DoE
EPA
United States
Department of Energy
Division of Solid Fuel
Mining and Preparation
Pittsburgh PA 15213
FE-11270-1
U.S. Environmental Protection Agency
Office of Research and Development
Industrial Environmental Research
Laboratory
Research Triangle Park NC 27711
EPA-600/7-79-007
January 1979
Management of Coal
Preparation Fine Wastes
without Disposal Ponds
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.
Price: Paper Copy $5.25
Microfiche $3.00
-------
EPA-600/7-79-007
January 1979
Management of Coal Preparation
Fine Wastes without Disposal Ponds
by
D.C. Hoffman, R.W. Briggs, and S.R. Michalski
Dravo Corporation
3600 Neville Island
Pittsburgh, Pennsylvania 15225
EPA/DoE Interagency Agreement No. DXE685AK
Program Element No. EHE623A
EPA Project Officer: D.A. Kirchgessner DoE Project Officer: Richard P. Killmeyer, Jr.
Industrial Environmental Research Laboratory Division of Solid Fuel Mining and Preparation
Research Triangle Park, NC 27711 Pittsburgh, PA 15213
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
and
U.S. DEPARTMENT OF ENERGY
Division of Solid Fuel Mining and Preparation
Pittsburgh, PA 15213
-------
Management of Coal Preparation Fine Wastes
Without Disposal Ponds
Final Report
July 1, 1977 to June 15, 1978
by
David C. Hoffman
Dravo Lime Company
Pittsburgh, Pa.
and
Robert W. Briggs
Stanley R. Michalski
Dravo Corporation
Pittsburgh, Pa.
Contract Number ET-79-11270
(Formerly USBM Contract Number J0177050)
This report represents work on a program that
was originated by the Interior Department's Bureau
of Mines and was transferred to the Department of
Energy on October 1, 1977.
Department of Energy
Branch of Procurement, Washington
2401 E Street, N.W.
Washington, D.C. 20241
-------
TABLE OF CONTENTS
Page
ABSTRACT 1
SUMMARY 2
CONCLUSIONS 8
ACKNOWLEDGMENT 10
DISCUSSION
Background 11
Sample Selection 13
Sample Collection 14
Laboratory Program 14
Physical/Chemical Data 15
Engineering Data 17
Stabilization Data 19
Data Correlation 22
Patents, Inventions, and Disclosures 24
BIBLIOGRAPHY 25
SUMMARY TABLES
Table I - Sample Preparation Plant Summary. ... 26
Table II - Sample Preparation Plant Disposal
Summary 27
Table III - Laboratory Testing Program Outline. . 28
Table IV - Chemical Analyses of Solids From Fine
Coal Refuse Samples 29
Table V - Chemical Analyses of Supernatant Liquid
From Fine Coal Refuse Samples 30
Table VI - Particle Size Analysis 31
Table VII - Settling Test Summary 32
Table VIII - Filter Leaf Test Summary 33
-------
TABLE OF CONTENTS (cont'd)
Page
Table IX - Rotary Vacuum Filtration Summary .... 34
Table X - Summary of Physical Properties - Untreated
Fine Coal Refuse 35
Table XI - Summary of Physical Properties -
Untreated Fine Coal Refuse 36
Table XII - Summary of Laboratory Strength Para-
meters - Consolidation Tests - Untreated
Fine Coal Refuse 37
Table XIII - Summary of Laboratory Strength Para-
meters - Direct Shear Tests - Untreated
Fine Coal Refuse 38
Table XIV - Summary of 40-Day Unconfined Compression
Strength. . . . ". . . . 3?
Table XV - Stabilization Rate, Days to Reach
4.5 TSF " 40
Table XVI - Stabilization Rate, Days to Reach
2.0 TSF 41
Table XVII-A - Particle Size Analysis 42
Table XVII-B - Rank - Grouping of Samples from Test
Results 43
APPENDIX (Test Method Description)
Particle Size Analysis i
Settling Tests ii
Leaf Filter Tests iii
Atterberg Limits and Indices iv
Specific Gravity Test v
Compaction . . vi
Permeability vii
•Direct Shear Test viii
Consolidation Test ix
Unconfined Compression Tests xi
Rate of Stabilization ..... xii
-------
ABSTRACT
In the beneficiation of coal, the common medium for
upgrading the mined coal is water. The end products of coal
beneficiation are generally three materials; (1) the clean
coal, (2) coarsely sized waste product, and (3) a finely
sized waste product. As the need for cleaner coal increases,
the amount of waste products generated also increases and the
importance of safe, environmentally acceptable disposal of
these wastes becomes more important.
Generally, the coarse refuse (plus 28 mesh) can be disposed
of safely because it is a solid with inherent engineering proper-
ties that can be utilized in engineered structures. However,
the fine refuse (minus 28 mesh) generally exists as a water-
based slurry devoid of significant engineering properties that
can be employed in engineering disposal methods.
The objective of this study is to ascertain some physical/
chemical properties of a diverse sampling of fine coal refuse
and to investigate the effect of chemical stabilization. Samples
of fine coal refuse were obtained from some of the major coal
seams in the Eastern bituminous coal fields. Two common and
one proprietary chemicals were employed in testing chemical
stabilization on nine samples of fine coal refuse. This study
indicated that chemical stabilization can be employed to dras-
tically improve some physical/chemical properties of the fine
refuse for better handling and disposal. The proprietary
chemical, Calcilox® additive, is the most effective over the
range of samples tested-. Portland Type I cement is also effec-
tive but greatly influenced by waste solids concentration. The
final chemical tested, lime, is inferior to the other two
chemicals and generally ineffective in improving the fine waste's
physical/chemical properties.
® - Registered trademark of Dravo Corporation.
- 1 -
-------
SUMMARY
With the increasing importance of bituminous coal as a
domestic energy source and/or chemical feedstock, the amount
of coal produced between now and 1985 is expected to dramatically
increase. In 1975, 648 million tons of coal were mined, and it
is projected that by 1985, 1.2 billion tons will be needed. In
1975, approximately 374 million tons of run-of-mine coal were
mechanically cleaned resulting in an estimated 15 million tons
of fine coal refuse (minus 28 mesh) and nearly 95 million tons
of coarse coal refuse (plus 28 mesh). This cleaning was accom-
plished in 388 cleaning plants which are primarily located east
of the Mississippi River. Nearly 80% of the cleaning plants are
in the states of West Virginia, Kentucky, Pennsylvania, and
Virginia; listed in decreasing number of plants per state.
The geographic location of these plants significantly
influences the economic, environmentally acceptable, and safe
disposal of the refuse due to the physical/chemical properties
of the refuse. The coarse refuse fraction has been studied and
disposed of quite safely over the years and sound engineering
can handle this waste. The fines fraction has also been studied
to some extent, and it has been generally concluded that this
material does not possess any definite engineering properties
that can be utilized by conventional civil engineering and con-
struction methods. The fine refuse consists of very fine
particles of coal, rock, and clayish substances in a water
slurry ranging in dry solids content from approximately 15% to
as high as 80% solids. These fines are very difficult to dewater
to higher solids concentration (>65% solid) and even then, they
still exhibit fluid-like properties. Even if the fluid proper-
ties can be temporarily overcome by high initial solids (-~ 75-85%
solids) for immediate handling, the material will readily reslurry
and the fluid properties will reappear. These slurries can
remain physically and chemically unstable for years. Thus,
topography that ranges from steeply mountainous to very hilly
is not conducive to above ground disposal of fine coal refuse.
Until recently, the commonly accepted method of fines dis-
posal was either a permanent or temporary settling pond or a
large impoundment of coarse refuse containing the fluid fines.
During the past 10 to 20 years, great emphasis has been placed
on various dewatering devices (vacuum disc filters, centrifuges,
and pressure filters) to decrease the moisture content of the
fine refuse. The primary objective has been to increase its
immediate handling properties for landfill-type disposal of the
fines with the coarse fines. These efforts have generally over-
looked the actual engineering stability of the fines and have
relied heavily on the overall stability of the combined coarse
and fines mixture. As it has been repeatedly shown, this small
amount of fines refuse can easily upset the whole disposal area.
- 2 -
-------
Since 1970, Dravo Lime Company has been engaged in research
into the chemical stabilization of inorganic industrial waste
slurries. As a result of this work, Dravo has developed a
material, Calcilox additive, that has been successfully applied
to the stabilization of flue gas desulfurization sludges in the
electric utility industry. Preliminary research has indicated
that Calcilox alone or in combination with other known stabi-
lizing materials can stabilize fine coal refuse as produced by
today's mechanical coal cleaning plants. Therefore, the objec-
tive of this study was to stabilize a wide sampling of fine coal
refuse with three stabilizing agents - Calcilox additive, lime,
and Portland Type I cement. Various laboratory tests were per-
formed to determine if these agents can impart definitive
engineering properties to these fine refuse solids.
Since each preparation plant waste disposal method might
be considered unique and chemical stabilization has not been
proven on a full scale basis, the scope of the study was to
investigate the possibility that chemical stabilization may have
universal applicability to all fine coal refuse streams. In
addition, the geographic area of concern would be the states
that have the most need for more viable alternatives for fines
disposal; mainly, West Virginia, Pennsylvania, Virginia, Kentucky,
and the mid-Western states of Ohio, Indiana, and Illinois. Thus,
sample selection was from these geographic areas with selection
parameters being coal seam, coal application, mining type, plant
circuitry, and fines handling. Important constraints in the
program were ease of plant participation, ease of sampling, and
budgetary considerations. As a result of these limitations,
nine plants were involved in the study; four from Pennsylvania,
two from West Virginia, and one each from Virginia, Indiana and
Illinois. The capacity of these plants represents nearly 10
million tons of clean coal which accounts for approximately
three percent of the 1975 clean coal production. The seams
mined represent some of the major steam and metallurgical seams
in the Eastern bituminous coal fields. Samples included both
strip and deep mines employing conventional, continuous, and
longwall mining methods. The preparation circuitry ranged from
simple jig plants to multi-circuit cleaning plants. Fines dis-
posal methods included unthickened plant discharge into abandoned
strip pits to vacuum filter cake disposal with the coarse refuse.
The laboratory objective was three-fold: (1) Determine
some engineering properties of the untreated fines; (2) Evaluate
the effective chemical stabilization of the fines with the use
of three substances (lime, Portland Type I cement, and Calcilox
additive); and (3) Determine the engineering properties of the
stabilized samples. Properties of the untreated fines that were
determined were:
1. Solids chemical analyses
2. Supernatant chemical analyses
3. Particle size analyses
4. Settling rate and settled solids concentation
- 3 -
-------
5. Atterberg limits
6. Solids specific gravity
7. Proctor densities
8. Permeabilities of settled slurry and filter ca' d
9. Direct shear properties, settled slurry and f~- Iter
cake
10. Consolidation properties
11. Vacuum filtration data
12. Unconfined compression data
Stabilization data were determined with the chemical addi-
tives at three dosage levels (5%, 10%, and 15%, on a dry solids
basis), and at three solids levels (25%, 35%, and filter leaf
test solids). Stabilization data were evaluated as a time rate
gain of penetration resistance (tons per square foot) for a
90-day stabilization period. More accurate stabilization data
were obtained by performing unconfined compressive tests on
specimens stabilized for forty days. For this study, improved
unconfined compressive strengths were regarded as the desired
result from chemical stabilization of the untreated fines. The
following is a brief description of the Summary Tables:
1. Solids Chemical Analyses - Table IV
Constituents are typical of fine coal refuse with ash
contents ranging from 24% to 59% and correspond with
calorific values ranging from 11,680 to 5,294 Btu/lb.,
respectively. Major constituents of ash are SiO^ ,
Al-O , and Fe-O.,. Table IV illustrates that consider-
able Btu are being wasted in fine refuse.
2. Supernatant Chemical Analyses - Table V
Major constituents are sulfate, calcium, magnesium,
and sodium in the liquid portion of the fine refuse
slurries. With the exception of one sample, the pH
values are in the pH 7-8 range.
3. Particle Size Analyses - Table VI
The samples exhibit a wide range of particle sizes.
However, five out of the nine samples have a large
weight portion below 20 microns.
4. Settling Test - Table VII
It appears that all the samples will settle to greater
than 40% solids in approximately two days. Review of
the settling curves illustrates that all the samples will
be positively affected by a thickener. That is the
solids from a preparation plant can be increased within
the retention time of most modern-day thickeners (8 to
12 hours). Also the various chemical additives will
affect the settled solids concentration.
- 4 -
-------
5. Vacuum Filtration Data - Tables VIII and IX
The filter leaf test summary (Table VIII) indicates
that all the slurries starting at 35% solids can be
filtered to 60-70% solids. Existing full scale
filters confirm this solids level to be realistically
achievable. However, it should be noted that there
are solid losses in the filtrate which will eventually
affect a full scale filter operation by lowering the
percent filter cake solids and/or yield. Utilizing
a pilot rotary drum filter (Table IX), higher filter
cake solids were obtained. It should be realized
that the starting solids were substantially higher
than typical of thickener underflow feeds to vacuum
filters. Continuous operation of a drum filter would
yield lower solids and yields due to cloth blinding.
6. Proctor Determination - Table X
Illustrating the problem of dewatering fines is the
comparison of optimum moisture content and that
achieved by filtering. It should be noted that the
filter cake solids are abnormally high because the
starting material is settled slurry. Real situations
will not achieve these high values. The Proctor
determined optimum moisture content is less, in all
cases, than can be achieved by conventional vacuum
filtration.
7. Permeabilities - Table XI
Permeabilities of the settled solids are quite low,
10~5 to 10~6 cm/sec., and are further reduced by
vacuum filtration.
8. Consolidation Data - Table XII
The low values for the compression index, C , imply
that the fine refuse is incompressible. In addition,
the recompression index, Ccr, is extremely low
revealing that there is no rebound in the fines after
the compression stress is removed or relaxed. The
water content is significant after testing and may be
a major contribution to the low values of compression
and recompression index. Another factor may be the
solid coal or waste particles in the fines.
9. Direct Shear - Table XIII
Primarily, the instability of fine refuse is due to
the lack of any cohesive strength in the untreated
slurries. This is shown in both the settled slurry
and filter cake. In no case does the cohesive strength
achieve 1 pound per square inch (psi).
- 5 -
-------
10. Unconfined Compressive Strength - Table XIV
The untreated fines were allowed to stand for 40
days after being mixed at 25%, 35%, and filter
cake solids consistencies. Temperature effects
were evaluated by placing duplicate samples at
two different temperatures; 35°F and 72OF. Only
tW.Q .f ilter cake specimens; attained any measurable
strength; 1105 and 1112. These strengths were
very low (4 to 6 psij and are probably attributed
to compaction strength attained while placing the
semi-rigid filter cake into the specimen mold.
The overall objective of this study was to determine if fine
coal- refuse can be chemically stabilized. The study utilized
three ^chemical additives and these were applied to three solids
concentrations; 25%, 35%, and filter cake levels. These solid
levels were chosen because Dravo's experience has shown that 25%
is the minimal solids level for economic viability of a chemical
stabilization approach. The 25% to 35% solids level would be
representative of a thickener underflow solids range, while the
higher filter cake values would be typical of disc filter and
centrifuge cakes.
Based on study limitations and Dravo's expertise, a 40-day
stabilization period was chosen as the maximum, curing time before
testing. The parameter chosen for evaluating stabilization was
unconfined compressive strength. Table XIV summarizes the data
pbtained. Briefly it illustrates:
1. Thickener underflow solids will not stabilize with
the additive dosages of 5% to 15%, by dry solids
weight, using either hydrated lime or Portland
Type I cement.
2. The addition of Calcilox to thickener underflows
will improve their unconfined compressive strength.
In 1 of 9 samples, the 5% addition did not produce
any strength improvement. At the 10% Calcilox level,
all solids levels and all samples showed strength
improvement.
3. The addition of 5% to 15% lime to filter cakes does
produce some strengths.
4. The addition of 5% to 15% Portland Type I cement to
filter cakes produces significant strength improve-
ments.
5. The addition of 5% to 15% Calcilox to filter cakes
produces significant strength improvements.
- 6 -
-------
6. With the use of Portland Type I cement and Calcilox
for stabilization, the effect of temperature is not
clearly illustrated. At the 15% Calcilox dosage
level to thickener underflows, increased temperature
does not always affect the stabilization rate. This
tends to imply that stabilization is not 100% chemical
but a combination of chemical and physical effects.
Correlation between various physical/chemical parameters and
stabilization results was attempted with mixed results. Manually
analyzing the various parameters obtained, a ranking of the various
samples was shown, Table XVII-B. Expansion of this approach via
a computer statistical package was performed attempting to corre-
late various parameters with unconfined compressive strength.
This was not successful and attributed to errors within the data.
- 7 -
-------
CONCLUSIONS
Based on this sampling of fine coal refuse, the following
conclusions can be made:
1. Major constituents of fine coal refuse are coal,
Si02, and A1203.
2."•. The.re is significant BTU value in the fine coal
refuse.
3. The pH of the fine coal refuse is generally
within a pH 7-8 range unless it contains highly
oxidized pyritic sulfur.
4. Liquid phases of fine refuse slurries do contain
appreciable dissolved solid species.
5. Particle size distributions are varied and can
contain appreciable quantities of fine particles
less than 325 mesh.
6. Fine coal slurries can be settled to higher solids
with the use of conventional thickeners.
7. Conventional dewatering devices (vacuum filters and
centrifuges) will not achieve the optimum fines
Proctor density. Typical field methods of compaction
for fine refuse (dozer treads) will also not achieve
this value.
8. Fine coal refuse settled solids and filter cakes
exhibit low permeabilities in the order of 10~5 to
10"' cm/sec.
9. Fine coal refuse settled solids and filter cakes do
not possess significant cohesive strengths, /~1 psi.
10. Fine coal refuse slurries may be considered incom-
pressible and exhibit virtually no rebound after
the compressive load is relaxed.
11. Calcilox additive will generally improve the unconfined
compressive strength of fine coal refuse materials
with greater than 25% solids concentration.
12. Overall, Calcilox additive stabilization is superior
to lime and Portland Type I cement in stabilizing
fine coal refuse.
- 8 -
-------
13. Lime and Portland Type I cement are only effective
on fine coal refuse materials of filter cake solids
consistency.
14. In comparing Portland Type I cement with Calcilox
on filter cake applications, Calcilox appears to
gain more strength. The rate of strength gain
appears to be comparable.
15. Temperature affects chemical stabilization, that is,
the warmer, the faster the rate of stabilization.
16. Technically, chemical stabilization will improve the
handling and disposal of fine coal refuse.
17. Statistical correlation of the data was not possible
although correlations were evident.
18. Additional data points are needed to gather statis-
tically significant information on chemical stabiliza-
tion.
- 9 -
-------
ACKNOWLEDGMENT
Dravo Lime Company and the authors take this opportunity
to thank the coal companies for their participation. For if
it were not for their help, this study would not have been
possible. Secondly, we wish to thank Mr. A. Deurbrouck of
the Department of Energy in Pittsburgh for his guidance,
cooperation, and follow-up in this study. Finally, the
authors wish to thank the numerous people involved at Dravo's
Research Center for their expedient and high quality work in
generating the data contained in this report. Without their
help, the project could not have been completed.
- 10 -
-------
DISCUSSION
Background
As previously stated in the Summary section, the United
States has an energy goal of producing 1.2 billion tons of
bituminous coal by 1985. This coal v/ill probably come from an
expansion of the existing Eastern bituminous coal fields and
from the new Western fields. The major consumers of bituminous
coal are the steel industry and the steam electric generating
industry. These two consumers are primarily located east of
the Mississippi River. The demand for coking coal is princi-
pally filled from the Eastern fields and the utility coal for
this area is still being supplied by the Eastern market. These
two industries are using two types of coal, metallurgical and
steam coal, that are different in their physical/chemical proper-
ties yet they have some things in common. As environmental and
economic pressures continue, beneficiation becomes more important
and more complex in order to supply these "spec" coals. Generally,
they both receive some degree of beneficiation even though it
may only be sizing.
Three solid product streams are generated when coal is
cleaned (beneficiated):
1. Clean Coal
2. Coarse Refuse (plus 28 mesh in size)
3. Fine Refuse (.minus 28 mesh in size)
The handling of the clean coal does not present any unmanageable
problems. However, the handling and disposal of the coarse and
fine refuse can and does present problems that can be so severe
as to cause loss of life, property, and production. The major
refuse problem is the fluidity of the fine refuse that can cause
it to flow like water. Increasing the problems with coarse and
fine coal refuse handling and disposal is the use of modern-day
high production mining methods such as continuous miners and
longwall mining. Overall, the handling and disposal problems can
be attributed to the mineralogy of the solids and their size
consist. Nearly 80% of the cleaning plants are located in the
states of West Virginia, Kentucky, Pennsylvania, and Virginia.
The topography of the coal sections in these states is steeply
hilly to mountainous and containment of a fluid mass such as
fine coal refuse is not easily done in an economic, environmentally
acceptable, and safe manner.
The extent of knowledge on the disposal aspects of the fines
fraction is quite limited, whereas t'^e coarse refuse has been
studied and definitive engineering properties on the general nature
of the material are available. Primarily, the fine refuse behaves
as a thixotropic substance that is a function of its percent solids
and method of handling. Because of this thixotropic behavior,
- 11 -
-------
the commonly accepted method of fines disposal is either a
permanent or temporary settling pond or a large impoundment
of coarse refuse in which the fines have been mixed. Currently,
these disposal methods may not be the most economic or environ-
mentally acceptable.
In recent years great emphasis has been placed on various
dewatering devices (vacuum disc filters, centrifuges, and
pressure filters) to increase the solids content of the fines
refuse. By doing so, the immediate handling properties are
improved but the actual stability of the fine coal refuse is
still unknown. Reportedly, fines stability can be achieved if
a solids threshold level for eliminating thixotropic behavior
is exceeded. For each refuse, this level is different but solids
values ranging from 85% to 92% have been reported as the stable
solids level. Currently, the only physical way of attaining
stability is by pressure filtration and/or combination with the
dry coarse refuse to achieve an overall refuse moisture content
in the acceptable range. Unfortunately, this method is quite
expensive and rarely achieves the desired stability. Several
methods of chemical stabilization of the fines have been inves-
tigated but have been unacceptable based on technical results,
economic value, or a combination of each. Overcoming the
economic barriers to chemical stabilization can be achieved with
commodity substances such as lime or Portland Type I cement.
However, technical problems still remain with the use of these
agents.
Dravo Lime Company has developed a chemical stabilizing
agent, Calcilox additive, and has successfully demonstrated its
effectiveness on flue gas desulfurization waste slurries. Pre-
liminary research has indicated that Calcilox additive alone or
in combination with other known stabilizing materials can
stabilize fine coal refuse as produced by today's mechanical
coal cleaning plants. Therefore, the objective of this study
was to treat fine coal refuse samples with lime, Portland Type I
cement, and Calcilox additive to determine if they would sta-
bilize. The test parameter for quantifying stabilization was
unconfined compressive strength. To measure the effects of these
chemical additives, unconfined compressive tests were also run
on the untreated fines. In addition, various other physical/
chemical tests were performed on the untreated material to gain
more insight into the engineering properties of the fine coal
refuse. An in-depth study was not undertaken since the wide-
ranging applicability of chemical stabilization of fine coal
refuse is yet to be proven. The area to be explored was the
universality of chemical stabilization to improve the handle-
ability and disposal of fine coal refuse.
- 12 -
-------
Sample Selection
The scope of the study might be considered a "shot gun"
approach in that we attempted to obtain samples that represent
major coal seams, various preparation circuitry, steam and
metallurgical applications, strip and deep mines, geographic
location, and various fine refuse disposal modes. Timing and
monetary constraints in the study dictated major reliance on
Dravo's experience within the coal industry for plant selection.
The plant qualifications that were initially predetermined were:
1. Plant must be in the Eastern bituminous coal field
and in a major coal producing area.
7.. Plant must be processing a major coal seam.
3. If possible, the plant should only be processing
one seam.
4. The final sample mix should contain a wide variety
of preparation circuitry.
5. The final sample mix should contain various
disposal modes.
6. The preparation plant should be a fixed plant of
reasonable size—greater than 200-300 tons per
hour (TPH) run-of-mine capacity.
Based on these initial qualifications, it was decided to
obtain samples from Central Pennsylvania, Western Pennsylvania,
Southern West Virginia—Virginia area, Alabama, and the Mid-
western coal field.
Initial screening of the plants in these localities was aided
by the 1976 Keystone Coal Industry Manual. Using the manual and
our experience, an initial list of twenty plants was formulated.
Letters of introduction and scope definition were sent to the
major operating executives of these twenty plants requesting an
initial site visit. Initial response was quite favorable with
only two plants refusing to allow an initial visit. The main
reason was their feeling that any participation, although actual
names would not be used by Dravo in reporting the data, would lead
to further Federal regulations and intervention into their par-
ticular problems. Twelve site visits were undertaken in July and
August, 1977. From these twelve plants, ten were selected for
plant sampling and study participation with the understanding of
complete anonymity in reporting the data. In return for the
plants' participation, their own results will be supplied to them
at the end of the study. The plants selected were:
1. Three from Western Pennsylvania
2. One from Central Pennsylvania
3. Three from West Virginia—Virginia region
4. Two from the Midwest
5. One from the Southeast
- 13 -
-------
Due to a prolonged wildcat strike that extended well beyond
our sampling phase, the plant from the Southeast was not sampled.
Table I, Sample Preparation Plant Summary, tabulates the actual
plant dataTThese mine plants produce an estimated 10,400,000
tons per year (TPY) clean coal. Compared with the 1975 statistics
for clean coal production of 374 million tons, this sample
represents 3% of the production in 1975. During the past few
years, clean coal production has not changed very much; therefore,
a 3% sample estimate is probably valid for 1977-78 production
figures.
Sample Collection
Samples were taken from the preparation plant discharges
prior to entering ponds, impoundments, or mechanical dewatering
devices. The sample was composited over a reasonable length of
time (typically four hours) during normal plant operations.
Table II, Sample Preparation Plant Disposal Summary, briefly
describes the fine coal refuse disposal at each location. As
expected, mines that have strip operations are utilizing them
for fines dispsoal, especially where land area is favorable and
abundant as in the Midwest. In addition, those plants with ample
land area do not employ thickeners but rely on large, permanent
settling ponds located in the stripped-out areas. Plants 1104,
1113, and 1114 are the plants pumping directly to abandoned strip
pits.
The only way that we have of surmising whether or not the
sample is representative of their normal operations is from
conversations with the plant personnel while sampling and our
experience in analyzing the data after the sample is taken.
Based on sampling conditions or our experience, the following
plant samples might be unrepresentative:
1106 - Sample is a grab sample due to sampling constraints
1111 - Sample is a grab sample due to preparation plant
upsets. Solids level, 8.8%, is very low for a
thickener underflow.
Laboratory Program
In conjunction with determining the effects of chemical
stabilization, an alternate objective was to gather physical/
chemical information on the untreated fine refuse. A goodly
portion of the laboratory program was devoted to this task.
Table III, Laboratory Testing Program Outline, lists the testing
program that was employed.The Index Properties (excluding
certain settling tests) Engineering Properties, and Filtration
Tests dealt with the untreated fines as settled solids on filter
cakes. The stabilization tests were of prime importance in
evaluating the effects of the chemical additives: lime, Portland
Type I cement, and Calcilox additive. Stabilization tests were
performed to evaluate treating thickener underflows at normal
underflow concentrations (25% to 35% solids, dry weight basis)
- 14 -
-------
and higher solids concentrations similar to that which is pro-
duced in vacuum disc filters and centrifuges. Since we were
concerned with a chemical reaction, temperature effects were
also considered. A brief description of the pertinent test
methods used is in the appendix.
Physical/Chemical Data
Table IV, Chemical Analyses of Solids From Fine Coal Refuse
Samples, lists various parameters of the fine refuse solids. In
the refuse solids, ash, fixed carbon, BTU/lb., percent solids,
and pH are important. The ash, fixed carbon, and BTU/lb. are
all related to the amount of coal that is in the waste. Dravo
has found that well run preparation circuits will discharge waste
with 40-60% ash and will possess a heat value of 5000 to 7000
BTU/lb. Four out of nine samples fell into this category: 1104,
1111, 1113 and 1114. Three of the nine samples, 1105, 1106, and
1112, contained a large amount of coal, greater than 10,000
BTU/lb. Plant 1105 personnel realize they discharge coal and
make an effort to reclaim the pond settled solids. Elimination
of the coal discharge can only be achieved by preparation circuit
modifications. On the contrary, 1106 is operated by the same
company and is a relatively new plant. Sampling and temporary
plant problems may have resulted in this non-representative,
high BTU sample. No explanation is readily apparent for the coal
losses at 1112. In general, significant BTU/lb. can be found in
the fine coal refuse and it may be worth recovering.
The pH values are within reason except for 1104 and this is
explained by their mining method. They are stripping pillars from
an old underground mine and the coal must be highly oxidized
resulting in acid mine drainage. In fact, the prep plant's major
problem is equipment corrosion. Experience indicates that plants
without thickeners will produce 10 to 15 percent solids discharge
whereas thickeners do concentrate the solids more. The exception
is 1111 with an 8.8 percent solids discharge. This is probably
due to the plant conditions and sampling.
Ash analysis of the refuse solids reveals several interesting
points. First, the ratio of Si07 to Al_0,. is farily consistent
between 2.0 to 2.5. Secondly, all the samples contain significant
portions of iron oxide. It appears that all the plants are
removing this from the coal in their attempt at reducing the sul-
fur content by removing pyritic sulfur. One interesting note is
sample 1112 and its high concentration of calcium oxide and mag-
nesium oxide. At these levels, the sample pH is not out of line
with the other samples. Thus, these substances may not be free,
but combined in a compound.
In conjunction with the solids analyses, the refuse liquid
was analyzed and the results appear in Table V, Chemical Analyses
of Supernatant Liquid from Fine Coal Refuse Samples. The major
constituent in the supernatant is the totalsulfur, reported as
sulfate. Other important parameters are chloride, calcium,
- 15 -
-------
sodium, magnesium, and hardness. These species are relatively
soluble and probably reflect the geologic conditions at the time
of the deposition of the coal-forming vegetation and subsequent
sedimentation of the overburden.
One of the more important physical parameters of the fine
refuse is the particle size consist and its ability to settle
to higher solids concentrations. Table VI, Particle Size Analyses,
tabulates the sieve analyses of the fine refuse solids as
samples. All of the samples have greater than 40%, by weight,
of particles finer than 100 mesh. More importantly, five of the
nine samples (1104, 1110, 1111, 1113, and 1114) have 50% or greater
minus 10 microns. In conjunction with these results, Table VII,
Settling Test Summary, can be coupled to obtain an insight into
the ability of these materials to be concentrated. This is
important in thickener design, pond or impoundment design, and
overall disposal considerations. Settling curves are used exten-
sively in designing thickeners and most thickeners have up to
12 hours operating retention time. Using this as a guideline,
review of the individual untreated settling curves in the appen-
dix reveals that thickeners operating at a 5 percent inflow
concentration will effectively concentrate the solids. Thus,
properly designed, installed, and operated thickeners can help
with the disposal problem by reducing the disposal volumes. For
example, a comparison of the settled solids from the sampled
materials and the 5 percent solids settling test for those plants
having existing thickeners is:
Settled Solids, % Solids
Plant As Received % 5% Initial Mixed Solids
1105 30.5 46.6
1106 20.9 50.5
1107 18.9 62.3
1112 30.0 59.9
As mentioned in the sampling section, 1106 may not be represen-
tative. Dravo's experience has shown that thickeners are con-
sistently operated below the solids levels that they are capable
of producing due to fears of pumping high solids. With today's
slurry pumps, this is not a problem and higher solids thickener
underflow operation should be investigated by the coal operator
to help reduce overall fines disposal volume. This is clearly
shown in the above table by noting the much higher level of
solids achieved in the lab settling test. Realistically, 30 to
45% thickener underflow solids can be achieved and pumped.
As a further indication of solids concentration for reduc-
tion of disposal volume, a laboratory simulation of vacuum
filtration was performed. Table VIII, Filter Leaf Test Summary,
tabulates the data from this test. An important point to note
is the prerequisite of a thickener prior to a vacuum filter in
order to assure that the solids are as high as possible prior to
- 16 -
-------
filtration. We employed a 35% initial solids concentration as
a reasonable approximation of a thickener underflow. Using this
starting point, all the samples filtered in the range of 60% to
70% solids, which is a realistic solids concentrate. Two other
results are important: 1) dry cake yield and 2) filtrate solids.
The filtrate solids indicate the build-up of fines in the
preparation plant circuitry which eventually must be controlled
so as not to choke the plant. The dry cake yield illustrates
the variation in filter size that must be considered when designing
a vacuum filter installation.
Because our laboratory program demanded large amounts of
high solids cake, we employed a pilot rotary drum vacuum filter
to generate the quantities needed. Table IX, Rotary Vacuum
Filtration Summary, contains the results of the test. It should
be noted that the filter feed was high solids settled slurry.
Consequently, the filter cake solids were higher than those
obtained from the filter leaf test. However, a very interesting
point is that the dry cake yield is very close to the yield
calculated in the filter leaf test. The values ranged from
5.5 Ib/sq.ft./hour to 18.8 Ib./sq.ft./hour. As can be seen from
these two tests, off-the-shelf installation of filtration equip-
ment without some degree of prior testing can greatly affect
solids handling and disposal.
Engineering Data
Thus far, the discussion has dealt with the more common
physical/chemical properties of the fines. The remainder of the
discussion will center on the engineering properties of the fine
coal refuse. Properties investigated and reported are:
1. Atterberg limits
2. Proctor density
3. Permeability
4. Direct shear
a. Cohesion
b. Angle of internal friction
5. Consolidation
6. Unconfined compressive strength
a. Untreated
b. Stabilized
Tables X and XI, Summary of Physical Properties Untreated Fine
Coal Refuse, contain the information on Atterberg limits,specific
gravity, effective grain size, settled solids, Proctor determina-
tion, permeability, direct shear, and consolidation.
Atterberg limit testing (indicative percent moisture levels)
is generally considered qualitative and not completely indicative
of a fine coal refuse since the particle size distribution employed
is not the complete size consist. One level, the liquid limit,
might be useful information in that it may indicate a critical
- 17 -
-------
moisture content where the fluid-like properties of the fine
refuse will predominate. Plastic limit is too qualitative to
be of any real value.
The specific gravity values and size cut values are very
interesting. The variation in specific gravity between plus 200
mesh and minus 200 mesh is important. Particle size analyses
indicated that the dividing line between fine refuse and
ultra-fine refuse is in the 100 to 200 mesh region; below 200
mesh being considered ultra-fine. Because the specific gravities
are lower in the coarser fraction, this indicates that the coal
is probably in the larger fraction and that modern cleaning
techniques employed on this small size consist would probably
yield a sizable coal fraction. Proctor density is another impor-
tant value for consideration in the disposal of fines. By taking
the dried fines, incrementally adding a known amount of water,
and compacting this mixture in a specific volume with a constant
amount of force, one obtains a curve of dry bulk density versus
moisture content. The curve is similar to a parabola with the
apex being the maximum bulk density obtainable with a minimum
amount of moisture. Realistically, this value means that a
cubic foot of refuse will contain the maximum amount of solids
possible thus minimizing disposal volumes. .Table X contains
this Proctor data. However, this does not say anything about
stability. Commonly employed dewatering devices, vacuum filters
and centrifuges do not achieve these Proctor values either.
Table XI contains some very important data; permeability and
cohesive strength of untreated fines. The permeabilities of all
the settled solids are low, 10 to 10 b cm/sec. The filter cake
permeabilities are slightly lower than those of the settled solids
cind rightly so because of the energy used to increase the solids
content; the lower permeabilities are due to less void space.
Since strength, i.e., stability is of prime concern, the measure
of cohesive strength is important. The data collected indicates
that the untreated fines, whether settled slurry or filter cake,
have virtually no strength. Values reported are all less than
1 psi cohesive strength.
Table XII, Summary of Laboratory Strength^Parameters -
Consolidation Test's - Untreated^ Fine Coal Refuse, contains an
important point. That is, the fines are not compressible. The
compression index, C, is a ratio of the void ratio and applied
pressure. High values, greater than .5, are indicative of
materials that compress whereas small numbers, .10 to .20, are
related to noncompressible materials. Secondly, the low recom-
pression index, C , indicates little or virtually no rebound of
the slurry when tne load is released. A possible relationship
may exist between the compression index, Btu/lb. value, and size
consist. Another important fact, the low cohesive strengths of
the untreated fines, is shown in Table XIII, Summary of Laboratory
Strength Parameters - Direct Shear Tests - Untreated Fine Coal
Refuse.This is true whether the fines are a settled solids or
a highly dewatered filter cake. In both cases, the fines have
- 18 -
-------
less than 1 psi cohesive strength. Recalling that the fines are
noncompressible, one sees that large impoundments of fine refuse
are only as safe as the impoundment itself. If released, the
material will not hold together and will flow when stresses are
applied. Flow movement can be caused by forces as small as
gravitational forces.
Stabilization Data
We have shown that untreated fine coal refuse lacks stability
and strength. The objective of the study is to improve the strength
with the addition of Portland Type I cement, hydrated lime, or
Calcilox additive. Stabilization is now defined as the gain of
unconfined compressive strength. Dravo Lime Company possesses
expertise in this technology and has developed a method of adding
stabilization chemicals to the slurry as a percentage of dry
slurry solids. To be effective, economically and technically, we
have shown that the slurry solids concentration must be greater
than 25%. Solids levels between 25%-35% would be representative
of thickener underflow discharges, whereas filter cake solids
would be indicative of mechanically dewatered slurries. Depending
on disposal mode (ponds, impoundments, or high solids cake)
chemical additive dosages fall within the range of 5% to 15%, by
dry weight of the slurry solids. The rate of strength gain is
monitored over time and plots of unconfined penetration resis-
tances (tons per square foot) versus time (days) are obtained and
can be used to determine optimum dosages for various strength
developments. The test method is described in the appendix with
the accompanying plots.
More accurate information is obtained at selected stabiliza-
tion times by performing unconfined compressive tests on the test
specimens. This test gives an accurate estimate of the strength
of the material since it has no lateral support to increase the
normal resistance. A relationship exists between unconfined
compressive strength (shear strength) and cohesive strength. Shear
strength is one-half of the unconfined strength, and shear strength
is related to cohesive strength as determined in the direct shear
test. Table XIV, Summary of 40-Day Unconfined Compression Strength,
tabulates the pertinent data.Forty days was arbitrarily chosen
to meet the timing constraints of the study and was an appropri-
ately long time to produce measurable strengths. Since we are
dealing with chemical reactions, temperature should have an effect.
The summer time temperature selected was 72 F, while 55 F was
selected as a yearly average temperature. A lower temperature,
39 F, was tested in rate monitoring as indicative of severe weather
conditions. Unconfined compression data at 39°F or below were not
obtained since results would only reflect freezing and not chemi-
cal stabilization.
- 19-
-------
Reviewing Table XIV, several specific results are apparent:
1. With the exception of three points, the untreated
underflow and filter cake do not possess any
measurable strength.
2. 5% Portland Type I cement addition improves filter
cake strengths.
3. 5% lime is effective on filter cake solids.
4. 5% Calcilox A is effective on slurries and filter
cakes, but slurries may be dependent upon slurry
solids.
5. 5% Caleilox B is not as effective as Calcilox A.
6. 10% Portland Type I cement is effective on filter
cakes.
7. 10% lime is only effective on filter cakes.
8. 10% Calcilox A and B are effective on slurries and
filter cakes.
9. 15% Portland Type I cement is slightly effective on
slurries.
10. 15% lime is not effective on slurries but may be
effective on filter cakes.
11. 15% Calcilox A and B are effective on slurries and
high solids cakes.
NOTE: Calcilox A and B vary only in the percent CaO.
Some general conclusions are:
1. Chemical stabilization of fine coal refuse is effective,
2. Calcilox is superior to lime and Portland Type I
cement for slurries and filter cakes.
3. Calcilox is effective on 25% to 35% slurries and
filter cake materials.
4. Portland Type I cement is not effective on 25% to •
35% solid slurries, but effective on filter cake
materials.
5. Lime is not effective on 25% to 35% solid slurries
but effective on filter cake materials.
6. Solids level is an important consideration for
chemical stabilization.
- 20 -
-------
7. Temperature has an effect on stabilization; the
higher the temperature, the more rapid the strength
development.
It should also be emphasized that these unconfined compression
data are only singular points. No duplication of test specimens
occurred and certain inconsistencies are present in the data
points although the method was employed correctly. For example,
one would expect 'higher strengths at 72°F than at 55°F. Generally,
this is true, but some data points do not show this. Also, higher
solids should produce high strengths. Again, some points appear
to show the reverse.
The unconfined compression data are very valuable; but other,
less precise information has also been gathered on the role of
stabilization. This is presented in Tables XV and XVI, Stabili-
zation Rate, where a 90-day stabilization period was monitored
by a hand-held penetrometer (Model CL-700, Soiltest, Inc.).* In
Table XV, the days to reach 2 TSF are presented. This penetration
resistance (2 TSF or 27.8 psi) is ample for handling by normal con-
struction means. Table XVI presents data on days to reach 4.5 TSF.
This value may be excessive and lesser strengths are adequate
for handling. Since the Portland Type I cement and Calcilox
reactions are cementitious in nature, it may be advisable to
handle a fine refuse at the minimum handleable strength and place
it in its final disposal site. At this site, the chemical reac-
tion will continue but at a slower rate. Final strengths are
nearly identical to those attained from non-disturbed samples.
Another data point that has been included in these tables is a
reading at 39°F and illustrates the rate difference between 39°F
and 55°F or 72°F. The difference between 55°F and 72°F is not as
promounced with this method as is shown in the unconfined results
due to the accuracy of the unconfined test method.
These rate measurements illustrate results similar to those
of the unconfined tests. Generally, they are:
1. Temperature affects the rate of stabilization, the
higher the temperature, the faster the reaction.
2. Overall, Calcilox is superior to lime and Portland Type I
cement in slurries and filter cake applications.
3. Lime additions appear to give better results than
Calcilox or Portland Type I cement in certain instances.
Comparison with unconfined strengths should be made
for realistic data.
4. Solids level affects chemical stabilization. The
higher the solids, the faster the stabilization.
*The monitoring method conducted with a hand-held penetrometer on
small samples is somewhat inaccurate. That is, the sidewalls of
the container exert a positive influence on the result. Secondly,
the error in the penetrometer itself is roughly ± 0.5 TSF.
- 21 -
-------
Data Correlation
Because of the scope of the study, the test results were,
of necessity, obtained from single sample testing. This con-
dition precludes an in-depth analysis of the test results.
However, there appears to be a definite relationship in the
physical properties of the various coal refuse samples. This
relationship can be expressed by ranking the coal refuse samples
in an order of high to low, or low to high, depending upon the
results from the various tests and how one physical property
relates to another.
When the graphs of the particle size analysis were combined,
they delineated a uniform broad band (Table XVII-A) covering about
50% of the finer, by-weight, scale. This combined graph showed
that there were distinctive differences in particle size from
sample to sample.
The particle size of a particulate solid is a determining
factor in practically all physical property tests that were
conducted. Large particles increase permeability, settling,
cohesion and filterability and decrease compaction, liquid limits,
and optimum water contents.
As a first ranking, the refuse samples were ranked according
to the percent retained on a 200 mesh sieve from low to high.
The 200 mesh division was chosen because specific gravities were
determined on the material as a whole and on the -200 mesh
material. Calculations to determine the specific gravities of
the +200 mesh material showed them to be in the coal range
indicating the +200 mesh material was mostly coal.
From the results of eight other tests, the samples were
ranked from low to high or high to low depending upon the effect
the particle size would have on the physical properties, and
these rankings were compared with the particle size ranking.
The samples were then re-ranked according to their average
ranking for all of the tests and separated into three groups
with three samples in a group (Table XVII-B). Over 65% of the
test results agreed with this group ranking.
With the use of a single test value, there are obviously
going to be some discrepancies and various preparation plant
processes, such as mixing with oil in the froth flotation process,
would also have an adverse effect. Sample 1107 in particular,
although ranked sixth in the average ranking, ranked high and low
in the individual tests, thus causing a lower percentage of agree-
ment. This non-conformity may have been due to the difficulty
in sampling from an almost inaccessible sampling point.
- 22 -
-------
Since there appeared to be some obvious rankings or
correlation, a further look at some of the data was undertaken
with the aid of a time share computer package called "Procap"
offered by CompuServe, Columbus, Ohio. This program performs
single step regression analyses on the variables and ranks them
in order of their importance in defining the dependent variable.
The dependent variable that was chosen as most important was
40-day unconfined strength.
In the first run utilizing Procap, certain fine refuse solids
physical/chemical characteristics were analyzed with the 40-day
unconfined strength of a 35% slurry treated with 15% Calcilox A
and cured at 72°F. Independent variables were chosen as volatile
matter, ash, percent carbon, BTU/lb., 50% weight diameter, particle
size, settling rate per day, and overall solids specific gravity.
After numerous runs with various combinations and eliminations of
variables, it was concluded that settling rate per day and specific
gravity had the greatest effect on stabilization strength. Since
the 40-day strengths at an initial 35% solids are low, a comparison
of 40-day strengths on filter cakes was made. The variables tested
were ash, volatile matter, carbon, BTU/lb., percent solids, 50%
weight diameter particle size, and solids specific gravity. The
dependent variable was again 40-day strength at filter cake solids
with 10% Calcilox A addition and cured at 72°F. Much to our
chagrin, no significant correlations were obtained with the filter
cake.
Because the computer program is capable of accepting many
independent variables, more physical/chemical data were inputed with
the independent variable still being the 35% solids mixture with
10% Calcilox A additive. Twelve independent variables were tested
resulting in less of a correlation than the first run with only
seven independent variables. New independent variables were
selected and tested against 40-day strength of a filter cake with
10% Calcilox additive. Again, no significant correlation was evident,
Intuitively, these results are not acceptable because experience
indicates that there are some correlations. It is felt that un-
confined compressive strengths may be in error since they are
singular points without any duplication. In reviewing the unconfined
data, there are obvious unexplained results although the test method
is correct. The sample 1107, the 10% lime addition to filter cake
is erroneous. The 55°F unconfined compression value is 290.9 psi
while the 72°F was only 15.2 psi. Comparing the other 10% lime to
filter cake values for the other samples, one sees that the 290.9
psi point is way out of range. Other questionable points probably
exist and if used in a sensitive statistical analysis, would produce
invalid results. It is suspected that this is the case in our com-
puter analyses since the ranking system produced reasonable results.
More accurate data points are needed to develop statistically valid
correlations.
- 23 -
-------
Patents, Inventions, and Disclosures
Based on prior patents and the state-of-the-art, we do
not believe that this .study produced any new information or
ideas of a proprietary or patentable nature.
- 24 -
-------
BIBLIOGRAPHY
Cooper, D. K., "Choosing Closed Circuits for Coal Preparation
Plants,", Proceedings of Second Kentucky Coal Refuse Dispos a 1
and Utilization Seminar, IMMR21-PD15-76 , pages 75-82 , •"'
University of Kentucky, Lexington, Kentucky; September 1976.
"1976 Keystone Coal Industry Manual", Mining Information Services,
New York, New York; 1976
"Minerals Yearbook - 1975", Volume 1, Metals, Minerals, and Fuels,
Goal - .Bituminous and Lignite, Table 31, United States
Governmerit Printing Office, 1978.
"Procap Process Capability Preliminary User's Manual - CS-376",
CompuServe, Columbus, Ohio; 1976.
- 25 -
-------
TABLE I
SAMPLE PREPARATION PLANT SUMMARY
Plant
Des ig-
nat ion
1 1 O't
1105
1 106
1 107
1110
1 1 11
1 1 12
1113
1 1 lit
State
Pa.
Pa.
Pa.
Pa.
W.Va.
W.Va.
Va.
Ind.
111.
County
Al legheny
Al legheny
Somerset
Greene
Fayette
Logan
Buchanan
Warrick
Will! amson
Seam
Pi ttsburgh
Upper
Freeport
Lower
Ki ttanni ng B
Pi ttsburgh
Coa 1 burg
Eagle
No. 2 Gas
Powel 1 ton
Five Block
Winifred
Chi 1 ton
Cedar Grove
Pocahontas
#3
Indiana V
1 1 1 i noi s 6
Mine
Type
Strip
Deep
Deep
Deep
Deep
and
Strip
Deep
Deep
Strip
Strip
Coal
Appl ica-
t ion
Steam
Steam
Metal lur-
gical
Metal lur-
gi cal
Metal lur-
gical
and
Steam
Metal lur-
gical
Metal lur-
g ical
Steam
Steam
Prep
Plant
Capaci ty
(TPH)
350
*tOO
350
950
600
800
625
1 ,800
900
C lean
Coal
(TPY)
600,000
800,000
600,000
1 ,250,000
800,000
1 ,000,000
1 ,250,000
3, 'tOO, 000
700,000
Prep Plant Operating Equipment
in
HI
0)
i_
O
l/>
/
/
/
/
/
/
'
l_
0)
_c
l/l
3
U
(_J
'
/
/
/
/
J
7
in
Oi
/
'
7
i_
4-» QJ
1/1 —
._ _Q
0> 0)
Q |-
/
/
/
/
/
01
C 1^
.-
OJ
C
o
-
<_J
/
/
/
/
/
rf)
w
CD
01
•^
C
'
/
/
/
/
/
y
i_
0 O
.— *~>
4-1 (U
-------
TABLE II
SAMPLE PREPARATION PLANT DISPOSAL SUMMARY
Plant Designation
1104
1105
1106
1107
1110
1111
1112
1113
1114
Disposal Summary
No thickeners ;are employed, plant
discharge is pumped to two large abandoned
strip pits:. A single point discharge
fiil-s one pit while overflow cascades
into adjoining pit which serves as final
settling pond. The p:H of final overflow
is acidic and is adjusted prior to reuse
in prep plant circuit.
The thickener underflow is pumped to one
of three large settling ponds. A single
point discharge fills one pond while.
others are being excavated. The super-
natant is discharged and the settled
solids excavated and saved for sale or
reprocessing. •
The thickener underflow is disposed of
either into two small settling ponds or
injected back into the mine for placement
and disposal. Because of the handling
problems associated with the settling
ponds and, above ground disposal, mine
disposal system is preferred.
The thickener underflow is pumped to a
large,- permanent coarse refuse impoundment.
The supernatant overflows into a clari-
fication pond for discharge.
The thickener underflow is filtered on
vacuum disc filters with transport of
the filter cake on the coarse reject belt.
The combined refuse is compacted with a
dozer.
The thickener underflow is pumped to a
series of permanent settling ponds that
are filled through a single point discharge.
The waste remains in these ponds and the
supernatant is discharged to receiving
streams.
The thickener underflow is vacuum disc
filtered and the resultant cake is
combined with the coarse on an aerial
tram. The trammed material is spread
and compacted with a dozer at the disposal
site.
The plant discharge is pumped to abandoned
strip pits and.the supernatant cascades one
into another for recycle back into the prep
plant. The settled solids are covered with
overburden.
The plant discharge flows by gravity into
abandoned strip pits and the supernatant
cascades one into the other for reuse back
into the prep plant. The /settled solids
are covered with overburden.
- 27 -
-------
TABLE III
LABORATORY TESTING PROGRAM OUTLINE
to
CO
I. Index Properties
A. Wet Screen Analysis
(to 10 Microns)
B. Settling Test
Starting at:
1) 5% Solids
2) 35? Solids
3) 35% Solids +
10% Calcilox A
4) 35% Solids +
107. Ca(OH)2
5) 35% Solids +
10% 'Portland Cement
Type I
C. -Liquid and; Plastic Limit
D. Specific Gravity
1) Entire Sample
2) -200 Mesh Material
E. Chemical Analyses
1) Refuse Solids
2) Supernatant
II. Engineering Properties
A. Permeability
1) Settled Slurry
2) Filter Cake
B. Consolidation
1) Settled Slurry
C. Penetration
1) Compacted Filter Cake
2) Compacted Filter Cake
(Lower Water Content)
3) Settled Slurry
D. Direct Shear @ 1 psi
1) Settled Slurry
2) Filter Cake
III. Filtration
&. Filter Leaf
B. Rotary Drum
IV. Stabilization Tests
Dosages
A. Untreated
B. 5% Portland Type
5% Ca(OH)2
52 Calcilox A
5% Calcilox B
C. 10% Portland Type
10% Ca(OH)2
10% Calcilox A
10% Calcilox B
D. 15% Portland Type
15% Ca(OH)2
15% Calcilox A
15% Calcilox B
Solids Level
25%
*x
I -
-
-
—
I *x
*x
*x
-
I *x
*x
*x
.
35%
*x
*x
*x
*x
*x
*x
*x
*x
*x
*x
*x
*x
*x
Filter Cake
*x
*x
*x
*x
—
*x
*x
*x
—
_
-
-
-
x - Mixed and monitored by penetrometer at
39°F, 55°F and 72°F for 90 days.
* - 40-Day unconflned tests performed at 55°F
and 72°F
-------
TABLE IV
Analysis of Refuse
AaV,
Volatile Matter . • .
Fixed Carbon ....
Sulfur, Total ....
Pyritic Sulfur . . .
Sulfate Sulfur . . •
Organic Sulfur . . .
RTU/Lb
Percent Solids. . . .
pH
Analysis of Ash
S109
re2u3
r2u5
JfoO
CHEMICAL ANALYSES OF SOLIDS FROM FINE COAL REFUSE SAMPLES1
(Analyses Reported in Weight % except BTU and pH)
Sample Number
1104
0.76
58.86
17.82
22.56
1.53
1.00
0.36
0.17
5294
10.6
4.4
60.79
24.41
1.26
6.81
1.12
1.07
0.38
0.88
0.11
2.97
1105
0.15
29.36
27.67
42.82
2.65
2.27
0.09
0.29
10416
30.5
7.7
45.88
20.93
0.96
12.85
8.50
0.81
0.45
6.07
0.14
2.26
1106
0.22
23.59
17.32
58.87
1.62
0.96
0.21
0.47
11680
20.9
8.1
50.68
25.12
1.60
12.36
3.57
0.72
0.29
2.36
0.49
1.88
1107
0.58
38.58
24.06
36.78
2.29
1.64
0.07
0.58
8640
18.9
7.9
51.14
21.07
1.10
12.65
5.95
1.23
0.42
2.20
0.18
2.47
1110
0.85
35.82
21.63
41.70
0.61
0.20
0.00
0.41
9078
24.2
7.9
58.05
26.65
1.23
5.55
1.52
1.41
0.42
0.84
0.01
4.00
1111
0.60
44.00
20.06
35.34
0.65
0.40
0.04
0.21
7814
8.8
8.0
58.35
24.85
0.99
6.05
1.82
1.74
0.47
1.18
0.02
4.40
1112
0.35
25.12
19.10
55.43
0.75
0.50
0.00
0.25
10844
30.0
7.8
39.50
18.62
0.81
6.96
17.08
9.45
0.39
4.61
0.01
2.51
1113
1.12
47.32
20.79
30.77
2.81
1.90
0.23
0.68
6415
14.9
8.0
57.15
24.28
0.88
8.36
2.95
1.44
0.45
1.30
0.13
3.02
1114
1.08
45.83
19.81
33.28
1.98
1.19
0.18
0.61
6901
15.2
7.7
63.02
17.56
0.81
5.94
5.98
1.43
0.47
2.06
0.05
2.56
As Received
-------
TABLE V
CHEMICAL ANALYSES OF SUPERNATANT LIQUID FROM FINE COAL REFUSE SAMPLES
(Analyses Reported in mg/L except pH and Specific Conductance)
Analysis
SAMPLE NUMBER
1104
1105
1106
1107
1110
1111
1112
1113
1114
o
I
pH 4.4
Alkalinity 0
Sp. Conductance ( u mhos\ .
cm ^b /u
Filterable Residue (TDS) . 4540
Total S (as SO^) 3010
Chloride 12
Total Iron 0.2
Calcium 650
Sodium 215
Potassium 8
Magnesium 250
Manganese 30
Total Hardness (as CaCO,) . 2660
7.7
120
3750
5330
2680
670
<0.1
470
965
11
90
2.6
1540
8.1
68
2670
3810
2530
13
<0.1
560
347
6
100
0.6
1800
7.9
114
1680
2280
1370
66
<0.1
300
290
5
29
0.5
8610
7.9
134
680
590
280
39
<0.1
100
80
6
11
0.4
3000
8.0
173
1480
1580
920
49
<0.1
120
300
10
27
0.3
10010
7.8
113
1520
1360
180
610
<0,1
140
290
4
25
<0.1
4000
8.0
103
1220
1620
980
53
< 0.1
140
211
7
56
1
6010
7.7
46
1970
3440
2260
18
<0.1
440
225
9
172
14
18020
As Received
-------
TABLE VI
PARTICLE SIZE ANALYSIS
PERCENT RETAINED
(WEIGHT %)
Sample
No.
1104
1105
1106
1107
1110
1111
1112
1113
1114
U.S. Standard Sieve No.
16
2.0
0.3
1.8
4.9
0.1
0.0
1.4
0.2
0.6
30
5.0
2.8
8.9
17.0
1.3
1.9
13.2
1.5
2.6
50
7.7
16.5
20.2
21.1
3.8
6.7
25.7
8.0
7.8
100
8.3
19.5
18.0
13.7
6.8
10.1
17.8
11.6
11.7
200
8.5
16.8
15.4
9.4
13.2
14.3
13.0
10.5
13.8
325
6.0
6.8
6.2
3.9
8.5
9.5
7.9
6.2
11.0
400
2.8
29.8
16.5
19.6
5.0
3.5
3.2
2.6
4.7
Microns
20
9.0
1.5
2.7
1.4
9.8
8.3
4.0
6.5
7.3
10
18.6
3.1
4.8
3.3
22.5
18.5
8.5
17.1
18.3
-10
32,1
2.9
5.6
5.7
29.0
27.2
5.3
35.8
22.2
I
u>
As Received
-------
TABLE VII
SETTLING TEST SUMMARY
SAMPLE NO.
1104
1105
1106
1107
1110
1111
1112
1113
1114
5% Solids
No Additive
Time
Days
1.1
0.2
0.1
0.6
2.0
2.0
0.9
2.3
1.3
Solids
Final %
44.4
46.6
50.5
62.3
45.1
51.7
59.9
40.3
35.5
35% Solids
No Additive
Time
Days
5.1
0.9
0.5
2.0
6.1
4.9
0.8
3.2
6.0
Solids
Final %
55.9
59.8
59.5
61.4
52.0
55.0
62.2
42.3
47.4
35% Solids
10% Ca(OH)2
Time
Days
2.7
0.8
2.9
0.7
2.3
3.0
0.9
3.2
2.7
Solids
Final %
39.8
59.2
50.5
39.9
42.3
42.6
61.8
43.5
41.9
35% Solids
10% Calcilon A
Tine
Days
1.2
0.2
0.5
0.3
3.0
1.0
0.3
0.9
0.9
Solids
Final %
41.1
57.6
53.4
42.0
42.1
43.4
61.3
38.6
39.1
35% Solido
10% Portland Coeaont
Ticc
Days
0.1
0.8
0.2
1.8
0.3
0.2
0.6
0.2
0.9
Solido
Final %
37.7
43.1
47.7
49.3
37.9
38.6
54.1
37.6
37.7
Ul
ro
-------
TABLE VIII
FILTER LEAF TEST SUMMARY
Slurry Solids. .
Filter Medium. .
Form Vacuum . .
Dry Vacuum . .
Dry Air Velocity
. 35%
. Eimco Polypropylene 853F
. 24 Inches of Mercury
. 24 Inches of Mercury
.0.5 Cubic Feet per Minute
1
co
LO
1
Sample
No.
1104
1104
1105
1105
1106
1106
1107
1107
1110
1110
1111
1111
1112
1112
1113
1113
1114
1114
Form Time
(Sec.)
15
45
15
45
15
45
15
45
15
45
15
45
15
45
15
45
15
45
Dry Time
(Sec.)
30
90
30
90
30
90
30
90
30
90
30
90
30
90
30
90
30
90
Dry Cake Yield
(Lb./Sq.Ft./Hour)
9.9
5.1
12.2
10.5
7.5
5.5
5.0
3.4
11.0
8.5
15.2
6.7
14.4
9.2
7.5
4.3
28.0
18.7
Filter Cake Solids
(Z)
67.3
68.6
63.9
67.6
65.5
67.8
69.1
75.0
62.4
64.2
64.6
65.0
69.2
70.5
64.8
65.1
66.9
69.9
Filtrate Yield
(Gal. /Sq. Ft. /Hour)
3.8
2.3
5.2
4.6
2.8
3.8
2.5
2.3
1.1
2.0
2.3
2.4
8.5
6.0
1.7
1.1
2.3
2.0
Filtrate Solids
(% )
.30
.47
.49
.48
.54
.49
.49
.44
.38
.27
.38
.40
.29
.26
.13
.38
.38
.43
-------
TABLE IX
ROTARY VACUUM FILTRATION SUMMARY
Filter Medium:. .
Filter Area • . .
Form Vacuum • • •
Dry Vacuum . . .
Dry Air Velocity .
. EiBco Polypropylene 853F
.4.7 Square Feet
. 20 Inches of Mercury
7 Inches of Mercury
. 4 Cubic Feet per Minute
Sample
No.
1104
1105
1106
1107
1110
1111
1112
1113
1114
Slurry Solids
(2)
51.6
59.9
62.4
65.0
49.4
40.8
56.7
43.3
43.8
Drum Speed...
(M.P.R.) u;
6
5
5
5
8
6
4
8
3
Form Time
(Sec.)
135
113
113
113
180
135
90
180
68
Dry Time
(Sec.)
180
150
150
150
240
180
120
240
90
Dry Cake Yield
tb./Sq.Ft./Hour
9.8
13.2
8.5
8.5
11.1
7.7
12.7
5.5
18.8
Filtrate Yield
Gal. /Sq. Ft. /Hour
.7
.45
.39
.32
.81
1.02
.80
.58
2.00
Filter Cake Solids
«)
75.0
72.0
81.2
81.5
70.3
73.7
81.0
70.5
71.6
I
u>
(1) - Minutes Per Revolution
-------
TABLE X
SUMMARY OF PHYSICAL PROPERTIES
UNTREATED FINE COAL REFUSE
Sampl e
No.
110*4
1105
1106
1107
1110
11 11
1112
1113
111*4
Atterberg Li mi ts
d^>
4->
._
E
I
-o
D
IT
_l
29-6
25.8
22. *4
25-5
3*»-5
3*4.2
2*4.1
36.2
37. *»
^e
** —
4->
'E
_i
0
4-1
U1
<0
Q_
2*4. *4
NP
NP
2*4.1
33.2
33.2
NP
29-7
3*».9
X
a>
-o
c
>~
4->
o
._
4->
l/>
OJ
a.
5.2
-
-
l.*4
1.3
1 .0
-
6.5
2.5
Spec! f i c Gravi ty
-o
0)
0)
o
0)
oc
(/>
<
2.09
1.53
1.58
1.81
1.65
1.76
1.3*»
1.80
1.83
Csj
T~
CO
LLj
2;
o
o
CXI
+
1.90
1.38
1.50
1.73
1.23
1.32
1.25
1.05
1.33
X
CO
LLJ
Z
O
o
exj
1
2.18
1.72
1-7*4
1.97
1.79
1 .98
1-55
2.15
2.12
Grain Size
0)
N
co
0)
o ~^
0)
y- c
M- —
UJ O
.006 *
.037
.018
.032
.007 *
.006 *
.016
.005 *
.006 *
y- ^
O CJ
4-J
C >~
0) 4->
'3 'I
*— L.
u- o
>4- U-
0) .-
o c
0 3
7.32*
3.95
11.11
10.31
5 . 00 A
8.33-
19.35
7 . 80 *
10.30*
Sol i ds Content
Settled Slurry
'~~
*^-^
(SI
T3
•—
^~
O
co
-o
•—
o
CO
(S-S1
LA
r^»
55-9
59.8
59-5
61. *4
52.0
55.0
62.2
*42.3
*47-*4
d~e
(U
4-1 0)
^ ^/
.- fD
LL- CJ
75.0
72.0
81 .2
81.5
70.3
73-7
81 .0
70.5
71.6
Proctor
)etermi rva-t ion
>-
4->
tn
c
a), — -
Qr<~>
E .-M
D U-
E~-^
'x -Q
03 —I
21 ,
86.0
6*4.3
72.1
81.6
67.7
73.2
71.0
73-9
78.0
0^—
16.6
23.9
1*4.6
1*4.5
22.0
18.8
15-0
23.3
17.0
NOTE: NP - Non-Plastic
* - Approximate Value
(1) - Initial Mixed Solids Content
(2) - Calculated Value
-------
TABLE XI
SUMMARY OF PHYSICAL PROPERTIES
UNTREATED FINE COAL REFUSE
Sample
No.
11 04
1105
1106
1107
1110
nil
1112
1113
1114
Permeabl 1 1 ty
Settled Slurry
cm/sec.
fi.Oxlo"6
l.OxlO"5
7.4xlO~6
2.6xlO~6
4.4xlO~6
4. 1xlO~6
I.9xl0"5
4.2xlO~6
S.Sxlo'6
i
Sol Ids
(*)
71.7
73.1
76.0
79.6
65.4
67.3
71.2
64.9
65-0
Filter Cake
cm/sec.
S.lxlfl"7
4. Ixio"6
4.8x!0"6
1.3x!0"6
1.6xlO~6
1.4xlO~6
6.2x!0"6
2.0xlO~7
8. Ixio"7
Sol Ids
(*)
76.8
77.0
79.6
82.5
73.0
75.9
82.0
74.5
74.0
Direct Shear
Settled Slurry
Consol 1 -
dated
Densl ty
(Lb/Ft3)
69.6
66.4
71.1
75.8
58.1
65.2
62.1
62.1
68.2
Degree
of
Satura-
tion (!)
84.7
76.7
98.9
78. -0
94.2
97.8
98.4
100.0
100.0
^L
10°
51°
53°
34°
20°
37°
22°
11°
8°
Cohes Ion
(psl)
0.018
0,218
0.173
0.036
0.073
0.045
0.327
0.100
0.055
Sol Ids
(*)
73-7
78.4
80.5
80.3
69.5
72.6
79.7
68.7
72.6
Fl Iter Cake
Consol 1-
dated
Density
(Lb/Ft3)
74.8
50.6
56.1
79-2
62.1
69.9
68.6
62.1
68.0
Degree
of
Satura-
tlon (I)
90.7
60.2
51.3
72.2
98.7
88.9
100.0
100.0
89.5
02
22°
36°
30°
38°
29°
42°
28°
14°
21°
Cohesion
(psl)
0.200
0.309
0. 182
0.791
0.045
0.245
0.255
0.218
0.082
Solids
CD
75.4
71.7
80.2
83.3
71.8
77.8
84.6
70.8
72.8
Consol 1-
dat Ion
Compres-
sion
Index
0 .19
0.12
0.09
0. 11
0.15
0.17
0.15
0.19
0.18
Sol Ids • I Dry Wgt.
Angle of Internal Friction
Settled Slurry
-------
TABLE XII
SUMMARY OF LABORATORY STRENGTH PARAMETERS
CONSOLIDATION TESTS
UNTREATED FINE COAL REFUSE
Sample
No.
1104
1105
1106.
1107
1110
1111
1112
1113
1114
Water Content
Before Test
(%)
47.9
35.7
24.4
22.4
59.0
53.4
39.1
66.1
56.3
After Test
(%)
21.2
24.7
13.6
15.5
32.1
31.6
21.2
29.9
30.4
Dry Unit Weight
Before Test
(Lb./Ft.3)
88.0
79.1
79.3
100.2
64.8
71.2
60.1
67.6
72.7
After Test
(Lb./Ft.3)
107.4
86.1
86.8
106.2
78.0
83.0
69.0
86.5
87.1
Initial
Void
Ratio
1.10
0.78
0.40
0.56
0.99
0.97
0.51
1.20
1.10
Initial
Compression
Index
(Cci)
.102
.050
.042
.015
.084
.070
.065
.113
.111
Compression
Index
(Cc)
0.19
0.12
0.09
0.11
0.15
0.17
0.15
0.19
0.18
Recompression
Index
(Ccr)
0.030
0.020
0.015
0.021
0.035
0.030
0.015
0.040
0.035
H?ater Content (%)
Wgt. of Water
Wgt. of Dry Solids
-------
TABLE XIII
SUMMARY OF LABORATORY STRENGTH PARAMETERS
DIRECT SHEAR TESTS
UNTREATED FINE COAL REFUSE
Samp 1 e
No.
11014
1105
1 106
1107
11 10
1 1 1 1
1 1 12
1113
1114
Preconsol i -
dat ion
Pressure
(psi)
SS
.9
.9
.9
.9
.9
.9
.9
.9
.9
FC
.9
.9
.9
.9
.9
.9
1 .0*4
.9
.9
Consol i -
dating
Pressure
(psi)
SS
.18
.36
.55
.73
_
.36
.55
.73
.18
.36
.55
.73
. 18
.36
.55
.73
. 18
.36
.55
.72
. 18
.36
.55
.72
.18
.36
.55
.72
.18
.36
.55
.72
. 1C
36
.55
.72
FC
. 18
-36
.55
-73
. 18
.36
.55
.73
. 18
.36
.55
.73
. 18
.36
.55
.73
. 18
.36
.55
-
.18
• ib
.55
.73
.09
.27
.45
.64
. 18
.36
.55
.73
-
.36
.55
.73
Shear
Stress
(psi)
SS
.05
.07
. 10
. 14
_
.66
.85
1 . 10
.40
.66
.88
1.24
.15
.26
.41
.53
.15
.21
.32
.35
.14
.37
.48
.62
.32
.48
.55
.64
.14
. 17
.21
.25
.07
. 1 1
.14
.16
FC
.28
.35
.44
.51
.44
.57
.7
.84
.29
.39
.50
.61
.79
1.10
1.23
1.36
.15
.25
-36
~
.41
.55
.73
.96
.26
.40
.50
.60
.25
.32
.36
.42
-
.23
.30
.36
Dry Wt.
After Test
(Lb./Ft3)
SS
69.6
66.4
71.1
75.8
58.1
65-2
62. 1
62.1
68.2
FC
74.8
50.6
56.1
79.2
62.1
69.9
68.6
62.1
68.0
Water Con-
tent After
Test. (%)'
SS
35.6
27-5
24.2
24.6
43.8
37.1
25.4
45.5
37-7
FC
32.6
39.5
24.7
20. 1
39.2
34.1
18.2
41 .2
37.4
Degree of
Satura-
tion 1%)
S3
84.7
76.7
98.9
78.0
94.2
97.8
98.4
100
100
FC
90.7
60.2
51.3
72.2
98.7
88.9
100
100
89.5
Sol ids Content - (&)
Before Test
SS
68.6
72.8
74.5
77.7
56.1
60.8
69.1
56.4
58.2
FC
75.0
72.0
81.2
81.5
70.3
73.7
81 .0
70.5
71.6
After Test
SS
73.7
78.4
80.5
80.3
69.5
72.6
79.7
68.7
72.6
FC
75-4
71-7
80.2
83.3
71 .8
77.8
84.6
70.8
72.8
Uni t Cohesion
(psi)
SS
.018
.218
.173
.036
.073
.045
.327
. 100
.055
FC
.200
.309
.182
.791
.045
.245
.255
.218
.082
1 nternal
Fr i c t i on
Angle
( Degrees)
SS
10°
51°
53°
34°
20°
37°
22°
11°
8°
FC
22°
36°
30°
38°
29°
42°
28°
14°
21o
Defi ni t ion of:
\ Water weight of water
we.ight of dry solids
' Sol ids = % Dry Weight
NOTE: SS - settled sol ids
FC - f i I ter cake
-------
TABLE XIV
SUMMARY OF ftO-DAY UNCONFINED COMPRESSION STRENGTH
U)
kC
1
Treatment
Addi t i ve
Untreated
5% Portland
Type 1
5% Lime
5% Calci lox A
SI Calci lox B
10% Portland
Type 1
10% Lime
10% Calci lox A
10% Calcilox B
15% Portland
Type 1
15% Lime
15% Calcilox A
15% Calci lox B
Mixed
Solids
%
25
35
FC
35
FC
35
FC
35
FC
35
25
35
FC
25
35
FC
25
35
FC
35
25
35
25
35
25
35
35
Sample Number
110ft
55o(D
*(2>
.1
*
103-1
*
30.0
j.
52.3
*
*
1.3
113-6
A
A
32. ft
S.lt
7.5
273-7
9.2
A
3.6
j.
A
13. ft
16.5
29. ft
72°
A
A
A
.A
95.9
A
37.2
j.
78.6
A
A
1.'6
I'd. 3
j.
A
39-8
6.1
15.1
368.6
10.3
A
ft. 5
j.
A
15. ft
2ft. ft
35.9
1105
55°
A
A
ft. 7
A
83.1
A
2ft. 2
10.6
123-5
9.8
A
A
158.8
A
17-7
fts.i
32.8
190.7
28.6
A
A
A
A
58.7
69.5
32. ft
72°
A
6.6
85.0
A
23.1
8.8
12ft. 9
7.2
A
0.8
165.3
A
A
12.6
39. ft
27.0
188.5
37.5
2.2
A
31.0
30.2
86.1
1106
55°
A
A
A
90.6
15.7
5.8
80. ft
13. ft
A
2.6
130.1
j_
A
20.3
18.1
25.9
295.5
37-5
A
8.1
A
A
51.0
55.3
ft9.2
72°
A
A
A
73.9
20.0
8.5
80.3
ft. 9
A
2.8
lftl.0
*
22.2
19.7
33-9
ftgft.ft
36.1
1.9
3-6
A
80.6
59. ft
80.3
1107
55°
A
A
A
A
112.1
A
2ft. 6
2.6
117-1
ft. 7
A
A
277.8
j.
*
290.9
216. ft
20.7
252.1
13-0
2.6
A
A
99.3
28.3
5ft. 8
72°
A
A
A
123-2
A
23.2
10.0
218.2
7.5
A
76.1
j,
A
15.2
20.6
38.9
233-7
32.8
A
A
A
ft7-7
35-7
87.9
1 1 10
55°
A
A
31.7
.<.
A
0.8
3ft. 2
A
A
111.0
A
A
13-7
5.8
2.2
70.0
ft. 9
A
0.8
A
A
6.8
ft.l
15.1
72°
A
A
A
18.7
.L
7.6
0.8
81.3
l.ft
A
J,
115.5
A
A
16. ft
5. ft
10.8
120.0
19.5
A
1.9
A
A
12. 1
17.1
3ft. ft
11 1
55°
A
A
A
62.2
16.9
0.8
117.7
0.8
A
J.
172.3
A
A
21.3
6.5
3.5
212, ft
3. ft
A
A
A
A
7-7
9.8
7.2
72°
A
A
67-6
A
A
1.6
50.7
2.6
A
Iftl .5
A
A
13.6
9.0
1ft. 1
116.6
12.2
A
A
A
A
15.5
21.0
26. ft
1112
55°
A
A
A
A
62.1
A
3-5
ft. 9
86.0
6.9
A
A
88. ft
A
20.7
7.1
6.9
85.8
13.3
A
A
A
A
3.1
7.6
19.9
72°
A
5.1
A
50.5
A
8.5
ft. 8
98.5
10.2
159- ft
A
A
13-3
7.7
ft. 5
69-9
33.7
A
A
A
*
6.6
13-3
39.3
1113
55°
A
A
A
A
ft7.7
*
8.7
A
ftft.8
A
A
0.2
igft. i
A
A
13-3
2.6
2.5
152.9
2.2
A
1.3
A
8. ft
11.8
12.2
72°
A
A
A
A
60. ft
A
9.6
l.ft
106.9
1.5
A
0.9
160.9
A
A
1ft. 6
3-7
8. ft
231.2
16. ft
A
1.9
A
*
10.3
2ft. 2
39.5
lllft
55°
A
A
*
A
57.7
A
Ift.ft
1.7
53-2
A
A
A
135.1
*
A
9.7
ft. 9
ft.l
207.0
ft.O
A
*
*
A
12.2
ft. 5
ft. 5
72°
A
A
A
A
67-1
A
15-6
2.0
89.1
2.0
A
A
Ift5.ft
A
A
16.1
ft. 7 '
ft. ft
22ft. 5
6.8
A
A
A
A
11.6
15. c
2ft. 0
, /Curing temperature measured in degrees Fahrenheit.
' '* Indicates specimen did not have a measurable strength at ftO days.
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
TABLE XVII-A
DRAVO CORPORATION
NEW TECHNOLOGY DEPARTMENT
Sample Description: Al1 Samples
Date: 12-5-77
Test No . :
Tested by :
100 r
90
80
.u —
"M 70 -<
• H
,-'
./ / •
/ ,
/ /
\
LO
c r o n
\
-
S
1
-
\
-
1
\
-—
0/01 0.005 O.C
0
10
n
30 ^
rt
40 o
w
U)
50 ">
cr
60 ^
CT
80
90
100
01
GRAIN SIZE IN MILLIMETERS
-------
TABLE XVII-B
Sample
1113
1104
1110
1111
1114
1107
1105
1106
1112
RANK - GROUPING OF SAMPLES FROM TEST RESULTS
TESTS
Particle
Sized)
3
2
1
4
5
8*
6*
7
9
Specific
Gravity
3
1
6*
5
2*
4
8
7
9
Liquid
Limit
2
5*
3
4
1*
7*
6*
9
8
Optimum
Water
Content
3
6*
1
4
5
9*
2*
8
7
Settling
1
6*
2
4
5
9*
7
8
3*
Consoli-
dation
Cc
1
2
5*
4
3*
8*
7
9
6*
Direct
Shear
Cohesion
6*
1
5*
3*
4
2*
8
7
9
Perme-
ability
3
5*
2
4
6
1*
8
7
9
Filtra-
tion
2
3
6*
5
9*
1*
8
4*
7
*Test Results that Did Not Conform to Ranking
(1) Ranked Fine to Coarse
-------
PARTICLE SIZE ANALYSIS
The grain size analysis is determined by both a wet and dry sieve segre-
gation.
Approximately 200 grams of the settled solids are washed through a
vibrating stack of sieves consisting of 200, 325, and 400 mesh sieves.
All of the wash water and its accompanying solids are retained. The wash
water is filtered and the solids retained by the filter and sieves are
oven-dried and weighed.
The dried solids retained on the 200 mesh sieve are.placed in a stack of
sieves containing 16, 30, 50, 100, 200, and 325 mesh sieves and placed on
the Ro-Tap for a period of approximately 10 minutes.
The solids retained by each sieve are weighed, and the percent solids
retained by each sieve determined by:
„ ,., ^ . , Weight of solids retained ,„„,,,
Percent solids retained = „,".-., ,. ... •- x 100%.
Total Weight of Solids
To obtain the 20 and 10 micron particle size fraction, 1 gram and 0.5 gram,
respectively, of minus 400 mesh solids are used for the determination.
The sieve with the sample is placed in a dish with dispersing liquid and
agitated with a pumping action causing the liquid to wash up and back
through the sieve. The dispersing liquid is changed as it becomes con-
taminated, and the agitation of the sieve continued until the dispersing
liquid remains clear.
The retained solids are then washed with methanol and dried.
The percent solids retained is calculated as before.
- i -
-------
SETTLING TESTS
The settling test determines the rate at which the slurry solids will
settle and the concentration to which they will settle.
To perform the test, a known weight of solids is placed in a 1000 milliliter
graduate and the volume adjusted to the 1000 milliliter mark by the
addition of water. The contents of the graduate are thoroughly mixed and
the graduate placed in a location where it will be undisturbed. Immediately
after mixing, a timer is started; and the settled volume of the slurry
portion is recorded with respect to the time to settle. Settled volume
and time readings are taken periodically up to 7 days.
A graph, plotting the change in settled solids height against the time for
settling, shows the settling characteristics of the solids. From the
height of the settled slurry and the weight of the solids, the settled
percent solids is calculated and then verified by an oven-dried sample.
- 11 -
-------
LEAF FILTER TESTS
The cycle time of a leaf filter test is analogous to the filter drum speed.
Generally .the faster the drum speed the higher the output, but the cake
may be thinner and sometimes wetter. The filter leaf test determines
the selection of the drum speed which is usually a compromise of these
conditions.
The filter cycle of a drum filter is the total time required for one
revolution of the drum which is four times the form time (cake forming)
or two times the dry time. The form time is dependent on the slurry's
ability to produce a dischargeable cake in a reasonably short time.
The leaf filter consists of a circular area of 1/10 square foot over which
a filter cloth can be mounted, and a vacuum outlet situated so that,
when a vacuum is applied, the solids in the slurry will be collected on
the filter cloth.
In operation, the filter leaf is submerged in a constantly stirred slurry.
The vacuum is turned on for the required form time, then the filter leaf is
removed from ,the slurry for the required dry time and the vacuum turned
off. The readings of vacuum, times, air flow during drying and other
pertinent information that relate to the drum filter operation are taken
during the immersion and dry times. The filtrate is collected and measured
along with the weight and the thickness of the cake on the filter leaf.
The cake is then dried and reweighed.
From the date, the percent cake moisture, dry pounds per square foot of
filter cloth per hour, and gallons of filtrate per square foot per hour
are calculated for the set of operating conditions.
The test is repeated with changes in the operating variables until the most
economical filtration rate is obtained. The set of operating conditions
that produce this filtration rate are then applied to the drum filter
operation.
- iii -
-------
ATTERBERG LIMITS AND INDICES
The Atterberg consistency limits of a particulate solid passing a No. 40
sieve provide indications of the surface chemical characteristics of the
material such as plasticity or friability, shrinkage and swell, and bonding
power. The Atterberg limits are the liquid limit, plastic limit, and
plasticity index.
The moisture content above which the material becomes liquid upon stirring
is called the liquid limit. The minimum moisture content at which the
material acts as a plastic solid is called the plastic limit. The numerical
difference between the liquid limit and the plastic limit is the plasticity
index.
To determine the liquid limit of a material, a portion passing a No. 40
sieve is mixed with water to a paste consistency. It is then placed in a
spherical shaped brass cup and the surface smoothed to provide a pat with
a maximum thickness of one centimeter. The pat is divided into two
segments with a grooving tool of standard shape and dimensions. The brass
cup is mounted in such a way that by turning a crank, it can be raised and
allowed to fall sharply onto a hard-rubber base. The shock produced by the
fall causes the adjacent sides of the divided pat to flow together. The
wetter the pat, the fewer shocks will be required to close the groove; the
drier the pat, the greater the number of blows. The moisture content at
which 25 blows causes the groove to close is defined as the liquid limit of
the material.
To determine the plastic limit of the material, a small quantity of the
material is rolled out with the palm of the hand on a flat surface until a
thread is formed. When the thread is rolled to a diameter of 1/8 inch and
becomes brittle so that it will no longer hold together in a .continuous
thread, the moisture content is determined and defines the plastic limit.
- iv -
-------
SPECIFIC GRAVITY TEST
The specific gravity of a particulate solid is the ratio of the weight
of a given .volume of solid particles to the weight of an equal volume of
water at a temperature of 4 C.
To perform a specific gravity test, a volumetric flask is filled to the
volume mark with the particulate solid and water, weighed to the nearest
0.1 g. (W1) and the temperature determined ( ).
The contents of the flask are removed, and the dry weight of the solids
(W ) is determined to the nearest 0.1 g.
s
The flask is refilled with water to the volumetric mark after adjusting
the temperature to that of the slurry, and weighed to the nearest
0.1 gram (W^.
W G
The specific gravity can be calculated from the formula: Gs = s T
Wg - W^ +
where GT = specific gravity of water at temperature T
W = dry weight of the solids
S
W = weight of flask, solids and water
W_ = weight of flask and water.
- v -
-------
COMPACTION
Compacting a particulate solids-water mixture is desirable for:
1) Decreasing future settlement, 2) Increasing the shear strength, and
3) Decreasing the permeability. The compaction test is a means of
determining the optimum water content of a compacted mixture that maximizes
these qualities.
The standard Proctor compaction test utilizes a 4 inch diameter by
4.6 inch high cylindrical mold with a capacity of 1/30 cubic foot. The
material is compacted into the mold in three layers with 25 blows per
layer from a 5.5 pound, hammer free-falling a distance of one foot, pro-
ducing a compactive energy of 12,400 foot pounds per cubic foot of material.
The compacted material is weighed, and a representative sample is removed
for a solids and water' content determination.
A series of tests are run on the same material with varying water contents.
A plot of the solids dry density mold versus the water content determines
a curve that, with an increase in the water content, shows a density
increase. As more water is added to the dry solids, the solids level in
the constant volume mold decreases. The water content at the maximum dry
density is the optimum water content.
- vi -
-------
PERMEABILITY
Permeability is a property which indicates the ease with which water will
flow, through the spaces between particulate solids in a mass.
In the laboratory method of determining the permeability of a particulate
solid-by the constant head method, a specimen of length "L" and cross-
sectional area "A", is placed in a tight-fitting tube. A standpipe from
an elevated constant level reservoir introduces water to the top of the
specimen, and the volume of water passing through the specimen is measured
with respect to time.
The flow of water "q" is directly proportional to the area "A" and to the
ratio of the height of the water column "H", and inversely proportional
to the length of the specimen "L". This relationship can be expressed by
HA
the equation, q ="— where "k" is the constant of proportionality called
L
the coefficient of permeability, or permeability with units of velocity.
The permeability, then, is a measure of the velocity of the water as it
flows through the specimen, expressed in centimeters per second.
-4
A permeability of 10 centimeters per second is frequently used as a
determination between pervious and impervious materials.
- vii -
-------
DIRECT SHEAR TEST
A direct shear test provides a means of determining the values of the
friction angle and the cohesion of a particulate solid.
In the direct shear test, a sample in the form of a low cylinder is placed
in a shear mold that permits the upper portion of the mass to slide in
relation to the lower portion. The plane through the sample on which the
sliding motion is produced is the shear plane.
A known load is applied in the direction normal to the shear plane and
the shearing load applied to the upper portion of the specimen. Most
tests are strain controlled, that is, the upper portion of the shear
mold is driven at a constant rate, and the shearing resistance is monitored.
Generally, the shearing force gradually increases to a constant ultimate
value.
When testing a material in direct shear, a minimum of three separate tests
are performed with a different normal load for each test. To provide
consistency in the test specimens, each specimen is consolidated under a
normal load, constant for all specimens,' that is greater than the testing
loads.
The normal load and the maximum applied shearing force are divided by the
cross-sectional area of the specimen at the shear plane to give the unit
normal pressure and the shear stress at failure. These values for each
test are plotted on a graph and generally the points will approximate a
straight line. The angle this line makes with the horizontal constitutes
the friction angle of the material, and the intersection at zero normal
stress is the cohesive stress.
- viii -
-------
CONSOLIDATION TEST
The purpose of a consolidation test is to obtain data used to predict
the rate and the amount of settlement of a particulate solid when
subjected to a load. The load may be the result of a structure or
overburden.
To perform this test, a specimen is encased in a ring to prevent
lateral movement and sandwiched between two porous plates. The
specimen is submerged in water, and vertical pressure or stress is
applied to the plates to compress the specimen. It is allowed to
remain until the compression virtually ceases, and then a much larger
stress is added. This is repeated for the range of stresses to which
the material is likely to be subjected. The amount of compression of
the material is measured with a micrometer dial.
Generally, the initial loading on the material is 1/2 kilogram per
square centimeter and progresses to 8 kilograms per square centimeter,
with each loading two times the preceding load. When 90% consolidation
has been reached at a particular loading, the load is changed.
After the consolidation is completed, the specimen is removed, and its
dry weight and dry density determined. The initial void ratio and the
void ratio at equilibrium at each load increment can be computed from
the volume of the solids and the volume of the specimen.
The compression strain is the result of a decrease in the volume of
the voids; therefore, it is convenient to express the stress-strain
relationship in terms of void ratio and unit pressure.
A plot of the equilibrium void ratio versus the log of the applied
pressure establishes a curve whose slope is numerically equal to the
compression index (C ), which is a measure of the compressibility of the
c
material.
- ix -
-------
The percent compression for each load can be determined by the change
in the vertical dial reading for each load increment.
The rate of consolidation is determined from the vertical dial readings
versus time for each load and is related to the permeability of the
material which controls the rate of flow of the escaping water.
The rate of consolidation under a load increment is represented by the
coefficient of consolidation, C . At a particular load increment
„ .0848 H2 , v
C = where
V ' fc90
H = average thickness per drainage surface
t = time for 90% compression.
The time - rate of settlement of a particular load for 90% consolidation
can be determined by „
= .848 H
C
v
t = time for settlement
H = maximum distance the pore water must flow to escape
C = coefficient of consolidation at the particular load.
- x -
-------
UNCONFINED COMPRESSION TEST
The unconfined compression test is used to measure the compressive strength
of a cylinder of material to which no lateral support is offered. Based
upon a combination of theoretical and empirical considerations, the ratio
of compressive strength to shear strength is equal to two. Therefore, the
test is also a simple and quick laboratory method of measuring the shear
strength of the material.
The unconfined compression test imposes uniform stresses and strains on
the specimen and causes the failure surface to develop in the weakest
portion of the specimen.
The uniformity of the prepared test specimen is very important for consistent
results; thus, a stabilized specimen is allowed to stabilize in a specimen
mold. The specimen is made cylindrical with a length to diameter ratio of
approximately two.
The ends of the specimen are trimmed to a plane normal to the longitudinal
axis, and the specimen is placed in the testing machine. A force is applied
to the longitudinal axis of the specimen until a failure or a strain of 20%
is reached. The strain is determined by a dial gauge that measures the
decrease in the length of the specimen as the force is applied.
The amount of strain at failure is used to calculate the average cross-
sectional area of the specimen at failure. The maximum force required to
fail the specimen divided by the average cross-sectional area at failure
gives the maximum compressive stress of the specimen. One-half of the
compressive stress is equal to the shear stress.
- xi -
-------
RATE OF STABILIZATION
Stabilization consists of changing a water-particulate solids mixture from
a slurry or soft mass to a hardened material exhibiting engineering
properties. This change is usually accomplished by the addition of
stabilizers that react with the particulate solids, or with themselves,
to produce the hardened condition.
The rate of stabilization is the degree of stabilization throughout the
time span in which the stabilization takes place.
Stabilization specimens are prepared by adjusting the percent solids of a
water-particulate slurry to a definite value and mixing with a stabilizer,
which is added on a weight percent of the slurry solids. The slurry-
stabilizer mixture is then stored at a constant temperature and allowed to
stabilize.
The rate of stabilization is determined by periodically measuring the
hardness of the specimen during the stabilizing period.
The hardness measurements are determined with a Soiltest Pocket Penetrometer
Model CL-700. This instrument consists of a flat-ended one-quarter inch
diameter (0.05 square inches) spring-loaded rod with a circular mark,
serving as a depth gauge, one-quarter inch from the end. In practice,
the rod is inserted into the material to the depth of the circular mark,
and the amount of spring compression is measured on a scale calibrated in
tons per square foot.
The penetrometer reading, measuring the resistance to penetration in tons
per square foot, at periodic time intervals is used to determine a
qualitative relationship between the hardening time and penetration
resistance (degree of stabilization) of the mixture.
- xii -
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/7-79-007
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Management of Coal Preparation Fine Wastes
Without Disposal Ponds
5. REPORT DATE
January 1979
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
D. C. Hoffman, R. W. Briggs, and S. R. Michalski
8. PERFORMING ORGANIZATION REPORT NO.
DOE FE-11270-1
9. PERFORMING OROANIZATION NAME AND ADDRESS
Dravo Corporation
3600 Neville Island
Pittsburgh, Pennsylvania 15225
10. PROGRAM ELEMENT NO.
EHE623A
11. CONTRACT/GRANT NO.
EPA Interagency Agreement
DXE685AK
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 ANT) PERIOD COVERED
Final; 7/77 - 6/78
14. SPONSORING AGENCY CODE
EPA/600/13
15. SUPPLEMENTARY NOTES project officers: D.A.Kirchgessner (IERL-RTP); andR.E. Hucko
(DoE), Div. of Solid Fuel Mining and Preparation, Pittsburgh PA 15213.
is. ABSTRACT rphe report gives results of a study to ascertain some physical/chemical
properties of a diverse sampling of fine coal refuse and to investigate the effect of
chemical stabilization. Samples of fine coal refuse were obtained from some of the
major coal seams in the Eastern bituminous coal fields. One proprietary and two
common chemicals were used to test chemical stabilization on nine samples of fine
coal refuse. The study showed that chemical stabilization can be used to drastically
improve some physical/chemical properties of the fine refuse for better handling
and disposal. The proprietary chemical Calcilox is the most effective over the range
of samples tested. Portland Type 1 cement is also effective, but greatly influenced
by waste solids concentration. The final chemical tested, lime, is inferior to the
other two, and generally ineffective in improving the fine waste's physical/chemical
properties.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDEDTERMS
c. COSATl Field/Group
Pollution
Coal Preparation
Wastes
Stabilization
Portland Cement Type 1
Calcium Oxides
Physical Properties
Chemical Proper-
ties
Pollution Control
Stationary Sources
Calcilox
13B
081
11B
07B
07D
18. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (This Report!
Unclassified
21. NO. OF PAGES
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
VU.S. GOVERNMENT PRINTING OFFICE: 1979-640-092/732
------- |