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 i»9
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 1100°C) 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 300°C. 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 430°C 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 F»rtltuUt*
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
t«r 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
S»tCwe.
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
1100°C, 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 (788°C) 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
20°C, 1x105 Pa
1,IOO°C, IxlO5 Pa
1,100°C, 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 CycloneC»5)
-------
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. (1»8)
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 65°C.
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 150°C, and to
decrease continuously as the temperature is either decreased or increased. Hot-
side precipitators which operate at temperatures up to 5*»0°C 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
1093°C 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 290°C.
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
(730°C) 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 480°C (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 (530°C,
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 74°C to 130°C 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 370°C.
Rexnord is currently seeking funding to develop a filter for high temperature and
pressure operation,(i.e. 870°C, 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 150°C, 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 (5^0°C) 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 430°C 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 900°C 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 200°C 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 980°C (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 r»n93
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
------- |