D A u-s- Environmental Protection Agency Industrial Environmental Research EPA
-• •» Office of Research and Development Laboratory _ ..
Research Triangle Park, North Carolina 27711 April
EPA-600/7-77-034
METHOD FOR ANALYZING
EMISSIONS FROM ATMOSPHERIC
FLUIDIZED-BED COMBUSTOR
Interagency
Energy-Environment
Research and Development
Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S.
Environmental Protection Agency, have been grouped into seven series.
These seven broad categories were established to facilitate further
development and application of environmental technology. Elimination
of traditional grouping was consciously planned to foster technology
transfer and a maximum interface in related fields. The seven 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
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 systems. The goal of the Program
is to assure the rapid development of domestic energy supplies in an
environmentally—compatible manner by providing the necessary
environmental data and control technology. Investigations include
analyses 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 environmental issues.
REVIEW NOTICE
This report has been reviewed by the participating Federal
Agencies, and approved for publication. Approval does riot
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 recommen-
dation for use.
This document is available to the public through the National Technical
Information Service, Springfield, Virginia 22161.
-------
-------
ABSTRACT
A methodology for comprehensive sampling and analysis of emissions
from an atmospheric fluidized-bed combustor has been developed and tested
experimentally. The methodology tested is a first attempt to develop an
approach to the Level 1 methodology as defined by TRW and is aimed at
providing a cost and information effective environmental assessment of
fluidized-bed combustion (FBC) units. Included in the report is a general
discussion of the pertinent areas likely to be encountered in sampling
and analyzing specimens from FBC units, as for example, the streams
encountered in FBC units, the selection of streams, procedures for sampling
the gaseous, solid, and liquid streams, and the multi-level analysis
approach to characterization of emissions from combustion units as
defined by EPA.
The adopted experimental methodology was put into practice, using
Battelle's 6-inch atmospheric FBC unit. The details involved in sampling
and analyzing the samples from the 6-inch FBC unit are discussed in relation
to (1) the preparatory work, i.e., background information, site survey,
development of a sampling and analysis plan, installation of equipment,
and shake down runs, (2) sampling procedures involved with the FBC unit,
and (3) analyzing the samples taken from the FBC unit. The analytical
data obtained from two runs made with the unit are presented and
discussed, mainly in terms of the trace element data obtained. These
data and other pertinent gaseous and particulate data are compared to
data obtained from other FBC and/or conventional fired units. The
report concludes with the presentation of a generalized sampling and
analysis plan to be used as a guide in characterizing the emissions
from FBC units.
iii
-------
CONTENTS
Abstract
Figures v
Tables v
1. Introduction and Objectives 1
2. Preliminary Plan Development 2
Technical Background - Review of Sampling Analysis
Procedures and Concepts 2
Selection of Influent and Effluent Streams .... 2
Sampling Procedures for Solid, Liquid, and
Gaseous Streams 5
Sampling and Analysis Strategy 8
Sample Preservation and Handling 10
Quality Control in the Sampling Program 12
3. Methodology Development 14
Development of a Comprehensive Sampling and Analysis
Plan 14
Background Information 14
Sampling and Analysis Plan 18
Site Survey 18
Installation and Operation of Equipment 22
Shake Down Runs 25
Collection and Preparation of Samples for Analysis . . 25
Sampling ..... 27
Sampling Problems 32
Labeling 32
Analyses of Samples 32
Additional Considerations in the Analyses of
Samples 32
Analyses Problems 35
4. Evaluation of Emission Data 36
General 38
Trace Elements 38
5. Recommended Sampling and Analysis Plan 45
Areas Needing Further Study and/or Consideration ... 47
References 49
Appendix 50
iv
-------
FIGURES
Page
1. Generalized Fluidized Bed Combustion System 4
2. Scale Drawing of 6-In. Fluidized Bed Combustor 16
3. 6-In. Fluidized Bed Combustor 17
4. Flue Gas Analyzer Flow Schematic 23
5. Typical Gas Absorption Train 26
6. Schematic Outline of Fluidized Bed Combustor and Sampling
Locations 28
TABLES
1. Stream Identification for Generalized FBC System 3
2. General Characteristics of Phased Level 1-Level 2 Strategy . . 11
3. FBC Sampling — Analysis Plan 19
4. Sample Identification and Analyses 33
5. Changes Made Between Run Nos. 1 and 2 37
6. Comparison of Data from Fluidized Bed and Pulverized Coal
Combustion 39
7. Comparison of Trace Element Data from Coal and Ash of Illinois
No. 6 Coal 40
8. Comparison of Trace Elements in Coals and Coal Products .... 41
9. Sampling and Analysis Matrix for Comprehensive Analysis of
FBC Streams 46
-------
SECTION 1
INTRODUCTION AND OBJECTIVES
Fluidized-bed combustion (FBC) offers specific advantages over other
coal-fired combustion processes in that it provides high thermal efficiency
and capability for in-situ antipollution control. Total environmental
assessments of FBC processes are only now being initiated, thus compre-
hensive procedures for collecting and analyzing FBC emission products as
well as reactant products are needed. The objective of this program was
to develop and test primarily a Level 1 methodology* for comprehensive
analysis of emissions from an atmospheric FBC unit and to consider the
application to pressurized fluidized-bed combustors as well as to other
coal-burning processes.
This report summarizes a program carried out under Task 33, EPA
Contract No. 68-02-1409. The emphasis in this program was mainly on
establishing a general methodology for comprehensive sampling-analysis
from FBC units and not to define the precision and/or accuracy of the
techniques employed. To achieve the overall objective of the program,
four distinct phases were considered. These were:
A. Developing a preliminary approach for comprehensive analysis
B. Designing and conducting a test program
C. Evaluating the emission data
D. Refining the preliminary comprehensive analysis approach.
Along with developing the general methodology, the development was
tested using Battelle's 6-inch fluidized-bed combustor.
* Level I methodology similar to that defined by TRW (1); all references
are on page 49.
-------
SECTION 2
PRELIMINARY PLAN DEVELOPMENT
At the time this program was undertaken two pertinent^ activities were
in progress. MITRE Corporation (2) was completing a state-of-the-art
survey of fluidized-bed sampling and analysis procedures and was developing
guidelines for carrying out these procedures, and IERL/RTP (3) was in
process of formalizing the Level 1-2-3 sampling-analysis concepts. Both
of these programs impacted on this program and provided useful background
information. Pertinent information in these program is reviewed here.
TECHNICAL BACKGROUND - REVIEW OF SAMPLING
ANALYSIS PROCEDURES AND CONCEPTS
The MITRE report* is pertinent to this sampling program in that it
presents a detailed up-to-date study of the tentative procedures for
environmental assessment of fluidized-bed units. The primary objective
of the work covered in the MITRE report was to develop sampling and
analysis procedures for use by source contractors in support of the environ-
mental assessment of fluidized-bed combustion technology. The review
presented here covers briefly the areas of interest to this program, i.e.,
sampling and analyses procedures and concepts, and constituents likely
to be encountered in fluidized-bed sampling. Pertinent information from
a Battelle planning study report (4) is included in the review.
Selection of Influent and Effluent Streams
In order to characterize the emissions from combustion processes, it
is important to consider all possible streams in the combustion system
which may lead to pollutant formation. In the case of the FBC units, a
total of 30 different streams have been recognized as possible sources of
pollutants. The 30 streams are shown in Table 1. The relationship of
the various streams to a generalized FBC system is shown in Figure 1.
* The review covered in this report is from the First Draft of the MITRE
report. A Second Draft of the report has been completed and some changes
are likely to have been made.
-------
TABLE 1. STREAM IDENTIFICATION FOR GENERALIZED FBC SYSTEM
Stream No. Stream Identification
1 Stack Gas from FBC
2 Particulate Removal Discard from FBC
3 Bed Solids Discard from FBC
4 Particulate Removal Discard—Regeneration Operations
5 Other Effluents from Regeneration and Sulfur Recovery Operations
8 Product from Sulfur Recovery (Sulfur or Sulfuric Acid)
9 Fugitives from Fuel Preparation
10 Fugitives from Sorbent/Additive Preparation
11 Raw Fuel to Preparation
12 Raw Sorbent/Additive to Preparation
14 Air to Combustor
15 Air/Steam to Regenerator
16 Fuel Feed to FBC
17 Fuel Feed to Regenerator
18 Start-Up Fuel Feed to FBC
19 Prepared Sorbent/Additive Feed to FBC
20 Bed Solids to Regenerator
21 Flue Gas from FBC to Particulate Removal
23 Recycle of Particulates from Particulate Removal to FBC
24 Recycle of sorbent from Regeneration
31 Fugitive or secondary Emission from Fuel Storage Facility
32 Fugitive or Secondary Emission from FBC Discard Bed Material
33 Fugitive or Secondary Emission from FBC Particulate Disposal
34 Effluent Gas from Secondary Stack Gas Cleaning Device (Similar to
Streams 25 & 26 in MITRE Study)
35 Discard from FBC Secondary Stack Gas Cleaning Device
36 Flue Gas from Regenerator to Particulate Removal
37 Fugitive or Secondary emission from Regeneration Bed Discard
38 Stack Gas from Regenerator to Sulfur Recovery Operations
39 Fugitive or Secondary Emissions from Regenerator Particulate Disposal
40 Fugitive or Secondary Emissions from Sorbent Storage Facility
-------
SORBENT/
ADDITIVE
PREPARATION
OPERATIONS
AND STORAGE
SECONDARY
STACK GAS
CLEANING
DEVICE
FUELS
PREPARATION
OPERATIONS
AND STORAGE
SULFUR
RECOVERY
OPERATIONS
PARTICIPATE
REMOVAL
OPERATIONS
(CYCLONES)
DISPOSAL
OPERATIONS
FIGURE 1. GENERALIZED FLUIDIZED BED COMBUSTION SYSTEM
-------
The numbering system has been adopted from MITRE. '
Many of the streams listed in the table are encountered in combustion
processes other than FBC units and would be considered there as appropriate.
On the other hand, some streams encountered in some FBC units have been in-
tentionally omitted from Table 1 on the basis of their low priority, or be-
cause they apply to units which were judged inappropriate for inclusion in
this study program, e.g., the CAFB units.
For each FBC unit sampled, certain criteria should be considered before
selecting or rejecting a sample stream. Overall, stream characterization
should be made when:
• A direct emission stream is representative of a FBC process
release to the environment
• A feed or supply stream will significantly affect direct
emissions
• A within-process stream might significantly affect an emission
stream or the performance of an emission control device
• A fugitive emission is identified, or suspected, which might
be characteristic of a reduced-to-practice FBC unit
• A secondary emission is simulated in a supplementary test
involving a supply or effluent stream.
Sampling Procedures for Solid, Liquid, and Gaseous Streams
Many of the selected streams will require special sampling techniques.
The following discussion is a synopsis of the recommended techniques to be
used in sampling solid, liquid, and gaseous streams, as well as suggestions
for sampling fugitive emissions. These are presently the tentative recommend-
ed procedures.
Solid Streams—
Recommended techniques for sampling solid streams include:
• Grab-full stream cut
• ASTM pulverized coal samples
• Automatic-full stream cut-Vezin type samples
• Stopped belt sampling method.
-------
In using these techniques, one should also take into consideration,
1. Heterogeneous nature of the material and degree of
stratification likely to be encountered
2. Tendency of different size fractions to concentrate
certain materials in sample, e.g., some trace elements
may tend to concentrate in the smaller size fractions.
With the above two points in mind, it would appear that the full stream cut
or the stopped belt methods would provide the best representative samples
from solids.
The following considerations should also be observed in sampling solid
streams to minimize contamination of the sample.
1. Sampling equipment surfaces in contact with sample
should be made of stainless steel to minimize erosion
effects.
2. Sample container material should have weak erosive
potential and be acid resistant and free of toxological
contaminant. Kapton is generally acceptable.
3. Air-tight stainless steel containers should be used for
high temperature samples.
Liquid Streams—
The recommended procedures for sampling liquid streams include:
• Grab-full stream cut
• Tap
• Automatic high volume samples
• Carbon absorber concentration techniques.
Sampling liquid streams is relatively straightforward if there are no sus-
pended particles or non-miscible liquids in the stream. Since this is not
generally the case, representative sampling of liquids becomes more difficult.
Point sampling and high non-isokinetic flow rate sampling are the prevalent
methods in water sampling. These techniques can be considered accurate for
most needs but may not suffice for trace sampling. Further work is needed to
verify the sampling techniques in this area.
In sampling liquids for trace metals, the liquid is filtered on site and
the filtrate stabilized with nitric acid. For trace organics, no filtering is
involved; the sample is preserved by refrigeration at 4° C. In sampling
liquid systems, Teflon, stainless steel and glass are the only materials
usually used.
-------
Suspended Particulate—
The Source Assessment Sampling System (SASS) train is the recommended
technique (5) for suspended particulate sampling of fluidized-bed units. This
unit and the Method 5 train fulfill the following criteria necessary for
efficient particulate sampling:
1. Provide sufficient quantity of sample for analyses
at each particle-size fraction desired.
2. Minimize number of sampling trains and avoid excessive
personnel interference.
3. Meet precision and accuracy goals defined for each
sampling level (Level 1 and Level 2 discussed later).
The SASS train meets all the above criteria for Level 1 particulate samp-
ling and criteria 1 and 2 for Level 2 sampling. The Method 5 rig with an
impactor and a high volume cyclone train meets criteria 3 for Level 2 partic-
ulate sampling, and in some cases may meet criteria 1.
The following considerations are important in particulate sampling for
minimizing contamination of either the trace organic, trace inorganic measure-
ments, or bioassay procedures.
• Stainless steel 316 should be maintained as the principal
material of construction with Viton-A used as gasket materials.
• Filters should be high-purity quartz material and should be
routinely analyzed to insure low inorganic concentration.
• Trains must be packed and unpacked in a clean room environment
to minimize contamination.
Gases—
The recommended techniques for sampling of gaseous components include:
• EPA Methods 3 through 7 (6), and 10 (7)
• ASTM methods
• Absorbing solution methods
• Condensation methods and others.
All of the chosen techniques have had proven field experience, except for the
porous polymer adsorber technique for nonvolatile hydrocarbons. Options to
some of the selected methods are available and are presented in the Mitre
report.
Other considerations in sampling gases are
1. Organic grab samples taken in glass bombs should
not be left in sunlight and should be analyzed as
soon as possible.
7
-------
2. Precautions must be taken to insure no stratification
of sample occurs.
3. Materials of construction must be inert towards
the sample material.
Fugitive Emissions—
Fugitive emission measurements are made on the basis of a background
versus operating plant method. Background ambient air quality measurements
are documented prior to the plant's construction or during a plant shutdown
for those plants already constructed. In the latter case, the aggregated
storage piles would have to be covered during the background measurements.
The background measurements are compared to measurements made during plant
operation and fugitive emission levels determined from the two measured levels.
A meteorologist should determine the duration of background and operation
fugitive testing.
Sampling and Analysis Strategy
A multilevel approach is used in sampling and analyzing samples from the
various selected streams of a combustion process. The levels are referred to
as Level 1, 2, and 3. Each of the levels is defined below, as taken verbatim
from reference (3). A more detailed report covering Level 1 sampling and
analyses has been issued recently by EPA (5) . Level 2 procedures are currently
being worked out. Level 3 will require further considerations.
Level 1 Sampling—
'Level 1 sampling stresses the concept of completeness by presuming that
all streams leaving the process will be sampled unless empirical data equiva-
lent to Level 1 outputs already exist. Further, Level 1 sampling is not pre-
dicated on a priori judgements as to the composition of streams. The
techniques prescribed presume that whatever prior knowledge is available is at
best incomplete. Predictive and extrapolation techniques employed during
environmental assessments serve as a check on the empirical data and not as a
replacement for it.
Level 1 sampling programs are therefore envisioned to permit detection of
the presence of all substances in the stream. They do not necessarily produce
information as to specific substances or their chemical form. For example, if
sulfur-containing gases are in the gas stream, Level 1 sampling will trap and
retain the sulfur. However, it is not designed to preserve the sulfur com-
8
-------
pounds as H2S, S02, COS, C^H^S, (CI^^S, etc. (In many cases a reasonably
specific compound Identification may result at Level 1, but conceptually it is
not necessary in judging the success of Level 1.)
Level 1 sampling programs are designed to make maximum use of existing
samples and stream access sites. While some care must be exercised to ensure
that the samples are not biased, the commonly applied concepts of multiple-
point, isokinetic or flow proportional sampling are not rigidly adhered to.
Normally, a single sample of each stream should be collected under average
process operating conditions, or alternatively, under each condition of in-
terest. These samples should be time-integrated over one or more process
cycles. When a series of discrete samples results, they are combined to pro-
duce a single "average" for analysis.
Level 2 Sampling—
Level 2 sampling programs are directed towards a more detailed charac-
terization of stream composition. They are not as "complete" as Level 1 in
that resources are expended to improve information on streams of a critical
nature. Additional sampling of other streams may be deferred because Level 1
information has indicated a potentially less-significant level of environ-
mental impact. Level 2 sampling is optimized for specific compounds or classes
of compounds contained in the streams sampled. It also provides a more
quantitative description of the concentrations and mass flow rates of the
various substances in the stream.
Level 2 sampling is considerably more refined than Level 1, since it is
being conducted on streams that have already been identified as having poten-
tially adverse environmental effects. In some instances, Level 2 will use
the same sampling techniques and equipment as Level 1. The primary difference
will be a more rigorous attention to selection and preparation of sampling
sites and adherence to procedures for acquiring a representative sample.
Level 2 sampling should also provide for replication of samples in order to
further improve on accuracy and representativeness.
In many cases, Level 2 sampling will utilize modifications of Level 1
techniques and/or the application of entirely new methods. Such cases result
from the necessity to identify more definitively the materials which produce
the adverse environmental problems. For example, if Level 1 has indicated a
-------
high concentration of sulfur-containing species, Level 2 sampling must be
specifically designed to provide isolation of the materials into appropriate
forms for subsequent analysis.
Level 3 Sampling—
At Level 3, emphasis is placed on the variability of stream composition
with time and process or control system parameters to define accurately the
range of values to be expected. An effective Level 3 sampling program is de-
signed to monitor a limited number of selected compounds or compound classes.
Level 3 sampling is designed to provide information over a long period
of time. To be cost effective, such programs must be tailored to the specific
requirements of each stream being monitored. Based on the information
developed at Level 2, specialized sampling procedures can be designed to track
key "indicator" materials at frequent intervals. Level 3 should also incor-
porate continuous monitors if at all possible.
During Level 3 programs, it is anticipated that more complete Level 2
type sampling will be conducted at predetermined intervals to check the limited
Level 3 information. Further, recommended procedures for compliance testing
should be introduced into the program at a time appropriate to the status of
process or control technology development."
Table 2 characterizes the Level 1-Level 2 strategy for combustion samp-
ling. Although the extent of the Level 2 and Level 3 sampling will depend on
results from Level 1 and Level 2 sampling, respectively, it is likely that
hybrids of these levels will be used in most combustion sampling programs. As
i
seen later in this report the present sampling program used a combined Level 1-
Level 2 approach. The extent to which this occurs depends largely on the in-
formation desired and the relative cost incurred in obtaining the extra data.
Sample Preservation and Handling*
Avoidance of contamination and degradation of samples is of prime impor-
tance in the comprehensive analysis of FBC units. Contamination considera-
tions extend from the preparative and packing stages, through the various
collection procedures, and to the analysis of the collected samples.
* Discussion taken from reference 4, page 33.
10
-------
TABLE 2. GENERAL CHARACTERISTICS OF PHASED LEVEL 1-LEVEL 2 STRATEGY (2)
Issue
Level 1
Level 2
GENERAL
• Goal
• Process operating conditions
• Streams considered
• Pollutant classes/species
addressed
Detection of potential pollutants and
planning basis for Level 2
Steady-state representative condition
All system Influents and effluents
All pollutant claaaes
Accurate measurement of specific
pollutants and determination of
mass emission rates
Steady-state representative condition
Selected system Influents and.effluents
Selected class/species
SAMPLING CHARACTERISTICS
• Representative of sample
• Sampling technique
• Sampling location
• Replications
• Sample quantity
Sufficient for detection of all
potential pollutants
Particulates: initially isokinetlc,
single point
Gases: single point grab unless stra-
tification exists (then full traverse)
Solids and liquids: partial stream cut
Level 2 sampling locations (aa incor-
porated into generic plant design) used
for Level 1 sampling where practical
Particulates: cross-section point of
average velocity
Gases: same location as for partlcu-
lates
Solids and liquids: most convenient
location consistent with obtaining
partial stream cut
Particulates: 3 minimum; composite
two samples and analyze
Gases, solids, liquids: none
Gases and particulates: dictated by
analytical method employed
Solids and liquids: per ASTM method
Consistent with Level 2 goals
Particulates: isoklnetic, full traverse
Gages: single point unless stratifica-
tion exists (then full traverse);
combination of integrated grab, vet
Federal Register methods, and
continuous on-line samplers
Solids and liquids: full stream cut
(minimum requirement)
Recommend Level 2 sampling capability
be incorporated in generic plant
design
Farticulates: full traverse of cross-
section; distances from disturbances
sufficient to minimize Irregular flow
patterns
Gases: same location as for particu-
lates
Solids and liquids: most convenient
location consistent with obtaining
full stream cut
Particulates: 3 minimum; separate
analysis for each sampling
Gases: complies with Federal Register
specs; statistically designed to
relate manual with continuous on-line
Gases and particulates: dictated by
analytical method employed
Solids and liquids: per ASTM method
and Federal Register specs
ANALYTICAL CHARACTERISTICS
• Sensitivity
• Accuracy
• Replication
• Specificity
• Gas analysis
• Elemental analysis
• Anlon analysis
• Organic solids and liquids
Highest sensitivity (consistent with
Level 1 resources)
Order of magnitude
None planned
Broad classes of organlcs
Some species (inorganic and organic)
Gas chromatography; absorption tubes
Spark source mass spectrometry
Not performed
Extraction, separation into functional
classes
Analysis by fourler transform infra-red
spectroscopy
Sensitivity requirements dictated
by Level 1 output
High accuracy
Statistically designed
Individual species corresponding
to class/species Identified at
Level 1
Primary reliance on GC (or best
method for Individual species
Atomic absorption spectroscopy (or
best method for individual species)
Method dependent on individual species
Extraction, separation by high
resolution
Liquid chromatography
Analysis by GC mass spectroscopy, IR,
NMR (or best method for individual
species)
BIOASSAY CHARACTERISTICS
• Analyses performed
Cytotoxicity, mutagenlclty
Cytotoxicity, mutagenlclty, carcino-
genlcity
11
-------
Containers used for packaging reagents for field use and storage of samples
should be dedicated to this purpose and scrupulously cleaned before put in
storage for subsequent use.
The collection and packaging of samples to be used for carcinogenicity
and mutagenicity should be handled in such a way as to minimize the amount of
degradation brought about by air, temperature, and light. Ideally, the
collected samples should be placed in brown bottles, purged with nitrogen or
argon, sealed, and frozen in dry ice. Brown bottles are often difficult to
obtain and cylinders of inert gas are frequently not available for field use.
When these adverse conditions prevail, the sample collection protocol would
be altered to use clear wide mouth jars or bottles. Immediately following
the collection of the sample, the container is sealed with tape around the
lip of the lid and wrapped in aluminum foil. Once sealed, the containers
should be carefully packed in dry ice of sufficient quantity to ensure arrival
of the frozen samples at their destination. Refrigeration at temperatures
near freezing or below is the best preservation technique available, but is
not applicable to all types of samples. Samples in water solutions cannot
be frozen in glass bottles due to possible breakage. At best, preservation
techniques can only retard the chemical and biological changes that inevi-
tably continue after the sample is removed from the parent source. Contami-
nation and degradation of samples can best be avoided by analyzing the
samples as soon as possible after collection.
Quality Control in the Sampling Program*
Quality control must be supplied in the sampling program to assure that
representative samples are obtained from the various process streams and
that the sample integrity is not compromised prior to delivery to the ana-
lytical laboratory. Obtaining representative samples requires that appro-
priate sampling locations and sampling technique be selected. These factors
will be especially critical when sampling the stack and flue gas streams for
particulates. Maintenance and calibration of equipment is also essential to
collection of representative samples. Calibration of flow meters and tempera-
ture measuring devices should be performed before each field trip. Pitot
* Discussion taken from reference A, page 35.
12
-------
tubes should be calibrated and leak tests made on the sampling trains on a
regular basis. Checks should be performed in the field to verify the equip-
ment is still within calibration. Spares should be available to replace
defective equipment.
Maintaining sample integrity demands that careful consideration be given
to materials used in the sampling systems or equipment. Materials should be
chosen to minimize the introduction of contaminants. Sample recovery from
particulate sampling trains must be performed in an area and using equipment
and techniques which will not contaminate the samples. All samples must be
stored in containers which are leak-tight and which do not introduce contami-
nation. Certain samples (for organic analyses and bioassays) must be pro-
tected from light and temperature extremes to maintain the integrity.
Continuous gas monitoring instrumentation presents special calibration
requirements. During a field program, these instruments should be calibrated
before and after each test using standard gas mixtures. Periodic analyses
should be performed to verify the concentrations of these calibration (span)
gases stated by the gas suppliers.
13
-------
SECTION 3
METHODOLOGY DEVELOPMENT
In consideration of the preceding discussions, the methodology develop-
ment in the present program was based around sampling and analysis experi-
ments with Battalia's 6-in. fluidized-bed unit. The Battelle unit provided
a means for evaluating the generality of proposed procedures in the laboratory.
DEVELOPMENT OF A COMPREHENSIVE SAMPLING AND ANALYSIS PLAN
A considerable amount of planning and preparation was required prior to
carrying out the sampling and analysis plan for the 6-inch FBC unit. The
preliminaries include, appropriate background information on the unit, develop-
ment of a sampling and analysis plan for the unit, site survey and planning,
installing and operating equipment, and shake down runs.
Background Information
Prior to the actual sampling of a FBC unit, it is essential that the
appropriate personnel become throughly familiar with the design and general
operation of the unit. Pertinent areas of interest include but are not
limited to
• Process description
• Process schematic flow sheets
• Plant layout
• Equipment specification
• Plant personnel data
• General range of firing conditions capable with the FBC
• Results of any previous emission testing.
Background information relating to the design and general operating fea-
tures of the Battelle unit is discussed below.
14
-------
Battelle 6-in. FBC—
The 6-in. FBC facility at Battelle can be fired in different velocity
modes with solid fuels at rates up to 50 Ib/hr. Preheaters enable the
incoming gases to be heated to nearly 1000 F, as desired, prior to
combustion. Fluidized-bed operating temperatures from 1400 to 1800 F
are established through control of the fuel feed-rate and preheat
temperatures.
Figure 2 shows a scaled drawing of the 6-in. FBC unit used in this
program. Figure 3 is a photograph of the unit. For this work, it was
desirable to fire the combustor in a low velocity mode typical of that
used in conventional FBC units. High sulfur Illinois Number 6 coal was
used for the fuel and grove limestone served as the bed material. These
are fed to the combustion chamber on a small conveyer belt. Cyclones are
used in the flue gas and sampling streams to fractionate particle sizes.
Normally, two cyclones are present in the flue gas stream from the FBC
unit, but for this sampling program, only one cyclone was used in this
stack gas region. This cyclone (No. 1) removes all particulates above
about 27 microns. A second smaller cyclone was installed in the Method
5 rigs to remove particulates between about 27 and 2.3 microns in the
sampled gas. This is discussed later.
Sampling ports are available at several different points in the
combustor. These are indicated in Figure 2 by the capital letters A, B,
C, D, and E. The legend in the figure describes the general type
sampling taking place at each port. An overflow tube (T) inside the
bed reaction chamber maintains the height of the bed at 4-ft and allows
continuous removal of the bed materials during a run. The overflow
tube acts essentially as a sampling port within the fluidized-bed region.
The flue gas, after passing the final sampling Port C, is directed
into a water scrubber where the gases are further cleaned before exiting
to the atmosphere. Solid material (sludge collected in the scrubber)
is removed from the scrubber by filtering.
The above background information provided a basis for choosing
sampling locations, proper sampling conditions, and selection of appropri-
ate sampling streams with subsequent development of a sampling and analysis
plan.
15
-------
58-
LEGEND
Continuous Monitoring
Sorption Monitoring
Particulate Monitoring
Gas (NO ) Monitoring
X
--T/Z.
~-Tio
FIGURE 2. SCALE DRAWING OF 6-IN. FLUIDIZED-BED COMBUSTOR
16
-------
FIGURE 3. 6-IN. FLUIDIZED-BED COMBUSTOR
17
-------
Sampling and Analysis Plan
The initial efforts in this program were directed to the development of
a comprehensive sampling and analysis (S&A) plan applicable to fluidized-bed
combustors and other coal burning units. The plan was developed around the
design and operating features of the Battelle FBC unit. Since the Battelle
unit is essentially a bench scale FBC model, many of the streams shown in
Table 1 were not applicable to this unit. Notably absent from the Battelle
unit were those streams relating to regenerative processes and fugitive emis-
sions. Other streams were omitted on the basis of low priority. The streams
chosen for sampling in this program were:
• Stack gas
• Particulate removal discard
• Bed solids discard
• Prepared fuel feed to combustor
• Prepared sorbent feed to combustor
• Scrubber discard.
The S&A plan derived from the Battelle unit was refined and later final-
ized into the form shown in Table 3. The tabulation shows the streams sampled,
the constituents measured in the streams, the analysis methods used, and the
procedure used to obtain the sample. This plan was later generalized to in-
clude all streams likely to be encountered in FBC S&A programs. The general-
ized plan is given in the final sections of this report (Table 9).
Site Survey
A site survey is conducted prior to carrying out the S&A plan developed
for the FBC unit. A site survey is an important part of the sampling program
in that it provides proper coordination of the plant's physical layout with
the test plan strategy. The more important points considered at the site
survey inspection for this program were:
a. Are the sampling sites accessible and is there proper space
allocation and supports (tables, platforms, etc.) for all
the equipment and supplies?
b. Are there a sufficient number of sampling ports and do
the sampling sites meet all the criteria for representative
sampling?
c. Do the sampling sites have access to utility outlets of the
number and type required by the sampling equipment?
18
-------
TABLE 3. FBC SAMPLING — ANALYSIS PLAN
Level 1 — Scanning for Classes of Compounds
Train and/or Sampling Method
Analysis Method
1976
Eat. Cost
COAL
Pollutant
Proximate/Ultimate
Moisture
Ash
Volatile matter
Fixed carbon (by difference)
C, H, N, S, 0 (by difference)
Sulfur - Total
Pyritic
Organic
S0,=
(525)
Obtain three or four grab samples,
-100 grams each, at intervals during
run and combine for representative
bulk sample
ASTM D291
ASTM D2492-68
Trace metal
Na, Ca
Particle size
Heating value
LIMESTONE
Particle size
Ca, Mg, CO', NO', N0~
Trace metals
OVERFLOW BED MATERIAL
Particle size
(Chemical identification)
Trace metals
SO", SO", S=
NO", NO,
Organic classes
ASH
-325 mesh material from overflow
bed material (weigh)
Trace metals
SO', SO", S-
NO-,
Obtain three or four grab samples,
-100 grams each, at intervals during
run and combine for representative
bulk sample
Collect all of overflow material;
sieve for -325 mesh material; weigh
-325 mesh sample and use for ash
Organic classes
SSMS
AA
Sieve
Calorimeter/ASTM D2015-66
Sieve
AA - Spectrophotometric, phenol desulfonic
acid
SSMS
Sieve
SSMS
Wet chemical methods
Extraction/liquid chromatography/IR
Sieve
SSMS
Wet chemical methods
Extraction/liquid chromatography/IR
(345)
(780)
(780)
-------
Train and/or Sampling Method
Analysis Method
1976
Est. Cost
t-o
O
SLUDGE
C, H, N, S, 0
so'
so'3
Trace metals
Organic classes
FLUE GAS STREAM
1. Participate >27 v
a. Trace metals
1) 60 (approx)
b. Minor elements (cations)
1) Fe, Al, SI, K
c. Anlons
1) Chloride, fluoride
2) CO-
3) NO", NO-
4) SO', SO'
d. Organics
1-8) Organic classes
9) POM
10) Organic-reduced sulfur
compounds (8 classes
combined)
e. C, H, N, S, 0
f. Size analysis
2. Partlculate (<27 u, >2.3 y)
a. Trace metals
b. Minor elements (cations)
c. Anions
d. Organics
e. C, H, N, S, 0
f. Size analysis
2a. Partlculate <2.3 u
a. (Same analyses as under
Section 2)
(790)
1 gallon sample from scrubber at end
of run; sample filtered and dried for
4 hrs at -160 F, remaining solid used
for sludge sample; solution stored in
capped bottle
Cyclone No. 1 emptied two or three
times during run and at end of run;
sample combined and stored in dark;
ahort-term exposure for high-level
material; long-term exposure for low-
level materials
Two M-5 trains, Isokinetlc sampling,
approximately 2 hrs; sample taken
from cyclones in M-5 rigs and stored
in dark
Sample from filters in M-5 rigs, or
cascade impactor; sample stored in
dark
Barium perchlorate tltratlon
H.O./Barium perchlorate titratlon
SSMS
Extraction/liquid chromatography/IR
SSMS
SSMS
SSMS
AA
Spectrophotometric, phenol disulfonic acid
Barium perchlorate titratlon (H_02 with SO
Extraction/liquid chromatogtaphy/IR
Fluorescence - fraction No. 2 above
GC/FPD
Sieve
(As described in No. 1 above)
(1160+)
(11604-)
(As described in No. 1 above)
(1160+)
TABLE 3. (Continued)
-------
Train and/or Sampling Method
Analysis Method
1976
Est. Cost
3. Gases
a. Acid gases*
1) C02
2) S02
3) NO - NO,
4) SO ^
b. Inorganic*
1) CO
2) 02
c. Organic
1) Total gaseous hydrocarbon
2) Organic classes:
alkanes, alkylbenzenes,
POM, thiophenes, carba-
zoles and esters,
aldehydes, PCB, ketonea,
and alcohols
3) Organics - reduced sulfur
(8 classes combined)
d. Hg, Cd, As, Pb, Se
e. HC1, HF, HCN, NH *
(755)
Continuous
Continuous
Continuous
Continuous
Continuous
Continuous or bag
Sample stored in dark
Bag or Tenax
10% KMnO, or K.Cr.Oy acid solutions;
use HO. prior to solutions to remove
SO ; solutions stored in dark
Gas absorption train (Figure 5 in
text); 0.01 N, NaOH used to collect
HC1 and HF, 5% KOH solution for HCN
and IN H,SO, to collect NH
NDIR
NDIR
NDIR - Chemiluminescence
Goks«Syr-Ross
NDIR
EC - Polarograph
FID
Tenax extraction/liquid chromatography/IR
CG/FPD
AA
Ion chromatograph (HC1, HF) and specific
ion electrode (HCN, NH,)
* Level 2 scanning for specific compounds.
-------
d. Do the sites meet critical OSHA requirements?
e. What arrangements must be made for transporting equipment
from the ground to the sampling site?
f. Does the plant have adequate laboratory space available;
if not, can the cleanup van be parked in close proximity
to the sampling area?
g. Does the plant have personnel available to assist in the
acquisition of solid and liquid samples?
h. What is the plants normal operating schedule and down-time
of equipment?
i. What type and condition is the on-line measurement
ins trumentation?
j. Are support services available to the field team such
as storeroom, welders, glass blowers, etc.?
Occasions may arise where other points may need consideration but these major
areas should suffice for most FBC sampling programs.
Installation and Operation of Equipment
A wide variety of instruments and sampling units are required in sampling
a FBC or other type combustion unit. Gas, liquid, and solid samples are
generally involved. Some instrumentation is likely to be available and in
operation at the plant site. If so, consideration should be made as to their
suitability for use in the comprehensive analysis program.
Continuous Gas Monitoring—
Continuous gas monitoring instruments were available and in operation at
the Battelle FBC unit and were used in this program for measuring CO, C02,
S02> NOX, 02 and total hydrocarbons from the combustor. The instruments used
for measuring each gas are given in the S&A plan shown in Table 3. Figure 4
is a schematic of the gas analysis system. The sample tubing to these in-
struments was made of stainless steel and/or Teflon and of proper diameter
to allow sufficient gas flow to the instruments. Stainless steel tubing was
installed at all sampling ports in preference to Teflon because of the some-
what high temperature for Teflon.
Calibration procedures were worked up for each instrument prior to the
sampling runs. Flow rates to the instruments for either sample gases or
span gases were adjusted to the needs of the instrument as specified by the
manufacturer. Stainless steel bellows pumps were used to provide sample gases
22
-------
N>
OJ
Particulant
Water traps trap _ ; n
««^i
N2 —
CO — $>
span
Icebath pirex wool f "\
^ I
vJ^ ATH
s
S.S. Bellows
pump
\ \
T • Y
V
V^vr
•••tr
H
=
IL
'
1
f Water, trap
dry icebath
5 1 1 jj- |
6b nn Y °
N2-<2>- N2-^~
Co2——
span span
L J-, -« -Sllicn gel
=^ ^i coiuim
O O
T ¥ Y
• i 1
S02 Anal. NO Anal. 02 Anal. 13
t t 1
flow nctcrs
FIGURE 4. FLUE GAS ANALYZER FLOW SCHEMATIC
-------
to the instruments. Room temperature fluctuations were not a problem in our
sampling, but temperature conditions must be considered if cold weather or
wide fluctuations in temperatures are involved. Traps were used to remove
particulates and moisture from the sample gas prior to entering the instru-
ments. Pyrex wool served as a filter to remove the particulates and a wet-
or dry-ice bath was used to effectively reduce the moisture content of the
sample gas stream to an acceptable level for operating the instruments. (A
word of caution here. While the removal of particulate and water vapor from
the sample line is imperative to proper operation of most of the instruments,
the total effect of the removal processes on the sampled constituents mainly,
N02 and SC>2, is questionable. As noted later, more consideration should be
given to the effects of the removal processes.)
Method 5 Sampler for Particulate and POM Samples—
Method 5 samplers, were installed to obtain particulate samples in this
program. Modifications were made to the Method 5 sampler as follows, (1) in-
stallation of a small glass cyclone inside the heated chamber prior to the
filter to further fractionate the particulates, and (2) installation of a
tenax column downstream of the filter outside the heated chamber to collect
POM samples.
A "Y" probe was constructed so that two Method 5 rigs could be used to
sample particulates simultaneously from a single sampling point within the
flue gas stream. Sample ports were installed so that samples could be taken
at about midstream of the duct at least 10 tube diameters downstream from any
elbows. Gas velocities were measured during the shake down run for isokinetic
sampling. Operation of the Method 5 rig was in accordance with the Federal
Register procedures (6), except no traverse of the duct was made.
Bubbler Samples—
In addition to the continuous monitors and the Method 5 rigs, sorption
trains were constructed and used in this program for collecting acidic (HC1,
HF, and HCN) and basic (NH3> gaseous components present within the gas stream.
Although not specifically required in a Level 1 analysis scheme, these sorp-
tion trains were installed to collect and analyze each of the gases for a more
complete characterization of the emissions from the FBC unit. An oxidizing
solution was also installed to collect certain trace metals that may be
24
-------
present in vapor form in the stream. The metals of interest were Pb, Hg, Se,
Te, Be, As, and Cd. Figure 5 shows a single typical absorption train used in
the program.
Although SO-j can also be collected in solutions such as alcohol-water
mixtures, it was decided that the Gtfksoyr-Ross method would be the better one
to use for 803 since much of the alcohol-water solution would evaporate over
a two-hour period. Therefore, the Goksoyr-Ross apparatus was installed in
place of a sorption train to collect 803. The manifold and sample line to
the Goksoyr-Ross unit beyond the manifold were heated with heating tape, to
prevent H2SC>4 condensation. It is important that these areas be heated to at
least 325 F to avoid loss of 803 (as H2S04).
Shake Down Runs
Unless the FBC unit has been run previously at the desired operating con-
ditions, it is generally worthwhile to make at least one preliminary or shake
down run prior to the actual sampling runs. In so doing, the technicians
operating the fluidized-bed unit are able to check their calculated values
for producing the desired bed temperature and gas flow rates, SC>2 levels, ex-
cess air, etc., and make appropriate adjustments where needed prior to the
actual sampling run.
Shake down runs were made prior to the sample Runs 1 and 2 in this pro-
gram. Fluidized-bed operation was checked out and proper adjustments made.
The technicians taking the samples were then able to make preliminary checks
on the sampling conditions of the unit under the desired operating conditions.
Appropriate temperatures of the various sampling ports were checked and gas
velocities measured, where necessary, to allow for isokinetic sampling during
the actual sampling runs. Instruments were checked out for proper operation
and checks were made for leaks in the sampling units. At the completion of
each shake down run, the FBC unit was prepared for the sampling run which
followed the next day.
COLLECTION AND PREPARATION OF SAMPLES FOR ANALYSIS
Details of the procedures involved in obtaining, handling, and preparing
samples for analysis from our fluidized-bed combustor are discussed in refer-
ence to the sampling and analysis plan shown in Table 3. This plan was
closely followed in each of the sampling runs made.
25
-------
SAMPLING
ZONE
PROBE
V
.
['//////7//J
V/M///A
HEATED
GLASS
FIBi.R
FILTER
....
^
»
r""
HEATED
MANIFOLD
*
1
1
I
^
•
.Fir .
1
1
1
. i
\ / EMPTY SILICA
• 1MPINGERS rr, , rr,
V*LZ L> L* lj LZ Iw
TO VENT —*-
i-QJ:
PUMP
DRY
GAS
METER
FIGURE 5. TYPICAL GAS ABSORPTION TRAIN
X
CHECK
VALVE
26
-------
Sampling
Samples were taken from the various selected streams after stabilizing
the FBC unit at the desired firing conditions. Sampling was started after
the SC>2 level in the gas stream was reduced to the desired level, approxi-
mately 700 ppm, via reaction with the limestone. Six major streams were
sampled as indicated earlier. The sampled streams are discussed in the order
given in Table 3. Figure 6 is a schematic outline of the sampling locations
in the FBC unit. Each sample stream discussed below is numbered in accordance
with the numbered locations in Figure 6.
(l)-Coal Feed Stream—
Three or four grab samples of coal, about 100 grams each, were taken at
about equal time intervals over the 2-hour sampling period. The crushed coal
was fed to the FBC unit on a small conveyor belt. Since the unit burned only
about 15 Ib/hr or less of coal, the coal stream to the unit was small and
cross-section samples of the stream were used as representative of the coal
mass. In larger units where much more coal is used, the procedure for col-
lecting a representative coal sample is more involved. The proper procedures
have been noted in an earlier section of this report.
The 3 or 4 grab samples of coal were combined, as the run progressed, in
a single plastic container and, at the end of the run, the total sample was
sealed in the plastic container and stored.
(2)-Limestone Feed Stream—
The limestone was also fed to the FBC unit on a small conveyor belt and
samples of limestone were taken in the same manner as for coal. The total
sample was sealed in a plastic container and stored. As with the coal, the
large combustors would require more detailed sampling procedures.
(3)-Overflow Bed Material—
This material consisted mostly of limestone and was continuously sampled
through the overflow pipe in the reactor bed during the entire 2-hour sampling
period. A representative sample was taken from the overflow container at the
end of the run. This was done by scooping samples from various part of the
mass bed material and combining them in a plastic container. The container
was sealed and stored in the dark. (All samples to be analyzed for POM and
other organics were stored in the dark to minimize decomposition.)
27
-------
Cyclone
#1
n
Bed /
•—s
3
z_
Air
Particulate
>2?K. (
Coal
Limestone ©
Ash
Flue Gas Analyzer
n - 1 — i - 1 - 1 i
CO C02 02 S02 NOX THC
Gas Absortion
Trains I
rtffi"
Scrubber
Sludge
xTenax Plug
(Pom, He)
Mod. Method 5 Trains
I I |
Particulate
Particulate
I
FIGURE 6. SCHEMATIC OUTLINE OF FLUTDI7.ED-BED COMBUSTOR
AND SAl-IPLTNG LOCATIONS
28
-------
(4)-Ash—
The ash sample referred to here was taken from the overflow bed material.
It is the small -325 mesh particles remaining in the overflow bed material at
the completion of the run. All of the overflow bed material was sieved at the
end of the run into several size fractions. The smallest fraction (-325 mesh)
was used as a representative ash sample from the bed material. Generally less
than 0.1 percent of the total overflow bed material remained in this smallest
fraction. However, sufficient sample was obtained in most runs for the proper
analysis. The ash sample was sealed in a plastic container and stored in the
dark.
(5)-Sludge—
The sludge sample is the solid material caught in the water scrubber
attached to the end of the flue gas stack. In obtaining a sludge sample, the
water used for scrubbing the flue gas was continuously removed from the
scrubber barrel and passed into a large centrifuge. The centrifuge was lined
with a fine filter which retained the solid material while passing the liquid.
The centrifuge was run during the entire sampling period. Several liquid
filtrate samples (about 1 gallon each) were taken from the centrifuge drain
at recorded time interval during the run and stored in large plastic bottles.
The solid material remaining on the filter was removed by scraping; the solid
sample was dried and sealed in a plastic container and stored with the other
samples in the dark.
(6) through (9)-Flue Gas Stream—
A large number of samples was taken from the flue gas stream. Samples
were obtained from the in-stream cyclone and from two Method 5 rigs, several
sorption trains and a number of continuous monitors. Particulate and gaseous
samples were involved. Sampling procedures are described below.
(6), (7), and (S)-Particulate—Particulate samples were obtained from two
sources (1) the solid material removed by the cyclone in the flue gas stream,
and (2) from samples taken by the Method 5 rigs. The cyclone and Method 5 rigs
provided three size fractions of particles. These were (approximately as cal-
culated from cyclone dimensions),
1. Particles greater than 27 y diameter
29
-------
2. Particles less than 27 y greater than 2.3 y diameter
3. Particles less than 2.3 y diameter (0.1 y limit).
The larger cyclone, which was a part of the flue gas stream, removed the
larger particles (>27 y ). A smaller cyclone installed in each Method 5 rig
'removed the intermediate size particles (<27 y , >2.3 y). The filter in the
Method 5 rig removed the remaining particles down to about 0.1 y.
Samples from the larger in-line cyclone (>27 y) were withdrawn from a
particulate collection chamber at the bottom of the cyclone at the end of the
2-hour run. The samples were sealed in a plastic bottle and stored in the
dark.
Particulate samples <27 y, >2.3 y were obtained from the cyclone in the
Method 5 rig under the following conditions.
1. Method 5 (isokinetic) sampling procedure was used.
2. Sample was collected at midstream of 4-in. diameter
stainless steel duct at a position about 1-ft from
end of duct (sample point C in Figure 2).
3. Heated stainless steel tubing, 1/2-in. I.D. was used
to withdraw sample from stream into Method 5 rig at
the rate of about 0.75 cfm (using appropriate size
nozzle to obtained desired isokinetic flow rate).
4. Particulate samples were collected for period of 2
hours.
5. Samples were taken from cyclone in Method 5 rigs
after run was completed.
6. Samples were stored in sealed glass containers in a
dessicator (in the dark and under N£ gas if stored
for several days). Particulate samples were combined
for analysis if more than one sample taken during run.
A problem was encountered in the separation of particle sizes. The designated
fractions were not always obtained from the Method 5 rig samples since the
smaller cyclone tended to clog rendering the separation meaningless. In those
instances, the filter and cyclone samples from the Method 5 rig were combined
to form a less than 27 y fraction. Particulates were also washed from the
sample probe with acetone followed by wash with methylene chloride. These
samples were dried and added to the <27 y catch. Filter samples in the Method
5 rig were removed when filter loading became sufficiently high. The filter
and loading were weighed and placed in a glass container, sealed and dessi-
cated. The samples were removed from the filter prior to analysis by flexing
30
-------
the filter material; this released most of the caked-on sample. The filters
were reweighed and particulate sample weight determined. All particulate
samples were stored in glass containers, and placed in dessicators in the
dark.
(9)-Impinger Samples—These samples were obtained by sorption of a particular
constituent in a solution which efficiently trapped the desired material. For
each sorption train, two solutions were used in series to insure more complete
removal of the desired constituent. The second solution collected only a
small fraction of the total sample (<10 percent). The impingers were immersed
in wet ice for better sorption of samples. Each unit (sorption train) was op-
erated independent of each other. Flow rates and total gas sampled were re-
corded in each sorption unit. Pumps were downstream of the bubblers and all
gases were dried with drierite prior to entering the pumps. Sampling rates
were about 0.2 cfm; the sampling period was usually 2 hours. Approximately
10 to 20 cubic feet of gas passed through each bubbler.
At the end of a 2-hr sampling period, the solutions were removed from
each impinger and placed in a separate glass bottle. Volumes were measured
and samples were stored in a cabinet at room temperature until ready for dis-
tribution and analysis.
The G8ksoyr-Ross unit was washed with distilled water then 3 percent
peroxide solution and acetone to remove the SO- (as H-SO,). The volume
of solution was recorded and the sample stored with the above impinger samples.
(9a)-POM—POM samples were collected on a tenax column attached to the samp-
ling line of the Method 5 rig downstream of the filter. The tenax column was
covered at all times to keep out light. At the end of the run, the column
was removed and filled with nitrogen, sealed and stored in a refrigerator
(kept in the dark at all times).
(lO)-Gases—
The remaining samples from the flue gas stream involved essentially the
gaseous species in the stream. The acid gases C02, S02, and N0-N02 and the
inorganic gases CO and 02 were monitored continuously as described earlier.
Total hydrocarbon content was also monitored continuously.
31
-------
Sampling Problems
A few problems were encountered in the sampling of the Battelle FBC unit.
These, however, were generally isolated events and produced no threat to the
completion of the sampling task. The specific problems encountered were:
• Small cyclones in Method 5 rigs tended to clog after several
minutes of sampling rendering particle size separation incom-
plete.
• Flue gas temperature at sampling port dropped below that
desired for good 803 sampling (325° F) for Runs 2 and 3.
• Filters would load heavily by end of a 2-hour run. Some
had to be replaced during the run (Method 5 rig).
Labeling
Proper labeling of samples is a very important part of the sampling pro-
gram. Each sample in each stream listed in Table 3 was labeled so as to
identify the stream sampled, material or sample collected, run number and
date of run. With proper sample labeling, no confusion will result in re-
lating the analytical results to the appropriate sample. It is equally im-
portant that starting and finishing times for all processes be noted and
coordinated with the individual collection processes for each sample category
so as to provide valid comparison of all pertinent data.
Analyses of Samples
Analyses charts were drawn up for each run relating the sample taken from
the FBC unit to the analyses to be made on the sample. An example chart is
shown in Table 4 for Run No. 2 samples.
As shown in Table 4 each original sample was divided into a number of
smaller fractions so that sufficient material was available for the different
analyses needed on each sample. The number of fractions of each sample needed
is indicated by the number of "forks" or divisions after each original sample
(shown in boxes) in Table 4. Each fraction of sample was weighed to the near-
est 0.1 gram before distribution for analysis. In the case of solutions, the
volume of each solution was recorded prior to distribution for analysis.
Additional Considerations in the Analyses of Samples
Three other analysis areas not shown in the analyses chart of Table 4
but which should be considered in a FBC sampling program are (1) biological
testing, (2) radioactivity measurements, and (3) noise measurements. Areas
32
-------
TABLE 4. SAMPLE IDENTIFICATION AND ANALYSES
Fluid-Bed Sample Number
Material Sample No. Designation
Illinois 16 coal
(-8 M) S-2-1
Grove limestone
• S-2-2
Bed material S-2O
Ash (-325 M) S-S-4
Sludge S-2-5
Flue gas stream
Particulate >27u 2
S-2-1-1 /
-/ S-2-1-2
\ S-2-1- 3
\5-2-l-4
S-2-2-1
\ S-2-2-2
\S-2-2-3
S-2-3-1
x^ s-2-3-'t
\ S-2-3-2
\i-2-1--}
S-2-4-1
-V^ S-2-4-2
\ S-2-4-3
V;_2-it-.l>
• S- 2-5-1
\.. S-2-5-?
\S-2-5-3
S-P-6-1
/ S-2-6-2
^\ S-2-6-3
\ S-2-6-4
Proximate /ultimate
/
^ Sulfur
Na, Ca
Heating value
Trace metals
Particle Size
Ca, Mg, CO" NO"
Trace metals
Particle Size
Trace metals
Organic Classes
SO" , S0j, S=, NO",
Particle Size
NO" NO" s", SO"
Fusion temperature
Trace metals
Organic Classes
C, H. N. S. 0, SO"
Trace Metals
Organic Classes
Ai
C~(l) Moistu
") (2) Ash
1(3) C.H.N.
V, 0(by d
Gl) Total
\(2) Pyrite
NOl
NO;
, and S0~3
, and SO*
Trace metals (approx 60 metals)
__ C, H, N, S, 0
Anione , NO", NO
Organic classes
-. so", so;
(5) Fixed carbon (by
difference)
;rence)
(3) Organic
(4) SO.
X
Partical Size
POM
Organic and reduced sulfur compounds
33
-------
TABLE 4.
Fluid-Bed Sample Number
Material . Sample No. Designation
Flue
Parti
(If t
frac
here
>2.3
sraal
desig
S-2-
yses
Flue
Gases
Cases
ooni
gas stream S-E
culaM? <21\l
rfo particulate
bions involved
, e. g. <2?
, and <2.3 ,
ler fraction is
nated Batch No.
i and same anal-
as S-2-7.)
gas stream S-£
continuoua S-2
:orine
S-2-7-1 Tra
7 ./ S-2-7-2
V
\\ S-2-7-3
C.
Ani
\ S-2-7-4 Orgi
Particle Size
(AL-3) S-2-9-1 s^'
/ ^-~^
-9 ~
-------
1 and 2 are included in the general sampling and analysis plan in Table 9 in
a later section of this report. Although samples were available for analysis
from our sampling program in areas 1 and 2 above, no testing or measurements
were made in our laboratories. However, samples were sent to EPA for biologi-
cal testing. On the other hand, noise measurements were made in the area of
the 6-in. FBC unit used in this program. The results are included in the
appendix. Further discussion of noise analyses and areas 1 and 2 above are
given in the "Recommended Sampling and Analysis Plan" section.
Analyses Problems
No serious problems were encountered in the analysis of samples taken in
this program. The determination of the concentration of a species "by dif-
ference" (e.g., the oxygen values in some of the samples in Table 4) is likely
to produce more error than the direct measurements made on other species.
But this would be expected from the procedure involved.
35
-------
SECTION 4
EVALUATION OF EMISSION DATA
Three runs were made in the course of this sampling program. These
are designated Run Nos. 1, 2, and 3. The firing conditions for each of
these runs differed as indicated in the following tabulation.
Run Run Run
No. 1 No. 2 No. 3
Coal feed rate, Ib/hr 15.9 8.8 9.3
Limestone feed rate, Ib/hr 15.4 4.3 8.1
Air feed rate, Ib/hr 145.0 87.3 84.2
Bed height: expanded, inches 48 48 48
settled, inches 12.6 21.6 21.6
Bed temperature, F 1538 1655 1490
Superficial gas velocity, ft/sec 9.1 6.0 5.3
Ca/S ratio 6.7:1 2.9:1 7.1:1
3
Particulate loading, g/m 1.44 1.64 NA
Sampling procedures were the same for Run Nos. 2 and 3, but different
slightly between Run Nos. 1 and 2, as did also some of the analyses
procedures between these latter two runs. These differences are brought
out in the list shown in Table 5. The changes were made to bring about
a more cost effective sampling and analysis plan while at the same time
completely defining the major pollutants emitted from the FBC unit.
Two of the three runs, Runs 1 and 2, were selected for analyses.
The results of the analyses of these samples are presented in the Appendix,
along with the firing conditions for each of the Runs 1, 2, and 3.
The primary purpose of the Runs 1 and 2 data presented in the Appendix
was to reduce the practice the various sampling and analysis procedures
under actual fluidized-bed operation. Although FBC emission data are
36
-------
TABLE 5. CHANGES MADE BETWEEN RUN NOS. 1 AND 2
Change Mode
Reason for Change
6.
7.
Proximate/ultimate analysis made on bed material -325 M
in Run No. 1; discontinued in Run No. 2
Fe, Al, Si, K, Cl~, F~ analyzed individually (e.g., atomic
absorption, ion electrode, etc.) in bed material samples
in Run No. 1; determined these elements by SSMS in Run
No. 2
NCL, SO , S , and CO anions added to the analysis of bed
material and particulate samples in Run No. 2
Analysis for POM increased to include bed material,
sludge, and particulate samples in Run No. 2
Organic class analyses increased to include bed material
and sludge samples in Run No. 2
Increased analysis of trace elements to include sludge
samples in Run No. 2
Increased particle size analysis to include bed material
and particulate (both >27 and <27 microns) samples in
Run No. 2
Analysis of little use
More efficient, eliminate dupli-
cation
More completely define anion concen-
trations in solid samples
More completely define the POM levels
in solid samples
More completely define organic classes
in solid samples
To better define trace elements in
effluent waste material
To define particle size ranges more
completely
Reduced quantity of solid sample for bioassay from 20
grams in Run No. 1 to 2 grams in Run No. 2
Only 2 grams of sample needed for
analysis
-------
currently quite limited, the data obtained in Runs 1 and 2 also merit
some evaluation in terms of other coal burning emissions — both fluidized-
bed and pulverized coal combustion emissions. Tables 6, 7, and 8 present
s uch comp ar is ons.
General
In Table 6 some average comparisons are made between the Run 2 FBC emis-
sion data and some small scale pulverized coal (PC) firings (8). The latter1
were data obtained by Battelle on another EPA program, Contract No.
68-02-2119, concerned with the combustion of specially treated coals. The
pulverized coal (PC) data shown are for raw (untreated) coals with sulfur
contents of about 2 percent. (We have directed our attention to Run 2, rather
than Run 1 here, because it was at a higher temperature, about 1655 F.)
As regards the gaseous emissions, the principal differences are in the
CO and S02 emission levels. S02 levels are noticeably lower in the FBC
operation, as expected, due to the capture of S02 by the limestone. The
limestone capture efficiency in Run 2 for a 4.1 percent sulfur coal was about
67 percent.
The high CO levels in FBC operation compared with pulverized coal firing
is another characteristic, so to speak, of FBC. The high CO levels in the
FBC run here also go hand-in-hand with high hydrocarbon emissions and
possibly with low NOX emissions. Hydrocarbon emission in PC firing are
usually quite low and therefore are often not even measured.
It is sometimes stated that NO emissions from FBC operations are lower
than from PC firings. The data in Table 5 neither support nor contradict
that statement. In a recent study on FBC NOX emissions under Contract No.
68-02-2138 we also point out this disparity (9).
Particulate loadings (particles <27 microns, particles that pass the
first cyclone) appear quite similar for FBC and PC firings. The same can be
said for POM loadings and for the ash analyses - within the limits that one
can compare data from a single FBC run.
Trace Elements
Table 7 shows a comparison of trace element data from an Illinois No. 6
coal (the same type coal as used in this study) and its ash. The analyses
38
-------
TABLE 6. COMPARISON OF DATA FROM FLUIDIZED-BED AND
PULVERIZED COAL COMBUSTORS
Gases
o2, %
co2, %
CO, ppm
S02> ppm
NO , ppm
A
HC, ppm(C)
Particulate
Loading, mg/m"
POM, ug/m3
Ash
C, weight percent
H
N
S
FBC
Run 2
3.6
17.3
2090
730
350
360
1640 (1500)*
5 (72)*
8-20%
.25%
0.1-0.3%
4-6%
PC
2-10%
13-15%
90-300
1200-1500
200-700
NA
1000-6000
0.1-65
high, to 50%
0.3%
0.1-0.4%
1-3%
* Run 1.
39
-------
TABLE 7. COMPARISON OF TRACE ELEMENT DATA
FROM COAL AND ASH OF ILLINOIS NO. 6 COAL
PPMW
Element
As
Be
Br
Ce
Co
Cr
Dy
Eu
F
Fe(xlo4)(a)
Hg
K(xl03)(a)
La
Mn
Na(xl02)(a)
Pb
Sc
Sm
Tb
Yb
ANL
2.1
1.6
4.1
3.2
22
0.2
79
1.2
1.2
1.5
3.9
19
3.0
8.0
2.1
0.005
1.4
COAL
BCL
Run
No. 1
2.7
0.17
4.7
4.0
13
110
>1
0.16
1.2
5.0
80
8.9
0.47
1.6
0.71
0.23
Run
No. 2
<3
2
10
10
100
<0.5
*3
*5
<2
•\,6
20
100
200
<1
3
<1
0.5
ANL
13
38
34
800
8.5
5.2
13
0.007
16
40
160
27
46
34
3.1
11
ASH
BCL
Run
No. 1
2.6
15
2.1
15
1.4
290
>1
<0.01
>5
6.4
60
^14
4.3
1.6
0.19
1.6
Run
No. 2
0.20
50
10
100
3
^5
^2
<0.3
M.O
50
200
30
20
30
0.5
1
(a) Indicates values in table that are to be multiplied by the
factor shown, e.g., 1.2 is 12000 ppm Fe.
40
-------
TABLE 8. COMPARISON OF TRACE ELEMENTS IN COALS AND COAL PRODUCTS (yg/g)
Element
As
Ba
B
Be
Br
Ce
Cd
Co
Cr
Cu
Dy
Eu
F
4\(d]
Hf
Hg
K(xl03)
La
Li
Mn
Mo
Ka(xl02)
Ni
Pb
Sb
Sc
Se
Sm
Tb
V
Yb
Zn
Zr
ANL BCL
Run . Run "
TR 3^' No.l
5 2.7
35
93
0.7 0.17
13 4.7
9.8
0.10
1.8 4.0
100 13
9.0
0.2 0.70
25 110
1 1 >1
0.18
0.15 0.16
5.8 1.2
4.2 5.0
0.13
26 80
2.9
6.9 8.9
33
29 0.47
0.3 0.56
1.7 1.6
. 0.29
0.8 0.71
20
<1 0.23
37
10
Coal
(b)
Run Power Mean
No. 2 Plant Value (c>
<3 14
200
<0.03
2 1.6
10 15
50
<30
10 0.6
100 14
30 9.6
<1
<0.5
<_3 61
-5 0.37 1.9
<2
<2 0.070 0.2
-6 1.6
20
20
100 49
30 0.99
200 5 '
<10
<1 35
<0.5 1.3
3
<5 1.9
<1
<0.2
500
<0.5
<10 7.3
300 13
Final
Bed
ANL BCL
Run Run
TR 3 No.l
3.5 0.67
180
33
0.8 <0.18
4.5
15
0.14
0.14
12 1.3
78
0.49
0.25
170
55 0.46
96 1.1
<0.005 <0.01
0.66 -1.6
3.7 6.4
5.7
39 26
1.2
13 3.2
2.3
51 1.9
0.7
1.8 0.16
0.1 0.83
7.0
5.2 <0.49
17
390 28
Run
No. 2
1.0
200
5
<0.005
1.0
5.0
<10
0.3
10
3
<0.2
<0.3
£3
-0.10
<0.5
<0.3
-2.0
2
1
20
<3
2
<2
1
<0.2
1.0
<2
<0.5
<0.1
10
<0.2
<3
20
Power
Plant
Sair.ple
BA
15
82
6.6
0.140
3.5
7.7
58
220
41
-------
Primary Cyclone
Element
As
Ba
B
Be
Br
Ce
Cd
Co
Cr
Cu
Dy
Eu
F
Fe(xlOA)
Hf
Hg
K(xl03)
La
Li
Mn
Mo
Na(x!02)
Ni
Pb
Sb
Sc
Se
Sm
Tb
V
Yb
Zn
Zr
(a) Run
(b) Run
coal
(c) Mean
ANL
Run
TR 3
25
350
2.6
19
11
180
1.8
20
(d> 5.9
2.9
0.46
3.7
31
110
41
95
3
9
4
Power
BCL Plant
Run
No.l
4.1
180
500
2.6
6.0
15
0.67
2.1
15
66
1.4
0.50
290
>1
1.8
<0.01
>5
6.4
60
3.3
-14
23
4.3
0.67
1.6
0.94
1.7
0.19
25
1.6
67
35
TR 3 bed temperature
No. 1 bed temperature
used.
analytical
(d) Recorded values
Run Sample
No. 2 MA
1.0 44
500
50
0.20
0.50
50
<3
10
.100
50 150
3
1
<5
-2 7.0
2
<0.3 0.026
-10
50
10
200
20 12
30
200
20
0.2
30
<5 4.1
5
0.5
500
1
<3 100
200 260
1560 F; Arkwright
1560 F; Run No.
values for constituents in
to be multiplied by 104, e
Secondary Cyclone
ANL
Run
TR 3
860
6.0
3
13
19
300
2.9
10
3.6
6
0.46
5
52
140
72
260
6.2
19
7.5
BCL
Run
No.l
6.1
320
2000
6.0
60
44
1.0
14
87
120
2.0
0.94
450
>1
1.8
14
>10
30
57
60
14
>50
99
43
1.4
7.3
9.4
3.6
0.48
140
2.5
140
120
Run
No. 2
3
300
300
2
1
20
<3
10
100
50
2
1
<_20
-2
<0.5
<0.3
-20
30
20
500
10
50
100
100
1
30
<5
3
0.3
1000
1
<3
200
Power
Plant
Sample
PA
120
230
6.9
0.310
41
27
250
210
coal used.
2 bed temperature 1655
101 different
.g. , reported
coals.
value of
F; Illinois
1 is really
ppmw.
42
-------
were determined by Argonne National Laboratory (ANL) (10) and BCL. The
ash from the burned coals were obtained from different combustion pro-
cesses. The ash material analyzed by ANL was from an unquenched gasi-
fier ash of the coal while ash analyzed by BCL was from the primary
cyclone of the FBC unit. The BCL analyses were spark source mass spec-
trometric analyses; the ANL group used different analytical techniques
in their analyses; these include wet chemical, atomic absorption,
fluorimetry, specific ion electrode, and neutron activation analysis.
Approximately 2/3 of the BCL and ANL trace element coal analyses agree
within a factor of three. Those elements showing deviation greater than a
factor of three include Be, Hg, Mn, Pb, Sm, Yb, Cr, F, La, and Na. Manganese
and lead are the only elements showing large deviations common to BCL Runs
1 and 2. The mercury values in the ANL data are reported as not representa-
tive of the coal seam.
Considering the possible difference in coal samples, the non-statistical
approach to these analyses, and the different analytical techniques used, the
agreement of data in the majority of the elements is satisfactory.
Most of the elements reported in the ash of BCL Run No. 1 and the ANL
run deviate by more than a factor of three. The elements showing large de-
viations are Be, Co, Cr, Dy, F, Fe, La, Pb, Tb, and Yb. On the other hand,
all but five of the elements reported in BCL Run No. 2 ash data agree within
a factor of three with the ANL data. Those elements showing greater than
factor of three deviation are Be, Cr, Fe, Tb, and Yb, all of which are also
common to the elements showing large deviations in Run No. 1
One would not necessarily expect the data from the various ash samples
to agree as closely as in the case of coal, since the ash samples were taken
from two different combustion processes (as mentioned above) and probably
are not comparable in particle size. The BCL ash sample contained particles
greater than 27 microns. The ash size is not reported for the ANL data.
Considering the possible difference in ash sample and the different analytical
techniques, the ANL and Run No. 2 data are in fairly good agreement.
43
-------
Table 8 offers an interesting comparison of trace element data from
fluidized-bed and pulverized coal combustors. These data are derived
from the ANL pressurized FBC study (10), a coal-fired power plant study (11)
and the present FBC study. The ANL study was carried out in a 6-inch
diameter pressurized fluidized-bed combustor, quite comparable to Battelle's
unit except the ANL unit was operated at about 8 atm absolute pressure.
The coal-fired power plant data were obtained from samples from the 180
MW Unit No. 5 of Public Service Company of Colorado's Valmont Power
Station near Boulder, Colorado.
In the Table 8 data, different coals were used by ANL and BCL in
firing the fluidized-bed combustors. ANL used an Arkwright coal and BCL
an Illinois No. 6 coal. However, the bed temperatures for ANL Run TR3
and BCL Run No. 1 were the same, 1560 F. Data from combustors other than
FBC units are also included in Table 8.
In summary, the comparison of the emission data in these studies
bears out the reduction to practice of the comprehensive sampling and
analytical techniques developed here for fluidized-bed combustors. The
data presented here however should only be used at this time for the
purpose of evaluating the sampling and analysis technique. This program
was not designed for statistical evaluation of the data. The data
presented in the two reference studies selected here for comparison
should also only be accepted at face value.
44
-------
SECTION 5
RECOMMENDED SAMPLING AND ANALYSIS PLAN
The generalized S&A plan for fluidized-bed units is presented in Table
9. This plan is similar to the one presented in Table 3 with appropriate
additions in Table 9 to include those sample streams not covered in sampling
the Battelle FBC unit. Table 9 lists all the main streams to be considered
in sampling a FBC unit, specific pollutants involved, collection technique,
analysis method and level approach.
Two areas of analysis shown in Table 9 which may require further brief
comment are the biological testing and radionuclide measurements. In refer-
ence to biological testing, currently it is felt that five tests of a
screening nature are available to evaluate cytotoxicity, mutagenicity, and
carcinogenicity. These tests are:
1. Ames' bacterial mutagenesis toxicity assays
2. BCL - prescreen toxicity assay (Mammalian cells - BALB/c 3T3)
3. BALB/c 3T3 clonal transformation assay (mouse fibroblast cell)
4. C3H 10T 1/2 mouse prostrate cell assay
5. Syrian hamster embroy clonal transformation assay.
A more detailed description of the assay options available is presented in
the Battelle report referenced earlier.
Radioactive content of samples should be considered in an overall samp-
ling program. It is suggested that a low background proportional counter be
used to make the measurements. The proportional counter gives a measure of
the gross a and 3 emissions from thorium, uranium, radium, and their decay pro-
ducts in the samples being analyzed. The instrument is highly sensitive and
-i r\
can detect radioactivity levels below a picocurie (10 x curies).
Some of the procedures involved in the analysis of other components
shown in Table 9 have been discussed previously. Other analytical procedures
45
-------
IT1
W
5
C/3
H
S
X
n
iS
33
M
25
Spaciei, Pollutants Sample Analysis Method^*' System Stream or Material
Collection f ) (Stream Number)
Techniques '
C02 Cw NDIR
S02 Cw Infrared or UV
N0 Cw NDIR or Chemiluminescence
N02 Cw NDIR or Chemi luminescence
CO Cw NDIR
°2 Cw Paramagnetic or Ft. ElectrodP
Integrated CQE Phase MM8,,re«enta
H2S IG GC
COS IG GC
DIsul fides IG GC
S°3 2S°A ^t Goksoyr-Ross/Ion chroma tography
NH3 St Kjeldahl
Cyanides St Coloriraetric
HC1 St Titration
Integrated Specimens Collected for Subsequent Group Analysis
Trace metals SASS/Gs SSHS
Major Elements (Fe, Al, Si, K, Ca) SASS/Cs OES
Organic, by class SASS/Gs Extraction/Liquid Chromatography/IR
Organic-reduced, sulfur compounds SASS/Gs GC/FPD (8 fractions combined)
POM SASS/Gs GC/HS
Proximate Gs ASTM D3172-73
Atmospheric Stack Gaa (la)
>10ti U3u
i
X
X
X
X
X
:
X
Ultimate SASS/CB ASTM PJ176-74 ' ' X
Sulfur forms Cs ASTM D2492-68 ' i
Biological SASS/Gs In vitro
Toxic Elements (Be,Cd,Hg,As ,Pb,Se,Sb,Te) -SASS/Gs AA
Cl SASS/Gs Color imetric |
F~ SASS/Gs Distillation/ Colorimetric
Na Gs AA
X
X
X
X
<3U
X
X
X
X
X
X
Filter
X
X
X
X
X
X
X
1
X
X
X
X
X
X
X
X
X
X
!
X
Pressur-
ized
Supply.
f!4)
X
X
Particulate
Removal
2nd Cyclone
X
X
X
X
X ' X
X
X
X
x
- 4
X
X
X
X
,X
X
X
: X
X
Bed
(3)
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Fuel
Feed
(16)
X
X
X
X
X
X
X
X
X
X
X
Sorbent
Feed
(19)
X
X
X
X
X
X :
X X
X
X
! x
X
i--:—
X
X
X
X
X
; x
x
X
X
X
X
X
X
X
Leachate
from Bed ,
(32)
\
ace
from
2no
cy-
mt '
X X
X
X
X X
X X
X X
X X
i r
i x
i x
1 ?_
J
M5 - EPA Method 5
(a) Cw - Continuous withdrawal through non-reactive line with mechanical filtration
Ci - i>s'-ade Impuctor in flowing stream.
Gs - Grab multiple samples riffled to reduce to 100 g representative sample.
IG — Integrated grab sample of gas in glass bulb.
St — Separate wet chemical train to collect gas (such as
SASS - Source Assessment: Sampling System.
-------
can be obtained from the reference methods given in the table; some are simply
standard calorimeteric or titration procedures for the listed component.
One other area to be considered in the environmental assessment of a
FBC unit is the noise generated by the unit and its associated components.
The analysis should include an inventory of all significant noise sources.
Noise emission data would be obtained for each source through direct measure-
ment and/or from available literature. An integrated noise effect, taking
into account transmission paths and isolation, would be generated from appro-
priate existing computer programs. One can develop a map of noise intensity
within and exterior to the plant.
Table 9 is suggested for use in planning an effective sampling and
analysis program for specific fluidized-bed units. Proper selection of
streams and pollutants to be measured can be obtained from Table 9 (with the
necessary prior information on the FBC unit). One can then proceed to
sample the FBC unit and develop an analysis scheme, such as shown in Table 4,
for each stream sampled.
AREAS NEEDING FURTHER STUDY AND/OR CONSIDERATION
In the course of this program, it became apparent that certain areas in-
volving (1) the sampling and analysis of samples, (2) the associated costs
and timing involved in carrying out a sampling program, and (3) the Level 1,
2, 3 approach to sampling should receive further consideration. Specific
areas of concern are presented below:
1. Two questionable areas were encountered in S03 sampling — (1)
the accuracy of the Gb'ksoyr-Ross method in determining 803
levels, and (2) the importance of flue-gas temperature in
sampling 803. The wide variation in S03 levels encountered
in this sampling program suggests that both the collection
method and flue-gas temperature effects should be examined
further to assess their relative importance in obtaining
representative 803 samples.
2. Further consideration should be given to the effectiveness
of sorption solutions in removing trace elements and the
acidic and basic components from the gas stream. Also, what
interference problems exist and how may they be overcome.
3. Cyclones were used to separate the particles into different
size ranges. The larger cyclone in the flue-gas stream of
our unit appeared to work properly (removing particulates
>27 ]i). The smaller cyclones in the Method 5 rigs produced
somewhat of a problem in that they tended to plug after
47
-------
several minutes of sampling. The small cyclones were heated
to about 375 F which should reduce sticking tendencies in the
particles. Nevertheless, the particles clung to the neck of
the small cyclones making a clean-cut separation impossible.
Enlarging the neck of the cyclone improved the separation but
Changed the size fraction range in the cyclone. Further con-
sideration should be given to these cyclones if used:.in the
Method 5 rig.
A further look at the effect on the sample of using cold
traps to remove moisture, and pyrex wool filters to retain
particles, would be worthwhile in the gas analysis part of
the sampling.
The use of stainless steel tubing in sampling N02 should re-
ceive careful attention since heated stainless steel can
reduce N02 to NO.
If time requirements for analysis is an important factor in
the sampling and analyses program, the following should be
considered (1) the actual time requirements needed to carry
out a complete analysis of all samples regardless of other
factors, and (2) the time required to have an analysis
carried out taking into consideration other activities of
the contractor's analytical laboratory. The latter area may
lead to the need of analytical assistance from outside
laboratories. The contractor should therefore have a clear
understanding of the identity of such laboratories, the
quality and reliability of their work, and the cost for
analysis.
An item requiring serious consideration in a sampling and
analysis program is the cost involved in carrying out the
analyses (as well as the sampling) part of this program.
Cost estimates should be made for each analysis carried
out on each sample. This requires a detailed analysis plan
such as shown in Table 4. Consideration of costs prior to
sampling can be an important factor in determining the ex-
tent of sampling and/or analysis carried out in the program.
The Level 1, 2, and 3 approach to sampling and analyzing a
FBC unit was not entirely satisfactory for our program. It
was felt that a combined Level 1-Level. 2 approach would be
more cost and information effective than strictly adhering
to Level 1 sample and analysis procedure. This is especially
true for certain group analyses where it would be better to
obtain sufficient and proper samples to analyze for indi-
vidual components in a group (e.g., 803 =, 804 = N02~ N03~)
than to analyze the group as a whole. Also it appeared
well worthwhile to spend extra time obtaining samples of
specific components, e.g., HCN, NH3, etc., than to return
later for these samples.
48
-------
REFERENCES
1. Ackerman, D. G., Ryan, L. E., Maddalone, R. A., and Flegal, C. H.,
Suggested Approaches for Environmental Assessment Level I Data
Reviews and Level II Analyses, TRW Defense and Space Systems Group,
Contract No. 68-01-3152, March, 1977.
2. Abelson, H. I. and Lowenbach, N. A. , Review of Sampling and Analysis
Techniques for the Environmental Assessment of Fluidized-Bed
Combustors, MITRE Corp., 1976, EPA Contract No. 68-02-1859. (Report
subsequently issued as EPA 600/7-77-009, January, 1977.)
3. IERL/RTP Environmental Assessment Guideline Document, First Edition,
Draft Report, January, 1976.
4. Allen, J. M., Howes, J. E., and Miller, S. E., Planning Study on
Comprehensive Analysis of Emissions from Fluidized-Bed Combustion
Units, Battelle's Columbus Laboratories, Contract No. 68-02-2138,
August, 1976.
5. IERL/RTP Procedures Manual: Level I Environmental Assessment, TRW
Systems Group, Contract No. 68-02-1412, June, 1976.
6. Standards of Performance for New Stationary Sources, Federal Register,
41 (111), June 8, 1976.
7. Ibid, December 23, 1971.
8. Stambaugh, E. P., Giammar, R. D., Merryman, E. L., McNulty, J. S.,
Sekhar, K. C., Thomas, T. J., Grotta, H. M., Levy, A., and Oxley,
J. H., Study of the Battelle Hydrothermal Treatment of Coal
Process, EPA Contract No. 68-02-2119, November 11, 1976.
9. Creswick, F. A., Engdahl, R. B., Levy, A., Nack, H., and Weller, A. E.,
Technical Program Planning — NO Formation and Control in Fluidized-
Bed Combustion Systems, EPA Contract No. 68-02-2138, September 27,
1976.
10. Swift, W. M., Vogel, G. J., Panck, A. F., and Jonke, A. A., Trace
Element Mass Balances Around a Bench-Scale Combustor, Proceedings,
Fourth International Conference on Fluidized-Bed Combustion, McLean,
Virginia, December 9-11, 1975.
11. Kaakinen, J. W., Jorden, R. M., Lawasani, M. H., and West, R. E.,
Trace Element Behavior in Coal-Fired Power Plant, Env. Sci. and
Tech., 9. (9), 862-9, 1975.
49
-------
APPENDIX A
FLUIDIZED-BED COMBUSTION SAMPLING AND ANALYSIS
DATA REPORT - RUN NOS. 1, 2, AND 3
50
-------
APPENDIX A
FLUIDIZED-BED COMBUSTION SAMPLING AND ANALYSIS
DATA REPORT - RUN NOS. 1, 2, AND 3
This appendix presents the sampling and analysis data obtained from
Run Nos. 1 and 2, carried out in January and April, 1976, respectively.
Firing conditions and gaseous components measured in Run No. 3 are also
included. It is the sole intent here to present the sampling and analysis
data. It is not the intent of this report to analyze the results in terms
of the overall fluidized-bed combustion process.
The purpose of this program was to develop comprehensive procedures
for collecting and analyzing fluidized-bed combustion reactants and
emission products. Battelle's 6-inch fluidized-bed combustion unit was
used to carry out Run Nos. 1, 2, and 3 toward accomplishing the objective
of this program.
Samples were collected and analyzed in accordance with the original
(Run No. 1) and revised (Run No. 2) Fluidized-Bed Sampling Plan submitted
to EPA. Figure A-l shows the sampling locations used to obtain data for
Run Nos. 1, 2, and 3. The sample identification numbers associated with
each location are derived as follows. Each sample is given a three-digit
number, e.g., S-l-6-3, where the first digit represents the run number
(here #1), the second digit gives the sample location (position 6 in the
example), and the last digit is the sample number for that location
(sample 3 at location 6 in the example).
Table A-l in Run Nos. 1 and 2 gives a breakdown of the samples taken
from each location and the analyses performed and is the key for identifying
all samples taken in the runs. No significant amount of reduced sulfur
compounds, b.P.<450 C, were found in the Run No. 1 samples. Therefore,
no table on reduced sulfur compounds is given in Run No. 1 (see Table A-ll
Run No. 2). Tables A-14 through A-19, Run No. 2, list the appropriate
51
-------
Cyclone
n
Bed /,
&
/
Air
Y
Participate
>27tt. ©
Cool
Limestone C?)
Flue Gas Analyzers
I 1 I I I I
CO C02 ,02 S02 NOX ,THC
Gas Absofetion
Trains C9)
Scrubber
-TenoxPlug &
(Pom, He)
Mod. Method 5 Trains
t I {
Parficulate
Porticulate
<27jj >2.3p
Ash
FIGURE A-l. SCHEMATIC OUTLINE OF FLUIDIZED BED COMBUSTOR
AND SAMPLING LOCATIONS, RUN NO. 1
52
-------
detection limits for the analysis procedures used in this program. The
tables in general are self-explanatory. Noise measurement data are given
in the following section. An estimated cost breakdown for groups of
samples taken in Run No. 2 are given in Table A-18 at the end of data
presentation.
Noise Measurements
Some acoustic measurements were made in the laboratory in which the
Battelle's Multisolid Fluidized-Bed Combustor is located. This particular
unit is very quiet; the support equipment makes more noise than the com-
bustion bed itself.
The following table shows the sound pressure levels as a function of
center frequency octave bands. These measurements were taken with the micro-
phone inside the enclosure, one foot away from the bed.
Center Frequency
Octave Bands
sec -1 31.5 63 125 250 500 1000 2000 4000 8000 16000
Sound Pressure
Flat Level, dB 72.5 64.5 68 68 66.5 66 61.5 58.5 56 50
The present OSHA requirements is to keep the A-weighted level* at 90 dBA
or less for an 8-hour per day exposure. The measured levels were well below
this level.
The sound pressure level at a distance of 2 ft away from the rotary feed
pump was 78 dBA A-weighted, and 82 dB on the flat scale. The background noise
was 68 dBA on A-weighted, and 79 dB on the flat side.
The acoustic radiation from this small model is not a valid indication of
the noise generation potential of a full-scale model. However, it is quite
possible that even in full-scale combustion beds, the support equipment will
make more noise than the combustion bed itself.
* An adjusted scale taking into account the response of the human ear at var-
ious frequencies; the A-weighted level (dBA) is always less than the flat
level (dB) at a given frequency.
53
-------
TABLE A-l. SAMPLE IDENTIFICATION AND ANALYSES, RUN NO. 1
Material
Illinois tt coal
(-8 M)
Fluid-Bed Sample Number
_ Sample No. Designation
S-l-1-1
S-l-1-3
Proximate/ultimate
(1) Moisture (3) Volatile Batter
Sulfur
' Pa, Ca
Heating value
Trace metals
(2) Ash
Fixed carbon (by
difference)
(1) Total (3) Organic
(2) Pyrites (4) SO
(5) C, H, N, S, 0 (b
difference)
01
Grove limestone
Bed material
Ash (-325 M)
Sludge
S-l-2*
S-l-3
S-l-4
\ S-l-2-2
S-l-3-1
\ S-l-3- 2
S-l-4-1
-*^ S-l-4-2
\ S-l-4-3
Ca, Mg, CO
Trace metals
Trace metals
C, H, N, S, 0, SO., and SO,
4 3
Fusion temperature
Trace metals
C, H, N, S, 0, S04> and
Flue gas stream
Particulars >27u
S-l-6-1
S-l-6*
/ S-l-6-2
\\ S-l-6-3
S-l-6-4
Trace metals (approx 60 metals)
Cations-Fe, Al, Si, K, and C, H, N, S, 0
Anions Cl~, F~, N0~ S0°
3 4
Organic classes
"POM
Organic and reduced sulfur compounds
-------
TABLE A-l.
Material
Flue gas stream
Particulate <27ii
Fluid-Bed
Sample No
Sample Number
Designation
S-l-7-1
/S-l-7-2
"\\ S-l-7-3
\ S-l-7-4
Analyses
Trace metals
Cations - Fe, Al, Si, K plus C, H, H, S, 0
Anions Cl , F , N0~, SO,
Organic classes
''POM
Organic and reduced sulfur compound*
(AL-3) S-l-9-1
Flue gas stream
Gases
Ui
Cases, continuous
monitoring
S-l-9*
S-l-10*
-«L S-l-9-2
W\ S-l-9-3
^\ S-l-9- 4
V\ S-l-9-5
\ S-l-9- 6
S-l-10-1
/ S-l-10- 2
4f S-l-10-3
\\ S-l-10-4
\\ S-l-10-5
\ S-l-10-6
^-~.
(AL-2)
(AL-2)
(AL-2)
CAL-4)
(AL-3)
^Org
HC1
HCN
NH3
Tra
so3
°2
co2
CO
so2
NO
X
HC
Organic classes
,POM
Organ!cs - reduced sulfur
Trace elements (solution)
(Gokaoyr-Ross )
* Composite samples of duplicate runs.
-------
TABLE A-2. SUMMARY OF RUN NO. 1 CONDITIONS
Run Number:
BCL Number:
Coal feed rate, Ib/hr
Limestone feed rate, Ib/hr
Air feed rate, Ib/hr'
Bed height: expanded, inches
settled, inches
Bed temperature, F
Superficial gas velocity,
ft/sec
Ca/S ratio
Particulate loading grams/in
1A
AL-2
15.5
11.2
139.9
48
11.8
1560
8.9
5/1
NA
IB
AL-3
16.9
17.9
144.4
48
13.3
1530
9.0
7.3/1
1.43
1C
AL-4
15.2
17.0
150.6
48
11.8
1525
9.4
7.7/1
1.40
56
-------
TABLE A-3. SIEVE ANALYSIS, RUN NO. 1
Illinois No. 6
Sieve No.
Wt. Percent
Grove Limestone
Sieve No.
Wt. Percent
Overflow Bed Material
Sieve No.
Wt. Percent
-8 + 16
-16 + 20
-20 + 50
-50 + 100
-100 + 200
-200 + 325
-325
28.0
17.4
32.6
9.7
4.9
1.3
6.1
-8 + 10
-10 + 12
-12 + 16
-16 + 20
-20
4.2
17.0
37.7
41.1
0
16
-16 + 20
-20 + 30
-30 + 40
-40 + 50
-50 + 100
-100
44.6
33.8
17.1
4.3
0.07
0.02
0.04
57
-------
TABLE A-4. PROXIMATE/ULTIMATE ANALYSES OF FLUIDIZED-BED SAMPLES, RUN NO. 1
Material
Illinois id coal
Bed material
Participate >27p
(-325 mesh)
On
00 Sludge
Paniculate >27y
Particulat* <27p
Weight Percent
Sample Volatile Fixed Sulfur
Number HjO Ash Matter Carbon C HNS 0 Total Pyrltic Organic Sulfate SOj
S-l-1-1 3.67 11.2 38.4 46.7 62.9 4.6 1.1 4.47 12.0 4.47 2.41 1.99 0.07 —
and
• S-l-1-2
S- 1-3-2 3.65
S-l-4-1 1.0 72.0 25.0 0.4 0.4 3.29 . 2.84 H.D.
S-l-5-1 1.24 61.7 26.3 0.9 0.4 1.31 8.2 0.45 H.D.
S-l-6-2 <0.05 77.3 20.5 0.3 0.3 4.47 1.20
and
S-l-6-3
S-l-7-2 0.28 88.9 8.2 0.2 0.1 5.93 3.24
and
S-l-7-3
-------
TABLE A-5. METAL AND ANION ANALYSES ON FLUIDIZED-BED SAMPLES, RUN NO. 1
Sample Weight Percent
Material
Illinois 86 coal
Grove limestone
Particulate >27p
Number Na Ca Mg CO Fe
S-l-1-1 0.23 0,41 0.06 0.68
S-l-2-1 37.1 0.51 57.7
S-l-6-2 5.42
and
S-l-6-3
Al Si K Cl F NO
________ _ __
2.53 6.00 0.47 0.12 0.005 0.008
Particulate <27p S-l-7-2 6.68 7.36 11.0 2.10 0.48 0.019 0.003
and
S-l-7-3
-------
TABLE A-6. ANALYSES OF ACIDIC AND BASIC GASES FROM FLUE
GAS SAMPLES WITHDRAWN FROM STACK AT 568 F, RUN NO. 1
Component
HC1
HF
HCN
NH3
so3*
Sample
Number
S-l-9-2
S-l-9-2
S-l-9-3
S-l-9-4
S-l-9-6
Collection
Method
0.01N NaOH
0.01N NaOH
5% KOH
IN H2S04
Goksjfyr-Ross
, 3
mg/m
62.9
0.065
0.077
1.24
58.9
ppm
43.5
0.082
0.070
1.8
18.6
* See page 47, Recommendation (1).
60
-------
TABLE A-7- TRACE ELEMENTS BY OPTICAL EMISSION
SPECTROSCOPY SAMPLE NO. S-l-9-5
(Collected in KMnO,), RUN NO. 1
(a)
Component ppmw
Hg
Cd <50
As <50
Se
Te
Pb <5
Ba
(a)
All components were near or below minimum
detection limit.
61
-------
TABLE A-8. POM ANALYSIS, RUN NO. 1
NAS(l)
Component Notation
Anthracene/Phenanthrene
Methyl Anthracenes ?
Fluoranthene -
Pyrene
Methyl Pyrene/Fluoranthene ?
Benzo(c)phenanthrene ***
Chrysene/Benz (a) anthracene *
Methyl Chrysenes ?
Benzo Fluoranthenes **
Benz(a)pyrene ***
Benz(e)pyrene
Perylene -
3-Methylcholanthrene ****
Indeno(l,2,3,-cd)pyrene *
Benzo (ghi)perylene -
Dibenzo (a, h) anthracene ***
Diebenzo(c,g)carbazole ***
Dibenz(ai and ah)pyrenes ***
Coronene
Total
3
Sample Weights, yg/m
Sample Number
, S-l-9-1
57.3
9.4
3.2
0.95
0.73
0.17
0.37
0.17
72.2
(1) Carcinogenicity rating as listed by National Academy of Sciences in
"Particulate Polycyclic Organic Matter", 1972.
Not carcinogenic.
* Carcinogenic.
**9 *A*S **** Strongly carcinogenic.
? Carcinogenicity not indicated by NAS.
62
-------
TABLE A-9. ORGANIC CLASS ANALYSES
RUN NO. 1
Organic
1
2
3
4
5
6
7
8
Sample Weight ug/gram
Sample No.
S-l-6-4 S
309
10.9
21.8
29.1
4.7
9.6
6.4
13.1
sample
-1-7-4
750
35
55
85
50
75
35
20
63
-------
TABLE A-10. LEVEL 1 ANALYSES OF ORGANIC CLASSES, RUN NO. 1
(Refer to Table A-9 for total mass of each fraction)
Fraction
S-l-6-4
S-l-7-4
2
3
4
5
6
Vinyl unsaturated hydrocarbons
Aliphatic esters
Aliphatic esters, ketone
Phthalate ester
Aliphatic hydrocarbons, vinyl
unsaturated hydrocarbons
Aliphatic esters
Conjugated ketone or quinone
Phthalate ester
64
-------
TABLE A-ll. ANALYSES OF GASEOUS COMPONENT IN
FLUIDIZED-BED SAMPLES, RUN NO. 1
Sample
Number
S-l-10-1
S-l-10-2
S-l-10-3
S-l-10-4
S-l-10-5
S-l-10-6
Component
0?, percent
C0?, percent
CO , ppm
S02, ppm
N0x, ppm
HC, ppmC
Average
Value
6.5
14.3
790
700
415
85
65
-------
TABLE A-12. TRACE ELEMENT ANALYSIS OF FLUIDIZED-BED SAMPLES,
RUN NO. 1, ppmw(a) (Except where designated percent)
Element
Li
Be
B
F
Na
Mg
Al
Si
P
S
Cl
K
Ca
Sc
Ti
V
Cr
Mn
Fe
Co
Ni
Cu
Zn
Ga
Ge
As
Se
Br
Rb
Sr
Y
Zr
Nb
Mo
Ag
Cd
Sn
Sb
Te
I
(111.
#6 Coal)
S-l-1-3
0.13
0.17
93
110
890
400
>1%
>1%
3.2
>0.5%
= 1100
=1200
=4000
1.6
240
20
13
80
>1%
4.0
33
9.0
37
1.7
2.0
2.7
0.29
4.7
21
80
9.4
10
3.1
2.9
—
0.10
0.28
0.56
<0.37
0.10
(Grove
Lime-
stone)
S-l-2-2
2.9
<0.18
6.2
140
140
>0.5%
= 3000
>1%
140
240
18
>0.5%
>1%
<0.11
87
7
0.87
15
=1200
0.14
0.69
2.2
4.0
0.23
—
<0.61
—
1.2
7.3
470
1.7
7.0
0.32
0.67
—
—
0.12
—
—
2.9
(Bed
Material)
S-l-3-1
5.7
<0.18
33
170
320
>0.5%
=3000
>0.5%
310
>0.5%
120
=1600
>1%
0.16
130
7
1.3
26
= 4600
0.14
23
78
17
1.1
1.4
0.67
—
4.5
7.3
470
3.6
28
0.69
1.2
—
0.14
0.55
—
0.29
(Partic-
ulate
>27 p
-325)
S-l-4-3
220
18
710
290
=3200
>0.5%
>1%
>1%
310
310
240
>0.5%
>1%
0.73
870
38
32
210
>1%
14
46
78
80
2.3
3.0
2.0
4.4
12
37
140
8.3
35
3.2
3.3
0.30
0.67
0.92
0.67
1.2
(Partic-
ulate
>27 y)
S-l-6-1
NR
2.6
500
290
=1400
>0.5%
>1%
>1%
310
310
240
>0.5%
>1%
1.6
870
25
15
60
>1%
2.1
23
66
67
2.3
3.0
4.1
0.94
6.0
20
470
3.6
35
3.2
3.3
0.67
1.2
0.67
0.58
(Partic-
ulate
<27 u)
S-l-7-1
57
6.0
= 2000
450
>0.5%
>0.5%
>1%
>1%
310
310
300
>1%
>1%
7.3
= 2200
140
87
60
>1%
14
99
120
140
11
14
6.1
9.4
60
37
250
29
120
16
14
0.30
1.0
12
1.4
29
66
-------
TABLE A-12.
Element
Cs
Ba
La
Ce
Pr
Nd
Sm
Eu
Gd
Tb
Dy
Ho
Er
Tm
Yb
Lu
Hf
Ta
W
Hg
Tl
Pb
Bi
Th
U
(111.
#6 Coal)
S-l-1-3
0.65
35
5.0
9.8
2.1
6.6
0.71
—
0.40
—
0.70
—
0.34
—
0.23
—
0.18
0.27
0.31
0.16
0.59
0.47
—
1.7
1.0
(Grove
Lime-
stone)
S-l-2-2
0.19
90
2.2
2.9
0.86
1.7
0.31
—
—
—
0.20
—
—
—
<0.21
<0.15
1.8
0.33
<0.22
<0.01
—
0.86
<0.24
0.34
0.34
(Bed
Material)
S-l-3-1
0.43
180
6.4
15
2.0
3.8
0.83
0.25
—
—
0.49
—
0.11
—
<0.49
<0.15
1.1
0.87
<0.32
<0.01
—
1.9
<0.24
1.1
0.80
(Par tic -
ulate
>27 y
-325)
S-l-4-3
4.4
180
6.4
11
2.0
7.5
1.7
0.50
0.32
0.25
2.0
0.39
0.28
—
1.6
0.38
0.79
0.43
0.36
0.01
1.0
7.2
<0.24
5.7
4.0
(Partic-
ulate
>27 y)
S-l-6-1
1.9
180
6.4
15
2.0
10
1.7
0.50
0.32
0.19
1.4
0.16
0.28
—
1.6
0.50
1.8
0.87
<0.32
<0.01
0.48
4.3
<0.24
2.4
1.7
(Partic-
ulate
<27 y)
S-l-7-1
4.3
320
30
44
15
21
3.6
0.94
0.69
0.48
2.0
0.61
0.56
0.14
2.5
0.50
1.8
0.87
1.1
14
10
43
0.80
11
8.0
(a) Rhenium and Indium used as internal standards. Gold, Platinum, Iridium,
Osmium, Lutecium, Thullium, Holmium, Palladium, Rhodium, Ruthenium pre-
sent at less than 0.1 ppm/wt.
67
-------
TABLE A-13. SAMPLE IDENTIFICATION AND ANALYSES, RUN NO. 2
cr>
CO
Fluid-Bed Sample Number
Material Sample No. Designation
Illinois 16 coal
(-8 H) S-2-1
Crove limestone
S-2-2
Bed material S-2O
Aah (-325 M) S-2-4
Sludge S-2-5
Flue gas scream ,
Parttculate >27p
S-2-1-1 /
-/ S-2-1-2
\ S-2-1- 3
\;-2-l-4
S-2-2-1
\ S-2-2-2
\S-2-2-3
S-2-3-1
^X 3,2-3-0
\S-2-3-2
^e-2-3-3
S-2-4-1
-*T S-2-4-2
\ S-2-4-3
\<5-'-4-4
• S-2-5-1
\ S-2-5-2
\£-?-5--5
S-2-6-1
/ S-2-6-2
~\V S-2-6-3
\ S-2-6-4
Proximate/ultimate j |;
/ (e
^ Sulfur ||J
^ Na, Ca
Heating value
Trace metals
Particle Size
_ Ca, Mg, CO* NO" NO"
trace metals
Particle Size
Trace metala
Organic Classes
_ SO", so^, S=, NO", NO"
Particle Size
Nor, NO", S*. SO", and
Fusion temperature
Trace metals
Organic Classes
C, H, N, S, 0, SO^, and
Trace Metals
Organic Classes
Trace metals (approx 60
C, H, N, S, 0
Aniona, NO", NO" SO
- 2" 3
Organic classes
A
) Moistu
) Aah
; oW
) Total
j Pyrite
so;
necals)
• «
4' S03
Moisture (k) Volatile matter
(5; Fixed carbon (by
difference)
»rence;
(3) Organic
(4) SO,
Partioai Size
Organic and reduced sulfur compounds
-------
TABLE A-13.
Material
Flue gas stream
Participate <27u
Fluid-Bed
Sample No.
Sample Number
Designation
S-2-7-1
S-2-7-4
Particle
Size
Analyses
Trace metals
C, H, N, S, 0
Anions, CO",
, So , SO
Organic classes
POM
Organic and reduced sulfur compounds
(AL-3) S-2;9-l
Organic classes
Flue gas stream
Cases
Gaaea, continuous
monitoring
s-e -9 j-
c j _-in _
/ S-2-9-2 (AL-2)
1\ S-2-9-3 (AL-2)
V\ S-2-9-4 (AL-2)
A S-2-9-5 CAL-4)
Vs-2-9-6 (AL-3)
S-2-10-1
/ S-2- 10- 2
/;/ s-2-10-3
\\ S-2-10-4
\\ s-2- 10- 5
Vs-2-10-6
^Organics - reduced sulfur
HC1, HF
HCN
Trace elements (solution)
SO. (Gokaoyr-Rosa)
°2
co2
CO
so2
NO
X
HC
-------
TABLE A-14. SUMMARY OF RUN NO. 2 CONDITIONS
Coal feed rate, Ib/hr 8.8
Limestone feed rate, Ib/hr 4.3
Air feed rate, Ib/hr 87.3
Bed height: expanded, inches 48
settled, inches 21.6
Bed temperature, F 1655
Superficial gas velocity ft/sec 6.0
Ca/S ratio 2.9
Particulate loading g/m 1.64
70
-------
TABLE A-15. SIEVE ANALYSIS, RUN NO. 2
Illinois Coal # 6
Sieve No.
+ 8
-8 + 12
-12 + 16
-16 + 20
-20 + 30
-30 + 50
-50 + 100
-100 + 200
-200 + 325
-325
Wt. %
0.14
13.93
18.61
14.32
12.91
17.92
9.28
5.14
5.42
2.32
Limestone Overflow Bed Material
Sieve No. wt. % Sieve No.
-8 + 12 32.53 20
-12 + 16 34.26 -20 + 30
-16 + 20 24.69 -30 + 40
-20 8.51 -40 + 50
-50 + 100
-100 + 200
-200
Wt. %
71.10
18.47
7.47
2.16
0.66
0.05
0.12
71
-------
TABLE A-16. PROXIMATE/ULTIMATE ANALYSES OF FLUIDIZED-BED SAMPLES, RUN NO. 2
Material
Illinois #6 coal
Sludge
Particulate >27V
Particulate >27y
Sample u n
No. 2
S-2-1-1 8.8
S-2-5-1 1.5
S- 2-6-2 1.6
S-2-7-2 0.6
Volatile
Ash Matter
10.5 36.4
83.9
62.7
88.8
Weight Percent ^
Fixed Sulfur(b)
Carbon C H . N 0 Total SO," Pyrites
44.3 62.5 4.5 2.3 7.3 4.07 — 1.65
— 13.0 0.4 0.3 0.6 0.25 none
detected
32.4 0.4 1.1 0.1 1.66
— 8.7 0.2 0.1 — 1.9
Organic Sulfates
2.10 0.32
0.09
See Table A-17
ditto
(a) Dashes in Table indicate no analysis made. Lower limit of detection for each component listed in Table is about 0.10 percent.
(b) All values reported as sulfur.
-------
TABLE A-17. METAL AND ANION ANALYSES OF FLUIDIZED-BED SAMPLES, RUN NO. 2
Weight Percent
Material
No.
NO
_/ \
C0
Na
Ca
Mg
Co
Illinois #6 Coal
Limestone
Overflow bed
material
Bed material ash,
-325 mesh
Particulate >27 y
Particulate <27 y
S-2-1-1
S-2-2-1 <0.0003 <0.0003
S-2-2-2 0.0012 0.0007
S-2-4-1 0.012
S-2-6-3 0.006
S-2-7-3 0.002
0.0043
0.04
0.16
0.0008 0.31
0.0006 <0.1
<.01
20.2
0.14 16.4
3.37
5.34
0.11 0.16
57.6 — 37.6 0.64
6.45
5.97
(a) Divide sulfate values by 3 for use in Table A-16.
-------
TABLE A-18. ANALYSES OF ACIDIC AND BASIC GASES FROM
FLUE GAS SAMPLES WITHDRAWN FROM STACK
AT 260 F, RUN NO. 2
Sample
No.
S-2-9-2
S-2-9-2
S-2-9-3
S-2-9-4
S-2-9-6
Component
HC1
HF
HCN
NH3
so3*
mg/m
55.3
0.57
0.20
4.87
0.46
ppm
37.3
0.70
0.18
7.05
0.14
* See page 47, Recommendation (1).
74
-------
TABLE A-19. TRACE ELEMENT ANALYSES OF FLUIDIZED-BED
SAMPLES BY ATOMIC ABSORPTION AND OPTICAL
SPECTROSCOPY, RUN NO. 2
Sample No. S-2-9-5
Component
Hg
Cd
As
Se
Te
Be
Pb(b)
Ba(b)
ppmw
0.007
<0.04
0.03
<0.03
<0.2
<0.02
<1.0
<1.0
ng/m
9 (a)
<52
39
<39
<260
<26
<1300
<1300
(a) Near lower detection limit. May be
some contribution from solvent.
(b) Determined by optical emission
spectroscopy; all others by atomic
absorption.
75
-------
TABLE A-20. POM ANALYSES, RUN NO. 2
NAS (1)
Component Notation S-2-3-4
Anthracene/phenanthrene - <"
Methyl anthracenes '
Fluoranthene "
Pyrene
Methyl pyrene/fluoranthene ?
Benzo (c)phenanthrene ***
Chrysene/benz(a)anthracene *
Methyl chrysenes ?
7 , 12-Dimethylbenz (a) anthracene ****
Benzo f luoranthenes **
Benz(a)pyrene **x
Benz(e)pyrene -
Perylene ~
Methylbenzopyrenes ?
3-Methylcholanthrene ****
Indeno(l,2,3,-cd)pyrene *
Benzo (gh i) peryiene
Dibenzo(a,h)anthracene ***
Diebenzo(c,g)carbazole ***
Dibenz(ai and ah)pyrenes ***
Coronene -
(b)
Minimum Detection Limit $
3(a)
ng/m
S-2-4-4 S-2-5-3 S-2-6-4
<0.04 470 <4
36.8
511
91 .9
20.4
16.3
81.7
20.4
<0.1
51.0
25.5
1.0
10.2
<0.1
2.0
2.0
<0.1
<0.1
<0.1
<0.1
0.04 o.l 4
(a) Use the following conversion factors to convert from ng/m to ng/gram sample
only, i.e., S-2-9-1 not included)
Sample No: S-2-3-4 S-2-4-4
S-2-5-3 S-2-6-4 S-2-7-4
S-2-7-4 S-
810
73.6
331
36.8
14.7
18.4
73.6
11.0
<0.7
58.9
36.8
<0.7
7.4
<0.7
7.4
7.4
<0.7
<0.7
<0.7
<0.7
0.7
(solid samples
2-9-1
2667
561
1404
211
28.1
14.0
35.1
3.5
<0.4
8.8
3.5
<0.4
1.8
<0.4
3.5
3.5
<0.4
<0.4
<0.4
<0.4
0.4
Multiply values by: 0.023 11.7 1.42 0.06
0.53
(b) Detection limits vary with sampling conditions encountered, i.E., in accordance with amount
of sample collected and gas volume involved. For solid samp}es the average detection limit
in units of ng/gram sample is about 0.27.
(1) Carcinogenicity rating as listed by National Academy of Sciences In "Particulate Polycyclic
Organic Matter , 1972. ' '
Not carcinogenic.
* Carcinogenic.
**, ***_ **** Strongly carcinogenic.
? CarcinogCTilcity not indicated by NAS.
76
-------
TABLE A-21. ORGANIC CLASS ANALYSES, RUN NO. 2
Organic
Class
1
2
3
4
5
6
7
8
3 (a)
yg/m
S-2-3-4
1288
41
275
170
135
381
381
52.7
S-2-4-4
33.5
2.5
6.3
5.6
3.0
2.3
7.2
3.3
S-2-5-3
247
231
214
184
91.9
150
124
28.7
S-2-6-4
372
422
144
112
73.8
66.8
162
38.7
S-2-7-4
144
83.2
150
118
70.7
141
536
8.1
S-2-9-1
4158
605
2364
1873
610
919
1033
10.9
(a) Multiplication factors given in Table A-20 can be used here to
convert from yg/m^ to yg/gram sample (solid samples only).
77
-------
TABLE A-22. LEVEL 1 ANALYSES OF ORGANIC CLASSES, RUN NO. 2
(Refer to Table A-21 for total mass of each
fraction)
Sample 2-3-4
Cut 1 - Aliphatic hydrocarbon containing a significant amount of vinyl
unsaturation.
Cut 2 through 6 - Contain only traces of hydrocarbon structure.
Quantity of material is very low.
Cut 7 - Material concentration extremely low. A trace of aliphatic and
carbonyl structure is present.
Sample 2-4-4
Cut 1 - Aliphatic hydrocarbon containing a small amount of unsaturation
including vinyl.
Cut 2 - Alphatic hydrocarbon containing a small amount of carbonyl.
Cut 3 - Aliphatic hydrocarbon containing some aliphatic ester.
Cut 4 - Carboxylic acid ester plus aliphatic ether groups, possibly
a vinyl ether.
Cut 5 - A small amount of phthalate ester.
Cut 6 - Nil
Cut 7 - Primarily aliphatic - 2 different carbonyls are present, one of
which is probably a ketone.
Sample 2-5-3
Cut 1 - Aliphatic and fused ring aromatic hydrocarbons; pyrene and
benzpyrene types are possible.
Cut 2 - Similar to 1 but concentration of fused ring aromatics is higher.
Cut 3 - Some of the fused ring aromatics of cuts //I and #2 but primarily
an aromatic ketone. Nitrile* groups are present. A small amount
of hydroxyl structure is present.
Cut 4 - Aromatic ketone and quinone structures. Small amounts of
nitrile* and hydroxyl.
Cut 5 - Aromatic ketone and quinone structures. A trace of nitrile. A
small amount of anhydride or other strained ring carbonyl is
probable.
Cut 6 - A complex mixture of many types of carbonyl, aliphatic, aromatic
structures and with a trace of nitrile*.
Cut 7 - Same as Cut 6.
Cut 8 - Only a trace of material; complex carbonyl structures.
78
-------
TABLE A-22. (Continued)
Sample 2-6-4
Cut 1 - Aliphatic hydrocarbons.
Cut 2 - Ester (very small amount of material).
Cut 3 - Trace of ester plus other carbonyl.
Cut 4 - Trace of material containing several carbonyls.
Cut 5 - Trace of material containing carbonyl.
Cut 6 - Trace of material containing carbonyl.
Cut 7 - Trace of material containing carbonyl.
Cut 8 - Nil
Sample 2-7-4
Cut 1 - Aliphatic and fused ring aromatics plus silicone.
Cut 2 - Aliphatic and fused ring aromatics plus a small amount of ester.
Cut 3 - A mixture of aliphatic and aromatic esters plus a trace of
nitrile*.
Cut 4 - Ester, ketone, and quinone are probable; both aromatic and
aliphatic structure are present.
Cut 5 - Aliphatic ester, probably unsaturated.
Cut 6 - Aromatic strained ring or halogenated carbonyl.
79
-------
TABLE A-23. REDUCED ORGANIC SULFUR ANALYSES, RUN NO. 2
oo
o
Component
Overflow bed material
-325 mesh overflow
bed material
_
Particulate >27u
Particulate <27u
Tenax
Sludge
Reduced Organic Sulfur
Sample With Retention Time Wit
No. Wt. Gram of Benzothiophene of
S-2-3-4 15.0 None
S-2-4-4 3.9
S-2-6-4 9.5 20
S-2-7-4 5.1 None
S-2-9-1 — 1500
S- 2-5-3 13.8 40
Compounds, ug
h Retention Time
Dibenzothiophene Miscellaneous
None None
1
5
5 2 cpds X. 5 ug
each
800 8 cpd's at 20
to 100 ug each
200 1 cpd — 100 ug
Remarks
No reduced organic sulfur
compounds found in S-2-3-4 or
S-2-4-4 above 5 ug. About one
dozen nonsulf ur compounds .
Major portion in C^& to €22
range; 25 to 500 ug quantities.
50 ug napthalene
10 to 40 u g nonsulf ur cpd's
in C10 t0 C16 range
2 non-sulfur compounds at leve!
of about 5000 V g and 15 at
200 to 1000 ug
About 20 nonsulfur compounds
at levels of 50 to 500 u g
(a) ug of material found in total sample weight given in Column 3. Minimum detection limit is about 0.5 yg for the above samples.
-------
TABLE A-24. ANALYSES OF GASEOUS COMPONENTS
IN FLUIDIZED-BED SAMPLES,
RUN NO. 2
Average
Component Value
0 , percent 3.6
CO , percent 17.3
CO, ppm 2090
SO , ppm 730
NO , ppm 350
X
HC, ppm C 360
81
-------
TABLE A-25. TRACE ELEMENT ANALYSIS OF FLUIDIZED-BED
COMBUSTION SAMPLES, RUN NO. 2, ppmw
(except where designated percent)
Grove -325M
111. Lime- Bed Bed
#6 Coal stone Material Material Sludge
Element S-2-1-3 S-2-2-2 S-2-3-1 S-2-4-3 S-2-5-2
Li
Be
B
F
Na
Mg
Al
Si
P
S
Cl
K
Ca
Sc
Ti
V
Cr
Mn
Fe
Co
Ni
Cu
Zn
Ga(a)
Ge
As
Se (a)
Br
Rb
Sr
Y
Zr
Nb
Mo
Ru
Rh
Pd
Ag (a)
Cd
In (a)
20
2
<0.03
<3
^2%
^2%
^5%
•^20%
100
5000
M3.5!
M).6:
5000
3
5000
500
100
100
^5%
10
<10
30
<10
<5
100
<3
<5
10
20
500
30
300
3
30
<1
<0.5
<3
<3
<30
<10
3
<0.005
1
27y <27y
S-2-6-1 S-2-7-1
10
0.2
50
<5
3000
5000
^5%
^20%
50
1000
800
. -VIZ
•v7%
30
1000
500
100
200
•^2%
10
200
50
<3
<5
3
1
<5
0.5
5
200
30
200
3
20
<0.5
<0.2
<2
<1
<3
<1
20
2
300
<20
5000
^1%
•v-5%
^20%
200
3000
^1%
-v2%
^7%
30
VL%
1000
100
500
•^2%
10
100
50
<3
<10
10
3
<5
1
10
300
200
200
5
10
<0.5
<0.5
<2
<1
<3
<1
82
-------
TABLE A-25.
Grove -325M
111. Lime- Bed Bed
#6 Coal stone Material Material
Element S-2-1-2 S-2-2-2 S- 2-3-1 S-2-4-3
Sn
Sb
Te
I
Cs
Ba
La
Ce
Pr
Nd
Sm
Eu
Gd
Tb
Dy
Ho
Er
Tm
Yb
Lu
Hf
Ta
W
Re
Os
Ir
Pt
Au
Hg
Tl
Pb
Bi
Th
U
<1
<0.5
<1
3
3
200
20
50
5
^
<1
<0.5
<1
<0.2
<1
<0.3
<0.5
<0.3
<0.5
<0.3
<2
<0.5
<1
<1
<\
<1
<1
<0.5
<2
<1
<1
<0.5
<1
<0.5
<1
<0.1
<0.3
1
<0.3
100
2
2
0.3
0.5
<0.3
<0.3
<0.3
<0.3
<0.2
<0.05
<0.2
<0.05
<0.2
<0.05
<0.5
<0.3
<0.3
<0.2
<0.2
<0.2
<0.3
<0.1
<0.3
<0.1
3
<0.1
<0.2
<0.5
<1
<0.2
<0.3
0.5
0.3
200
2
5
0.3
0.5
<0.5
<0.3
<0.3
<0.1
<0.2
<0.05
<0.2
<0.1
<0.2
<0.05
<0.5
<0.3
<0.3
<0.2
<0.2
<0.2
<0.3
<0.1
<0.3
<0.2
1 '
<0.1
0.2
0.5
200
2
<0.3
3
5
300
20
50
5
5
5
2
5
0.5
2
0.5
2
0.2
0.2
<0.1
1
<0.3
1
<0.2
<0.2
<0.2
<0.3
<0.1
<0.3
5
300
1
10
5
Sludge
S-2-5-2
100
1
<0.3
1
3
5000
200
500
20
50
30
10
20
2
10
3
5
1
5
0.5
10
<1
2
<0.2
<0.2
<0.2
<0.3
<0.1
<0.3
10
200
10
50
20
Partic- Partic-
ulate ulate
27 27
S-2-6-1 S-2-7-1
<2
0.2
<0.3
0.5
0.5
500
50
50
5
20
5
1
1
0.5
3
0.5
0.5
0.3
1
<0.1
2
<0.2
0.3 .
<0.2
<0.2
<0.2
<0.3
<0.1
<0.3
5
20
<0.1
5
2
10
1
<0.5
5
2
300
30
20
5
20
3
1
3
0.3
2
0.5
1
0.1
1
<0.1
<0.5
<0.2
<0.3
<0.2
<0.2
<0.2
<0.3
<0.1
<0.3
10
100
1
5
2
(a) Memory from previous sample.
83
-------
TABLE A-26. APPROXIMATE DETECTION LIMITS FOR SAMPLES RUN ON
SPARK SOURCE MASS SPECTROMETER, RUN NO. 2
(refer to Table A-25)
Element
Li
Be
B
F
Na
Mg
Al
Si
P
S
Cl
K
Ca
So
Ti
V
Cr
Mn
Fe
Co
Ni
Cu
Zn
Ga
Ge
As
Se
Br
Rb
Si
MDL*
(PPMW)
0.005
0.005
0.03
0.5
0.1
3.0
10.0
2.0
0.3
0.5
0.5
0.5
0.5
0.3
0.3
0.3
0.3
0.3
0.3
0.3
2.0
0.3
1.0
0.3
0.2
0.3
1.0
0.3
0.3
0.3
Element
Y
Zn
Nb
Mo
Ru
Rh
Pd
Ag
Cd
In
Sn
Sb
. Te
I
Cs
Ba
La
Ce
Pr
Nd
Sm
Eu
Gd
Tb
Dy
Ho
Er
Tm
"Yb
Lu
MDL
(PPMW)
0.3
0.3
0.3
0.3
0.3
0.3
0.3
1.0
3.0
1.0
1.0
0.1
0.3
0.3
0.3
0.5
0.3
0.5
0.2
0.5
0.3
0.3
0.3
0.1
0.2
0.05
0.2
0.05
0.2
0.05
Element
Hf
Ta
W
Re
Os
Ir
Pt
Au
Hg
Tl
Pb
Bi
Th
U
MDL
(PPMW)
0.3
0.2
0.3
0.2
0.2
0.2
0.3
0.1
0.3
0.1
0.2
0.1
0.1
0.1
Minimum detection limit for samples of approximate composition studied in program.
84
-------
TABLE A-27. DETECTION LIMITS FOR METALS ANALYSIS
BY ATOMIC ABSORPTION, RUN NO. 2
(refer to Table A-19)
Element
Hg
Cd
As
Se
Te
Be
Pb(b)
Ba(b)
ppm
0.005
0.04
0.01
0.03
0.2
0.02
1.0
1.0
MDL(a)
(ng/m )
6.5
52
13
39
260
26
1300
1300
(a) Minimum Detection Limit for
samples of approximate com-
position analyzed in this
study
(b) Determined by Optical
Emission Spectroscopy.
85
-------
TABLE A-28. DETECTION LIMITS FOR ACID AND BASIC GASES,
RUN NO. 2 (refer to Table A-18)
Compound
HC1
HF
HCN
NH,
Solution Analysis
Method of
Analysis
Ion Chromato graph
Ion Chromatograph
Ion Selective Electrode
Ion Selective Electrode
MDL , ppm
0.06
0.06
0.03
0.03
(a)
Ion Chromatograph 0.01
(a) Minimum Detection Limits for samples
collected in this program.
86
-------
TABLE A-29. DETECTION LIMITS FOR ANIONS AND METALS,
RUN NO. 2 (Solid Samples, refer to Table A-17)
Compound
co3=
NO
NO
so4=
so "
Analysis Method
Titration
Colorimetric
Colorimetric
Gravimetric
Titration
MDL
(Wt. Percent)
0.05
0.0003
0.0003
0.05
0.05
s
Ca
Mg
Na
0.01
Atomic absorption .01
.01
.01
87
-------
TABLE A-30.SAMPLES FROM RUN NO. 2 AND ESTIMATED COST FOR ANALYSES
Analysis
Run No. 2
Sample Number
Estimated
Cost, 1976
Particle size
Trace element
Minor elements
Organic classes/POM/
organic sulfur
CHNSO
Anions
(a) SO^, S0=3, S=, N0~,
and N0~
(b) SO,, SO,, S N0
t J ,
and NO,,
S-2-1-4
S-2-2-3
S-2-3-3
S-2-6-5
S-2-7-5
S-2-8-5
S-2-1-3
S-2-2-2
S-2-3-1
S-2-4-3
S-2-5-2
S-2-6-1
S-2-7-1
S-2-8-1
S-2-3-4
S-2-4-4
S-2-5-3
S-2-6-4
S-2-7-4
S-2-8-1
S-2-1-1
S-2-5-1
S-2-6-2
S-2-7-2
S-2-8-2
S-2-3-2
S-2-5-1
S-2-6-3
S-2-7-3
S-2-8-3
$ 400
3600
1400
6240
750
440
660
88
-------
TABLE A-30.
Analysis
(c) SO^, Soij
(d) N0~, N0~, CO^
S03
Elemental Analysis
(AA)
(a) Hg, Cd, As, Pb,
Se, Te, Be
(b) Ca, Mg
(c) Na, Ca
Proximate
Run No. 2
Sample Number
S-2-5-1
S-2-2-1
S-2-9-6
S-2-9-5
S-2-2-1
S-2-1-1
S-2-1-1
Estimated
Cost, 1976
$ 120
150
60
420
120
120
90
Sulfur (total, pyritic,
organic, and sulfate) 180
S-2-1-1
89
-------
-C
en
QJ
0
^
E
o
99.8
99.5
99
98
90
80
70
60
50
40
30
10
0.5
0.2
0.6 08 I
t
-------
yy
98
95
90
80
? 70
o>
* 60
>.
^ 50
§ 40
£ 30
O)
•| 20
o
1 10
o
5
2
1
0.5
0.2
n i
^
-U
y>
/
3
/
X
/
j
CS-?-fi-T>
20 30 40 60 50 100 200 300 400 600 800 1000
Particle Diameter, microns (i)
FIGURE A-3. PARTICLE SIZE ANALYSIS, RUN NO. 2
(S-2-6-3 Particulate >27 y, Cyclone No. 1
(1) Aerodynamic size, i,e., equivalent spherical particles of unit density.
2000 3000 5000
-------
TABLE A-31. SUMMARY OF RUN NO. 3 CONDITIONS
Coal feed rate, Ib/hr 9.2
Limestone feed rate, Ib/hr 8.1
Air feed rate, Ib/hr 84.2
Bed height: expanded, inches 48.0
settled, inches 21.6
Bed temperature, F 1490.0
Superficial gas velocity, ft/sec 5.3
Ca/S ratio 7.1
92
-------
TABLE A-32. SIEVE ANALYSIS, RUN NO. 3
Overflow Bed Material
Illinois No.
Sieve No .
+8
-8 + 12
-12 + 16
-16 + 20
-20 + 30
-30 + 50
-50 + 100
-100 + 200
-200 + 325
-325
6 Coal
Wt. %
0.14
13.93
18.61
14.32
12.91
17.92
9.28
5.14
5.42
2.32
Limestone Run AL8
Sieve No. Wt. % Sieve No.
-8 + 12 32.53 20
-12 + 16 34.26 -20 + 30
-16 + 20 24.69 -30 + 40
-20 8.51 -40 + 50
-50 + 100
-100 + 200
-200
Wt. %
76.80
13.96
5.21
2.36
1.49
0.12
0.07
Run AL9
Sieve No.
20
-20 + 30
-30 + 40
-40 + 50
-50 + 100
-100 + 200
-200
Wt. %
77.68
14.90
4.85
1.79
0.72
0.03
0.03
93
-------
TABLE A-33. ANALYSES OF GASEOUS COMPONENTS IN
FLUIDIZED-BED SAMPLES, RUN NO. 3
Component
Average
Value
02, percent
CO,.,, percent
CO, ppm
S02, ppm
NO , ppm
HC, ppm C
3.3
18.3
4790.0
620.0
300.0
900.0
94
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/7-77-034
2.
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Method for Analyzing Emissions from Atmospheric
Fluidized-Bed Combustor
5. REPORT DATE
April 1977
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
E.L. Merry man, A. Levy, G.W.Felton, K.T.Liu,
J.M.Allen, and H. Nack
8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Battelle-Columbus Laboratories
505 King Avenue
Columbus, Ohio 43201
10. PROGRAM ELEMENT NO.
EHB536
11. CONTRACT/GRANT NO.
68-02-1409, Task 33
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
Task Final; 9/76-3/77
14. SPONSORING AGENCY CODE
EPA/600/13
^.SUPPLEMENTARY NOTES T£RL-RTP project officer for this report is Walter B. Steen,
Mail Drop 61, 919/549-8411 Ext 2825.
16. ABSTRACT Tne repOrj- describes an experimentally developed method to comprehen-
sively sample and analyze an atmospheric-pressure fluidized-bed combustion (FBC)
unit. The method is aimed at providing a cost and information effective environmen-
tal assessment of FBC units. The report includes a general discussion of the perti-
nent areas likely to be encountered in sampling and analyzing specimens from FBC
units; for example, streams encountered in FBC units, the selection of streams,
procedures for sampling gaseous, solid, and liquid streams, and the multilevel
analytical approach to emission characterization defined by EPA for combustion units.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
P Dilution
Fluidized-Bed Processors
Atmospheric Pressure
Sampling
Analyzing
Emission
Pollution Control
Stationary Sources
Fluidized-Bed Com-
bustion
Environmental Assess-
ment
13B
07A
04B
14B
13. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (This Report)
Unclassified
!1. NO. OF PAGES
97
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
95
------- |