AEPA
United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park Nr. 2771 1
EPA 600 7 79
July 1 ' ' •
Environmental
Assessment of Coal
Cleaning Processes:
Selection of Test Sites
for Source Test Program
Interagency
Energy/Environment
R&D Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wide range of energy-related environ-
mental issues.
EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect
the views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/7-79-073d
July 1979
Environmental Assessment of Coal
Cleaning Processes: Selection of Test
Sites for Source Test Program
by
D. A. Tolle, R. E. Thomas,
R. K. Markarian, and V. Q. Hale
Battelle-Columbus
505 King Avenue
Columbus, Ohio 4320I
Contract No. 68-02-2163
Task No.421
Program Element No. EHE623A
EPA Project Officer: James D. Kilgroe
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
FOREWORD
Many elements and chemical compounds are known to be toxic to man and
other biological species. However, our knowledge concerning the levels and
conditions under which these substances are toxic is extremely limited.
Little is known concerning the emission of these pollutants from industrial
processes and the mechanisms by which they are transported, transformed,
dispersed, or accumulated in our environment.
Portions of the Federal Clean Air Act, the Resource Conservation
Recovery Act, and the Federal Water Pollution Control Act require the U.S.
Environmental Protection Agency (EPA) to identify and regulate hazardous
or toxic substances which result from man's industrial activities. Industrial
pollutants are often identified only after harmful health or ecological
effects are noted. Remedial actions are costly, the damage to human and
other biological populations is often irreversible, and the persistence of
some environmental contaminants may endanger future populations.
EPA's Office of Research and Development (ORD) is responsible for health
and ecological research, studies concerning the transportation and fate of
pollutants, and the development of technologies for controlling industrial
pollutants. The Industrial Environmental Research Laboratory, an ORD
organization, is responsible for development of pollution control technology
and conducts a large environmental assessment program. The primary objectives
of this program are:
o The development of information on the quantities of
toxic pollutants emitted from various industrial
processes—information needed to prioritize health
and ecological research efforts.
o The identification of industrial pollutant emissions
which pose a clearly evident health or ecological
risk and which should be regulated.
o The evaluation and development of technologies for
controlling pollution from these toxic substances.
The coal cleaning environmental assessment program has as its specific
objectives the evaluation of pollution and pollution control problems which
are unique to coal preparation, storage, and transportation. The coal
preparation Industry is a mature yet changing industry and in recent years
significant achievements have been made in pollution abatement.
In focusing on the effectiveness and efficiency of coal cleaning processes
as methods of reducing the total environmental impact in the use of energy
derived from coal, this report deals with the selection of sites at which
experimental measurements will be performed. The information derived will
be used for the environmental source assessment of physical coal cleaning
facilities.
ii
-------
ABSTRACT
The objective of this subtask is to select coal cleaning facilities
on which field testing and sampling programs will be conducted in support
of the overall program directed at making an environmental assessment of the
pollution potential of various coal cleaning processes. The approach developed
has been to select a small number of coal cleaning plants from which data may
be obtained. The plants have been selected to represent extremes in variables
considered important in evaluating the pollution potential of these types of
facilities.
The site selection procedure developed here consists of two steps. The
first classifies coal cleaning plants into a number of site categories and
identifies a sampling sequence for these categories. The second consists of
gathering additional detailed information on specific plants and applying
secondary constraints as a means of selecting a single suitable site for
sampling within each of the site categories identified by the first step. In
the first step it has been assumed that all sites are equal within each category:
in the second step, however, the best or most representative site within a
category is to be selected.
The classification of coal cleaning facilities into various site categories
has been based on four criteria. These criteria are represented by four
variables: (1) acid neutralization potential of the soil surrounding the facility,
(2) pyritic sulfur content of run-of-mine (ROM) coal, (3) average annual
precipitation, and (4) coal cleaning process technology. Taking these variables
in combinations as extremes, high and low levels for each, there are 16 possible
combinations; however, six do not occur for actual sites (e.g., high"pyritic
sulfur content of ROM coal does not occur in combination with low average annual
precipitation).
iii
-------
Since the variable of coal cleaning process technology did not
fall naturally into a high and low range, it was necessary to select the
kinds of cleaning processes that would be studied. Therefore, coal prepara-
tion was classified into four levels according to the coal sizes being
washed. For the site selection process, it was decided that Level 1 and 2
facilities, which wash the coarse coal sizes only, would constitute low
(simple) technology, while Level 4 facilities, which wash the very finest
coal sizes, would constitute high (complex) technology.
Using statistical principles, a sampling sequence for the ten available
site categories was established so as to give first priority to investigation
of main effects of a single variable and second priority to investigation of
Interactive effects of variables. The rankings of priority were based on
answers to the following questions:
If only one site category were to be sampled, which
combination of variables should it represent?
If two site categories were to be selected, which
combination should be second?
This procedure was followed for all of the ten site categories.
The selection of sites within each site category will be accomplished by
imposing a series of constraints to assure that specific kinds of questions
are answered about pollution potential and pollution control for the coal
cleaning processes. These contraints are grouped in order of importance
as being primary, secondary, or tertiary in importance.
An initial sorting of the more than 400 known coal cleaning plants, using
information available in the literature, produced lists of facilities which
correspond to each of the ten site categories. For categories which included
a large number of cleaning plants three secondary constraints were imposed
that eliminated plants considered undesirable from a field sampling viewpoint.
These abbreviated lists include a total of 47 facilities. Site visits need
to be made to those listed facilities with cooperative management to obtain
information not available in the literature. This information is required
before a final selection of sampling sites can be made.
iv
-------
TABLE OF CONTENTS
Page
FOREWORD ii
ABSTRACT Hi
ACKNOWLEDGEMENTS viii
INTRODUCTION 1
EVOLUTION OF SITE SELECTION PROCEDURE 3
RECOMMENDED PROCEDURE 5
Selection and Definition of Variables 5
Low and High Levels for Each Classification Variable ... 9
Site Categories 11
Complete Factorial Listing 11
Sequential Sampling Procedures 12
A Statistical Rationale for Sequential Sampling 13
Recommended Procedure for Selection of Site Categories .... 13
Recommended Sampling Design 14
Assessments of the Recommended Sampling Design 14
Conditional Main Effects 16
Conditional Interaction Effects 16
Some Limitations 19
Crude Benefit/Cost Ratios 20
A Review of the Optional Sampling Plans 20
Ranking by Panel Members 23
Site Selection 23
Primary Constraints Classed as Essential 25
Secondary Constraints 25
-------
TABLE OF CONTENTS
(Continued)
Tertiary Constraints ................ 27
Potential Sampling Locations ............... 28
Information Needed for Final Site Selection ....... 31
REFERENCES ............................. 32
APPENDIX A
LISTS OF CANDIDATE SAMPLING SITES BY SITE CATEGORY ......
APPENDIX B
COAL PREPARATION PLANT INFORMATION FORM
vi
-------
LIST OF TABLES
Page
1. Generic Types of Coal Preparation Plants 8
2. Classification Variables and Associated Levels
Used to Define Site Categories 10
3. Factorial Listing of Site Categories 10
4. Recommended Sequential Sampling Design for
Coal Cleaning Plants 15
5. Conditional Main Effects and Conditional Interactions
Associated with the Recommended Sampling Design 18
6. Crude Benefit/Cost Ratios 21
7. Final Site Selection Procedures, Use of
Prioritized Constraints 24
8. Ranking of Pollution Control Technology for
Cleaning Plants and their Associated Refuse Disposal Areas • • • • 26
9. Summary of Potential Sampling Locations
by Site Category 29
A-l. Candidate Sites for Site Category 1 (1111)—Pennsylvania 35
A-2. Candidate Sites for Site Category 2 (1110)—Ohio 37
A-3. Candidate Sites for Site Category 3 (1011) 38
A-4. Candidate Sites for Site Category 4 (0111)—
Kentucky (Western) ..... 40
A-5. Candidate Sites for Site Category 5 (1010) 41
A-6. Candidate Sites for Site Category 6 (0110) 42
A-7. Candidate Sites for Site Category 7 (0011) 43
A-8. Candidate Sites for Site Category 8 (0010) 44
A-9. Candidate Sites for Site Category 9 (0001) 45
A-10. Candidate Sites for Site Category 10 (0000) 46
B-l. Coal Preparation Plant Information Form 49
vii
-------
ACKNOWLEDGEMENTS
This study was conducted as a Task in Battelle's Columbus Laboratories
ongoing program, "Environmental Assessment of Coal Cleaning Processes", which
is supported by EPA IERL/RTP. The contributions of the Program Manager,
Mr. G. Ray Smithson, Jr., and the Deputy Program Manager, Mr. Alexis W.
Lemmon, Jr., are gratefully acknowledged.
Significant contributions to the body of this report and critical
reviews were made by Drs. Peter Van Voris, current Task leader, and Beverly S.
Ausous. Steven E. Pomeroy assisted with Appendix A.
The advice and counsel of the EPA Project Officer, Mr. James D. Kilgroe,
and others at the IERL/RTP facility were invaluable in performance of this work.
viii
-------
INTRODUCTION
Battelle's Columbus Laboratories has contracted with the Industrial
Environmental Research Laboratory, Research Triangle Park, (IERL-RTP), North
Carolina, of the U.S. Environmental Protection Agency (EPA) to perform an
environmental assessment of coal cleaning processes. The broad objective of
Battelle's program with EPA is to perform a comprehensive assessment of the
environmental pollution potential resulting from coal cleaning (physical and
chemical), transportation, storage, and refuse disposal. In addition, combined
techniques to achieve environmental goals, such as coal cleaning with stack
gas scrubbing to reduce the pollution potential of coal-fired power plants, will
be evaluated. Increased use of both physical and chemical coal cleaning techniques
for the removal of sulfur and ash-forming constituents is of paramount importance
in an era of growing dependance on coal as a major domestic source of energy.
The advantage of coal cleaning processes is the removal of sulfur, ash, and
a variety of other unwanted materials from the raw coal prior to any oxidation
or reduction reactions.
In order to optimize the overall control of emissions from plants that
use coal, it is necessary to assess the pollution potential of coal cleaning
processes, allied operations, and — in certain cases— the end uses of coal.
This optimization process will also include an economic evaluation of the
benefits derived from coal cleaning, e.g., reduced cost of coal and ash
transportation, simplified boiler design, less boiler downtime, smaller dust
collectors and ash-disposal systems, and reduced costs for SO- control.
The broad goal of this program is to establish a strong base of
engineering, ecological, pollution control, and cost data which can be used to
determine which coal cleaning processes or pollution abatement devices are
most acceptable from both an environmental and economic viewpoint. This
information could also be used to identify needs for the development of new
or improved pollution control technology.
-------
In support of the broad goal stated above, a subtask has been
established to select several coal cleaning facilities, for evaluation of
their pollution potential. This evaluation will involve sampling of process
and effluent streams for laboratory analysis using the EPA Phased Approach.
A variety of selection schemes theoretically could be applied to
determine categories appropriate for study sites. Because the complexity of
the technology employed at a coal cleaning facility is dictated by both the
raw coal and the desired product, a number of possible schemes dealing with
a variety of variables were examined prior to selecting the scheme presented
in this report. The selection schemes examined ranged from simple random
selection from the approximately 480 coal cleaning facilities by a single
variable (i.e., geographic area) to more complex approaches dealing with
sophisticated multivariate statistical techniques, such as permutation design
or fractional factorial design. Basic treatment of the selection schemes
considered may be found in Hansen et al. , Cochran and Cox'*), and
(3)
Snedecor and Cochran . Examination of this full array of schemes allowed
selection of the approach that would limit the number of sites necessary for
the environmental assessment and yet assure examination of the major variables
thought to influence the pollution potential of a facility.
In the final development of an effective site-selection procedure,
several constraints were imposed to assure that the requirements of the
overall environmental assessment program would be satisfied and the program
would be cost effective. These constraints resulted, in part, from a decision
to select plants which represented extremes (high and low) in values of the
variables to be tested (for example, pyritic sulfur levels, environmental
settings, and various kinds of coal cleaning and pollution control technologies
employed) rather than average or typical conditions. It was felt, that
by examining the extremes in the variables, it would be possible to investigate
the contribution of each Individual variable in affecting the total environ-
mental burden imposed by the cleaning of coal.
This report delineates the selection procedure, presents the statistical
rationale in support of this procedure, and gives a preliminary listing of
candidate sites for the environmental assessment of coal cleaning.
-------
EVOLUTION OF SITE SELECTION PROCEDURE
There are over 400 coal cleaning plants listed in the 1977 Keystone Coal
(4)
Industry Manual , ranging in application of technology from simple crushing
and screening operations to sophisticated systems which clean and recover very
fine coal. Because the number of plants from which to choose test sites is very
large and because these potential test sites represent a wide variety of coal
types, environmental settings, and cleaning processes, early in the program it was
decided that the selection procedure should consist of two steps:
(1) classification of the plants into groups and subsequent reduction of
large groups to a manageable size by imposing constraints associated with
data available in the literature, and (2) final selection of sampling sites
from each group based on information obtained during site visits.
In the first scheme proposed (not ulitmately utilized) for plant classifica-
tion, the variables were based mainly on differences occurring as a consequence
of geographical location. These variables were applied in sequence. The
first variable was acid or alkaline mine drainage as defined in the effluent
guidelines report. ' (Here the term "acid mine drainage" is also used to describe
the water component of coal cleaning process streams.) This divided the U.S.
into two areas: an acid area which comprised northern Appalachia and an
alkaline area which comprised the rest of the country. This was in agreement
with the system described in the effluent guidelines report. The second
variable was average annual precipitation. This further divided the alkaline
area into southern Appalachia and Alabama (high rainfall), the Midwest
(intermediate rainfall), and the West (low rainfall). Each of these areas
was then divided according to the low and high pyritic sulfur content of
the coals produced. This division did not affect the southern Appalachian,
Alabama, or western areas since none of them report high-pyrite coals. At
this point there were six plant categories.
Each of these plant categories was then divided into three groups
according to complexity of the coal cleaning process, using the Stages I,
II, and III processes as defined in the effluent guidelines report.
-------
Using this system, the variables produced 18 categories into which coal cleaning
facilities might be classified. Unfortunately, the system requires the inclu-
sion of all categories in any evaluation in order to produce any statistically
rfgorous evaluation of coal cleaning. The problem associated with the above
selection scheme is that it cannot tolerate an incomplete design for site
sampling without significantly affecting the results. Preliminary evaluation
of sites for this scheme revealed that facilities did not exist in all categories;
thus this approach failed. Other forms of this scheme were evaluated during
the selection process, such as assigning numerical values to the variables,
or introducing other variables such as plant production capacity, but the
basic problem remained and thus this approach to site selection was abandoned.
A new scheme was then devised for the classification step of the site
selection procedure. This scheme employed the same variables as the previous
scheme and assigned two levels of each (high and low). These variables were
then considered in combination rather than applied sequentially. When these
four variables were considered each at two levels, 16 possible combinations
were produced of which only 10 exist in the U.S. Each of the coal cleaning
plants was then to be classified into one of these existing four-variable
combinations and at least one plant from each group was to be included in
the field program.
This scheme has had to be modified and refined through a number of stages
to tailor it to fit the specialized needs of the program. One of the most
important innovations was the creation of a panel consisting of nine Battelle
researchers who are active in coal cleaning technology and related environmental
research. The job of the panel was to rank the 16 possible combinations
according to what they perceived as the potential of each to produce environ-
mental impacts and then to assign numerical ratings to each rank. This
scheme was an attempt to make selections on a system which would confound
subjectivity as much as possible and be statistically defensible. The panel
members worked independently and the results were tabulated and compared.
Agreement among panel members was not very good, possibly because each
considered his own special interest, and this approach was abandoned.
However, the idea of having a panel make the evaluations was good, not
only because it brought in the experience from several disciplines, but also
because it tended to confound each member's personal biases. Consequently,
the panel was retained and their opinions were used as the site selection
procedure developed.
4
-------
RECOMMENDED PROCEDURE
The site selection procedure consists of two steps. The first step
consists of classifying coal cleaning plants into site categories and identify-
ing the sampling sequence; i.e., which site category is to be sampled first,
second, etc. This step includes gathering information from the literature
on all plants listed and applying constraints relevant to the available data
to eliminate plants considered undesirable for field sampling. This first
step has been successfully accomplished, and the results are described below.
The second step is to gather detailed information from plant management for
the plants on the abbreviated lists and apply additional constraints to
select the most suitable sampling site within each of the site categories.
The constraints used in both steps are defined below.
Selection and Definition of Variables
The purpose of the initial step is to classify coal cleaning plants
into groups that are as much alike as possible in selected characteristics,
but with the differences among the groups representing the ranges of each of
those characteristics. The environmental and process characteristics desired
to be tested led to the selection of the variables, which were combined to
establish the original 16 possible combinations. These variables were chosen
by the panel as those having the greatest influence on the kinds of pollution
controls needed for coal cleaning operations. The number of variables was
limited to four to keep the combinations to a manageable number. The number
of combinations which result from considering each variable at two levels is
equal to 2 raised to the power of the number of variables. Thus, with
four variables there are 16 combinations and with five variables there
are 32 combinations and so on. The four variables selected were
-------
neutralization potential of soil surrounding the coal cleaning plants,
pyritic sulfur content of ROM coal, average annual precipitation, and
coal cleaning process technology. The reasons for selecting these are
discussed below.
• Neutralization potential (N). This is related to the presence
or absence of calcareous materials in the spoils, refuse,
and soil which can neutralize acid wastewater and is most
easily measured as soil pH. As such, the concept trans-
cends that of acid or alkaline mine drainage, though geo-
graphically they fall in the same areas. The intent
here is to consider the capacity of the natural environ-
ment to absorb and counteract the impact of acid waste-
waters. Thus, a high level of N constitutes a low pollu-
tion potential.
• Pyritic sulfur content of ROM coal (S). Areas of both
high and low neutralization potential produce coals which
vary widely in their content of ash, sulfur, iron, and
other metals. There is a high positive correlation between
pyritic sulfur concentrations and other coal contaminants that
have high pollution potential. Refuse piles are
sources of heavy metals, iron, and acid-soluble leachates
which are a continuing threat to environmental systems
and water supplies.
• Average annual precipitation (R). Precipitation is the
most Important single factor in creating an environmental
setting. Precipitation, particularly rainfall, is also the
single environmental event most affecting the pollution
potential from coal cleaning operations. It governs (1) rate
of runoff from refuse and coal piles, (2) deposition of airborne
contaminants, (3) erosion or dissolution of partlculates and salts
into natural waters, (4) sediment contaminant loading, and
(5) acidification of soils and waterways.
• Coal cleaning process technology (T). Characteristics of the
wastes from a coal preparation plant are highly dependent on
the processes utilized, which are in turn dictated by the raw coal and
final product. As processes become increasingly more complex
-------
and finer fractions of the coal are cleaned and collected, the
pollution potential changes because (1) complex plants frequently
use thermal driers which are a source of gaseous and particulate
air pollution, (2) there is greater opportunity during
processing for the soluble pollutants to be contacted by a
solvent, and (3) chemical additives are used in static
thickeners and froth flotation cells, thus increasing the
number of potential pollutants. Since this variable did not
fall naturally in a high (complex) and low (simple) range, it
was necessary to select the kinds of cleaning processes
that would be studied. The kinds of unit operation combina-
tions typically employed in different plant types are shown in
Table 1. In order to provide a systematic method for select-
ing plants based on technology extremes, coal preparation has
been classified into four levels according to the coal sizes
being washed. The coal preparation plants are then categor-
ized into nine generic types based on coal cleaning processes
employed. These levels of preparation and types of plants
are summarized in Table 1 and defined in more detail as
follows:
• Level 1 - Crushing and Sizing
• Type A: Crushing for top size control with limited
removal of coarse refuse and trash by scalping
screen and/or rotary breaker,
• Level 2 - Coarse Size Coal Beneficiation
Type B: Type A followed by dry screening at 3/8
inch and wet beneficiation of plus 3/8-inch material
only with jig or dense-medium vessel. Minus 3/8-inch
material is mixed with coarse product without washing.
Simple mechanical dewatering for plus 3/8-inch material.
• Level 3 - Medium Size Coal Beneficiation
Type C: Type B plus dry beneficiation of minus 3/8-
inch material with air table.
-------
TABLE 1. GENERIC TYPES OF COAL PREPARATION PLANTS
Coal Size and Unit Operation
Level (a)
1
2
3
3
3
4
4
4
4
Plant
Type
A
B
C
D
E
F
G
H
I
Coarse
(3 x 3/8 in.)
CS
CS 4- J/DMV + MD
CS 4- J/DMV 4- MD
CS 4- J/DMV 4- MD
CS 4- J/DMV 4- MD
CS 4- J/DMV 4- MD
CS 4- J/DMV 4- MD
CS 4- J/DMV 4- MD
CS 4- J/DMV 4- MD
Medium
(3/8 in. x 28 M) (28
* Ai
< WT + MD —
< DMC 4- MD —
WT 4- MD HC 4-
WT 4- MD F 4-
DMC 4- MD HC 4-
DMC 4- MD F 4-
Fine
M x 0)
— >
— *•
MD 4- TO
MD 4- TO
MD 4- TO
MD 4- TD
(a) Level 1 - Crushing and Sizing
Level 2 - Coarse Size Coal Beneficiation
Level 3 - Medium Size Coal Beneficiation
Level 4 - Fine Size Coal Beneficiation
(b) CS - Crushing and Sizing Devices
J - Jigs
DMV - Dense-Medium Vessels
DMC - Dense-Medium Cyclones
AT - Air Tables
WT - Wet Concentrating Tables
HC - Hydrocyclones
F - Froth Flotation Units
MD - Mechanical Dewatering Devices
TO - Thermal Dryers
-------
Type D: Type A followed by wet screening at 3/8 inch
and Type B beneficiation of minus 3/8-inch material
with concentrating table.
Type E: Same as Type D except that heavy-medium
cyclone is used for minus 3/8-inch material beneficiation.
• Level 4 - Fine Size Coal Beneficiation
Type F: Type D followed by wet beneficiation of minus
28-mesh material with hydrocyclones. Thermal drying
for fine coal product.
Type G: Same as Type F except that froth flotation
circuit is used for minus 28-mesh material beneficiation.
Type H: Type E followed by wet beneficiation of minus
28-mesh material with hydrocyclone. Thermal drying
for fine coal product.
Type I: Same as Type H except that froth flotation
circuit is used for minus 28-mesh material beneficiation.
The two levels of plant types selected from Table 1 to represent
technology extremes are (1) Level 1 which represents a simple
(low) technology group including plant types A and B, and
(2) Level 4 which represents a complex (high) technology group
including plant types F, G, H, and I*. Process stream samples
representative of the other plant types can be obtained, if
desired, by sampling upstream of specific unit operations or
pollution control devices.
Low and High Levels for
Each Classification Variable
To better identify suitable site categories, two levels for each
of the N, S, R, and T variables have been defined. Table 2 lists the
classification variables and their associated low (0) and high (1) levels.
This selection was made based on the complexity of the technology ««ed
and not necessarily on degree of pollution potential Level 3 facilities,
for example, may have the greatest pollution potential, since they make
no attest ^Incorporate ?he fine-fines (28M x 0) back into their product.
-------
TABLE 2. CLASSIFICATION VARIABLES AND ASSOCIATED
LEVELS USED TO DEFINE SITE CATEGORIES
Variable
N (Neutralization
potential)
S (pyritic
sulfur)
R (average annual
rainfall)
T (coal cleaning
process technology)
Low Level (0)
PH> 7.5(a)
<. 1.0%
<_ 15 in./yr
Plant Types
A & BW
High Level(l)
pH <_ 6.0(a)
>_ 2.0%
>_ 25 in./yr
Plant Types F,
G, H. llb'
(a) pH of soil in the receiving environment. As defined,
low N actually refers to a low pollution potential or
high soil alkalinity which is, in fact, a high ability
to neutralize acid streams.
(b) See page 7 for definitions of plant types.
TABLE 3. FACTORIAL LISTING OF
SITE CATEGORIES
(N,S,R,T)(a) (N,S,R,T)
(1111) (0111)
(1110) (0110)
(1101)(b) (0101)(b)
(1100) 0>) (0100)(b)
(1011) (0011)
(1010) (0010)
(1001)(V (0001)
(1000) (0000)
(a) See Table 2, above, for definitions of low (0)
and high (1) levels of pollution po-
tential as expressed by neutraliza-
tion potential, N; pyritic sulfur, S;
average annual rainfall, R; and coal
cleaning process technology, T. The
combination (1010), for example, de-
notes a site category with N and R at
Level 1 and S and T at Level 0.
(b) Site categories marked with an asterisk
are excluded from further consideration,
since no cleaning plants exist with
these combinations of variables.
in
-------
The low and high levels are intended to reflect relatively small or large
pollution potentials, respectively, for the first three variables and simple
or complex process technology. Thus, (0) indicates low sulfur and rainfall
but high neutralization potential. It should be noted that intermediate ranges
of N, S, R, and T have been excluded in the definitions of low and high levels.
This means that the sites ultimately selected will have relatively extreme
values (low or high) for the classification variables. Such sites are de-
sired in order to reflect the range of possible conditions among coal
cleaning plants, and are neither expected nor intended to be "statistically
representative" or "typical" with respect to the population of coal cleaning
sites.
Site Categories
On the basis of low (0) and high (1) levels for each variable shown in
Table 2, a site category can be represented by an ordered sequence, such as
(0110) for N, S, R, and T, respectively. Such a sequence represents a site
category characterized by a neutralization potential, N, at Level 0; pyritic
sulfur, S, at Level 1; annual rainfall, R, at Level 1; and process tech-
nology, T, at Level 0. Each four-component vector, consisting of zeros and
ones, serves to identify a particular combination of low and high levels
for the classification variables N, S, R, and T.
Complete Factorial Listing
With the two levels defined for each of the four variables, N, S, R, and
T, a complete factiorial listing of the 16 conceptually possible combinations
of levels can be generated.
The 16 combinations of low and high levels are listed in Table 3. (Page 10)
The combinations marked with an asterisk denote site categories that were
subsequently excluded from consideration. It is seen that the excluded
categories are all associated with low rainfall (R=0) and are therefore
represented by western sites. Four of these site categories (0100), (0101),
(1100), and (1101) have relatively high sulfur (S=l) and high sulfur is
not characteristic of western coal. The combination of low rainfall and
11
-------
high sulfur is taken to be an inconsistency and is the basis for exclusion
of these four categories from the list. In a similar manner, the site categories
(1000) and (1001) are excluded from the list because the low level of
neutralization potential (K-l) is not characteristic of low rainfall (R=0)
western sites. The remaining ten site categories are subsequently examined
to assess their relative importance.
Sequential Sampling Procedures
Sequential sampling procedures are often appropriate in those settings
where time or cost constraints may cause the sampling procedure to be
terminated prior to the sampling of all units. In the context of sampling coal
cleaning plants, for example, it is possible that sufficient resources may
not be available to permit sampling at ten different sites, with one site
selected from each of the ten site categories. In this setting one could
raise the following questions: If only one site category could be sampled,
which would it be? If an additional site category could then be sampled,
which would it be? This could be applied for all ten site categories. Clearly,
it would be desirable to sample ten different sites, one from each of the
ten site categories. However, to protect against the possibility that this
cannot be done, it is prudent to identify an optimal sampling sequence.
The optimal sequence would be one wherein the sites providing the most
desired information are sampled early in the sampling program, with the
sites yielding the less desired information being scheduled for sampling
later in the program. This procedure would permit the most desired infor-
mation to be obtained even though the sampling is terminated before all
sites have been sampled.
In the statistical literature, sequential sampling designs based on
factorial combinations, as shown in Table 3, are usually called one-factor-
at-a-time designs. These designs have been examined by Webb as a
contractible permutation invariant design and by Rechtschaffner as a
saturated fractional factorial. The design is said to be contractible
because the sequence of tests can be stopped after specified groups
of tests have been run; the design is saturated because no degrees of
freedom are available for estimating error.
12
-------
A Statistical Rationale for
Sequential Sampling
A basic statistical rationale for identifying sampling sequence for
factorial designs may be stated as follows:
• Early in the sampling sequence, sample those site
categories that permit the "main effect" of each
classification variable to be investigated; next
include those site categories that permit investi-
gations of the "interaction effect" associated with
two classification variables, then three classifi-
cation variables, etc.
This rationale is based on the assumption that the most desired information
is that associated with the main effects of single variables. After
obtaining information on the effect of each classification variable, the
next priority is then associated with information on the combined effects
of two classification variables, etc.
Recommended Procedure for
Selection of Site Categories
For the selection of site categories Battelle recommends that a sequen-
tial sampling procedure for factorial designs be used, with the investigation
of main effects of single variables given first priority and interaction
effects among pairs of variables given second priority.
In comparison with the designs discussed in the literature, the
sampling designs resulting from this procedure are unavoidably degraded
by the deletion from the factorial listing of the six site categories in
which cleaning plants do not exist. A complete factorial design, consisting
of all 16 possible site categories, would permit the consideration of all
possible unconditional main effects and interactions of variables. As
discussed below, the degraded design permits only certain conditional main
effects and interactions to be considered. Nevertheless, the following
discussions show that the degraded design retains a number of desirable
statistical properties, and it is generally supported by individual, independent
evaluations made by the nine panel members.
13
-------
Recommended Sampling Design
Battelle has applied the procedure recommended above to the selection
of site categories. The application was made among the ten existing site
categories represented by the factorial listing shown in Table 3.
The sequential sampling design for the ten coal cleaning site categories
resulting from the site selection procedure is shown in Table 4. The table
shows three sampling groups of site categories. The first group to be
sampled consists of four site categories: (1111), (1110), (1011), and (0111).
Site Category (1111) represents the "worst" case with all four classification
variables at high levels of pollution potential. Group 2 would be sampled
next and consists of four site categories. Group 3 would be sampled last
and consists of two low-rainfall, western site categories. The geographic
regions associated with the site categories are shown in the last column
as Northern Appalachia (NA), Southern Appalachia (SA), the Midwest (MW),
Alabama (A), and the West (W).
The recommended sampling design calls for sites to be sampled in the
sequence 1 through 10 as shown in Table 4. On the basis of the statistical
rationale described above, the sites within each of the three groups may be
sampled in any order. However, on the basis of engineering judgment, Battelle
recommends the exact sampling order for each group indicated in Table 4.
Further Justification for this sampling order is given following a more
detailed discussion of the statistical properties of the recommended sampling
design.
Assessments of the Recommended
Sampling Design
As noted above, the recommended selection procedure for site categories
assigns first priority to investigation of single-variable main effects,
followed by investigation of two-variable interaction effects. Because of
the limited number of site categories, these main effects and interactions
are conditional (rather than unconditional). The following discussion
makes those conditions explicit.
14
-------
TABLE 4. RECOMMENDED SEQUENTIAL SAMPLING DESIGN
FOR COAL CLEANING PLANTS
Group
Number
1
2
3
Site
Category^3'
1
2
3
4
5
6
7
8
9
10
Site Characteristics
(N, S, R, T)(b)
(1111)
(1110)
(1011)
(0111)
(1010)
(0110)
(0011)
(0010)
(0001)
(0000)
Regions
NA
NA
NA
MW
NA
NA
MW, SA,
MW, SA,
W
W
(c)
A
A
(a) One site from each category is to be sampled in the se-
quence from 1 through 10, with Site Category 1, defined
as (1111), sampled first and Site Category 10, defined as
(0000), sampled last. The sampling may be terminated
after sampling four site categories (Option C), eight
site categories (Option B), or ten site categories
(Option A).
(b) See Table 2, page 10 for definitions of low (0) and high (1)
levels for neutralization potential, N; pyritic sulfur, S;
average annual rainfall, R: and coal cleaning process techno-
logy, T. The combination (101U), for example, denotes a
site category with N and R at Level 1 and S and T at
Level 0.
(c) Regions associated with the site categories are denoted by
Alabama (A), Northern Appalachian (NA), Southern Appala-
chian (SA), Midwest (MW), and West (W).
15
-------
Conditional Main Effects. To obtain Information on the main effects
of neutralization potential, for example, data from two sites must be compared,
with one site having relatively good and one site having.relatively poor
neutralization potential; the comparison sites should be "matched" with
regard to their levels of sulfur, rainfall, and technology. Such a compari-
son yields a conditional main effect. The word "conditional" is used to
highlight the fact that a main effect of N is determined under the conditions
specified by the levels for S, R, and T.
To indicate how the recommended sampling procedure generates conditional
main effects, note that the first two site categories, (1111) and (1110),
are matched with respect to the levels of N, S, and R. In particular, N, S,
and R are each at level 1 for both site categories. The site categories
differ only with respect to the levels of technology, T. Thus, the sampling
results from the sites selected from those two site categories would permit
differences attributable to technology to be investigated. Such differences
would be labeled as conditional main effects of technology given relatively
matched, high level conditions for N, S, and R. These conditional main
effects are denoted by the symbol (HIT). In a similar manner, after a site
from Site Category 3 is sampled, conditional main effects of sulfur would be
obtained by comparing data from Site Category (1011) with data obtained from
Site Category (1111). This would yield conditional main effects of sulfur,
given relatively matched, high levels for N, R, and T, and is represented
by the symbol (1S11). Similarly, after sampling Site Category 4, a conditional
main effect for N, given high levels for S, R, and T, would be obtained by
comparing results obtained from sites selected from Site Categories (0111)
and (1111) to give (Nlll). In this way, it is seen that after sampling
one site from each of the first four site categories—1,2,3,4 of Table 4—
a conditional main effect could be investigated for N, S, and T under the
condition that all matched variables are at relatively, high levels.
To summarize, Option C (see footnote (a) to Table 4) of the recommended
sampling procedure is intended to yield data that will permit conditional
main effects of N, S, and T to be investigated, under the condition that
all matched variables are at relatively high levels.
Conditional Interactive Effects. Some of the effects related to com-
bined changes in the levels of two variables can also be investigated by
16
-------
means of the recommended sampling design. For example, after sites from the
first five site categories shown in Table 4 have been sampled, comparisons
among results obtained for Site Categories 1, 2, 3, and 5 will permit investi-
gations to be made of conditional interactions between technology and sulfur
(TxS) with both N and R at high (1) levels. To see this, it is necessary
to note only that in these site categories- (N, S, R, T) = [(1111), (1110),
(1011), and (1010)]- the levels of N and R (first and third positions) are
all equal to 1, whereas the levels of S and T (second and fourth positions)
take all possible combinations of levels: (11), (10), (01), and (00).
It should also be noted that after Site Category 5 has been sampled,
a second conditional main effect of technology may be investigated by
comparing the results obtained from Site Category 5, (1010), with the results
obtained from Site Category 3, (1011). This comparison yields a main effect
of technology given that N, S, and R have levels 1, 0, 1, respectively.
At this stage of the sampling procedure, the two resulting conditional main
effects of technology are symbolized by (HIT) and 101T).
It can be seen that there are eight conceptually possible conditional
main effects for technology, associated with the eight possible combinations
of zeros and ones for the first, second, and third positions of the symbol
(xyzT). Theoretically, the average of these eight conditional main effects
would give the unconditional main effect of technology. However, because
six of the 16 conceptually possible combinations do not represent actual
coal cleaning sites, unconditional main effects cannot be obtained from the
sampling procedure. More generally, it should be noted that the sampling
procedure can yield only conditional main effects and interactions. Because
there are many possible conditions associated with each main effect and
interaction, the exact condition must be explicitly indicated.
Table 5 summarizes the conditional main effects and interactions that
may be investigated as data are successively obtained from Site Categories
1 through 10, as required by the recommended sampling designs. As shown in
the table, no main effects or interactions are obtained from sampling at
Site Category 1 (1111). After sampling from Site Category 2 (1110), a
conditional main effect of technology, T, given that N,S, and R are all at
Level 1, is obtained by comparing results for Sites 1 and 2, as shown by the
last column in Table 5. This conditional main effect is symbolized in
17
-------
TABLE 5. CONDITIONAL MAIN EFFECTS AND CONDITIONAL INTERACTIONS
ASSOCIATED WITH THE RECOMMENDED SAMPLING DESIGN
Croup
Option
C
B
A
Site
Category (a)
1(1111)
2(1110)
3(1011)
4(0111)
5(1010)
6(0110)
7(0011)
8(0010)
9(0001)
10(0000)
(a) Obtained from Table
(b) The
symbol (1S10),
Conditional
Effects
—
T
S
N
T
S
SxT
K
T
NxT
N
S
NxS
N
S
T
SxT
NxT
NxS
R
R
T
RxT
4.
for example, is
Conditions^)
(N.S.R.T)
(1111)
(HIT)
(1S11)
(Nlll)
(101T)
(1S10)
(1S1T)
(N110)
(OUT)
(NUT)
(N011)
(OS11)
(NS11)
(N010)
(OS10)
(001T)
(OS1T)
(N01T)
(NS10)
(OOR1)
(OORO)
(OOOT)
(OORT)
ordered according
Comparison of
Categories*"^
—
1,2
1.3
1,4
3,5
2,5
1.2,3.5
2,6
4.6
1.2.4,6
3,7
4,7
1.3,4,7
5.8
3,5
7,8
4,6,7,8
3,5,7.8
2.5.6,8
7.9
8,10
9,10 .
7,8,9.10
to (N.S.R.T) and
(c)
indicates that a conditional main effect of sulfur can be investigated,
the condition being that N and R are at Level 1 and T is at Level 0.
The two-variable Interactions are indicated by symbols having two let-
ters, so that (1S1T) indicates a conditional interaction between sulfur
and technology given N and R are at Level 1.
The entry 1,2,3,5 shown Tor (1S1T) indicates, for example, that a con-
ditional interaction effect can be investigated by comparing results
obtained from sites selected front Site Categories 1(1111), 2(1110),
3(1011), and 5(1010).
18
-------
column 3 as (HIT). Subsequent sampling of Site 3 then allows the conditional
main effect of sulfur (1S11) to be obtained by comparing results for Sites 1
and 3, with N, R, and T at Level 1. Table 5 shows that after obtaining data
from one site in each of the first four site categories conditional main
effects for N, S, and T can be investigated.
The second group of sites is to be selected from Site Categories 5, 6,
7, and 8. The table shows that after sampling Site Category 5, a second
conditional main effect for T (101T); a second conditional main effect for
S (1S10); and a conditional two-variable interaction (1S1T) between sulfur
and technology may be investigated. It should be noted that this two-variable
interaction (1S1T) requires comparisons among data from the four site
categories—1, 2, 3, 5—as shown in the last column of the table. In general,
the table shows that, as sites are successively sampled, more and more condi-
tional main effects and conditional interactions can be investigated.
Some Limitations. It is important to note that most of the statistical
principles described above are strictly applicable to controlled experimenta-
tion performed in a laboratory. Such principles not only serve as a basis
for defining experimental conditions (sampling design) but also provide
appropriate methods for analysis of the resulting data. The sampling design
and the method of data analysis are mutually dependent. The statistical
properties of the sequential sampling design shown in Table 5 are discussed
as though the recommended sampling of the coal cleaning plants were per-
formed in a controlled laboratory environment. This, of course, is not the
case. Many uncontrolled influences, such as recent weather conditions, may
serve to contaminate the data obtained at the sampled sites. These uncon-
trolled influences would be expected to complicate the interpretation of
measured differences between two sites.
As an example, suppose that a particular environmental measures
are found to be markedly different for Site 1 (1111) and Site 2 (1110).
According to Table 5, such a difference is a measure of the conditional main
effect of technology (HIT). However, such an interpretation may be totally
erroneous because the difference, in fact, may have resulted from variables
not controlled at the two sites. From this perspective it is seen that
the only conditions at the two sites that are "controlled" are those condi-
19
-------
tions associated with the "matched" levels of N, S, and R, and Table 2
shows that even those variables are only loosely constrained to lie in
relatively broad "high" ranges.
These considerations indicate that the interpretation of every quanti-
tative comparison involving different sites must be separately examined for
validity. The recommended sequential sampling design does not, in itself,
guarantee that such comparisons will yield valid measures of conditional
main effects and interactions. The recommended sampling design simply
serves to identify which conditional main effects and interactions should
be investigated and which site comparisons are required to make such
inves tigations.
Crude Benefit/Cost Ratios. A further assessment of the sampling design
is provided by examining the average benefit/cost ratio as successive sites
are sampled. These ratios may be crudely approximated by assuming a unit
cost for sampling each site. It is also assumed that the benefits are
measured by the number of main effects and interactions that can be in-
vestigated.
Table 6 shows the benefit/cost ratios based on the sampling sequence
given in Table 4. The table shows an average benefit/cost ratio of 0.75 after
the four sites of Option C have been sampled. This ratio increases to 2.25
after the eight sites of Option B have been sampled. A small decrease In
the ratio occurs with the addition of the western sites of Option A
because not many additional effects can be investigated by sampling these
sites. This results from the previous exclusion of the six western site
categories.
A Review of the Optional
Sampling Plans
On the basis of the preceding discussion, a summary review of the
sampling Options A, B, and C can be made. Option C consists of sampling one
site from each of the first four site categories shown as Group 1 in
Table 4. This option yields data on Site Category (1111) that has the
potential for the greatest pollution severity. In addition, data on condi-
20
-------
TABLE 6. CRUDE BENEFIT/COST RATIOS
Cumulated
Option Sampled Sites
C 1
2
3
4
B 5
6
7
8
A 9
10
Cumulated
Benefits (a)
0
1
2
3
6
9
12
18
19
22
Cumulated
Costs 0>)
1
2
3
4
5
6
7
8
9
10
Average
Benefit/Cost
Ratio
0.00
0.50
0.67
0.75
1.20
1.50
1.71
2.25
2.11
2.20
(a) Assumes that each conditional main effect and interaction shown in
Table 5 is of unit benefit.
(b) Assumes unit cost for each sampled site.
21
-------
tional main effects of technology, sulfur, and neutralization potential are
obtained from this option. Although Option C provides information on
separate conditional main effects for each of these classification variables,
it does not provide any information on the combined effects of two classifi-
cation variables. Such information is provided by Option B which involves
sampling from Groups 1 and 2 shown in fable 4. In particular, under Option
B, in addition to the information obtained from Option C, conditional inter-
action information is obtained for combined effects of sulfur and
technology, neutralization potential and technology, and neutralization
potential and sulfur. Option B also yields data on main effects of tech-
nology, sulfur, and neutralization potential under conditions different
from those obtained under Option C. In addition to the three conditional
main effects of Option C, Table 5 shows that Option B provides information
on a total of nine conditional main effects and six conditional two-variable
interaction effects. Option A consists of sampling sites from all ten site
categories in Table 4. This option yields additional information on two
conditional main effects of rainfall, a conditional main effect of technology,
and a conditional interaction effect between rainfall and technology. The
two additional site categories associated with Group 3 are placed at the
end of the sequential sampling plan because: (1) Site Categories 9 and 10
require sampling at western sites for which unit cost may be greater than
for the other site and (2) as shown in Table 5, the benefit/cost ratio
passes through a maximum for the eight sites represented by Option B. In
short, Option A yields the maximum information obtainable from the ten site
categories; Option B yields over 80 percent of the possible conditional
main effects and interactions, with the effects of rainfall excluded; and
Option C yields information on the main effects of technology, sulfur,
and neutralization potential.
There are several characteristics of western coal producing areas which
are important to consider. Largely because of the low rainfall, the environ-
ment of the West is fragile; that is, it is easily damaged and slow to
recover. With the expansion of coal production in that part of the country
as predicted, the protection of that environment will be of Increasing con-
cern. In addition, there are metal constituents in western coals which
differ from those in other parts of the country. Such things as uranium
22
-------
as well as heavy metals (e.g., antimony, cadmium, and selenium) as effluent
/Q\
products should be considered. Finally, the only opportunity to consider
rainfall as a variable and study its effects as a part of this program is
to include western sites. Option A is therefore recommended.
Ranking by Panel Members
The desirability of sampling at each of the ten site categories was
independently assessed by each of the nine panel members. The agreement
among the nine resulting rankings was assessed using Kendall's coefficient
(3)
of concordance. On a scale between 0 (no agreement) and 1 (perfect
agreement), the panel rankings yielded a coefficient of concordance equal
to 0.78 when the site categories were grouped into the three categories defined
in Table 4. The concordance measure of 0.78 is taken to indicate a reasonable
measure of agreement for the preference rankings of the three groups. How-
ever, considerable disagreement among panel members was found for the
relative rankings of the site categories within the three groups. In short,
the panel agreed dh the overall rankings of the three groups but did not
agree on the relative rankings within groups. The disagreement may be
partly due to the differing technical specialties among the panel members.
Site Selection
The purpose of the following selection procedure is to provide a
mechanism for selecting sites within each of the ten site categories by
imposing a series of constraints. Abbreviated lists of candidate sites from
each site category chosen by imposing some of these constraints are presented
in Appendix A. These lists are based solely on information readily available
in the literature and previous surveys of coal cleaning plants. Telephone
contacts and site visits with plant management at sites which seem to be
prime candidates for study will provide the additional information needed
to make the final site selection from the abbreviated lists.
The constraints described below are not all equally binding and may
be classified in order of importance as primary, secondary and tertiary as
shown in Table 7. Plants are expected to be found which meet most, but not
23
-------
all, of these constraints. Primary constraints are those essential criteria
that must be met in order for a facility to be considered for the sampling
program. Plants that also meet some of the secondary constraints will be
considered highly desirable sampling locations, but the status of these
constraints will not necessairly result in inclusion or exclusion of a
facility from the list of sampling locations. If several plants can be
found in a given site category which meet all of the primary and all or
most of the secondary constraints, then the tertiary constraints will be
imposed in the final selection of a facility for testing.
TABLE 7. FINAL SITE SELECTION PROCEDURES,
USE OF PRIORITIZED CONSTRAINTS
Constraints
Primary
(1) Most efficient pollution control
technology
(2) Plant management cooperation
Secondary
(1) Two ages of refuse piles
(2) Greater than mean plant capacity
(3) Baseline data available
Tertiary
(1) Minimal coal industries in area
(2) Nonmixed refuse piles
(3) Uniform coal source
24
-------
Primary Constraints Classed as Essential
Primary constraints are described as listed in Table 7.
(1) Most efficient pollution control technology. Since the overall
goal is to assess pollution potential, we plan to assess the pollu-
tion loading of one of the coal cleaning facilities in each site
category which is currently serving as a benchmark in pollution
control technology for that particular type of coal cleaning process.
Selection of the facility within each site category which
has the most efficient pollution control technology will be
accomplished by using the ranking information in Table 8. After
site visits have provided details on the pollution control
technology at each site on the abbreviated lists, a score will be
tallied on the basis of Table 8 and the number of points awarded
for each type of pollution control technology as follows: (a)
high efficiency techniques - 3 points, (b) medium efficiency
techniques - 2 points, and (c) low efficiency techniques - 1 point.
Facilities in each site category with the highest score (most
efficient pollution control technique) will be given first pri-
ority in the final site selection process.
(2) Plant management cooperations. The willingness of plant management
to let us carry out field studies is a necessity for a successful
sampling and testing program.
Secondary Constraints
Secondary constraints are described as listed in Table 7.
(1) Two ages of refuse piles. These are required to initially
discern the effects of leaching upon the area. One pile
should be approximately 10 years old for these processes
to be easily defined.
25
-------
TABLE 8. RANKING OF POLLUTION CONTROL TECHNOLOGY FOR CLEANING
PLANTS AND THEIR ASSOCIATED REFUSE DISPOSAL AREAS
Coal Cleaning
Ale Pollution Control Technology
Partlculate Control
Devices
Gaseous Pollutant
Control Devices
Plants
Hater Pollution Control Technology
Suspended Solids
Control Devices
Dissolved Solids
Control Devices
Disposal Area for Cleaning Plant Refiise
Air Pollution Control Technology Hater Pollution Control Technology
Fugitive Dust
Control Techniques
Pico Control Erosion Control
Techniques Techniques
Leachate
Control Techniques
HUh Efficiency
Bog (fabrU; filters
Virturl acrubtaers
Self-induced scrubbers
S02 scrubbing
systeat*)
High-efficiency cyclone*
iBplngcoent scrubber*
Centrifugal scrubber*
Closed loop using
thickeners with
f locculants
Open loop using
Blcro screen
units, vecuua
filters, settling
ponds
Closed process
water loop
Hedlun
Open process weter
loop with
neutralisation,
aeration , and
sedimentation
Sealing agent
Soil covering
R* vegetation
Efficiency
Compacting
low-wind- velocity
location
Soil covering Diversion ditches
and stltratlon
basins for
runoff
Mlnlul grade of Coapactlng
eabankaent Soil covering
CoBpactlng
Sealing agent
•
Ion exchange treatment of
leachate pond water'*'
Reverie osmosis tieetaant
of leachate pond uiter^
lisachaie collection pond(i)
with neutralisation,
aeration, and sedimen-
tation
{•pervious liner for
disposal area
De water refuse slurries
Grading refuse to prooote
runoff
X* vegetal, ion
tow Efficiency *
louver (gravity)
settling chanters
Conventional cyclones
tt'ater spray
Dust enclusuraa
Vsnturl scrubbers
taping eaent
scrubbers
Centrifugal
scrubbers
Self-induct d
scrubbers
Opan loop using
centrifugal
Opan process water
loop with
neutralisation
Hater spray
Hater spray
Diversion ditches for
Slurry Impoundments
Soil covering to clnialts
exposed refuse area
(a) Mot presently In use at a coal cleaning facility.
-------
(2) Greater than mean plant capacity. A plant selected
from any site category will be one whose production
capacity is greater than the mean capacity of the plants
listed in that site category. Plant capacity is important
for two reasons: larger plants have a greater
pollution potential and are more likely to yield useful
information on effectiveness of complex pollution controls
and the greatest effects on ecosystems will result
from these plants because of the volume of wastes.
(3) Baseline data available. The best way to determine the
effects of coal cleaning pollutants on environmental
quality is to have environmental data that extend over
a period of time, preferably starting from before the
CCP began operation. It will be more efficient to study
plants where air and water quality data are already in existence.
Tertiary Constraints
There are three desirable criteria which can be applied if several
sites are left following application of primary and secondary criteria.
These are as follows.
(1) Minimal coal industries in the area. This criterion is
intended to minimize the compounding or the confounding of
pollution effects by other related industries whose
effluents will likely be similar and which may overshadow
those of the CCP.
(2) Nonmixed refuse piles. This criterion is intended to
enable sampling of segregated solid wastes from the various
sizes of cleaned coal, recognizing that the polluting
constituents may not be uniformly distributed. Such
segregation may be difficult to find.
(3) Homogeneous coal source. To assure that effluents are
relatively constant in their chemical characteristics,
the ideal situation would be to select plants with a
single source of coal of uniform quality, However,
27
-------
it is recognized that coal seam characteristics vary
both laterally and vertically and thus some variability
is inevitable. The attempt will be merely to choose
plants which receive coals with as uniform quality as
possible.
Potential Sampling Locations
An abbreviated list of preparation plants which fits each of the
ten site categories (see Table 9) is included in Appendix A along with a
summary of their characteristics as reported in the open literature. These
47 potential sampling locations represent the second step in the site
selection process and must be condensed through telephone calls and
site visits with plant management. The lists in Appendix A should be con-
sidered tentative, since process equipment and pyritic sulfur information
is not sufficiently detailed in the literature to make positive Judgments
for all of the sites listed. In addition, the management of the plants on
the abbreviated lists have not yet been contacted to determine their
willingness to cooperate din a testing program.
Initially, the 4004- coal cleaning plants listed in the 1977 Keystone
(4)
Coal Industry Manual were separated according to the ten site categories
on the basis of the four major variables (i.e., neutralization potential of
the soil, pyritic sulfur content of ROM coal, average annual precipitation,
and coal cleaning process technology). To shorten this process, it was
assumed that all of the sites located in northern Appalachia had a low
neutralization potential in the refuse and surrounding soil. Thus, sites
in northern Appalachia were considered to have a high pollution potential.
All of the other coal cleaning sites in the U.S. were considered to have a
high neutralization potential (low pollution potential). This assumption
about soil pH is in agreement with the effluent guidelines report.
Data on the pyritic sulfur content of ROM coal were also used to
organize facilities according to the ten site categories. This information
(4)
was obtained from the 1977 Keystone CoalIndustry Manual and an EPA
(9)
report on sulfur in U.S. coals . The best information these two sources
could provide is still only an estimate of the pyritic sulfur content of
28
-------
TABLE 9. SUMMARY OF POTENTIAL SAMPLING LOCATIONS
BY SITE CATEGORY(a)
Sampling
Option
C
B
A
Total sites
Site
Category
1
2
3
4
5
6
7
8
9
10
Site
Characteristics
(N.S.R.T)
(1111)
(1110)
(1011)
(0111)
(1010)
(0110)
(0011)
(0010)
(0001)
(0000)
to be considered further . . .
Number of
Potential
Sites(b)
3
6
11
2
3
4
8
4
3
_3_
. . 47
(a) See Tables 1, 2, and 4 for details on site cate-
gory and characteristics.
(b) See Appendix A for actual listing of potential
sites.
29
-------
coals actually cleaned by a specific facility. For most facilities,
however, this estimate of pyritic sulfur levels should be sufficiently
accurate to categorize the ROM coal as either low or high in pyritic sulfur.
Information on process technology which was used to organize cleaning
plants according to the ten categories was obtained primarily from the list
(4)
in the 1977 Keystone Coal Industry Manual . Since this process information
was not sufficiently detailed, it was supplemented by information from
journal articles and the results of other cleaning plant surveys. These
additional sources are reported as footnotes to the lists in Appendix A.
However, detailed information was not available for some facilities in any
of the articles or surveys. Therefore, some questions about the equipment
in use at a particular site were not resolved. For example, the description
of equipment at a plant may have indicated that driers were used when it
would have been preferable to specify the type of drier, such as centrifugal
or thermal.
The initial sorting of the more than 400 known coal cleaning plants
using the four variables described earlier resulted in up to 40 facilities
listed in some plant categories and as few as two facilities in other plant
categories. The large lists were reduced to the more manageable short lists
found in Appendix A by imposing several different constraints. The constraints
used to select the best plants for sampling purposes include (1) plants
cleaning coal from only one seam, (2) plants with a capacity above the mean
for that category, and (3) plants not located near coal-fired power plants.
This information was available in the literature and did not require a site
visit. Verification or investigation of some of the constraints listed in
previous sections of this report will require site visits. Lists for some
plant categories were already extremely limited in number, so none of the
above constraints were used to narrow the lists further. Instead, footnotes
were used to indicate the occurrence of undesireable constraints from a
sampling viewpoint.
30
-------
Information Needed for Final Site Selection
Some of the information crucial in the final selection of one sampling
site for each site category was not available in the literature. This in-
formation must be obtained by site visits to some of the 48 cleaning plants
listed in Appendix A. The initial step should be to determine whether the
management of these plants will be willing to cooperate in a field sampling
program. Site visits should be made to plants where the management has
given a positive response to a request for a sampling program. Information
that needs to be obtained during these site visits is outlined in
Appendix B. On the basis of the information obtained during site visits,
a final decision can be made regarding the most appropriate sampling sites.
31
-------
REFERENCES
(1) Hansen, M.H., Hurwitz, W.N., and Hadow, W.G., Sample Survey Methods and
Theory, Vols. 1 and 2, John Wiley & Sons, Inc., New York (1953), 332 pp.
(2) Cochran, W.G., and Cox, G.M., Experimental Design, John Wiley & Sons, Inc.,
New York (1968), 611 pp.
(3) Snedecor, G.W., and Cochran, W.G., Statistical Methods, The Iowa State
University Press, Ames (1973), 593 pp.
(4) 1977 Keystone Coal Industry Manual, F.G. Nielsen (Ed.), McGraw-Hill, Inc.,
New York (1977), 1,184 pp.
(5) Effluent Guidelines Division, Development Document for Interim Final
Effluent Limitation Guidelines and New Source Performance Standards for
the Coal Mining Point Source Category, EPA 440/1-76/057-a (1976).
(6) Webb, S.R., "Design, Testing, and Estimation in Complex Experimentation;
Expansible and Contractible Factorial Designs and the Application of Linear
Programming to Combinational Problems", Aerospace Research Laboratories,
Report No. 65-116, Part 1 (June, 1965).
(7) Rechtschaffner, R.L., "Saturated Functions of 2n and 3n Factorial Designs",
Technometrics, 9, 569-576 (1967).
(8) Zubovic, P., "Geochemistry of Trace Elements in Coal", Symposium Proceedings
Environmental Aspects of Fuel Conversion Technology, EPA 600/2-76-149 (1976).
(9) Cavallaro, J.A., Johnston, M.T., and Deurbrouck, A.W., "Sulfur Reduction
Potential of U.S. Coals: A Revised Report of Investigations", EPA 600/2-
76-091, Bureau of Mines RI 8118, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina (April, 1976), 323 pp.
32
-------
APPENDIX A
SHORT LISTS OF CANDIDATE SAMPLING
SITES BY SITE CATEGORY
33
-------
TABLE A-l. CANDIDATE SITES FOR SITE CATEGORY 1 (1111) ^--PENNSYLVANIA
OJ
Cn
Company Name
Pennsylvania Electric
Company
Managed by:
Consolidation
Coal Co.
Owned by:
Hational Mines
Corporation
(subsidiary of
National Steel)
Duquesne Light Co.
Mine or Capacity,
Plant Name Town/County tpd
Homer City Cleaning Homer City/
Plant(c) Indiana
Interim Configu- 21,600
ration^1*)
Final Configure- 28,600
tion
Laurel Mine*f) Reel's Corner/ 3,300
Somerset
Warwick Mine Greensboro/ 18,000
Nos. 2 & 3 Greene
Seam
Name(s)
Upper
Freeport
(from 2
nines),
Trucked-
in coal
Ditto
Lower Kit-
tanning
Sewickley,
Pitts-
burgh
Pla?h
Type(b) Sulfur
. *
Sc-CY-CT-T-W 1.44-4.95 P
TD-Fil Avg 2.52 P(8>
Add. : HMC-Mag
HMC-F-T-CY- 2.47
CT-Sc-CD-
Fil-Mag-Br
HMC-CY-CT- 2.19 P
T-Fil-
Br-Sc-Mag
p(h)
(h)
Footnotes appear on following page.
-------
Footnotes for Table A-l:
(a) Represents low soil neutralization potential, high pyritic sulfur, high rainfall, and high technology.
(b) Abbreviations for cleaning equipment in Appendix A are as follows: A = air tables, Br = breakers,
CD = centrifugal driers, Cr = crusher, CT = centrifuges, CY = cyclones, D = driers, F - flotation
units, Fil = filters, H = heavy media washer, HMC = heavy media cyclones, HMV - heavy media vessel,
J = jigs, Mag = magnetic separators, PT = pick tables, Sc = screens, T = thickeners, TD » thermal
driers, W = washing tables.
(c) See the following articles for additional information:
McConnell, James F., and Statler, Charles W., "Multi-Stream Coal Cleaning Strategy for Control
of Sulfur", AIME Transactions, 260 (3), 237-241 (1976).
<*> Anonymous, "Multi-Stream Coal Cleaning System Promises Help With Sulfur Problem", Coal Age
(January, 1976).
(d) Equipment used corresponds to a Type F plant.
(e) Equipment used corresponds to a combination of Plant Types F and H.
(f) See the following article for additional information:
Mason, Richard H., "New Laurel Mine Features Refined Preparation Techniques", Coal Mining
and Processing (February, 1976).
(g) p = pyritic sulfur; determined from Battelle's analysis of coal from the Helen Mine, Helvetia Mine,
and trucked-in-coal sources.
(h) P = pyritic sulfur; determined from Cavallaro, J. A., Johnston, M. T., and Deurbrouck, A. W.,
"Sulfur Reduction Potential of U. S. Coals: A Revised Report of Investigations", EPA-600/2-76-091,
U. S. Burea of Mines RI 8118, U. S. Environmental Protection Agency, Research Triangle Park, North
Carolina (April, 1976), 323 pp.
-------
TABLE A-2. CANDIDATE SITES FOR SITE CATEGORY 2 (1110) w—Ohio
Company Name
North American
Coal Corporation
North American
Coal Corporation
Central Ohio Coal
Mine or
Plant Name
Powhatan No.
Mine
Powhatan No.
Mine
1
3
Muskingum Mine
Town/ County
Powhatan Point/
Belmont County
Powhatan Point/
Belmont County
Cumberland /Noble
Capacity, Seam
tpd Name(s)
5,000 Pittsburgh
(No. 8)
5,000 Pittsburgh
(No. 8)
11,000 Meigs Creek
Plant * v
Type Sulfur lc;, %
HMV(f)-Br-Sc-
Cr-Ct
HMV-Br-Sc-Ct
J-CT-T-Cr-Br-Sc-
2.88 P
2.56 P
4.7 T
Company
Southern Ohio Coal
Company
Southern Ohio Coal
Company
Meigs Mine No. 1 At hens /Meigs
18,850
Raccoon Mine Athens '/Vinton 7,000
No. 3
Peabody Coal Company Sunnyhill Mine New Lexington/Perry 6,000
(e)
(e)
(No. 9) Mag
Clarion H-CT-T-Fil-Cr-Br- 2.03 P
(No. 4-A) Sc-Mag
Clarion J-CY-CT-T-Fil-Cr- 2.03 P
(No. 4-A) Br-Sc-Mag
Middle Kit- J-Cr-Sc-D-PT 0.11-3.58 P
tanning Avg 1.84 P
(No. 6)
(a) Represents low soil neutralization potential, high pyritic sulfur, high rainfall, and low technology.
(b) For definition of abbreviations see Footnote (b), Table A-l.
(c) T « total sulfur; P = pyritic sulfur. Sulfur values determined from the following two sources:
1977 Keystone Coal Industry Manual, G. F. Nielsen (Ed.), McGraw-Hill, Inc., New York (1977), 1184 pp.
Cavallaro, J. A., Johnston, M. T., and Deubrouck, A. W., "Sulfur Reduction Potential of U.S. Coals:
A Revised Report of Investigations", EPA-600/2-76-091, U.S. Bureau of Mines RI 8118, U.S. Environ-
mental Protection Agency, Research Triangle Park, North Carolina (April, 1976), 323 pp.
(d) Located in general vicinity of E, M. Poston Power Plant, RFD 2, Athens, Ohio.
(e) Pyritic sulfur levels reported by North American Coal Corporation.
(f) The HMV at Powhatan No. 1 mine is a chance cone.
-------
TABLE A-3. CANDIDATE SITES FOR SITE CATEGORY 3 (1011)
00
Company Name
National Mines
Corporacion
(Sub: National
Steel Corp.)
Blue Diamond Mining
Island Creek Coal
Company
Virginia Pocahontas
Company
Westmoreland Coal
Company
Itmann Coal Company
Maple Meadow Mining
Company
Westmoreland Coal
Company
Southern Ohio Coal
Company
Mine or
Plant Name
Beaver Creek Div/d'
(Stinson Mines)
Leatherwood Mine 1
Virginia Pocahontas
No. 1 Mine
Virginia Pocahontas
No. 4 Mine*6'
Virginia
Pocahontas
No. 2 U>
Bullitt Mine(e.g)
Itmann Mine(h»O
Mine No. 1
Hampton Div. Mine
No. 3
Ferrel Mine
Martinka Mine
No. 1
Capacity,
Town/County tpd
Kentucky (Eastern)
Kris/Knott 10,000
Leatherwood/ 6,000
Perry
Virginia
Keen Mountain/ 8,000
Buchanan
Keen Mountain/ 8,000
Buchanan
Keen Mountain/ 8,000
Buchanan
Big Stone Gap/ 8,000
Wise
West Virginia
Itroann/Wyoming 12 ,000
Fairdale/Raleigh 6,000
Clothier/Boone 5,900
Clothier/Boone 8,000
Fairmont/Marion %8,000
Seam
Name(s)
Elkhorn
No. 3
Leather-
wood
•
Pocahontas
No. 3
Ditto
Ditto
Dorchester
Pocahontas
No. 3
Beck ley
Cedar
Grove
Ditto
Lower Kit-
tanning
Plant
Type<»>) Sulfur^), Z
H-F-CY-CT-T-
W-Fil-Cr-
Br-Sc-TD-
Mag
H-J-F-CT-T- 0
D-Cr-Br-
Fi 1-Mag-Sc
H-F-CT-W-D- 0
Fil-Sc
H-F-CT-W-D-
Cr-Sc
H-F-CT-W-D- 0
Fil-Cr-Sc-
Mag
H-F-CY-CT-T-
W-Sc-Mag
H-F-CY-CT-T- 0.
W-TD-Fil-
Cr-Br-Sc-CD
HMV-HMC-F-CT-
TD-Cr-Mag-T-Sc
H-F-CY-CT-T-W-
D-Cr-Br-Sc-
PT-Mag-Fil
H-F-CY-CT-T-
W-D-Cr-Br-
Sc-Mag-Fil
HMV-W-CY-Fil-
CT-T-TD-F^^
0.67 P
.6-0.9 T
.3-0.8 T
0.85 T
.3-0.8 T
0.25 P
4-1.0 T
0.8 T
0.15 P
0.145 P
2.46T
Footnotes appear on following page.
-------
Footnotes for Table A-3:
(a) Represents low soil neutralization potential, low pyritic sulfur, high rainfall, and high technology.
(b) For definitions of abbreviations see Footnote (b), Table A-l.
(c) See Footnote (c), Table A-2.
(d) For additional information, see the following article:
Sisti, D. C., "Sulfur Removal at Beaver Creek Consolidated Coal Company's Stinson Plant", Society of
Mining Engineers, AIME Trans., 258. 95-97 (1975).
(e) See Footnote (d), Table A-2.
(f) For additional information, see the following article:
Valentine, William E., "Preparation Plants for Buchanan County Coal", Mining Congress Journal, pp 53-55
and 58 (September, 1969).
(g) For additional information, see the following articles:
O'Brien, E. J., and Walker, J. L., Jr., "Environmental Layout and Process Innovations in the Bullitt
Plant", Mining Congress Journal, pp 37-42 (May, 1973).
Lowraan, Stephen G., "Westmoreland Coal's Bullitt Plant Upgrades Steam Coal Quality", Coal Age, M (3).
70-75 (March, 1973).
(h) For additional information, see the following articles:
Anonymous, "Consol Preparation ... Quality for All Markets", Coal Age, 6J2, (10), 158-170 (1964).
Anonymous, "Environmental Protection—A Consol Objective Since 1948", Coal Age, pp 122-126 (October, 1972)
(i) Data available from EPA (Office of Air Programs) study (unpublished).
(j) See flowsheet on pages 312-313 of Keystone Manual. [Reference is on Table A-2, Footnote (c).]
(k) Personal communication with American Electric Power, December, 1977.
-------
TABLE A-4. CANDIDATE SITES FOR SITE CATEGORY 4 (0111) '—KENTUCKY (WESTERN)
Company Name
The Pittsburgh &
Midway Coal
Mining Company
Peabody Coal Company
Mine or
Plant Name
Colonial Mine(d>
Ken, Ken 4, and
Ken 4-N Mines
Town/County
Madisonville/
Hopkins
Beaver Dan/
Ohio
Capacity,
tpd
10 ,000
10,000
Seam
Name(s)
Nos. 9
and 14
No. 9
-------
TABLE A-5. CANDIDATE SITES FOR SITE CATEGORY 5 (1010)***
Mine or
Company Name Plant Name
Town/County
Capacity, Pla?h'»
tpd Seam Name Type^b'
Sulfur (O, Z
Kentucky (Eastern)
Indian Head Mining Company Mine NOB. 4
and 5
Hazard/Perry 2,500
Pennsylvania
Greenwich Collieries
North and South Edensburg/Canbrla 13,000
Mines
West Virginia
Island Creek Coal Company Elk Creek No. Emmett/Logan
10
Hazard No. 7
Lower Freeport J-Sc-CY-CT
2,500 Cedar Grove
J-Sc
0.7-0.9 T
0.34 P
0.36 P
(a) Represents low soil neutralization potential, low pyritic sulfur, high rainfall, and low technology.
(b) For definition of abbreviations see Footnote (b), Table A-l.
(c) See Footnote (c). Table A-2.
(d) For additional information, see the following article: Anonymous. "Island Creek ... 'Precisioneered' Coal
Preparation". Coal Age, 10 (10), 130-141 (196S).
-------
TABLE A-6. CANDIDATE SITES FOR SITE CATEGORY 6 (0110)
to
Company Name
Peabody Coal Company
Island Creek Coal Company
Mine or
Plant Name
Mine No. 10
Fies Mine(d'e>
Hamilton Mine
No. l(
-------
TABLE A-7. CANDIDATE SITES FOR SITE CATEGORY 7 (0011)
(a)
*>
OJ
Company Name
Mine or
Plant Name
Town/County
Capacity,
tpd
Plant
Seam Name(s) Type
Sulfur , *
Alabama
Calvert & Marsh Coal
Company, Inc.
Mead Coal, Industrial
Products Croup, The
Mead Corporation
Jim Walter Resources, Inc.
Berry Mt.
Mines
Mulga Mine
Bessie
Mine''''
Blue Creek
No. 3
Blue Creek
No. 4
Oneonta/
Blount
Mulga/
Jefferson
Birmingham/
Jefferson
Adger/
Jefferson
Brookwood/
Tuscalooaa
1,000-1.600
""""
2,600
10,600
10,000
Rosa Sc-H-CY-W-Cr
Pratt Cr-Br-Sc-D-H-F-
CY-T-Fil
Mary Lee Cr-Br-Sc-Mag-HMV-
F-CY-W-Fil
Blue Creek Fil-Br-Sc-Mag-H-
F-CT-T-W
Blue Creek Fil-Br-Sc-Mag-H-
F-CT-T-W
0.65 P
1.4 f
0.13 P
0.8 T
0.72 P
National Mines Corporation Sugarloaf
(subsidiary of National Mine
Steel)
Arkansas
Ft. Smith/ %600
Sebastian
Hartshorne
Illinois
Old Ben No.
Old Ben Coal Company
Freeman United Coal Mining Orient
Co. (Dlv. Material Mine No. 3
•Service Corp.)
Sesser/
Franklin
Waltonville/ 14,000
Jefferson
800 tph No. 6
No. 6
Br-Sc-Mag-H-CY- 1.0 T
Ct-W
Br-Sc-HMC-Fll-TO- 0.8 P
J-F-CY-CD-A
H-F-J-Cr-Sc-D-Mag 0.87 P
(a) Represents high soil neutralization potential, low pyritic sulfur, high rainfall, and high technology.
(b) For definition of abbreviations see Footnote (b). Table A-l.
(c) See Footnote (c), Table A-2.
(d) For additional Information see the following article: Jackson, Daniel, Jr., "Mining and Preparing a
High-Reject Seam", Coal Age, _73 (6), 58-67 (1968).
(e) Data available from Effluent Guidelines Survey (unpublished).
(f) Plant flowsheet available from EPA (Office of Air Programs) study (unpublished).
(g) For additional Information, see the following article: Anonymous. "New Coal for Midwest Steel", Coal Age,
65 (7), 100-105 (1960).
-------
TABLE A-8. CANDIDATE SITES FOR SITE CATEGORY 8 (0010)
Company Name
Mine or
Plant Name
Town/County
Capacity,
tpd
Seam
Name(s)
Plant
TypeCb)
Sulfur
-------
TABLE A-9. CANDIDATE SITES FOR SITE CATEGORY 9 (0001)***
Company Name
Mine or
Plane Name
Town/County
Capacity, Seam
tpd Name(a)
Plant
Type(b) Sulfur*"). X
Kaiser Steel
Corporation
York Canyon
Hine(e)
New Mexico
Raton
-------
TABLE A-10. CANDIDATE SITES FOR SITE CATEGORY 10 (0000)
Company Name
The Imperial
Coal Company
Bract ah
Plateau Mining
Company
Mine or
Plant Name
Eagle Mine
Bract ah Mine
NOB. 3, 5
Star Point Mine
Nos. 1 and 2
Capacity,
Town/County tpd
Colorado
Erie/Weld %1,000
Utah
Helper/Carbon %2,500
Price/Carbon 5,200
Seam
Name(s)
Laramie
No. 3
Castle Gate
D, Subaeam
No. 3
Hiawatha,
Wattia
Plant
Type(b) Sulfur, X
H-CT-Mag-Cr- 0.06 P
Sc-PT
H-Cr-Sc-Mag 0.53 T
H-CT-CT-Fii- 0.16 P
Cr-Sc
(a) Represents high soil neutralization potential, low pyritic sulfur, low rainfall, and low technology.
(b) For definition of abbreviations see Footnote (b), Table A-l.
(c) See Footnote (c), Table A-2.
-------
APPENDIX B
COAL PREPARATION PLANT
INFORMATION FORM
47
-------
TABLE B-l. COAL PREPARATION PLANT INFORMATION FORM
i.
WO
a.
i.
4.
Xu
flani Baa*» Clt*
Mai
Yaar
County
Lllof Addraaat Coatacti TtUphona Ho.
Straac
Plant Coda No.
City Stata Up Coda
Claanad Coil Tonnaui
IM
op.
klin CaeacltTi
PIAJIT pUIGg. AKD OPEIATUK DATA
iraelu Capacity. (TfD) (TTY)
S
0
u
It
c
c
n
I
K
I
C
A
L
C
b
A
It
A
C
T
E
II
If
H
V
S
1
C
A
L
C
N
A
K
A
C
T
E
*
Kln»(a)
Dlatrlet
SiaaN)
RiRlon
::*«k
T>flM
:•,,!. tur.
Vulatlla Mattar
Aah
rix.d Carbon
Total Sulfur
leu
Pyrttie Sulfur*
orgunlc Sulfur
SuUuca Suifur
other Crucify)
*Thl> tnfomatlon la
T«i> Sl.u
Batten Siia
fr.o Sw.lllni lmt*x
Crlndablltty Iruf«»
(Htrdjcovi)
Spucltle Gravity
fUO COAL(S)
vary lapercant to th* protraal
normal Oparation Schadulti
Hr/day
Oava/wk
Wka/yr
CIZAN COAL(S)
\ /
\ /
\s
/\
/ \
/ \
\ /
\ /
\/
/\
/ \
/ \
REFUSE(S)
k /
\ /
\/
/\
/ \
/ \
\ /
\ /
\/
/\
/ \
/ \i
-------
PLANT EQUIPMENT INFORMATION
Name of Company
Plant Code No.
U1
o
EQUIPMENT
6. Breaking and Crushing
_ Sizing, Screening,
Blending, Splitting
TYPE
DESCRIPTION/MANUFACTURER
NUMBER
FEED COAL
SIZE
TONNAGE
HANDLED
-------
PLANT EQUIPMENT INFORMATION
Name of Company: Plant Code No.
EQUIPMENT
8. Wet Washing
-
9 . Dewatering
TYPE
DESCRIPTION/MANUFACTURER
NUMBER
FEED COAL
SIZE
TONNAGE
HANDLED
0
-------
PLANT EQUIPMENT INFORMATION
ro
Name of Company: Plant Code No.
k
EQUIPMENT
10. Dry Concentrators
11. Driers
TYPE
DESCRIPTION/MANUFACTURER
f
NUMBER
FEED COAL
SIZE
TONNAGE
HANDLED
-------
12. Details on Thermal Driers 'if present)
(a) Describe the type of fuel used in dmier furnace. If coal is used
indicate the source.
Furnace operation:
(d)
(e)
Type Fuel
Type of
Firing
Fuel Rate,
Ib/hr
% Sulfur
in Fuel
Caloric Value
Btu/lb
(b) Range of Normal Operating Temperatures in System:
Combustion zone: to °F (in furnace)
Fluid bed: to °F (above bed)
to °F (below bed)
(c) Dry Precleaners:
No. of Cyclones/Drier
Manufacturer and Type
Wet Scrubbers:
No. of Scrubbers/Dr;ler
Manufacturer & Type
Exhaust System Data:
Design basis, emissions from dry collector;
[SCF = ft3 dry air at 70°F, 1 atm pressure]
Design basis, emissions from scrubber outlet;
(f) Exhaust System:
Gr/SCF
Gr/SCF
Exhaust Locations
No.
(g) Sampling Ports: Indicate the location of all sampling ports on the
thermal dr.ier and its associated pollution control devices on an
attached diagram. Attach the results of previous monitoring efforts
on the thermal drier.
53
-------
13. Facility Diagrams:
(a) Supply schematic representations of the preparation plant and
ancillary facilities (e.g., coal conveyors, coal storage areas,
refuse disposal areas, settling pond, wastewater treatment facili-
ties, etc.)
(b) Supply sketch or schematic details of any thermal dryer system,
including the air pollution control equipment.
14. Refuse Disposal Area:
(a) Describe the type and size of refuse generated from cleaning opera-
tions, and indicate whether the sizes are segregated into different
piles at the disposal area.
(b)
(c)
Describe the method of wa.ste disposal and type of disposal area
(e.g., dump, sanitary landfill, incineration, open burning,
salvaged, etc.)
Indicate the number of years refuse has been dumped at the disposal
area. If more than one dumping area has been used in the past,
indicate the age of each separate refuse pile
Age of 1st pile
yrs. Age of 2nd pile
yrs.
(d) Indicate the name of the refuse dump operator.
(e) If soil analyses have been made at the refuse dump area, indicate
the average soil pH (i.e., the neutralization potential of the soil)
If no data is available on soil pH from the plant, obtain this in-
formation from the county Soil Conservation Service.
Soil pH
(average)
15* Coal Industries in the Area:
List all coal mining, processing, or burning industries within a 5-mile
radius of the coal cleaning plant and estimate the distance of each
industry from the plant.
Coal Industries in Area
Distance From
Cleaning Plant
54
-------
16. Availability of Baseline Data:
Do you presently, or have you in the past, monitored air or water quality
in the vicinity of the coal cleaning plant or associated facilities (e.g.,
coal storage pile, refuse disposal area, etc.) Q yes Q no
If the above answer is yes, please indicate the location of your sampling
sites and the parameters measured.
Sampling Locations
Parameters Measured
Please attach example copies of the results of several sampling runs or
copies of any summaries of the air and/or water quality monitoring
analyses.
Have any environmental assessments (e.g., EIS or NPDES) ever been made
of the coal cleaning plant or its associated facilities. Q yes Qno
If the answer above is yes, please attach copies of these assessments.
55
-------
TECHNICAL REPORT DATA
(Please reed Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/7-79-073d
2.
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Environmental Assessment of Coal Cleaning Pro-
cesses : Selection of Test Sites for Source Test
Program
5. REPORT DATE
July 1979
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
D.A.Tolle, R.E.Thomas, R.K.Markarian, and
V.Q.Hale
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING OROANIZATION NAME AND ADDRESS
Battelle-Columbus
505 King Avenue
Columbus, Ohio 43201
10. PROGRAM ELEMENT NO.
EHE623A
11. CONTRACT /GRANT NO.
68-02-2163, Task 421
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 277U
13. TYPE OF REPORT AND PERIOD COVERED
Task Final; 12/77 - 12/78
14. SPONSORING AGENCY CODE
EPA/600/13
15. SUPPLEMENTARY NOTES TjERL-RTP project officer is James D. Kilgroe, Mail Drop 61,
919/541-2851.
i6. ABSTRACT
rep0rt describes the selection of coal cleaning plants at which field
testing and sampling will be conducted to support the environmental assessment of
the pollution potential of various coal cleaning processes . The approach was to
select a few plants (representing extremes in variables considered important in
evaluating the pollution potential of these types of plants) from which data may be
obtained. Site selection involved: (1) classifying plants into a number of site catego-
ries; and (2) gathering additional detailed information on specific plants and applying
secondary constraints to a single suitable site for sampling in each site category
identified in step (1). Step (1) assumes that all sites are equal within each category;
in step (2), however, the best or most representative site within a category is to be
selected. Classifying plants into various site categories is based on four criteria
(variables): the acid neutralization potential of the soil surrounding the plant, the
pyritic sulfur content of run-of-mine coal, the average annual precipitation, and the
process technology. An initial sorting of the more than 400 known coal cleaning
plants produced lists of plants which correspond to each of the 10 site categories. Im-
posing constraints reduced the number of potential test sites to 45. Site visits , to ac-
Suire additional information, are necessary before the final sampling sites can be
alerted _ ; _ " _
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
COSATI Field/Group
Pollution
Coal Preparation
Test Facilities
Sampling
Soil Chemistry
Sulfur
Pyrite
Pollution Control
Stationary Sources
Coal Cleaning
SoilpH
Pyritic Sulfur
13B
081
14B
08G,02A
07B
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report/
Unclassified
21. NO. OF PAGES
67
20. SECURITY CLASS (Thit page)
Unclassified
22. PRICE
EPA Form 2220-1 (»-73)
56
------- |