable.

#   ESP is not applicable to a fixed bed gasifier due to high carbon content and
    low resistivity of particles.
                                         77

-------
                                   SECTION 6

                                FATE OF  POLLUTANTS


           In  the  previous sections, technologies for controlling  the  particulate
 and  tar  levels of the converter product gases have been  discussed.  Each control
 technology  identified,  in turn,  generates solid, liquid, and/or gaseous wastes that
 also must be  disposed of  in a proper manner.   In addition, precautions must be
 taken to prevent  environmental  damage resulting from the combustion of the product
 gas  or tars.

          To  assess the pollution potential of controls and product/by product
 utilization,  the  "fate of pollutants" must be understood.  Composition of waste
 and  product streams must be quantified  for trace elements, sulfur compounds,
 organics, and any other potential pollutants.  By determining  those streams   in which
 certain  pollutants tend to concentrate, and thus present potential pollution  pro-
 blems, proper disposal and control technologies can be selected to minimize environ-
 mental degradation.

          The following discussion on the fate of these pollutants is based upon
 limited  data  from experimental  and commercial installations obtained  during various
 test programs.  The data provide insight into where potential  problems exist.  How-
 ever, due to  the  fact that most of the available data are preliminary and limited,
 caution  must  be taken in comparing data from different tests or assuming that infor-
 mation collected  is necessarily  representative of either a spe.cific process or a
 generic  type  of gasifier.

          It  is well documented that coal contains a vast array of trace elements
 that are present  in concentrations of less than 1%.  During gasification, many of
 these elements may volatilize.  Some, such as mercury, selenium, arsenic, molybdenum,
 lead, cadmium, beryllium and fluorine are recognized as toxic  substances.  Unlike
 the  product gases of coal combustion, coal gasification takes place in a net reducing
 atmosphere.   Volatile compounds such as carbonyls, hydrides, and sulfides are likely
 to be formed.

          Jahnig  (70)estimated  the quantities of certain trace metals likely  to be
volatilized during coal gasification.  His estimates are presented in Table 24
Assuming that these elements are carried overhead with the product gas, the poten-
 tial  magnitude of the problem can be realized.  Thus, these volatilized elements
must be collected and properly  disposed of.  If these contaminants should remain in
the product gas when it is combusted, severe environmental consequences may occur.

          Jahnig concluded that "the combined amounts of all volatile portions of

                                        78

-------
                   TABLE  2k

       TRACE ELEMENTS - ESTIMATED VOLATILITY  (70)
       Hypothetical
         Coal ppm         % Volatile*     kg/day (lb/day)**
Cl
Hg
Se
As
Pb
Cd
Sb
V
Mi
Be
Zn
B
F
Ti
Cr
1500
0.3
1.7
9.6
5.9
0.8
0.2
33
12
0.9
44
165
85
340
15
90+
90+
7^
65
63
62
33
30
2k
18
10
10
10
10
nil
25,000
4.5
23
110
67
9.1
1.4
180
52
3.2
80
300
154
617

(54,000)
(10)
(50)
(250)
(148)
(20)
( 3)
(397)
(115)
( 7)
(177)
(660)
(340)
(1360)
nil
*   Volatility based mainly on gasification experiments but
    chlorine is taken from combustion tests, while zinc,
    boron, and fluorine were taken at 10 percent for illustration
    in absence of data.

**  Estimated volatility for 1.8xl07 kg/d (20,000 TPD)  of coal to
    gas i fication.
                        79

-------
trace elements can present a formidable disposal problem."  Some of the volatile
matter which  is carried out of the gasifier with the raw product gas will be
removed when  the gas  is cooled and scrubbed.  Some compounds such as cyanides can
be destroyed  by recycling to the gasifier, but elements such as arsenic, chlorine,
and  lead cannot be recycled, and must be separated and recovered or deactivated
prior to disposal.

          Particulates collected by the various control technologies discussed in
this report must be disposed of, or recycled  in an economical and environmentally
sound fashion.  In some cases, these wastes will be in the dry solid form.   In
other instances, however, such as in the case of venturi scrubbers, the collected
particulate matter is removed by a wet system.  In such cases, the contaminated
water may require treatment prior to discharge.

          The particulate control technologies evaluated in this report include-'

          o   venturi  scrubber

          o   electrostatic precipitator

          o   conventional cyclone

          o   rotary cyclone

          o   fabric filter

          o  granular bed filter

Of these six  technologies, only the venturi scrubber is not a dry process.  The
other five processes generally produce a dry, granular or powdery, solid waste.
In the case of a venturi or other wet scrubber, the collected fly ash will be wet,
complicating  disposal of the ash while also necessitating wastewater treatment.
Liquid waste streams from scrubbing or quenching operations must be treated prior
to final disposal or discharge to surface or ground waters.  Present and proposed
regulations for liquid discharges generally require a high degree of water recycle
and  reuse within the plant, thereby minimizing the amounts of liquid to be
released from the plant.

          The collected ash, whether wet or dry, must be disposed of in a landfill
or other environmentally acceptable manner.  Undesirable elements can sometimes be
leached from the collected particulate matter.  Even if a dry collection system is
used, the solid wastes will ultimately be exposed to leaching by ground water if
they are disposed of as land fill or returned to the mine.   Use of liners and
entrapment of runoff and drainage water will minimize the likelihood of ecological
degradation.

          The standard technique for removing tars (as well as ammonia and some
other pollutants)  from the raw product stream Is by quenching.  The water that is
used to quench the gas then becomes a gas liquor.   This quench liquor cools the
crude gas mixture to a temperature at which it is saturated with water.  The gas
liquor is then flashed and tar removed from the bottoms.  The remaining water

                                        80

-------
is then sent to water purification.   The recovered  tars  can  be  used  as  a  boiler
fuel, incinerated, or sent to a landfill.   It  is  unlikely  that  recovering of
specific compounds, such as phenols,  will  prove to  be  an economically viable  alterna-
tive.

          The limited data on the fate of pollutants presented  in  this  chapter does
provide insight into the need for proper control  techniques. Additional  data is
essential, however, in order to clearly define the  fate  of these pollutants.

6.1       FIXED BED GASIFIERS

          The raw product gases from fixed bed gasifiers usually contain  both tars
and particulate matter.  The ultimate fate of these pollutants  is  dictated by both
the removal technology and the physical and chemical characteristics of the con-
taminants.

          The standard technique for removing tars  from the raw product gases is  by
quenching.  The gas liquor and condensate produced  is sent to the liquor treating
area for separation of the tars (as well as other hydrocarbons, ammonia,  carbon
dioxide, and hydrogen sulfide).  The gas liquors are cooled and, in the case of
Lurgi, treated in three steps.  The first step employs physical separation vessels
for the recovery of the tars and tar oils present in the gas liquor.  The second
and third phases  remove phenols, ammonia, and acid gases, respectively.

Table  25  shows  the organic composition of quench liquor extract from a fixed bed
gasifier  (73)- The high concentrations of phenolic compounds should be noted.
Forced evaporation  of  large quantities of this spent quench  liquor could cause
emissions of significant amounts of organics  into the atmosphere.  Page  (72)  has
reported  that the concentration of sulfur  in  the by-product tar is dependent upon
the nature of the coal  feed stock.  The sulfur concentration is two to three times
greater in tar produced from lignite than from high volatile bituminous  coal.  It
is probable that  the sulfur content of  the tar originating  from bituminous coal
would require SQ-2 scrubbing of the flue gases  resulting from the combustion of the
tars.   In  this case, recycling of the  tar to  the gasifier or landfill ing may be a
more viable alternative.

          Table 26   presents the trace element analyses of cyclone dust, quench
liquor and tar.   It can be seen that the concentration of Pb, Se, Cs and Fl were
highest in the cyclone dust, while the Hg concentration was greatest in the tar.
Table  27  compares the analyses of trace elements in the quench liquor with
various water quality standards.  Selenium concentration in the quench liquor is
reported to be ^00 times greater than the selenium standard for surface and intake
water and approximately 80 times greater than the satisfactory level for irrigation
water.  Nearly all other elements equal or exceed Federal Water Quality  Standards.

          Even after initial treatment  to remove oils,  tar  and other pollutants,  the
gas  liquor will likely  still contain small levels of heavy  hydrocarbons, trace
inorganics, and dissolved gases.  This stream  is, therefore, a potential pollution
source, and may require further processing or  disposal  in solar evaporation  ponds.


                                          81

-------
TABLE 25  ORGANIC  COMPOSITION OF QUENCH LIQUOR - FIXED BED GASIFIER  (73)
% of
Fraction Total
Number Extract
1 0.01
2 0.08
3 0.22
4 0.76
5 7-31
6 7*. 18
7 3.09
8 8.34
Functional Groups Present
Aliphatic hydrocarbons
Alkyl-aryl hydrocarbons
Alkyl-aryl hydrocarbons, trace
carbonyl, possible polycyclics or
multi -substituted aroma tics
Alkyl-aryl hydrocarbons, possible
polycyclics, -OH present (possibly
atmospheric moisture)
Phenol + alkyl/dialkyl phenols
Principally Phenol + other phenols
Pheno 1 s
Alkyl-aryl CH stretch, -OH, -C=0,
Methyl Bending Vibration, Primary OH,
possible inorganic sulfur + other
ionic compounds
                                  82

-------
                                     TABLE 26

                        TRACE  ELEMENTS  IN GRAB  SAMPLES BY SSMS (73)

                                        Cyclone

Uranium
8 ismuth
Lead
Mercury
Barium
Antimony
Cadmium
Molybdenum
Selenium
Arsenic
Zinc
Copper
Nickel
Chromium
Vanadium
Ti tani urn
Chlorine
Sulfur
Fluorine
Boron
Beryl 1 i urn
L i th i urn
Ash
56
0.4
7
NR
MC
1
3
22
20
k.
26
540
120
510
MC
MC
230
250
56
130
22
190
Bottom
Dust

<2
60
0.01
460
8
<2
14
2k
27
85
130
30
90
100
MC
720
MC
270
70
6
27
Quench
Liquor

__
0.04
0.007
0.1
0.1
<0.02
0.06
4
0.2
0.07
0.1
0.07
0.03
0.004
0.05
0.3
MC
2
2

0.2
Tar

__
10
0.12
27
0.8
--
1
3
4
7
3
5
3
0.8
29
6
520
22
19
0.1
4
Note:  All values expressed as ppm except liquor in which values are expressed as
        M9/ml.
MC = Major Component
                                         83

-------
                    TABLE  27

LEVELS OF TRACE ELEMENTS IN LIQUIDS FROM THE QUENCH LIQUOR
    AND BY-PRODUCT TAR VERSUS WATER QUALITY STANDARDS (73)
Element
Antimony
Arsenic
Barium
Bery 1 1 ium
Boron
Cadmium
Ch rom i urn
Fl uorine
Mercury
Lead
Manganese
Molybdenum
Nickel
Seleni urn
Vanadi urn
Zinc
Copper
Surface
Water
_ _
0.05
1.0
--
1 .0
0.01
0.05

--
0.05
0.05
--
--
0.01
--
5.0
1.0
1 rrigat ion
Water

1 .0
--
--
0.75
0.005
5.0
--
--
5.0
2.0
0.005
0.5
0.05
10.0
5.0
0.2
Public
Water
Intake
__
0.1
--
--
1.0
0.01
0.05
--
0.002
0.05
0.00
--
--
0.01
--
5.0
1.0
Liquor
ju.g/1
0.1
0.2
0.1
--
2.0
10.02
0.03
2
0.007
O.Qk
0.03
0.06
0.07
k
--
0.07
0.1
Tar
ppm
0.8
k
27
0.1
19

3
22
0.12
10
0.9
1
5
3
--
7
3

-------
          The particulate matter collected from the raw product gases will have
varying particle size distribution, bulk density, and ash and carbon content,
depending upon the coal type and the gasifier operating conditions.   Apparently
the levels of organic compounds are much higher in the particulate matter than in
the bottom ash.  Approximately 400 ppm of organics have been founrd in the parti-
culate matter, as compared with 20 ppm in the bottom ash.  Table  28  indicates the
organic composition of the cyclone dust.  Table 29   indicates other characteristics
of the particulate matter such as ash content, size and density.

          Assuming a 7-1 x 10  m /day (250,000 scfd) production of medium Btu-gas,
it is estimated that for fixed bed gasifiers, up to 4.5 x 10^ kg (100,000 Ibs) of
flyash per day would be collected and then disposed of or recycled.

6.2       FLUID BED GASIFIERS

          Several experimental studies (75,76,77,78) have been performed using the
Pittsburgh Energy Research Center's (PERCJ 10 cm (4-inch) diameter Synthane gasifier.
Samples taken do provide some understanding of the fate of pollutants in fluid bed
gasifiers.  However, the extrapolation of pilot plant data to commercial scale situa-
tions must be exercised with caution.  Also,  the accuracy of the data is question-
able as indicated by inconsistencies  in material balances. Still, the data does give
insight into the fate of the pollutants.

          Table  30 shows the trace element analysis for a typical test  run of the
Synthane gasifier.  Other runs performed during this study tended to substantiate
these data.  As can be seen, many of  the trace elements  tend to  concentrate  in the
particulate and char.  Some of the more volatile elements such  as arsenic,,   lead,
cadmium and mercury also are found  in  the tars  in  levels  that are potentially  harmful,
Almost all of the chlorine concentrates  in the water.

          Analyses were performed on  tars from various coals as  shown below:

                                        Illinois  #6   Illinois #6  Lignite

               Carbon

               Hydrogen

               Nitrogen

               Sulfur

           This  analysis is  on a moisture and ash free basis.

           Table   31   presents  the organic  analysis for  the benzene soluble tar from
 the  Synthane gasifier. Apparently,  the tar  is predominantly made up of  napthalenes,
 acenapthenes,  aromatics,  and  phenylnapthalenes.

 6.3        ENTRAINED BED GASIFIERS

           Data on the fate  of pollutants from entrained bed gasifiers is somewhat

                                          85
82.6
6.6
1.1
2.8
83.4
6.6
1.1
2.6
83.8
7-7
1.0
1.1

-------
                                   TABLE  28

                    ORGANIC COMPOSITION OF CYCLONE DUST EXTRACT
                            FIXED BED GASIFIERS (73)
Fract ion
1
2

3

J
5
6


7
% of
Total
Extract
35-2
5.8

0.3

0.9
1.0
36.5


14.3
Assignment
Pa raff in ic hydrocarbons, considerable branching
Paraffinic functional groups, traces of substituted
aroma tics
Split carbonyls, esters (formate or butyrate) ,
methyl, isopropyl and tributyl branching, primary
alcohols)
Aromatics, carbonyls, al iphatic hydrocarbons
Split carbonyls; methyl, isopropyl and tributyl
branching; secondary alcohols; esters, possible 5C
ring lactones; branched cyclic alcohols
(No assignment made)
                                    TABLE 29
     CHARACTERISTICS OF PARTICULATES - FIXED BED GASIFIER (73)
     Collected by Cyclone
Collected by Cyclone
Suspended in Tar
Coal
Type
Bit.
Bit.
Anthr.
Lign.
Average
dp(>tm)
170
95
200
70
Ash
Content
(wt*)
10.2
15.4
47.3
23.0
Bulk
Density
0.40
0.53
0.93
 
Average
dp (/I m)
--
20*
<1*
 -
Ash
Content
(wt*)
--
10.4
54.7
-
Bulk
Dens i ty
--
0.31
--
 -
Average
dp (Xm)
2-20
--
--
--
* Agglomerated
                                       86

-------
                                 TABLE  30

                      LABORATORY  SYNTHAME  GASIFIER
                TRACE  ELEMENT ANALYSIS OF  ALL STREAMS (77)

Run #     Feed Coal      Filter  Fines    Char        Tar            H20
162       PPM Cug/g)     PPM  Gug/g)      PPM  Gug/g)  PPM  Cug/g)     PPM Gug/g)

Ag         0.01           <0.01          < 0.05
Al        >0.5           540           1800          29             0.007
As         0.87            3-7             6.5         0.71          0.001
B         86              64            380          12            43
Ba       170             130              98           3.6           0.10
Be         1.5             7.2             4.6         0.03
Bi        <0.10           <1.7             <0.44        0.20
Br         0.23            0.65            1.6         0.02          0.001
Ca        >1%             >0.5%           >\%         450            2.4
Cd         0.097           0.88            1.6
Ce        47              25              54           0.29
Cl        93              11              33            1.5          300
Co        14              17              95           0.09          0.002
Cr       170              47             240            7.1           0.043
Cs         0.26            1.2              0.65
Cu        39              70              40           0.74          0.003
Dy         1.4             3.9              1.6
Er         2.1             0.41            0.80
Eu         0.55            0.39            0.65
F        490             610             150           0.97         39
Fe        >]%             >0.5%           >\%         240            0.081
Ga         8.3             3.6              4.5          0.08
Gd         1.9             1.2              0.48
Ge         1.1             1.3              5.4          0.08
Hf         0.83            3.5             11
Hg         0.10            0.20           *            1.2          0.027
Ho         0.43            0.16            0.45
 I          0.4             1.9              0.27         0.02
K         >U             190            5400          14            0.31
La        22               6.7            17           0.03
Li         0.8            34              67           0.51          0.001
Lu         0.085          tOJ8            0.40
Mg      2800            4600           3500          240            0.57
Mn        160              48            240            2.2           0.20
Mo        15              21              14            0.31
Na      1900              >U          4700          360             6.6
Nb         4.7              7              13            0.08


Continued


                                        87

-------
                              TABLE  30   (Continued)
 Run #     Feed Coal    Filter Fines      Char         Tar            H20
 162       PPM Oug/g)   PPM  (jjg/g)        PPM (/Jg/g)   PPM (>ig/g)     PPM (jjg/g)

 Nd           23             19              11           0.06
 Ni           43             12              25           1.2            0.018
 P           130           460             460          14              o.04
 Pb            0.55          2.2            21           0.22           0.003
 Pr            7.3           7-5             4.2         0.02
 Rb          180             36              27           0.10
 S            7U         7700            2100         120              1.6
 Sb            0.18          0.04            1.9
 Sc            5.3           6.4            17           0.02
 Se            2.2           15               4.7         0.23           0.14
 Si           71*            >}%             >]%        500              2.8
 Sm            2.7           0.30            0.86        0.01
 Sn            0.6           0.75            1.9         0.03
 Sr            3.3           ^4              70           4.5            0.12
 Ta            0.73          0.64            1.2
 Tb            0.2           0.20            0.89
 Te           <0.29         <0.19            0.15
 Th            3             5.8             4.3         0.06
Ti          880          1800            3300           8.4            0.003
Tl          <0.12        <0.19          <0.25        0.11
Tm            0.24          0.10            0.20
U             1.4           5.6             5.4         0.01
V           100             44             190           0.21
W             0.08          2.2             4.8         0.09
Y            21              37              48           0.10
Yb            0.35          2.7             2.2
Zn           25             11              100           0.48           0.13
Zr           10             22              28           0.26
* Insufficient results

-------
                                TABLE  31
             MASS SPECTROMETRIC ANALYSES OF BENZENESOLUBLE
               TAR  FROM SYNTHANE GASIFIER "VOL. %  (76)
Structural type
(includes alkyl
derivatives)
Benzenes
Indenes
Indans
Naphthalenes
Fl uorenes
Acenaphthenes
3-ring arc-mat ics
Phenyl naphthalenes
4-ring peri condensed
4-ring catacondensed
Phenols
Naphthols
Indanols
Acenaphthenols
Phenanthrols
Dibenzofurans
0 i benzoth i ophenes
Benzonaphthoth iophenes
N-heterocycl icsc
Average molecular weight
Run HP-1
No. 92,
Illinois3
No. 6 coal
2.1
3.6b
1.9
11.6
9.6
13.5
13.8
9.8
7.2
4.0
2.8
( )b
.9
-
2.7
6.3
3.5
1.7
(10.8)
212
Run HPL,
No. 3k,
1 ignite
4.1
1.5
3-5
19.0
7-2
12.0
10.5
3.5
3.5
1.4
13.7
9.7
1.7
2.5
-
5.2
1.0
-
(3.8)
173
Run HPM No. Ill,
Montana
subb i turn! nous
coal
3.9
2.6
4.9
15.3
9.7
11.1
9.0
6.4
4.9
3.0
5.5
9.6
1.5
4.6
.9
5.6
1.5
-
(5.3)
230
Run HP-1 18
No. 118, a
Pittsburgh
seam coal
1.9
6>
2.1
16.5
10.7
15.8
14.8
7.6
7.6
4.1
3-u
( )b
.7
2.0
-
4.7
2.4
-
(8.8)
202
a Spectra indicate traces of 5-ring aromatics.
  Includes any naphthol present (not resolved in these spectra).
                             isotope corrections were estimated.
c Data on N-free basis since
                                        89

-------
more  limited than that available on fixed or fluid bed gasifiers.  However, in a
study by Lee et al.  (7*0,effluents from a high temperature entrained flow gasifier
were analyzed.  In this study, tests were run on an experimental gasifier.  The
authors report:  "This gasifier is a pressurized, entrained-flow gasifier that has
a capacity of k.S x  lO^gOOO pounds) of coal per hour,and has a down flow configura-
tion with some similarity to an entrained flow gasifier operated by the Bureau of
Mines during the period 1952-1963.  It also has some similarity to the Texaco
entrained flow gasifier configuration."   In these tests a high volatile, non-caking
Utah bituminous coal was used.  Among the streams sampled were the scrubber effluent
water and the gas evolved on depressurization of the scrubber water.  The scrubber
is used for final  cleaning of soot and flyash particles.  It follows a quench sec-
tion, where molten ash droplets are solidified, and a heat exchanger.

          Results of the test data are shown in Tables  32,  33     and  31*  .  Since
only O.*5g (1 x 10~3 Ibs) of organic material was found in the scrubber water, it is
speculated that organics concentrated in the particulate matter.  This was substantia-
ted by a napthalene absorption test run by Brigham Young University, in which naptha-
lene was added incrementally to scrubber water containing particulates.  As shown
in Figure 21, napthalene tended to be absorbed by the particulate matter.  In terms
of trace elements, apparently little or no Hg, Se or As was absorbed in the scrubber
water.  In the case of the scrubber flash gas, it should be noted that the total
sulfur emissions are insignificant  when compared with current EPA standards for
S02 emissions from coal-fired  boilers (1.2 lb/106 Btu).

          For entrained bed gasifiers, total particulate collection will typically
range from 1.8 x 10? to 1.6 x 10b kg/day (200 to 1800 tons/day) for a system producing
250,000 scfd of medium-Btu gas.

          Oldham,  et al. reported that typically, in the Koppers-Totzek process, the
product gas passes through a venturi washer and gas cooler.   The flyash scrubbed
from the gas is removed as a 50 wt. % solids slurry from a clarifer.  Both the solid
and liquid phases  of the flyash sludge contain trace elements.  Ammonia, nitrates,
sulfates and cyanides are all  present in the sludge liquor.   Polynuclear aromatic
hydrocarbons in the gasifier product are also likely to condense in the ventur!
scrubber water and exit the scrubbing system with the flyash sludge (79).
                                         90

-------
VjD
                      TABLE  32   SCRUBBER WATER EFFLUENT  INORGANIC  SPECIES



                            ENTRAINED BED TEST GASIFIER (7*0





                    Species        ppm                 g/kg   (ibs/ton of coal)
HCO,"
cf
F "
NO,"
3
NO '
210
9.
1.
0.

0.

2
2
i,

3
6.
0.
0.
0.

0.
6
36
0*.
02

02
(3.3)
(0.18)
(0.02)
(0.01)

(0.0-1)
                       2          2-
                    SO. "  and SO    detected at same level both before



                    and after scrubbing





                    Carboxylic acids, phenols, and amines not detected

-------
tsi
                  TABLE  33    SCRUBBER WATER EFFLUENT ELEMENTAL COMPOSITION

                              ENTRAINED BED TEST GRSIFIER  (74)


                  Element               ppm       g/kgx1p3 (Ibs/ton of coal xlO3)
Ca
S
Si
Fe
Cl
Zn
K
Sr
Cu
50
6.0
3-0
2.3
1.4
1.1
1.0
0.35
0.01
1880
240
120
92
56
42
38
14
0.4
(940)
(120)
(60)
(46)
(28)
(21)
(19)
(7.0)
(0.2)
                  Tl, Mn, Ni, Br, Hg, Se, and As detected at same level  both before
                  and after scrubbing

                  Other Elements (I4i 7^40) not detected (^O.Sppm), where Z is the
                  atom!c number

-------
TABLE 34   SCRUBBER WATER FLASH GAS COMPOSITION




              ENTRAINED BED TEST GASIFIER  (7!,)
FLASH GAS                     g/kg of  coal  (ib/ton  of  coal)






  C02                             5.6   (2.8)




   N2                             3.6   (1.8)




  CO                              3-6   (1.8)




  CH^                             0.24  (0.12)




  H2S                             0.058 (.029)




  S02                             0.042 (0.021)




   H2                             0.016 (0.008)




  HCN                             0.0008 (0.0004)
                      93

-------
                      100
                       30
                  -e
                   o
                   in
                       60
vo
                   <-

                   &
                   <0
                   8

                   01
                       20
                                                   I
I
I
_L
                          0           0.10         0.20        0.30         0.^0        0.50



                                             Participate Matter Concentration (PPT)



                          FIGURE  21.   Absorption of Naphthalene By Participates   (71*)
                                    0.60

-------
                                  REFERENCES


I.  Spaite,  P.W.  and  Page,  G.C.,  "Low-and  Medium-Btu  Gasification Systems:
    Technology Overview",  EPA-600/7-78-061, March  1978.

2.  Cavenaugh, E.G.,  Corbett,  W.E.,  and  Page,  G.C., "Environmental Assessment Data
    Base for Low/Medium-3tu Gasification Technology,  Volume  II, Appendices A-F",
    EPA 600/7-77-1256,  November 1977.

3.  Dravo Corporation,  "Handbook  of  Gasifiers  and  Gas Treatment Systems", FE-1772-
    11, February 1976.

k.  Becker,  D.F., and Murthy,  B.N.,  "Feasibility of Reducing Fuel Gas  Clean-up
    Needs",  FE 1236-15, June 1976.

5.  Meyer, J.P., and Edwards,  M.S.,  "A Survey  of Processes  for High  Temperature  -
    High Pressure Gas Purification", ORNL/TM-6178, November 1978.

6.  Sinor, J.E., "Evaluation of Background Data Relating to New Source Performance
    Standards for Lurgi Gasification", EPA-600/7-77-057, June 1977.

7.  WESCO, "Final Environmental Impact Statement", WESCO Coal Gasification  Project,
    1975.

8.  Moore, A.S., Jr., "Cleaning Producer Gas  from  MERC Gasifier",  U.S. ERDA,
    May 1977.

9.  Murthy,  B.N., Klett, M.G., Becker, D.F.,  Szwab, W., and Fischer, W .  H.,
    "Fuel Gas Cleanup Technology for Coal  Gasification", FE 2220-15, March  1977.

10. Forney,  A.J., Haynes, W.P., Gasoir, S.J.,  Johnson, G.E. and Stakey, J.P.,
    "Analysis of Tars, Chars, Gases, and Water Found  in Effluents  from the  Synthane
    Process", U.S. Bureau of Mines,  TPR %, January  197^.

11. Conoco Coal Development Corp. and Stearns-Roger Engineering Company,  "Commercial
    Plant Conceptual Design and Cost Estimate - C02 Acceptor Process Gasification
    Pilot Plant", Vol. 10, FE/173^3, August 1976 -  December  1977.

12. Zabolothy, E.R., "Purification of Hot Fuel Gases  from Coal or Heavy Oil",
    EPRI-2^3-1, Interim Report, November
                                     95

-------
                             REFERENCES (cont'd)


13.  Squires, A.M., "Gasification of Coal  in High Velocity Fluidized Beds and (I I)
     Hot Gas Cleaning", Presented at I GT Clean Fuels from Coal, Symposium II Papers,
     June 23-27, 1975-

14.  Whiteacre, R.W. , Koppers-Totzek, Personal Communication, August 1978.

15.  Parker, R., and Calvert, S., "High-Temperature and High-Pressure Particulate
     Control Requirements", EPA-600/7-77-071 , July 1977.

16.  Woodall-Duckham Ltd., "Trials of American Coals in a Lurgi Gasifier at
     Westfield, Scotland", ERDA R&D Report #105, 1975.

17.  Jahnig, C.E., "Evaluation of Pollution Control in Fossil Fuel Conversion
     Processes; Gasification; Section 1: C02 Acceptor Process", PB 2^*1  141,
     December 1971*.

18.  Magee, E.M., C.E. Jahnig, and Shaw, H., "Evaluation of Pollution Control in
     Fossil Fuel Conversion Processes;  Gasification; Section 1 ; Koppers-Totzek
     Process, EPA-650/2-7A-009a, January 197*t.

19.  Jahnig, C.E., "Evaluation of Pollution Control in Fossil Fuel Conversion
     Processes; Gasification; Section 8, Winkler Process", EPA-650/2-74-009-J ,
     September 1975.

20.  Kalfadelis, C.D., and Magee, E.M., "Evaluation of Pollution Control  in Fossil
     Fuel Conversion Processes; Gasification; Section 1: Synthane Process",
     EPA-650/2-71-009-b, June
21.  Jahnig, C.E., "Evaluat ion of Pollution Control in Fossil Fuel Conversion
     Processes; Gasification; Section 7, U-Gas Process", EPA-650/2-7i-009-i ,
     September 1975-

22.  Jahnig, C.E., "Evaluation of Pollution Control in Fossil Fuel Conversion
     Processes; Gasification; Section 5, BI-GAS Process", EPA-650/2-7^-009-g,
     May 1975.

23.  Shaw, H., and Magee, E.M. ,  "Evaluation of Pollution Control  in Fossil
     Fuel Conversion Processes;  Gasification; Section 1: Lurgi Process",
     EPA-650/2-7*t-009-c, July
2k.  Jahniq, C.E., and Magee, E.M., "Evaluation of Pollution Control in Fossil
     Fuel Conversion Processes; Gasification; Section 1: CQ-2 Acceptor Process",
     EPA-650/2-7^-009-d, December 197**.

25-  Robson, F.L., Giramonti, A.J., and Blecher, W.A. ,  "Fuel Gas Environmental
     Impact: Phase Report", EPA-600/2-75-078, November 1975.


                                     96

-------
                                 REFERENCES  (cont'd)


26.  White, J.W.,  Sprague,  R.,  and McGrew, W.,  "Papers-Clean  Fuels  from Coal
     Symposium II", Institute  of Gas  Technology,  June  1975.

27.  Ferrel, J.,  and Poe,  G.,  "Impact of Clean  Fuels  Combustion on  Primary  Parti-
     culate Emissions from Stationary Sources", EPA-600/2-76-052, March 1976.

28.  Cavanaugh, E.C., Corbett,  W.E.,  and Page G.C.,  "Environmental  Assessment  Data
     Base for Low/Medium-Btu Gasification Technology:  Volume  1,Technical  Discussion",
     EPA-600/7-77-125a,  November 1977.

29.  Robson, F.L., Blecker, W.A., and Colton, C.B.,  "Fuel  Gas Environmental  Impact",
     EPA-600/2-76-153,  June 1976.

30.  Szwab, W., Fischer, W.H.,  Wheelock, B.R.,  and Murthy, B.N., "Gas Cleanup
     Systems for Application to the MERC Stirred/Fixed Bed Gasifier", FE-1236-13,
     March 1976.

31.  Ayer, F.A.,  and Massoglia, M.F., "Symposium Proceedings: Environmental  Aspects
     of Fuel Conversion Technology III", EPA-600/7-77-08,  April 1978.

32.  Curran, G.P., Clancey, J.T., and Fink,   C.E., et al., "Development of the CQ.2
     Acceptor Process Directed Towards Low-Sulfur Boiler Fuel", PB  210 840,
     November 1971.

33.  Howard-Smith, I.,  and Werner, G.J., "Coal  Conversion Technology", Noyes Data
     Corporation,  Park Ridqe, New Jersey, 1976.

34.  Ayer, F.A., "Symposium Proceedings: Environmental Aspects of Fuel Conversion
     Technology (May 197*0", PB-238 304, October 1974.

35-  Ayer, F.A., "Symposium Proceedings: Environmental Aspects of Fuel Conversion
     Technology II (Dec. 1975)", PB-257-182, June 1976.

36.  National Research Council, "Assessment of Low-and  Intermediate-Btu Gasi f i catior.
     of Coal", FE/1216-4, December 1977.

37.  Waitzman, D.A., Faucett, H.L., and Kindahl, E.E., "Evaluation of  Fixed-Bed,
     Low-Btu Coal  Gasification  Systems  for Retrofitting Power Plants", PB-241-672,
     February  1975.

38.  National Academy of Engineering, "Evaluation of Coal  Gasification Technology,
     Part  Il-Low-and Intermediate-Btu Fuel Gases", PB-234-042, March  1974.

39.  Hall,  E.H.,  Peterson,  D.B.,  Foster, J.F., Kiang, K.D.,  and Ellzey, V.W.,
     "Fuels Technology, A State-of-the-Art Review",  PB 242-535, April  1975.
                                     97

-------
                                REFERENCES (cont'd)


bQ.   Institute of Gas Technology, "Environmental Assessment of the HyGas Process",
      Quarterly Reports, 2, 3, and 4, 1976-77-

41.   Lowry, H.H., "Chemistry of Coal Utilization", John Wiley 6 Sons, Inc.,
      New York, N.Y., 1963

42.   Fink, C., Curran, G., and Sudbury, J., "CO- Acceptor Process Pilot Plant -
      1975 Rapid City, South Dakota", Presented at Seventh Synthetic Pipeline Gas
      Symposium, Chicago,  Illinois, October 27, 1975-

43.   Forney, A.J., Haynes, W.P., Gasoir, S.J., Kornosky, R.M., Schmidt, C.E., and
      Sherky, A.G., "Trace Element and Major Component Balaces Around the Synthane
      PDU Gasifier", PERC/TPR-75/1.

44.   Liptak, B.C., "Environmental Engineer's Handbook", Vol.11, Air Pollution,
      Chilton Book Co., Radnor, Pennsylvania, 1974.

45.   Calvert,S., and Parker, R., "Effect of Temperature and Pressure on Particle
      Collection Mechanisms:  Theoretical Review", A.P.T., Inc. for EPA,  NTIS
      PB-264 203, 1977.

46.   Phillips, K.E.,  "Energy Conversion From Coal Utilizing CPU-400 Technology",
      Final Report, Vol. 1, for ERDA, FE-1536-30 (Vol. l), 1977.

47.   Keairns, D.L., et al.,  "Fluidlzed Bed Combustion Process Evaluation, Phase II,
      Pressurized Fluidized Bed Coal Combustion Development", Westinghouse Research
      Laboratory, for EPA, NTIS PB-246 116, 1975.

48.  Walker, K.A., "Multiclone Collectors", Bull. No. 7010, Environmental Elements
      Corp., Ramsey, New Jersey, 1977.

49.   Gordon, M. "Aerodyne Series "SV" Dust Collector", Bull. No.  1275-SV, Aerodyne
      Development Corp., Cleveland, Ohio, 197B.

50.  Klett, M.G.,  Szwab, W., Clark, J.P.,  "Particulate Control for Pressurized
     Fluidized-Bed Combustion", Gilbert/Commonwealth, R  D Division, for ERDA,
     FE-2220-16, January, 1977.

51.  Shannon, L.J., Gorman,  P.G., and Reichel, M., "Particulate Pollutant System
     Study , Vol.  II:  Fine Particulate Emissions",Midwest Research Institute, for
     EPA,  NTIS PB-203  521, 1971.

52.  Shannon, L.J., "Control Technology for Fine Particulate Emissions", Midwest
     Research Institute, for EPA, NTIS PB-236 646, 1974.
                                    98

-------
                                   REFERENCES  (cont'd)


53.   Bush,  J.,  Feldman,  P.,  and  Robinson,  M.,  "Development of A High-Temperature/
     High-Pressure ElectrostaticPrecipitator",  Research Cottrell,  Inc.  for EPA,
     EPA-600/7-77-132,  1977.

54.   Ensor, D.S.,  Hooper,  R.,  Scheck,  R.W. ,  and Carr,  R.C.,  "Performance and
     Engineering Evaluation  of The  Nucla  Baghouse",     Proceedings of  EPA
     Symposium on  Particulate  Control  in  Energy-Processes, Acurex  Corp., for  EPA,
     EPA-600/7-76-010,  1976,  pp.  377-399.

55.   Spagnola,  H., Turner, J.H.,  "Operating  Experience and Performance  at the
     Sundbury Baghouse",     Proceedings  of  EPA Symposium on Particulate Control
     in Energy Processes,  Acurex Corp.,  for EPA,  EPA-600/7-76-010,  1976, pp. 401-
     428.

56.   Shackleton, M., Kennedy,  J., "Ceramic Fabric  Filtration at  High  Temperatures
     and Pressures",     Proceedings of EPA/DOE Symposium on High  Temperature ,High
     Pressure Particulate  Control,   Acurex Corp.,  for EPA/DOE,  EPA-600/3-7-3-004,
     1977,  pp.  133-234.

57.   Drehmel, D.C., and Ciliberti,  D.F.,  "High Temperature  Fine  Particle  Control
     Using Ceramic Filters", Westinghouse Research Labs,  for EPA,  EPA-600/2-77-207,
     NTIS PB 274485, 1977.

58.   Mills, K., Brunswick  Corp., One Brunswick Plaza, Skokie, Illinois, private
     communicat ion.

59.   Murthy, B.N., Klett,  M.G.,  Becker, D.F., Szwab, W.,  and Fischer, W.H.,  "Fuel
     Gas Cleanup Technology  For Coal Gasification", Gilbert/Commonwealth  R   D Div.
     for ERDA, FE-2220-15, 1977.

60.   Kalen, B., and Zens,  F.A.,  "Ppllution Control Operation: Filtering Effluent
     from a Cat-Cracker",  Chemical  Engineering Progress,  69(5):6?-71, 1973-

61.   Hoke, R.C., and Gregory, M.W., "Evaluation of a Granular Bed Filter,  For
     Particulate Control  in  Fluidized Bed Combustion",     Proceedings of EPA/DOE
     Symposium on High Temperature High Pressure Particulate Control, Acurex
     Crop, for EPA/DOE, EPA 600/9-78-004, 1977, pp. 111-132.

62.   McCain, J.D., "Evaluation of Rexnord Gravel Bed Filter", Southern Research
     Institute, for EPA,  EPA-600/2-76-164,  1976.

63.  Lee,  K.C., Rodon,  I., Wu, M.S., Pfeffer,  R., and Squires, A.M., "The Panel
     Bed Filter", The City College of  the City University of New York, for EPRI ,
     EPRI  AF-560,  1977.
                                     99

-------
                                REFERENCES (cont'd)


6k.  Wade, G.L., "Performance and Modeling of Moving Granular Bed Filters",
     Proceedings of EPA/DOE Symposium on High Temperature High Pressure Particulate
     Control,  Acurex Corp., for EPA/DOE, EPA-600/9-78-00^, 1977, pp. 133-192.

65.  Calvert, S., Patterson, R.G., and Drehmel, D.C., "Fine Particle Collection
     Efficiency  in the A.P.T. Dry Scrubber",    Proceedings of EPA/DOE Symposium
     on High Temperature High Pressure Particulate Control, Acurex Corp., for
     EPA/DOE, EPA-600/9-73-004, 1977, pp.
66.  Moore, R.H., Schiefelbein, G.F., Stegen, G.E., and Ham, D.G., "Molten Salt
     Scrubbing For Removal of Particles And Sulfur From Producer Gas",
     Proceedings of EPA/DOE Symposium on High Temperature High Pressure Particulate
     Control, Acurex Corp., for EPA/DOE, EPA-600/9- 73-00^, 1977, pp. ^30-^63.

67.  Zahedi, K., and Melcher, J.R., "Electrof 1 ui di zed Beds in the Filtration of A
     Submicron Aerosol", Journal of APCA, 26 (k) , 3^*5-352, 1976.

68.  Kirsten, L., Apitron Division, American Precision Industries, Inc., private
     communicat ion.

69.  Robson, F.L., et al . , "Fuel Gas Environmental Impact: Phase Report", United
     Technologies Research Center, for EPA, EPA-600/2-75-078, 1975.

70.  Jahnig, C.E., "Evaluation of Pollution Control in Fossil Fuel Conversion
     Process; Gasification: Section 8, Winkler Process",  EPA-650/2-7A-009-J ,
     September 1975.

71.  Attari , A., "Fate of Trace Constituents of Coal  During Gasification", EPA-
     650/2- 73-004, August, 1973.

72.  Page, Gordon C., "Fate of Pollutants in Industrial Gasifiers",  Presented at
     EPA Symposium, Environmental  Aspects of Fuel  Conversion Technology III,
     Hollywood, Florida, September 1979, EPA-600/7-78-063.

73.  Bombaugh, Karl J., "Analyses  of Grab Samples  From Fixed Bed Gasifiers",
     EPA-600/7-77-1**!, December 1977.

7*.  Lee, M.L., Hansen, L.D., Ahlgren, R., Phillips,  L. .Mangelson, N., and Eatough,
     D.J., "Analytical Study of the Effluents from a  High Temperature Entrained
     Flow Gasifier", Presented at  ACS  Symposium on Environmental  Aspects of Fossil
     Fuel Processing, Anaheim, California, March 1973.

75-  Nakles, D.V., Massey, M.J., Forney, A.J.,  and Haynes, C.P., "Influence of
     Synthane Gasifier Conditions  on Effluent and  Product Gas Production", PERC/RI-
     75/6, December 1975.
                                    100

-------
                              REFERENCES  (cont'd)


76.   Forney,  A.J.,  Haynes,  W.P.,  Gasoir,  S.J.,  Johnson,  G.E., and Stakey, J.P.,
     "Analysis of Tars,  Chars, Gases,  and Water Found  in Effluents  from the Synthane
     Process", U.S.  Bureau  of Mines,  TPR  76,  January  197^.

77.   Forney,  A. J.,  Haynes, W.P., Gasoir, S.J., Kornosky,  R.M.,  Schmidt,  C.E., and
     Sharkey  A.G.,  "Trace Element and Maj'or Component  Balances Around the Synthane
     PDU Gasifier",  PERC/TPR-75/1.

78.   McMichael, W.J.,  Forney, A.J.,  Haynes, W.P.,  Strakey,  J.P.,  Gasoir,  S.J., and
     Kornosky, R.M., "Synthane Gasifier Effluent Streams",  PERC/RI-77/k.

79.   Oldham,  G. and Wetherold, R.G.,  "Assessment,  Selection and  Development of
     Procedures for Determining the  Environmental  Acceptability  of  Synthetic  Fuel
     Plants Based on Coal", FE 1975-3(PT  1),  May 1977-

80.   Sinor, J.E., "Evaluation of Background Data Relating to New Source Performance
     Standards for Lurgi Gasifiers",  PB-269-557, June  1977.

81.   Pellizzari, Edo D., "Identification  of Energy-Related Wastes and  Effluents",
     EPA 600/7-78-00'*, January 1970.
                                      101

-------
                                TECHNICAL REPORT DATA
                          (Please read Ixttmctioru on the reverse before completing)
 1. REPORT NO.
 EPA-600/7-79-170
     2.
                                                       3. RECIPIENT'S ACCESSION-NO,
4. TITLE AND SUBTITLE
 Control Technologies for Particulate and Tar
  Emissions from Coal Converters
                                5. REPORT DATE
                                 July 1979
                                6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
 C.Chen, C.Koralek, and L.Breitstein
                                8. PERFORMING ORGANIZATION REPORT NO.
. PERFORMING ORGANIZATION NAME AND ADDRESS
 Dynalectron Corporation/Applied Research Division
 6410 Rockledge Drive
 Bethesda, Maryland  20034
                                 10. PROGRAM ELEMENT NO.
                                 EHE623A
                                 11. CONTRACT/GRANTNO.

                                 68-02-2601
 12. SPONSORING AGENCY NAME AND ADDRESS
 EPA, Office of Research and Development
 Industrial Environmental Research Laboratory
 Research Triangle Park, NC 27711
                                 13, TYPE OF REPORT AND PERIOD COVERED
                                 14. SPONSORING AGENCY CODE
                                  EPA/600/13
. SUPPLEMENTARY NOTES ffiRL-RTP project officer Is Robert A. McAllister, Mail Drop
 61, 919/541-2134.
ic. ABSTRACT
                      gives res ults of a characterization of solid and tar particulate
 emissions in raw product gases from several types of coal gasifiers , in terms of
 their total quantities , chemical composition, and size distribution. Fixed-bed gasi-
 fiers produce the smallest particulate loadings, about 3 g/cu m. Entrained-bed gasi-
 fiers produce the largest, about 110 g/cu m. Control technologies for particulate
 emissions were assessed with respect to the limitations of the control device  and to
 existing and proposed regulations. Fabric filters were not suitable where tar  parti-
 culates were found or at higher than 300 C.  Electrostatic precipitators operated as
 high as 1100 C.  Rotary cyclones showed the widest range of applicability.  Conventio-
 nal cyclones were most economical for particles larger than 50 microns.  Solid and
 tar particulate  emissions collected for 250,000 scfd of a medium- Btu gas contained
 up to 1. 6 million kg of particulate containing about 400 ppm of organic compounds
 which were benzene extractable. Naphthalenes and three-ring aromatic compounds
 each showed compositions of about 15 ppm. More than a dozen other classes of com-
 pounds were identified in the analyses of the organic material deposited on the parti-
 culate matter.
 7.
                             KEY WORDS AND DOCUMENT ANALYSIS
                DESCRIPTORS
 Pollution
 Coal Gasification
 Tars
 Dust
 Fabrics
 Gas Filters
Electrostatic Preci-
 pitators
Cyclone Separators
                                          b.lDENTIFIERS/OPEN ENDED TERMS
Pollution Control
Stationary Sources
Particulate
Fabric Filters
                                             c. COSATI Field/Gioup
13B
13H
07C
11G
11E
13K
07A
IS. DISTRIBUTION STATEMENT

 Release to Public
                    19. SECURITY CLASS (TM* Report)
                    Unclassified
                         21. NO. OF PAGES

                            112	
                    20. SECURITY CLASS (TU* page)
                    Unclassified
                                             22. PRICE
PA Perm 2MO-1 (-7J)
                   102

-------