United States Industrial Environmental Research
Environmental Protection Laboratory
Agency Research Triangle Park NC 2771 1
EPA-600/7-78-150
July 1978
Assessment of
Coal Cleaning
Technology:
First Annual Report
Interagency
Energy/Environment
R&D Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wide'range of energy-related environ-
mental issues.
EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect
the views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/7-78-150
July 1978
Assessment of Coal Cleaning Technology:
First Annual Report
by
Lee C. McCandless and Robert B. Shaver
Versar. Inc.
6621 Electronic Drive
Springfield, Virginia 22151
Contract No. 68-02-2199
Program Element No. EHE624A
EPA Project Officer: James D. Kilgroe
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
ACKNOWLEDGMENT
The preparation of this Coal Cleaning Technology Development Annual
Report was accomplished through the efforts of the staff at Versar, Inc.,
Springfield, Virginia, under the direction of Dr. Robert G. Shaver, Vice
President and Mr. Lee C. McCandless, Operations Manager.
Acknowledgment is given to the staff of Denver Equipment Division, Joy
Manufacturing Company, Denver, Colorado, under the guidance of Dr. Thomas
Plouf for their work on the program.
Mr. James D. Kilgroe, Project Officer, Fuels Process Branch, Energy
Assessment and Control Division, through his assistance and direction, made a
valuable contribution to the preparation of this report.
Also our appreciation is extended to the Versar and Joy/Denver staff
with special thanks to:
Versar, Inc. Denver Equipment Div., Joy/Denver Co.
Mrs. Gayaneh Centos Mr. Les Apodaca
Dr. Irwin Frankel Mr. Williams Higgins
Mrs. Deborah Guinan Mr. Roger Harkins
Mrs. Jean Moore
Mrs. Judi Robinson
Mr. Donald Sargent
Appreciation is also given to the secretarial staff of Versar, Inc., for
their efforts in typing this document.
11
-------
CONTENTS
Acknowledgment ii
Figures V
Tables vii
1. Introduction 1
Project scope 2
2. Coal Characteristics 3
Types of sulfur 3
Coal washability data 4
Equipment performance 9
Development of ash-sulfur correlations from
washability data 11
3. Current Process Technology For Physical Coal Cleaning 22
Introduction 22
Breakers and crushers 26
Screens 33
Jigs 37
Wet concentrating tables 44
Hydrocyclones 48
Heavy media cyclones 51
The Otisca process 56
Cone concentrators 60
Vor-Siv 61
Spiral concentrators 62
Vorsyl 63
Flotation 64
4. Current Process Technology For Chemical Coal Cleaning 70
Magnex **-? process 74
Syracuse process 74
Meyers process 77
Ledgemont process 77
ERDA process 80
GE process 80
Battelle process 83
JPL process 83
IGT process 86
KVB process 86
Summary of minor and miscellaneous processes 89
Process and cost comparison 93
5. Current Process Technology For Fine Coal Dewatering
And Drying 104
Introduction 104
ill
-------
CONTENTS
(continued)
Mechanical dewatering devices 107
Thermal drying 127
Oil agglomeration 132
6. Coal Slurry Sampling 135
Summary of slurry sampling information 135
Other samplers and sampling systems 136
Bias test plan for two slurry samplers at Homer City
coal preparation plant 138
7. Coal Preparation Requirements For Synthetic Fuel
Conversion Processes 140
Particle size 140
Total moisture content 142
Ash content 144
Sulfur content 145
Heating value 145
Effect of coal beneficiation on coal conversion
and utilization 146
Summary of coal preparation requirements for
commercial and federal government sponsored
synthetic fuel processes 147
iv
-------
FIGURES
Nunber Page
1 Washability curve of an "easily cleanable coal" 6
2 Washability curve of an "average cleanable" coal 7
3 Washability curve of a "difficulty cleanable" coal 8
4 Distribution curve illustrating specific gravity
of separation 10
5 Generalized distribution curve for data represent-
ed in figure 4 10
6 Change in sulfur/ash ratio with gravity cleaning
of 75 selected coal samples 14
7 Fraction of original pyrite content in coal
floating at 1.40 S.G 15
8 Coal constitutent data for Northern Appalachian Region .... 18
9 Sectional view of hammermill 31
10 A typical ring crusher . 32
11 Baum Jig cross section 41
12 Generalized distribution curve for Baum Jig treating
a 15 cm (6 in.) by 48- rossh composite feed 43
13 Batac Jig cross section 45
14 The distribution of table products by particle
size and specific gravity 47
15 A typical hydrocyclone 51
16 A dense medium cyclone 53
17 Basic-media-only coal beneficiation plant 59
18 Relationship between pH and product recovery 66
19 Flotation cell 68
20 Magnex ™ process flow sheet 75
21 Syracuse coal comminution process flow sheet 76
22 Process flowsheet for the (Meyers') process 78
23 Ledgemont oxygen leaching process flow sheet 79
v
-------
Nunber
FIGURES
(continued)
24 ERDA process flow sheet 81
25 General Electric microwave process flow sheet 82
26 Battelle hydrothermal process flow sheet 84
27 JPL process flow sheet 85
28 IGT process flow sheet 87
29 KVB process flow sheet 88
-------
TABLES
Number Page
1 Relation of pyrite to other minerals in coal
separated at 1.40 S.G. (float fraction) 17
2 Surmary of physical coal cleaning unit operations 23
3 Vibrating screen equipment for coal preparation
service 38
4 Screen equipment for coal preparation service . . 39
5 Heavy media cyclone operating results for 31.75 x
0.75 ram. raw coal feed (1 1/4 x 1/32 in.) 55
6 Process information summary of major chemical
coal cleaning processes 71
7 Process information summary of minor chemical
coal cleaning proccesses 90
8 Process information summary of miscellaneous
chemical coal cleaning processes 92
9 Process performance and cost comparison for major
chemical coal cleaning processes 95
10 Operating cost comparisons for major chemical
coal cleaning processes 98
11 Cost effectiveness and other comparisons of
chemical coal cleaning processes 101
12 Fine coal dewatering and handling equipment 106
13 Summary status of coal preparation for high
BTU gasification processes 148
14 Summary status of coal preparation for low
BTU gasification processes . 149
15 Summary status of coal preparation for lique-
faction processes 151
16 Metric units conversion table 153
Vll
-------
SECTION 1
INTRODUCTION
Older the Clean Air Act of 1970 and its subsequent Amendment of 1977,
the U.S. Environmental Protection Agency is charged with promulgating stand-
ards and implementing state and federal plans for reduction of sulfur dioxide
emissions to the atmosphere. In 1974, sulfur dioxide emissions from coal
combustion were in excess of 18.6 million metric tons (20.5 million tons).
With a large projected increase in U.S. use of coal under the National
Energy Plan, improved sulfur dioxide controls are quickly needed. Industrial
coal consumption is predicted to quadruple from 62 million metric tons (68
million tons) in 1975 to 251 million metric tons (277 million tons) in 1985.
Utility coal consumption is projected to increase from 366 million metric
tons (404 million tons) to 707 million metric tons (779 million tons).
Near-term control options consist of flue gas desulfurization and
physical coal cleaning. Future options appear to be chemical coal cleaning,
use of synthetic fuels and fluidized bed combustion. Physical coal cleaning
may prove to be the most cost-effective method for reducing sulfur dioxide
emissions from small boilers, since many small industrial users will find
flue gas desulfurization too costly. Large industrial users and electric
utilities may realize substantial savings by using both physical coal
cleaning and flue gas desulfurization.
An evaluation of washability data published by the U.S. Bureau of Mines
in RI 8118 indicates that specific gravity separation conditions now commonly
used for removal of ash from coal can remove 25 to 55 percent of the pyrite
from U.S. coals. Advanced physical coal cleaning techniques may be capable
of removing up to 90 percent of the pyritic sulfur in coals, corresponding
to a total sulfur reduction in coal ranging from 30 to 60 percent. The
degree of desulfurization depends on the initial sulfur level in the raw
coal, the ratio of pyritic to organic sulfur, the size distribution of the
-------
pyrite in the coal and the preparation techniques vised. Based upon specific
gravity separation methods, the desulfurization potential of U.S. coals varies
among coal regions and also among coal beds within the sane region.
PROJECT SCOPE
This report covers the initial ten months of a three^ear contract with
the EPA Industrial Environmental Research Laboratory, Research Triangle Park,
North Carolina. The primary objective of this project is to evaluate physical
and chemical coal cleaning technology for removal of sulfur from coal.
The program contains six major technical tasks:
• The collection of existing data on sulfur removal by physical
coal cleaning equipment;
• The generation of new data and the evaluation of physical coal
cleaning technology for sulfur removal;
• The evaluation of equipment and costs for fine coal dewatering
and handling;
• The assessment of coal preparation requirements for synthetic
fuel conversion processes;
• The performance of studies on physical coal preparation processes
to evaluate the trade-offs between sulfur removal and costs; and
• The evaluation of chemical coal cleaning processes.
Task methodology includes literature searches and manufacturer contacts,
compilation of data from many representative sources (Bureau of Mines and
other governmental organizations researching this area, industrial research
facilities and commercial sources), testing and evaluation of currently
operational equipment and cost evaluation of various processes. In addition,
recent technological developments in fine coal dewatering and coal cleaning
are being evaluated.
-------
SECTION 2
COAL CHARACTERISTICS
A major portion of the sulfur dioxide pollution problem in the U.S.
is attributable to the combustion of coal. Removing sulfur dioxide from
stack gases by scrubbing has proven to be an effective, but expensive
process. By removing the sulfur from the coal prior to combustion, the
need for scrubbing can be reduced, or in some cases eliminated. The effect-
iveness of methods for removal of sulfur from coal prior to combustion depends
upon the relative percentages of the types of sulfur in the coal.
TYPES OF SULFUR
There are three general forms of sulfur found in coal: organic,
pyritic, and sulfate sulfur. Sulfate sulfur is present in the smallest
amount (0.1 percent by weight or less). The sulfate sulfur is usually
water soluble, originating from in-situ pyrite oxidation and can be
removed by washing the coal. Mineral sulfur occurs in either of the two
dimorphous forms of ferrous disulfide (FeSa) - pyrite or marcasite. The
twD minerals have the same chemical composition, but have different
crystalline forms. Sulfide sulfur occurs as individual particles (0.1
micron to 25 cm. in diameter) distributed through the coal matrix. Pyrite
is a dense mineral (4.5 gm/cc) compared with bituminous coal (1.30 gm/cc);
and is quite water-insoluble, thus the best physical means of removal is
by specific gravity separation. The organic sulfur is chemically bonded
to the organic carbon of the coal; and cannot be removed unless the
chemical bonds are broken. The amount of organic sulfur present, defines
the lowest limit to which a coal can be cleaned with respect to sulfur
removal by physical methods. Chemical coal cleaning processes, currently
in the developmental stage, are designed to attack and remove up to 40%
of the organic sulfur.
-------
The various physical cleaning processes enployed are based upon
physical or physical-chemical differences between coal and its associated
iirpurities. The nost coimonly employed property is density: "pure"
bituminous coal having a specific gravity of about 1.3, "shale" about 2.0
to 2.7, and "pyrite" about 4.8 to 5.0.
COAL WASHABILITY DATA
Ihe potential for improving the quality of a coal through physical
coal cleaning is determined by a series of tests called the washability
of that coal. To determine the preparation method and the equipment used
to clean the coal, the preparation engineer must conduct physical and
cheinical tests to obtain washability data. The coal is split into sub-
samples, by size and specific gravity distribution, which are then analyzed
for moisture, ash, heating value, pyritic and total sulfur, and other
characteristics. The test procedure may embrace all or only several of
the above characteristics, depending on the information required. Wash-
ability studies are conducted primarily to determine the yield and quality
of clean coal produced at a given specific gravity. These data are for a
specific coal at a specific size and are often presented in the following
tabular format.l
WASHABILITY DATA
Individual Fractions
Cumulative Float
Specific Gravity
Wt % Ash% Ash Prod. Wt % Ash Prod. Ash
SINK FLOAT
1.27
1.27 X 1.30
1.30 X 1.38
1.38 X 1.50
1.50 X 1.70
1.70 X 1.90
1.90
34.5
28.4
16.9
5.4
3.3
3.0
8.5
2.8
3.9
8.8
16.9
30.6
46.2
71.3
96.6
110.8
148.7
91.3
101.0
138.6
606.1
34.5
62.9
79.8
85.2
88.5
91.5
100.0
96.6
207.4
356.1
447.4
548.4
687.0
1293.1
2.8
3.3
4.5
5.3
6.2
7.5
12.9
-------
These results can be plotted in a number of ways to produce a set of curves
which are characteristic of the coal.
Three such curves are reproduced and shown on the following pages
as Figures 1, 2 and 3. They represent run-of-inine samples with varying
degrees of washability. In each figure three separate curves are given:
curve A is the curve of instantaneous ash content of the float fraction,
which is coitinonly called the characteristic curve; curve B shows the
cumulative ash content of the float fractions; and curve C shows the
cumulative ash content of the sink fraction. Similar curves can be drawn
for pyritic sulfur and total sulfur content.
Figure 1 — This is the curve of a readily cleanable coal. Curve A, the
characteristic curve, flattens abruptly at a specific gravity of 1.5, and
at this natural "cut point" an effective separation between coal and ash
could be made easily. At this specific gravity a product yield of about
67 percent clean coal with an ash content of about 10 percent could be
expected.
Figure 2 — This is the curve of a coal of average cleanability, the gradual
flattening of the characteristic curve showing no well-defined cut point. At
any specific gravity in the range of 1.5 to 1.6 the separated fractions would
contain a certain proportion of "middlings"; that is, particles consisting
of both coal and ash. In other words there would be some ash in the coal
and some coal in the ash.
Figure 3 —This is the curve of a difficult coal to clean by specific
gravity methods. The characteristic curve shows no cut point at all but a
gradual increase in yield and ash content from pure coal to pure ash. In
such a case the specific gravity at which separation was made would have to
be an arbitrary compromise based on market conditions. The figure shows a
coal yield of 40 percent with about 8 percent ash, with the middlings amount-
ing to 35 percent.
Middlings can be handled in two ways. The coal and ash can be broken
apart by crushing and then separated by re-washing; though in some types of
-------
FIGURE 1 WASHABILITY CURVE OF AN "EASILY CLEANABLE" COAL
JO « 50 60
ASH CONTENT %
70
COAL.
MIDDLES
DIRT
>*.*
Float-sink analysis
Floats at 1 -35 sp. gr.
,. 1-40 „
„ 1 -45 „
„ 1 -SO „
„ 1 -55 ,.
1 -60 .,
1-65 „
„ 1 -70 ,.
Sinks a: 1-70 „
Weight %
59-4
2-6
1-7
1-5
1-5
1-4
1-5
1-4
29-0
Elementary ash %
5-8
15-3
16-7
18-9
22-1
40-0
60-5
65-5
73-3
Cumulative ash %
5-8
6-2
6-5
6-8
7-1
7-8
8-9
10-0
28-4
-------
FIGURE 2 WASHABILITY CURVE OF AIM "AVERAGE CLEANABLE" COAL
o
10
20
30
•40
1-35 50
tl.
Si i-*
t-7
60
70
100
COAL
MIDDLES
H
DIRT
-
30 10 50
ASH CONTENT
70
90 100
Float-sink analysis
Floats at
it
,,
tt
II
tt
if
ft
Sinks at
1-35 sp.gr.
1-40 „
1-45 „
1-50 ..
1-55 „
1 -60 „
1-65 „
1-70 „
1-70 „
Weight %
49-5
8-6
4-7
4-2
4-2
3-9
2-6
3-2
19-1
Elementary ash '/,
8-3
16-2
19-4
24-7
35-5
52-0
65-0
70-4
76-1
Cumulative ash "/„
8-3
9-5
10-2
11-1
12-6
14-6
16-3
18-4
29-4
-------
FIGURE 3 WASHABILITY CURVE OF A "DIFFICULTLY CLEAIMABLE" COAL
COAL
MIDDLES
30 HO SO 60
ASH CONTENT •/,
DIRT
n
Float-sink analysis
Floats at
»
it
..
,.
tt
tt
..
Sinks at
1 -35 sp. gr.
1-40 „
1 -45 ,.
1 -50 „
1-55 ..
1-60 .,
1 -65 .,
1 -70 „
1 -70 .,
Weight %
29-7
9-5
5-6
6-2
5-1
6-1
5-7
6-2
25-9
Elementary ash yn
9-5
20-7
28-3
33-3
39-0
45-2
52-4
60-3
72-5
Cumulative ash '/,
9-5
12-2
14-2
16-5
18-6
21-2
23-8
26-9
38-7
-------
middlings the coal and ash are so closely combined that this is not practical
economically. Alternatively, the middlings can be disposed of as low-grade
fuel.
The washability results, for each size fraction, represent the ultimate
in quality improvement possible, using the physical process of specific
gravity separation. When performance curves ("distribution" curves) for
continuous commercial machinery are applied to the washability data for a
coal, the practical quality improvement may be predicted for a combination
of machine and that particular coal.
EQUIPMENT PERFORMANCE
To characterize the performance of any unit of equipment using the
float-sink principle, a distribution, or performance curve is developed.
This is a plot of percent distribution of the raw coal to the clean coal
product versus the specific gravity of the raw coal slurry. There is a
different distribution curve for each size increment of coal. These
curves are also called "partition" or "separation" curves. The curve
indicates the percentage of material in the feed, having a given specific
gravity d, that will be distributed to the clean coal product.
The term "specific gravity of separation" or "650" is the specific
gravity fraction at which the feed is partioned into equal fractions
of refuse and clean coal; this is the specific gravity that corresponds
to the midpoint (50 percent point) of the ordinate of the distribution
curve. This 650 specific gravity is considered a control variable. Actual
separations follow a curve which approximates that of Figure 4.
A coal washing device can be adjusted physically so that the value
of 650 can be decreased or increased; the distribution curve and the
resultant yield and quality then shift. An increase in d50 causes an
increased quantity of the "float" fraction and hence an increased yield
of this clean coal fraction, but it decreases its quality because more
impurities are also floated.
-------
100
1.4 1.6 1.8
SPECIFIC GRAVITY, d
2.0
Figure 4. — Distribution curve illustrating specific gravity
of separation.
100
*
50-
0.8 0.9 1.0 1.1 1.2
REDUCED SPECIFIC GRAVITY, x
1.3
Figure 5. - Generalized distribution curve for data rep-
resented in figure 4.
10
-------
A normalized distribution curve has been developed that it is indepen-
dent of the d50 specific gravity. "* The normalized curve is constructed by
dividing each point on the distribution curve by the specific gravity of
separation (d50) to produce a "generalized distribution curve". The curve
thus expresses the performance of a cleaning device (for a given coal
particle size increment and type of equipment) in terms of an abstract
number rather than actual specific gravities. Figure 5 shows
the generalized distribution curve for the data of Figure 4.
This normalization technique thus enables the entire range of
distribution data, with respect to specific gravity, to be generated from
one single set of normalized data which itself is independent of specific
gravity.
Washer equipment performance is predicted by using a unique general-
ized distribution curve for each of several size increments. The washer's
effect on each size and specific gravity fraction is determined by use of
each of the generalized distribution curves. dice the weight distributions
are established, the BTU, ash, and sulfur content of each product stream
can easily be determined. This is possible because the composite yield
of clean coal and its quality is the arithmetically weighted sum of the
yield and quality of clean coal of each size fraction. The yield and
quality of clean coal for each size fraction is obtained from a simple
calculation involving the unique distribution curve for that size and
the specific gravity analysis of the raw coal for that size.2
DEVELOPMENT OF ASH-SULFUR CORRELATIONS FEOM WASHABLLITY DATA
This work is concerned with evaluation of existing data for correlations
between sulfur and ash removal by physical separation. Historically, wash-
ability studies have been concerned with ash content, and little attention
11
-------
was given to sulfur removal. More recently, additional work has been done
which includes data on pyritic sulfur and total sulfur. With this information
it is possible to identify similarities and differences in the efficiency of
removal of pyritic sulfur, total sulfur, and ash by standard tests. The
existing data base has been analyzed for correlations of sulfur to ash as
functions of specific gravity and size reduction. Also the data were analyzed
to study the reduction of ash and sulfur by specific gravity separation after
crushing to different particle size. The data base for these calculations is
principally that of Bureau of Mines RI 8118.3 An internal draft report of the
Bureau of Mines was also used for data on washability at smaller particle
sizes. Both RI 8118 and the detailed washability data concentrate mainly on
Eastern and Midwestern coals. Western coal data are less plentiful. Many
of the Western coals, however, meet current EPA standards for sulfur emissions
without further cleaning.
Analysis of Ash-Sulfur Relationships Based upon a Sample of Washability Data
Covering all U.S. Coals
Initially seventy-five coal sanples of the 455 coal samples listed in
RI 8118 were selected and studied. The basis of this selection was a pyritic
sulfur content of at least 0.4 percent to focus on coals potentially best
beneficiated by pyrite removal. Representative samples from all such coal
beds were selected. The data from these included: state, county, coal bed,
size of crushed sample, % pyritic sulfur at 1.40 float, % total sulfur at
1.40 float, % ash at 1.40 float, % ash total, % pyritic sulfur raw coal and
% sulfur raw coal. A specific gravity of separation of 1.40 was selected
arbitrarily as indicative of the effects of deeply cleaning coal.
Based upon the data compiled, the following ratios were calculated as
appropriate for the 3.8 cm x 100 mesh (1 1/2" x 100), 0.95 on x 100 mesh
(3/8" x 100) and 14 mesh by 0 sizes:
12
-------
% Sulfur (at 1.40 float); a clean coal sulfur-ash ratio,
* %~Ash
% Sulfur (Raw Coal) ; the raw coal sulfur-ash ratio, and
% Ash
% Pyritic Sulfur at 1.40 float ; the fraction of pyrite remaining.
% Pyritic Sulfur (Raw Coal)
Ihese calculated ratios and their standard deviations were plotted vs.
particle size for each coal sanple. The purpose of these calculations was
to measure relationships between sulfur and ash removal.
Figures 6 and 7 show the average of the results obtained from these
computations. Figure 6 shows that the sulfur to ash ratio increases as a
result of crushing to successively smaller top sizes and floating at 1.40
specific gravity. An increase in sulfur to ash ratio occurs at all specific
gravities which were analyzed from the laboratory data (1.30 and 1.60 specific
gravities). In general, mechanical washing of coal results in an increase
in the sulfur to ash ratio of the coal. This implies that ash-forming
minerals are removed more progressively than pyrite. However, since the
pyritic minerals are also ash-forming constituents, the significance of
this conclusion is not clear. For this reason, further computations were
made to separate the reduction of pyritic minerals and the reduction of
other minerals (e.g., silicates) in these washability experiments. This is
discussed later.
In Fiqure 7 the pyrite remaining in the clean coal is plotted as a
function of coal size for the overall data. A clear progressive reduction
trend is shown as coal size is reduced from 3.8 cm x 14 mesh (!%" x 14 mesh)
top size.
The next step was to analyze the available coal cleaning data region-
by-region. To do this, a larger data base was used, essentially the whole
data base available from RI 8118. Calculations, as described before, were
performed on the coals from each region and the results averaged for each
region. Similar trends were seen as those shown for the data in Figures
6 and 7, so a more detailed analysis was attempted to gain insight into the
rates at which the various mineral constituents of coal are reduced by
size reduction followed by gravity separation.
13
-------
0.8
0.7
0.6
EC 0.5
O
u.
_i
(A
-I 0.4
.
O 0.3
O
<
oc
0.2
0.1 —
.0
•O
•0
RAW COAL 11/2"x100M 3/8" x 100 M
14M xO
FLOAT AT 1.40 S.G.
FIGURE 6. CHANGE IN SULFUR/ASH RATIO WITH GRAVITY CLEANING
OF 75 SELECTED COAL SAMPLES
14
-------
DATA POINTS REPRESENT AN AVERAGE
OF 75 SELECTED COAL SAMPLES
x 10OMESH v
3/8 IN. x 100 MESH
PARTICLE SIZE
14 MESH xO
FIGURE 7
FRACTION OF ORIGINAL PYRITE CONTENT IN COAL FLOATING AT 1.40 S.G.
15
-------
Analysis of Ash-Forming Mineral Seduction on a Regional Basis
A method of separating the differential effects on coal quality of
rerroval of pyrite versus other ash-forming minerals was needed. So the
following computations were performed on the data base to obtain values
for ash corrected for the amount of iron oxide that would result from the
combustion of pyrite:
j_ ,_,, molecular wt.Fe203
ash attributable to pyrite = pyrite sulfur content X
4 X atomic wt.S
"non-pyritic" ash = ash content - pyrite attributed to ash
These values of "non-pyritic" ash have been compared to ash and pyritic sulfur
in the 1.40 float fraction of the data base for various coals, coal regions,
as well as overall.
The principal conclusion that was derived was that with the possible
exception of the Northern Appalachian coals the modified sulfur to ash
ratios defined as
pyritic sulfur
"non-pyritic" ash
in the float fractions are relatively constant with top size. However,
there are regional or geographic differences. Table 1 lists the averages
of this ratio for each of the six regions at 1.40 float and for the raw coal.
A comprehensive analysis of the ratio at specific gravities other than
1.40 was not undertaken. Primary calculations on several coals showed that
the trend in sulfur-ash ratios with specific gravity is essentially the
same as with particle size. Deeper cleaning of coal by virtue of lower
specific gravity or smaller top size produces the same trend in the ratio.
The principal exception to this conclusion is the data from the Northern
Appalachian region, where the "non-pyritic" ash content of the clean coal
was found to be relatively constant with coal size whereas the pyritic sulfur
was markedly reduced with particle size. This is shown in Figure 8 as a
logarithmic plot that demonstrates the greater removal of pyrite than ash
and clearly indicates that the pyritic sulfur/non-pyritic ash ratio de-
creases significantly with particle size for the 1.40 float fractions of the
16
-------
TABLE 1. RELATION OF PXRITE TO OTHER MINERALS IN COAL
SEPARATED AT 1.40 S.G. (FLOAT FRACTION)
Ratio Of Ratio of
Data Base Pyritic Sulfur To Total Sulfur
ion No. of Sanples Coal Top Size Non-Pyritic Ash To Ash
Northern Appalachian, 326 Raw Coal 0.16 0.20
All levels of Sulfur 3.8 on ( 1 1/2") 0.14 0.27
0.95 cm (3/8n) 0.11 0.27
14 M 0.09 0.27
Northern Appalachian 32 Raw Coal 0 51 0 38
High Sulfur (5% av.pyritic) 3.8 on ( 1 1/2") 0.26 0*35
0.95 on (3/8") 0.17 0.31
14 M 0.12 0.31
Southern Appalachian 44 Raw Coal 0.04 0.10
3.8 cm ( 1 1/2") 0.05 0.20
0.95 on (3/8") 0.05 0.22
14 M 0.05 0.23
Alabama 10 Raw Coal 0.08 0.14
3.8 cm ( 1 1/2") 0.09 0.22
0.95 cm (3/8") 0.09 0.23
14 M 0.08 0.24
Eastern Midwest 98 Raw Coal 0.20 0.28
3.8 cm ( 1 1/2") 0.17 0.38
0.95 on (3/8") 0.17 0.41
14 M 0.16 0.44
Western Midwest 65 Raw Coal 0.30 0.32
(55) 3.8 on ( 1 1/2") 0.29 0.43
0.95 cm (3/8") 0.29 0.48
14 M 0.28 0.51
Western 44 Raw Coal 0.03 0.08
3.8 cm ( 1 1/2") 0.02 0.10
0.95 cm (3/8") 0.02 0.11
14 M 0.02 0.13
17
-------
20
O
cc
LL.
u.
«s-
10
9
8
7
S
5 •
2
o
cc
LU
0.
1.0
0.9
0.8
0.7
0.6
0.5
0.4
RAW COAL 11/2 IN. x 100 MESH 3/8 IN. x 100 MESH
14 MESH xO
PARTICLE SIZE
DATA POINTS REPRESENT AVERAGES OF 32 HIGH SULFUR COAL SAMPLES
FIGURE 8 COAL CONSTITUENT DATA FOR NORTHERN
APPALACHIAN REGION
18
-------
high sulfur coals from this region. The difference in this respect with
the other regions can be seen in Table 1 which lists the ratio of pyritic
sulfur to non-pyritic ash for each coal top size, as well as for the original
sample, region by region. Only in the Northern Appalachian region is there a
significant change in this ratio from size to size and from the original
content. In the coals of this region, pyrite is more effectively released
from the coal by grinding to reduce coal top size than other minerals. Compu-
tations were made on both the average of all 326 coal samples from this
region, and for the 32 selected high sulfur samples. The same general
effect was found in both.
Table 1 also shows the total sulfur/ash ratios for these regions. It
should be noted that these ratios generally increase with decreasing top
size, unlike the pyritic sulfur/non-pyritic ash ratio.
Additional studies of these relationships, particularly the extension
of data into finer particle sizes, may provide a basis for classification
and selection of area geological, or mineralogical/identification of coal
reserves according to potential sulfur cleanability.
Regression Analysis of Sulfur and Ash Eelationships
The raw coal data of the Bureau of Mines contained in RI 8118 was
analyzed by linear regression for evidence of correlation between total
sulfur and ash. The apparent correlations between percentages of total
sulfur and ash differed greatly from region-to-region as follows:
correlation number of degree of
region slope intercept coefficient Samples certainty
N. Appalachian
S. Appalachian
Alabama
E. Midwest
W. Midwest
West
All Data
19
0.03
-0.01
0.06
0.05
0.19
0.00005
0.12
2.58
1.15
0.74
3.24
2.28
0.67
1.33
0.11
-0.08
0.42
0.21
0.56
o.noi
0.08
227
35
10
95
44
44
455
90%
99%
< 80%
< 90%
99.9%
« 80%
90%
-------
In most cases a moderate degree of correlation between these two coal
quality parameters is indicated. HDwever, for the Southern Appalachian and
Vfestern regions, the relation between the parameters is even less than
indicated by the linear regression correlation coefficients because the
slopes of the regression lines are close to zero. Ihis means that total
sulfur is virtually independent of ash content in these cases. For the data
set as a whole a moderate degree of correlation was found (90% certainty).
This is not of sufficient significance to predict sulfur from ash content for
individual seams. Western Midwest coals may be exceptions to this generaliza-
tion because of the high degree of certainty (99.9%) found as well as the
high average sulfur content (3.6%) and high slope of the sulfur-ash linear
regression.
Linear regression analysis of the whole data set was also carried out
for evidence of correlation between total sulfur and non-pyritic ash, since
this latter parameter is a measure of mineral content that is not sulfur-
containing. A poorer correlation was expected than with the unmodified ash
values, and this was found. The degree of certainty of correlation was much
less than 80% (approximately 55%) based on a correlation coefficient of 0.04.
Correlation between total sulfur and non-pyritic ash was not calculated on a
region-by-region basis.
A positve correlation between organic and pyritic sulfur contents with
a high degree of certainty (99%) was calculated for the whole data set.
Because of this, it was expected and found that there is a low degree of
correlation between pyritic sulfur and non-pyritic ash similar to that
found for total sulfur-to-non-pyritic ash. She calculated correlation
coefficient was 0.04 (less than 60% certainty). Thus it is concluded that
pyrite content is essentially unrelated to other mineral content in U.S.
coals.
20
-------
SECTION 2
REFERENCES
1. Leonard, J.W. and D.R. Mitchell (Eds,). Coal Preparation Third Edition,
AIME, New York, 1968. pp. 4-23 through 4-24.
2. Jacobsen, P.S. Coal Preparation Computer Simulation of Plant Perform-
ance: Mineral Industries Bulletin, Colorado School of Mines, Vol. 21,
Jan. 1978, No. 1
3. Cavallaro, J.A., M.T. Johnston and A.W. Deubrouck. "Sulfur Reduction
Potential of U.S. Coals: A Revised Report of Investigation," Bureau of
Mines RI 8118 or EPA-600/2-76-091, Washington, B.C., April 1976.
4. Gottfried, B.S., and Jacobsen, P.S., 1977. Generalized distribution
curve for characterizing the performance of coal cleaning equipment:
U.S. Bureau of Mines RI 8238, 21 p.
21
-------
SECTION 3
CURRENT PROCESS TECHNOLOGY FOR
PHYSICAL COAL CLEANING
INTRODUCTION
At least fifty-five percent of the coal mined in the United States
is subjected to some physical cleaning process. The degree of cleaning
varies widely and the process technologies used range from simple mechani-
cal removal of rock and dirt, to complex coal beneficiation plants for the
removal of heavier contaminants and non-combustible minerals in coal.
Many cleaning processes are well established, large scale and commercially
operating. Some newer cleaning processes for the recovery of coal fines and
removal of pyrite are either in the developmental or demonstration stage.
The type of cleaning technology and the extent of cleaning depends mostly
on the type of coal, the method of mining and the end use of the clean coal.
All commercial coal cleaning plants utilize physical cleaning techniques to
separate coal from the heavier ash forming minerals. At the present time
these plants are primarily designed for the removal of ash not pyrite
(mineral sulfur) in coal. Reduction of ash minerals in coal is cost
effective since higher heating value results in decreased storage and coal
handling requirements and lower transportation cost.
A list of major physical coal cleaning unit operations are given in
Table 2. Existing coal cleaning plants use one or more of these operations
to prepare the coal for a given market.
In a complete coal preparation facility coal from the mine is crushed
and screened to remove oversize material. Secondary breaking or crushing
is sometimes necessary to ensure good separation of coal from impurities in
the subsequent processing plant. Classifying screens separate coal particles
22
-------
TABLE 2 SUMMARY OF PHYSICAL COAL CLEANING UNIT OPERATIONS
Unit Operation
Jigging
Description
A pulsating fluid stratifies
coal particles in increasing
density from top to bottom.
The cleaned coal is overflowed
at the top.
Efficiency Remarks
N.A. Most popular and least ex-
pensive coal washer avail-
able, but may not effect as
accurate a separation as
desired. Sizes: 3.4 to 76 ran.
(6 mesh to 3 in.)
Tables
(jJ
Dense Media
Pulverized coal and water are
floated over a table shaken with
a reciprocating motion.
The lighter coal particles
are separated to the bottxm of
the table, while the heavier,
larger, impure particles nove
to the sides.
N.A.
Coal is slurried in a medium with
a specific gravity close to that
at which the separation is to be
made. The lighter, purer coal
floats to the top and is con-
tinuously skimmed off.
Sizes: .15 to 6.4 mm
(100 mash to 1/4 in.)
N.A. Advantages: Ability to make
sharp separations at any
specific gravity within the
range normally required;
Ability to handle wide range
of sizes; Relatively low
capital and operating costs
when considered in terms of
high capacity and small space
requirements; Ability to
handle fluctuations in feed
quantity and quality. Sizes:
.59 to 200 mm (28 mesh to 8 in.)
(continued)
-------
TAELF 2. (Continued)
Unit Operation Description
Hydrocyclones The separating mechanism is des-
cribed as taking place in the
ascending vortex. The high and
low specific gravity particles
noving upward in this current
are subjected to centrifugal
forces effecting separation.
NJ
Humphrey
Spiral
Efficiency Remarks
N.A. If maximum pyrite reduction
and maximum clean coal yield
are to be obtained, supple-
mental processes such as cy-
clone classifying, fine mesh
screening and froth flotation
are necessary (on stream pro-
cess) . Hydrocycl°nes present-
ly are used in the U.S. to
clean flotation - sized coal,
but can be used for coal as
coarse as 64 x 0 mm. (1/4 x 0 in.)
Coal-water slurry is fed into a
spiral conduit. As it flows down-
ward, stratification of the solids
occurs with the heavier particles
concentrated in a band along ihe
spiral. An adjustable splitter
separates the stream into 2 pro-
ducts—a clean coal and the mid-
dlings .
N.A. Has shown significant ash and sulfur
reduction on .42 x 0 mm. (35 x 0
mesh) Middle Kittanning coal.
Launder-type
Washer
Raw coal is fed into the high end
of a trough with a stream of
water. As the stream of coal and
water flows down the incline, par-
ti cles having the highest set-
tling rate settle into the lower
strata of the stream. These are
the middling or refuse particles.
The clean coal particles gravi-
tate into the upper strata before
se£jaration.
N.A. Three types of launders are
recognized based upon mode
of transport. Sizes:
4.76 to 76 mm. (4 mesh to 3 in.)
(continued)
-------
TABLE 2. fcontinued)
Unit Operation
Description
Efficiency
Remarks
Pneunatic
Coal and refuse particles are
stratified by means of pulsating
air. The layer of refuse formed
travels forward into pockets or
wells from which it is withdrawn.
The upper layer of coal travels
over the refuse and is removed at
the opposite end.
N.A.
Most acceptable preparation
method from the standpoint
of delivered heating value cost.
Sizes: up to 6.4 mm.
(1/4 in.).
Froth
Flotation
to
Ul
A coal slurry is mixed with a
collector to make certain
fractions of the mixture hy-
drophilic. A frother is added
and finely disseminated air
bubbles are passed tlvrough the
mix. Air-adhering particles
are floated to the top of the
remaining slurry and are then
removed as a concentrate.
N.A.
Froth flotation is used to re-
duce pyrite in English coals;
The flotation of coal refuse
to obtain salable pyrite is
uneconomical in view of todays
poor sulfur market; If ethyl-
xanthate is used as the col-
lector, it is absorbed onto
coal pyrite in such a manner
as to make it ineffective for
flotation. Fizes: 1.17 to .044 mm
Two stage Flotation
for Pyrite
Experimental coal flotation pro- Pyritic sul-
cess in which the coal is floated fur reduced
while high ash impurities are re- 9°* with 80%
jected. The froth concentrate recovery
is then repu]ped in H2O, treated
with an organic colloid to depress
coal. A xanthate collector and al-
cohol frother are added and then
refloated.
Frothing agent—methylisobutyl
carbinol; pH regulators—NaOH
and HC1; Coal Depressant—Aero
Depressant 633; Pyrite flo-
tation collector—potassium am-
ylj
-------
by size and route them to various coal cleaning unit operations which may
be either wet, dry or a combination of both.
Current technologies and equipment for physical coal preparation and
cleaning are discussed in detail in this section.
BREAKERS AND CRUSHERS
Ihere are two fundamental objectives in crushing coal: (1) to reduce
run-of-mine (RCM) coal to sizes suitable for cleaning and (2) to meet the
market specifications for specific sizes. Production of fines in the
crushing process has traditionally been considered undesirable, therefore,
crushers are designed to produce a minimum amount of undersize material.
KM coal is typically broken into increasingly smaller sizes by staged
reduction. Primary breaking is the first stage, where raw coal is reduced
to a top size of 100 to 200 itm. (4 to 8 in.). Various sizes may then be
screened and sent to washing units or to secondary crushing. Coal that
is already smaller than the primary breaker product size is often screened
out ahead of the breaker and sent to processing; or it is screened out
prior to the primary breaker, routed around the breaker and then combined
with the product from the breaker.
Secondary crushers perform the second step in the reduction of coal
and reduce the products of primary crushers and washers to a top size of
about 45 rrm. (1 3/4 in.).
Types of crushers covered here are rotary breakers, single and double
roll crushers, hammermills and ring crushers.
Rotary Breakers
The rotary breaker is often called a Bradford breaker. It is
essentially a large rotating cylinder powered by an electric motor via a
chain and reducer drive. The cylinder diameter ranges from 1.8 to 3.6
meters (5.9 to 11.8 ft.) with a length of 1-1/2 to 2-1/2 times the diameter.
26
-------
Rotary Breaker Operation—
Coal is introduced through one end of the rotating cylinder and is
broken by impacting other coal particles or the bottom of the cylinder. A
typical cylinder speed is 12 to 18 revolutions-per-minute. The cylinder
walls are heavy screen plates through which the sized material may pass to a
conveyor below. Rock, slate, and other materials which resist breakage
are retained within the cylinder. Internal lifting shelves and deflectors
carry debris along the length of the machine to the discharge end, where
a continuously rotating plow scoops out the debris.
Rotary Breaker Performance--
Bradford type breakers are usually quite dependable and require little
maintenance because of their design simplicity. Feeds having a high clay
content cause the perforations in the cylinder screens to become clogged
causing equipment shutdown.
Capacities of rotary breakers range up to 1,800 kkg (2,000 tons) per
hour for the larger units. The data in the table below cover sizes and
capacities for reducing ROM coal to a 160 x 180 mm. (6.3 x 7.1 in.) product.
Data presented below are for Pennsylvania Crusher Corp. Rotary breakers.
SIZES AND CAPACITIES OF ROTARY BREAKERS
Bearing Load
Size Casing
Diam,x Appox.
with Coal,Lbs*
Length,Fi* WtfLbs* Feed End
6 x
7 x
9 x
Hftx
12 x
8
14
17
19
22
7,300
9,100
14,000
22,300
32,700
13
17
35
51
95
,000
,300
,000
,625
,500
Disc'g End
11,
15,
21,
48,
77,
450
600
400
625
500
Maximum
Weight,
Motor,
Appox.
Piece, Lbs* Hp*
32,
37,
42,
15,
17,
150
500
160
050
400
10
15- 20
40- 50
60- 75
100-150
Capacity Type
Appox. of
Tph* Coal
75- 150
125- 250
275- 450
500- 750
1,000-1,
Soft
Soft
Medium
Medium
500 Hard
* Conversion factors from English to metric units are available in Table 16.
A Variation: The Rotary Breaker Hammermill
In the trunnion mounted breakers, the addition of a hammermill rotor in
the discharge end makes the unit more effective in handling harder coals and
can add 20 to 25 percent production capacity to the unit.
27
-------
Single-Roll Crushers
The single-roll crusher is one of the oldest types of comminution equip-
ment. It consists primarily of a heavy frame on which are mounted a toothed
crushing roll and a stationary breaker plate.
Single-Poll Crusher Operation—
The single crushing roll usually has long teeth equally spaced
around and along the roll with short teeth inserted in the spaces between the
long teeth. Coal is caught between the revolving roll and the breaker plate
where the long teeth act as feeders and do the initial breaking of large lumps
and the short teeth accomplish the final size reduction. Variation of tooth
design helps to tailor the crusher to a particular feed. Product size adjust-
ment is made by changing the clearance between crushing roll and breaker
plate.
Single-Roll Crusher Performance—
A typical single-roll crusher is capable of handling RDM coal, reducing
it to 40 x 0 mm. (1-1/2 x 0 in.) without stalling. Capacities of some
single roll crushers are shown on the next page.
Double-Roll Crusher Operation—
Most of the crushing action of the double-roll crusher is by impact of
roll-teeth against coal, rather than by compression between the rolls. The
fact that compression plays a secondary role is evidenced by the production
of a relatively small amount of fines.
Adjustment of the spacing between the rolls is an important factor.
Some units are adjustable by means of a traveling bevel gear and bearing
box while others utilize a special gear train or chain drive, driving both
rolls to allow adjustment during operation. One company has developed a
special gear train which allows adjustment of up to 28 cm. (11 in.) while
the crusher is in operation. The system uses pinion yokes which change
their angle as the movable roll position is changed, thus keeping the
pinion gears in mesh with the gears.
28
-------
CAPACITIES OF SINGLE-ROLL CRUSHERS
*
Tons per Hour
Roll Size Max. Size
Diam.x of Feed,
Width, In.* In.*
Product Size, In
1
ih
2%
3 3/4
5
6
.*
1% 8%
9 3/4
Hard Bituminous Coal
18 x 18
24 x 36
30 x 45
36 x 45
Hp*per Ton*
Crushed
8
14
20
22
16
55
155
150
7/8
27
85
170
230
5/8
65
160
300
380
1/3
85
190
330
450
1/3
110
220
350
550
1/6
260
425
625
1/6
500
720 850
1/6 1/9
1000
1/9
Medium Hard Bituminous Coal
18 x 18
24 x 36
30 x 45
36 x 54
Hp*per Ton*
Crushed
18 x 18
24 x 36
30 x 45
36 x 54
Hp *per Ton *
Crushed
10
16
22
24
12
18
24
30
17
55
120
160
3/4
Soft
22
75
145
195
1/2
32
105
200
275
1/2
75
190
350
420
2/9
Bi-cuminous
40
130
250
335
1/3
85
200
380
460
1/6
100
230
400
500
2/9
Coal
115
250
450
550
1/6
140
275
425
600
1/9
155
300
480
700
1/12
325
500
720
1/9
350
550
850
1/12
575
800 1000
1/9 1/12
625
950 1100
1/12 1/16
1200
1/12
1300
1/16
*Conversion factors from English to metric units are available in Table 16.
Double-Roll Crusher Performance—
In time, normal wear of the roll teeth will reduce the double-roll
crushers capacity and increase its horsepower requirement. The capacity of
a unit depends on the roll speed, diameter and length of the rolls, and their
set. Capacities of typical double-roll crushers are given below.
29
-------
CAPACITIES OF DOUBLE-ROLL CRUSHERS
Tons* per Hour
Poll Size Max. Size
Diam.x of Feed,
Width, In.* In*
Speed of
Rolls
RPM
Ik
2
Hard Bituminous
18 x 15
20 x 18
24 x 36
30 x 48
Hp*per Ton*
Crushed
18 x 15
20 x 18
24 x 36
30 x 48
Hp*per Ton*
Crushed
4-8
4-10
6-16
8-20
4-10
4-14
6-18
8-24
150
150
130
115
Medium
150
150
130
115
20
30
65
110
1/2
Hard
25
40
80
140
1/3
30
40
90
150
1/2
Product Size, In.*
3
Coal
35
50
115
190
1/3
4
50
70
170
250
1/3
5
60
85
200
330
1/6
6
270
400
1/6
7%
300
450
1/6
Min.
Motor
Hp*
5
lh
15
25
Bituminous Coal
35
50
115
185
1/3
Soft Bituminous
18 x 15
20 x 18
24 x 36
30 x 48
Hp*per Ton*
Crushed
4-12
4-16
6-20
8-24
150
150
130
115
30
50
100
175
1/4
40
60
140
215
1/4
45
65
140
235
1/4
Coal
50
80
180
280
1/6
65
85
200
300
1/4
70
100
220
375
1/6
70
100
260
390
1/8
90
130
290
470
1/10
290
460
1/8
350
550
1/10
350
575
1/8
450
700
1/10
5
lh
15
25
5
lh
15
25
*Conversion factors from English to metric units are available in Table 16.
Hamrnermills
Hamtermills consist of a heavy frame supporting a rotor to which hammers
are attached. On one side of the rotor is the feed opening; on the other
side are grate bars whose function is to fix top product size. A metal
trap is usually included to prevent damage to the hammers or cage bars by
"uncrushable" material. A sectional view of a representative haimiermill
is shown in Figure 9.
30
-------
FIGURE 9
SECTIONAL VIEW OF HANMERMILL
Hammermill Operation—
In all hainnermills, the material is broken by impact from hammers strik-
ing the material and reducing it until it is fine enough to pass through
openings in grates (cage bars) or screen plates. Generally speaking, heavy
hammers operated at a slower speed are best for coarse crushing operations
while a greater number of lighter hammers operated at a higher speed are best
for crushing to fine sizes.
Some manufacturers offer reversible hammermills in which the direction
of rotation may be reversed to increase hammer life. Reversible mills have
twin breaker blocks attached syrmetrically around the top half of the rotor
and twin cages attached symetrically around the bottom half.
The cage position may be adjusted to compensate for hammer wear.
Cage bar spacing may be varied for adjustment of product size.
31
-------
Hanmermill Performance—
Banners, breaker plates, and grate bars must be in good condition to
maintain proper product size and efficient operation.
Capacities for one
SIZES AND CAPACITIES OF HAW1ERMELLS
Capacity, Tph* - ROM Coal
Diam.x
Length, In.
20 x
24 x
36 x
36 x
42 x
12
20
36
60
66
t Approx.
12- 20
30- 40
75-100
150-175
175-250
Weight, Opening,
Lbs* In.*
2,200
4,200
11,800
17,800
36,000
11 x
13 x
18 x
18 x
22 x
11
20
36
60
66
Speed
RPM
1,500-1
1,400-1
900-1
900-1
700-1
,800
,600
,200
,200
,000
Floor
Space*
2 '8"
3'2"
4'4"
4'4"
4'10
x 4'6n
x 5 '10"
x 8'7"
xl2'4n
"xl3'0"
l/8"*Bar
Opening
2-3
6-8
20-25
45-50
75-85
l"*Bar
Opening
6- 8
18-20
75-90
140-160
175-220
* Conversion factors from English to metric are available in Table 16.
Ring Crushers
Ring Crushers are similar to haimennills,but use circular rings in
place of swing hammers; both have breaker plates, rotor assemblies and
cage assemblies.
Figure 10 is a sectional view of a typical ring crusher.
FIGURE 10. A TYPICAL RING CRUFHER
32
-------
Ping Crusher Operation—
Rings may be of toothed or smooth design and are often used in combina-
tion. Much of the breaking action is by impact; compression crushing ac-
counts for the remainder. A mechanism for cage position adjustment provides
a method to compensate for ring wear. Cage bar spacing may be altered for
adjustment of product size.
Ring Crusher Performance—
As with the hammermill, the various parts which are subject to wear must
be kept in good condition or replaced when necessary to ensure constant pro-
duct size and optimum machine efficiency.
Capacities and related data for representative ring crushers are given
below.
CAPACITIES OF RING CRUSHERS
Diameter
Hammer Circle
In. *
24
30
32
38
54
Maximum
Feed Size
In. *
10
16
18
22
28
Approx.
Hp*
20
100-125
200-225
250-275
500-600
Approx.
Capacity
Tph*
40
175- 225
350- 400
500- 550
1,050 - 1,200
* Conversion factors from English to metric units are available in Table 16.
Capacities based on handling run of mine bituminous coal of average
hardness and moisture content for reduction to a nominal 3/4 in. and under
product.
SCREENS
Approximately half of the 1975's coal production of 584 million metric
tons (644 million tons) has passed over screens within a preparation plant;
and the remainder of that production has, at some point, gone over screens
for scalping or sizing.
The screens used in coal preparation circuits fall into six major
categories:
33
-------
• Scalping screen - for separating refuse and fines prior to size
reduction;
• Raw coal sizing screen - for separating the raw coal into coarse
and fine size for further processing;
• Pre-vet screen - to remove fines prior to the mechanical cleaning;
• Heavy Media Recovery Screens - for recovery and reuse of Magnetite;
• Desliming Screens - to remove extreme fines; and
• Dewatering Screens - to remove water.
The number, style and types of screens will vary by plant depending
on:
• nature of the coal deposit;
• method of coal mining;
• method of cleaning;
• tonnage;
• crushing requirements;
• desired product; and
• availability of water.
Scalping
The run-of-mine (KM) scalping screen can play an important role in
a plant by reducing the size and number of crushers required. Since
efficiency is critical to crusher sizing, the vibrating scalper is normal-
ly preferred over the stationary bar screen.
Normally, minus 610 mm. (24 in.) RDM coal, which can include rock,
slate and other foreign materials, is brought from the mine by trucks
or conveyors and fed to the scalper, with oversize going to the crusher
and fines by-passing the crusher.
The method of mining affects the material size analysis of the feed to
the POM scalper, thus affecting the screen size. The screen size is also
affected by the percentage of moisture in the ROM coal. Moisture can create
a balling situation, with fines sticking to large lumps.
34
-------
The RCM scalper is usually a heavy-duty, mechanically vibrated, single
deck, inclined-type, with circular stroke, and with a feed box having liners
to receive initial material impact. The feed box on a screen can be import-
ant in reducing decking replacement costs, and also help spread material over
the deck for more efficient operation.
The decking must be capable of handling impact and abrasion, so scalper
decking is generally composed of skid bars, tapered and self-relieving, made
of manganese steel, abrasion resistant (AR) steel or polyurethane. A per-
forated plate is also widely used with skid bars.
Deck openings are usually from 15 to 30 cm (6 to 12 in.) depending on
size analysis of ROM coal. The coal passing through the scalper bypasses
the crusher, oversize going to the size-reduction machine and both materials
combining for conveyance to the raw coal screen or to storage or loadout if
no further sizing or coal processing is required.
Raw-Coal Sizing
In some plants, due to the method of mining, a higher percentage of
coal reports as fines, the top size usually being less than approximately
250 mm. (10 in.) - The RCM scalper is not required in this case, so the
first station will be the raw-coal sizing screen.
The raw-coal sizing screen receives combined feed from the primary
crusher and/or the ROM scalper screen, usually minus 250 mm. (10 in.)
coal, and separates the raw coal into coarse and fine sizing for further
processing. It is normally a heavy-duty, mechanically vibrated, double-
deck, inclined type, circular stroke machine with a lined feed box to
absorb impact.
The top deck is usually a steel perforated plate. Recent legislation
on noise emissions strongly suggests that polyurethane and rubber be
considered for its noise-deadening effect. The top deck size has approxi-
mately 38 mm. (1/2 in.) openings.
The bottom deck is most commonly of wire cloth with approximately 8 mm.
(5/16 in.) openings; but polyurethane and rubber are coming into use to
cut noise.
35
-------
The plus 8 mm. (5/16 in.) coarse material is directed to the ooarse
ooal cleaning plant, where the first station is the pre-wet screen. The
minus 8 mm. (5/16 in.) material goes to the fine cleaning plant and the
desliming screen. In plants where no further cleaning is needed, the coal
goes to the final sizing screen, loadout or storage.
Pre-Wet Screen
Ihe pre-wet screen employs a water spray to remove fines before the
coal is sent to the mechanical cleaning process. Ihis screen receives a
feed of coarse ooal from the raw coal screen. Usually, 10 to 20 liters/
minute (3 to 5 gpm) per ton per hour (of coal) spray water is directed
onto the coal. Fines are washed through the screen and directed to the
fine coal desliming screen.
Normally the pre-wet screen is a heavy-duty, mechanically vibrated,
double-deck, horizontal machine with a straight line stroke at 45°, and a
lined feed box with water spray bars directed to both decks. Other acces-
sories required are sideplate drip angles to prevent water spillage outside
the through hopper.
The purpose of a double-deck is to reduce the depth-of-bed on the screen
to accentuate feed stratification and thus assist the fines in being washed
through the bottom deck. Top decking is of wire cloth or polyurethane
rubber with approximately 25 mm. (1 in.) openings. The bottom deck is
usually 1^-304 stainless steel profile decking or polyurethane slotted with
a 1 mm. clear opening.
The coarse raw coal oversize is now directed to the cleaning device
thich is either a jig or heavy-media bath. The fines washed through are
directed to the fine cleaning plant. (It should be noted that in some
plants the raw coal sizing and pre-wet screens can be combined into one
common, inclined-type machine).
Haavy Media Recovery Screens
The purpose of the heavy-media recovery screens is to receive the
coal, middlings, and refuse from the heavy-media bath and, with water
sprays, wash the magnetite off larger particles and lumps for recovery and
36
-------
revise. (Magnetite is expensive and recovery is critical to plant operating
costs.)
The heavy-media, drain-and-rinse screen is usually a heavy-duty, mecha-
nically vibrated, double decked horizontal-type machine with straight line
stroke at 45°, and sprays directed to both decks. A cross-flow screen or
sieve bend installed ahead of the screen is commonly used in place of a feed
box. The sieve bend can reduce the required screen size up to 20 percent.
The top deck is of wire cloth, polyurethane or rubber, with an opening
approximating 25 mm. (1 in.). Ihe bottom deck is usually T-304 stainless
steel or polyurethane slotted to effect a 1 mm. (1/32 in.) separation that
will pass the magnetite, but retain the coal. Coarse clean coal, coarse
middlings or coarse refuse directed to dewatering screens.
Desliming Screens
The first station in a fine-coal cleaning plant is the desliming screen,
which is normally preceded by a stationary sieve bend or cross-flow screen.
The desliming screen receives a feed of minum 8 mm. (5/16 in.) raw coal
from the pre-wet screen.
The desliming screen is usually a heavy-duty, mechanically vibrated,
single-^deck machine with a straight line stroke at 45°. The screen's depth
of bed is held to a minimum so that fines can be easily washed through.
Dewatering Screens
Washed coal from wet cleaning operations contains large volumes of
water which must be removed from the coal to meet market requirements.
Dewatering screen selections are based on average particle size and
the bed depth of the coal. A detailed discussion of dewatering screens
is given in Section 5 of this report.
Tables 3 and 4 summarize screen usage.
JIGS
Jigs rely on the process of particle stratification due to the range
of specific gravities represented by the coal and impurities. The strati-
fication results from repeated expansion and compaction of a bed of particles
37
-------
TABLE 3
VIBRATING SCREEN EQUIPMENT FOR COAL PREPARATION SERVICE
00
Coarse Coal Cleaning Plant
Type Number of Decks Installation Aperture (s)
Angle
ROM Scalper Single 17-25° 15.2 on
Raw Coal Double 17-25° 3.8cm
Sizing
Screen
8 iron
Prewet Double Horizontal 3.C cm
Screen
T rnn ""
Heavy Media Double Horizontal 2.5 cm
Drain
and
Rinse Screen 1 ram
Dewatering Single Horizontal 1 ram
Screen
Screen Section
Type
Manganese Skid
Bars AR Perfor-
ated Plate with
Skid Bars
AR Steel Pefor-
ated Plate Poly-
urethane Rubber
Polyurethane
Wire 304 Stain-
less Profile
Rubber
Wire
Polyurethane
Rubber
Stainless Steel
Profile Deck
Polyurethane
Wire
Polyurethane
Rubber
304 Stainless
Steel Profile
Decks
Polyurethane
304 Stainless
Steel Profile
Deck
Polyurethane
-------
CO
vo
TABLE 4
SCREEN EQUIPMENT FOR COAL PREPARATION S"""ICE
Fine Coal Cleaning Plant
Type Number of Decks Installation Aperture
Angle
Deslirne Single Horizontal 1/2 ran
Screen
Heavy Media Single Horizontal 1/2 ran
Drain and
Rinse
Screen
Dewatering Single Horizontal 1/2 ran
Screen
Screen Section
Type
304 Stainless
Steel Profile
Deck
Polyurethane
304 Stainless
Steel Profile
Deck
Polyurethane
304 Stainless
Steel Profile
Deck
Polyurethane
-------
by a pulsating fluid, usually water. In the pulsation stroke, water moves
upward through the particle bed and causes it to expand. Water moving
downward through the bed causes compaction (suction stroke). During the
pulsation stroke the particles move upward in relation to their specific
gravities, i.e., fragments of lighter specific gravity material move a
greater distance. During the suction stroke the lighter specific gravity
materials descend at a rate slower than their heavier counterparts. After
such cycles, particle stratification within the bed results.
Aside from the mechanical system designed to produce and control the
pulsation-suction cycles, two other systems comprise the typical jig. They
are:
• a system to convey the unsorted material into the machine; and
• a system to separate the stratified bed into product and refuse
and to remove each of these fractions.
Differing methods for producing the pulsation-suction cycles distinguish
the three main types of jigs:
• plunger-type - water movement is caused by the reciprocating
motions of a plunger in a compartment of the jig;
• basket-type - in which the box containing the bed is moved up
and down in standing water; and
• air pulsated-type—in which the tank is constructed in the
shape of a "U". Water reciprocation is caused by applying
and releasing compressed air to the closed end of the "U".
Baum Jigs
By far the most popular of the three jig types in the coal industry is
the air pulsated-type, called Baum Jigs. A typical Baum Jig is shown in
Figure 11. Since their development in Germany more than 75 years ago and
their subsequent introduction into the United States in 1928, the use of
Baum Jigs has risen to the point where they now have 75 percent of the coal-
jig market.
40
-------
Baum Jig Operation—
The Baum jig is essentially a "U" shaped box having one end of the "U"
sealed. On the other side, near the top, are perforated screen plates,
whose function is to support the bed of coal. The screen plates have 6.4 ram.
(1/4 in.) holes, although openings up to 13.8 mm. (1-1/4 in.) have been used.
The area below the screen in the lower part of the "U" is called the "hutch
conpartment."
Most Baum jigs have a screw conveyor situated at the bottom of the
hutch compartment to move fine refuse, which has fallen through the screen
plates, to a bucket elevator.
Air Inlet
Air Chamber
Water Pulsation
Screen
FIGURE 11. BAUM JIG CROSS SECTION.
On the side opposite the screen plates is a sealed compartment where the
water pulsations are initiated. This chamber is fitted with intake and ex-
haust valves for the admission and exit of compressed air. When the intake
valve opens the water is forced upward through the screen plate and the bed
expands. With the exhaust valve in the open position, the water returns to
its original level. The principal reason for the remarkable success of the
Baum jigs over the plunger-type jigs is the control capability of the "back
suction", which is unique to the Baum machine. The expanded bed returns to
its ccnpacted state under the influence of gravity; the rate of particle
descent being controlled by varying the release rate of air from the pulsa-
41
-------
tion chamber. Thus, optimum conditions are obtained much more readily than
with the plunger-type.
The jig box is divided horizontally into compartments by fixed vertical
weirs designed to control the flow of the bed laterally across the machine.
Each oorpartment is actually a jig within a jig; as each one has the means
for discharge of refuse and exit of clean coal, either to the next compart-
ment or to a conveyor.
To separate the strata into refuse and product, the bed is divided along
a horizontal plane, by refuse ejectors. The refuse is sent to a bucket
elevator either via the final refuse ejector or via the screw conveyor men-
tioned earlier.
Finally, Baum jigs have a mechanism for controlling the relative amount
of refuse withdrawn, therefore controlling the quality of the cleaned coal.
This is necessary because of the lack of a sharp interface between refuse and
product; rather, there is a gradational contact. The mechanism consists of
a float immersed in the coal-water bed near each point of refuse removal.
The float acts as a hydrometer, thus determining the amount of refuse at that
level. The float height level varies with specific gravity of the bed at the
set location and actuates mechanical devices to change the rate of refuse
withdrawal.
Baum Jig Performance—
The jigging process works well provided that:
• The feed is not closely sized;
• The jig is not loaded so heavily that particle interaction
prevents stratification; and
• Ultra-fine coal or mineral particles do not greatly effect
the natural viscosity of the water and hence the upward and
downward movements of particles within the bed.
Efficiency of a jig is often depicted by a distribution curve as
described in Section 2 of this report. A generalized distribution curve
for a typical Baum-type jig is shown in Figure 12.
42
-------
r 100
GO
O
03
Q.
o
o
UJ
o
o
h-
(T
h-
l/}
Q
80
60
40
20
0
0.7
I
0.8 0.9 1.0 I.I 1.2 1.3
REDUCED SPECIFIC GRAVITY
1.4
1.5 1.6
FIGURE 12 GENERALIZED DISTRIBUTION CURVE FOR BAUMJIG TREATING
A 15 cm (6 in.) BY 48 - MESH COMPOSITE FEED
-------
Itecent technology—Perhaps the most significant advancement in recent
jigging technology is the joint development of the Batac jig by the Germans
and Japanese (see Figure 13) for fine coal preparation. This device pro-
duces water pulsations directly beneath a feldspar bed screen in multiple,
independently controlled chambers, which are uniformly distributed through-
out the jib. The jib utilizes electronically controlled, flat-disc design
valves for sharp cutoff of input and exhaust. These valves can be adjusted
for speed and length of stroke, thereby permitting the specific gravity of
separation to be varied from 1.60 to 2.00. Bed widths are available up to
5 meters (16 feet). Capacities for processing fine coal, 13.7 mm. (1/2 in.)
range up to 360 kkg (400 tons) per hour and for processing coarse coal,
127 x 10 mm. (5 x 3/8 in.), up to 720 kkg (800 tons) per hour.
Advantages of the Batac Jig over conventional jigs are claimed to be:
• reduced buildingwspace requirement;
• lower equipment weight; and
• cleaning ability extending over a wider range of particle sizes.
WET C3CNCENTRATING TABLES
Today's wet concentrating tables are an outgrowth of an evolutionary
process that began years ago. The introduction of suspended, multiple-deck
tables in the late 1950's and early 1960's is the most recent development
of real significance in the manufacture of concentrating tables. To a
large extent, this development has eliminated two of the primary disadvant-
ages of concentrating tables, the need for large amounts of floor space
and the need for massive concrete foundations to absorb the impact of the
drive mechanisms.
Concentrating tables flow a slurry of pulverized coal and water over
an inclined riffled deck which shakes rapidly effecting a separation of
the particles—by size and specific gravity.
Table Operation
Because of the reciprocating action of the table and the transverse
flow of water, the water and coal pulp fans out upon contacting the table
44
-------
WATER PULSATION
AIR INLET
'AIR CHAMBER
SECTION A-B
FIGURE 13. BATAC JIG CROSS SECTION.
surface. The upward slope of the table toward the refuse end, and the
retaining effect of the pool riffles cause the slurry to form a pool near
the feedbox. In the pool, the bed of material is several particles deep
and substantially above the standard riffles; this area becomes the zone of
primary stratification. In this zone the shaking motion of the deck, com-
bined with the cross current of water, stratifies the particles by density,
similar to the action of a jig.
The particles that make up the feed become arranged so that the
coarsest and heaviest particles are at the bottom and the finest and
lightest particles are at the top. The heaviest particles are carried by
the table movement towards the refuse end at a faster rate. The lighter
pieces ride on the top layer of particles and flow down the slope of the
deck as a result of the cross flow of wash water which is at right angles to
45
-------
the shaking movement of the table. Since stratification and separation of
particles are not complete as a result of any one riffle, a series of riffles
is used, repeating the cycle of stratification and hindered settling from
riffle to riffle, thus obtaining purer refuse products as the particles fan
out and progress forward and downward over the table. Conversely, the
purer, cleaner coal is discharged at the drive end of the table.
As shown in Figure 14, successive samples collected along the side and
end of the table, starting at the head-motion end, show a steady increase in
ash content and a steady decrease in the average particle size for each in-
dividual specific-gravity fraction.
Concentrating tables are provided with a number of adjustments which
should be used to obtain the best possible operation. Among these are:
• speed;
• length of stroke;
• feed rate;
• amount and distribution of wash water;
• water-to-solids ratio of the feed pulp;
• uniformity of feed size;
• riffle design;
• side tilt; and
• end elevation.
The oscillation of the deck usually is 260 to 290 strokes per minute,
depending on the characteristics of raw coal and the feed rate. If there
are high percentages of refuse in the raw coal or if the feed rate is high,
an increase in the frequency is required.
Closely related to the frequency is the amplitude, or length of the
stroke. The amplitude and frequency are varied to maintain the mobility of
the bed necessary for stratification while retaining the coal on the deck long
enough for proper separation. To move large quantities of refuse material
along the deck, an amplitude as long as 32 mm. (1 1/4 in.) may be re-
quired. The amplitude and frequency of the stroke are decreased as the
amount of near-gravity material in the feed increases. Generally, a fine
feed will require a higher speed and shorter stroke than a coarse feed.
46
-------
LOW GRAVITY CLEAN COAL
MIDDLING (HIGH SULFUR COAL)
HIGH GRAVITY REFUSE
FIGURE 14 THE DISTRIBUTION OF TABLE PRODUCTS
BY PARTICLE SIZE SPECIFIC GRAVITY
47
-------
The cross slope and amount and distribution of wash water to the table
can be changed easily and quickly to compensate for minor variations in feed
rate and composition.
Perhaps the most important of all table adjustments is the end elevation
or the amount of upward inclination of the deck measured along the line of
motion from the feed end to the discharge end. By creating a moderate slope
that the high specific gravity particles will climb more readily than the low
specific gravity minerals, separation is greatly improved.
Table Performance
Table capacity varies with the size consist, the percentage of reject
contained in the feed and the washability of the table feed. Coarser feeds
are processed at higher rates than are finer feeds and feed rates will be
limited by the percentage of rejects above 25 percent. The majority of
table installations utilized in bituminous coal are fed 9.5 x 0 mm.
(3/8 x 0 in.), or 6.4 x 0 mm. (1/4 x 0 in.) or deslimed fractions of a
larger top size. The capacity per double-deck table is 23 kkg/h (25 tph)
feed; i.e., 11.2 kkg/h (12 1/2 tph) per deck. For 19.1 or 12.7 mm.
(3/4 or 1/2 in.) top size, commonly handled when cleaning steam fuels,
a capacity of 27 kkg/h (30 tph) per twin-deck table can be expected.
HYDRQCYCLONES
Hydrocyclones separate coal from impurities by radially accelerating
a stream of slurry and causing a centrifugal as well as gravitational
force to act on the material. A separation results producing clean coal
(overflow) and refuse (underflow). The overflow will have a relatively
low sulfur content and a correspondingly small percentage (3 to 9%) of
misplaced material. The underflow will have a relatively high sulfur
content and a high percentage (often around 50%) of misplaced coal.
Hydrocyclones, when used to treat fine coal of less than 6.4 imu (1/4 in.),
are operated at low pulp densities of 5 to 20% solids. In the United
States, hydrocyclones are presently used to clean flotation-size coal of
.42 mm. and smaller (35 mesh), but they can be used on coal as coarse as
6.4 mm. (1/4 in.). When the top size of the feed exceeds 6.4 mm., the
reduction of ash and sulfur in the minus .149 mm. (100 mash) fraction is
minimal.
48
-------
Ihe hydrocyclone (waters-only cyclone) differs from the conventional
dense-medium cyclone as follows:
• hydrocyclone does not employ an artificial gravity suspension, but
separation takes place in an autogeneous water medium;
• hydrocyclones have an apex angle of up to 120°, much greater than
the apex angle of the dense-medium cyclone;
• the hydrocyclone has a longer vortex finder than does the
dense-media cyclone; and
• the hydrocyclone uses much larger underflow and overflow
orifices than does a dense-media cyclone of equal diameter.
The hydrocyclone, as a fine coal cleaning device, has advantages over
other cleaning devices. Among these are:
• Efficient cleaning of oxidized raw coal of plus .149 mm. (100 mesh).
• Higher pyritic sulfur reduction in .595 x 0 mm. (28 x 0 mesh) coal
than that normally accomplished lay flotation;
• Simple design incorporating no moving parts and requiring minimal
maintenance;
• low capital investment on a cost-per-ton-processed basis;
• Very little plant space required relative to capacity.
Hydrocyclones have some disadvantages in that:
• separations obtained in a hydrocyclone are not as sharp
as those obtainable by certain other processes.
• clean refuse and clean coal cannot be simultaneously produced.
Ihe clean coal may be contaminated by high ash fines. In such
cases, the hydroclone product must be desldmed in a cyclone and
the fines cleaned by flotation.
• hydrocyclones are not applicable to processing difficult-to-clean
coal. For an easily cleaned coal it can do as good a job as a
table, but can not compete with a feldspar jig. For good clean-
ing, two-stage separation must be utilized; and
• the large quantity of water required for proper cyclone perform-
ance means that a hydrocyclone circuit will be energy intensive.
49
-------
Hydrocyclone Operation
Figure 15 shows a typical hydrocyclone.
The hydrocyclone body consists of a short cylindrical section which
is attached to an inverted truncated conical section. The underflow orifice
is at the apex of the conical section and the overflow orifice (vortex
finder) is centrally located at the top of the cyclone. A slurry under
pressure is introducted tangentially into the cylindrical section of the
cyclone where centrifugal forces act on the particles in proportion to
their mass. The centrifugal forces acting on the particles increase with
decreasing radii, as the slurry moves downward into the conical section of
the cyclone. The heavier, more dense (heavy mineral matter and pyritic
sulfur) particles of a given size move outward toward the descending vortex
more rapidly than their lighter, less dense counterparts. As the lighter
particles approach the apex of the cone, they are drawn into the upward-
flow vortex next to the central air core where they exit through the over-
flow as clean coal. The heavier particles, which flow along the cyclone
wall, exit through the underflow orifice as refuse.
The cyclone manufacturer will generally determine the cyclone diameter
and apex angle for a particular situation. The orifice sizes will influence
flow rate, retention time, and solids concentration in the overflow and
underflow. The length of the vortex finder is the determining factor in
establishing the specific gravity separation point between the underflow
and overflow.
Due to the large amounts of relatively low specific gravity, misplaced
particles in the underflow, a two-stage or even three-stage hydrocyclone
treatioant can be very effective in improving the overall efficiency of the
circuit. When staging, the underflow product (refuse) from the primary
hydrocyclone is the feed to the second cyclone. The overflow from the
secondary hydrocyclone joins the washed coal from the primary unit to form
the final washed-coal product; the refuse from the secondary hydrocyclone
is the final refuse material. In three-stage processing the procedure
is repeated between a second and third hydrocyclone.
50
-------
FIGURE 15. A Typical Hydrocydone
FEED SLURRY
VORTEX
OVERFLOW
UNDERFLOW
Hydrocyclone Performance
Generally, hydrocyclone diameters range from 50 to 610 imu (2 to 24 in.)
Feed rates can reach 61 kkg/hr. (68 tph). Up to 25 hydrocyclones can be
arranged in a single bank with a common manifold in the feed and overflow
lines. Generally smaller feed particle sizes require a smaller hydrocyclone
diameter.
HEAVY MEDIA CYCLONES
The first heavy media cyclone plants were installed in the United
States in 1959. Their use has spread rapidly in the preparation of
premium metallurgical coals. New areas for application are being investi-
gated because of the intense interest in preparation of low ash and low
sulfur fuels from coal for power generation. Heavy media cyclones effect
the low gravity separations required to produce these fuels. They will
also be used in preparation of fine coal of minus .595 mm. (28 mesh) where
they have been only sparingly applied in the past.
51
-------
Heavy itedia cyclones are the most efficient devices available for
cleaning of intermediate size coal. In addition, their ability to produce
precise separations makes them almost a necessity for cleaning coals with
a high percentage of near-gravity material (±0.10 specific gravity range).
Cyclones for heavy media operation are 300 to 760 ran. (12 to 30 in.)
in diameter and have high capacities. A 660 ram. (26 in.) cyclone can
process up to 108 kkg/hr. (120 ton/hr.) of feed.
Ifeavy Media Cyclone Operation
When coal is crushed below about 6.4 mm. (1/4 in.), the difficulty of
gravity separation increases, because the time required for any particle to
settle in water is dependent upon its specific gravity and the resistance of
the water to the settling of the particle. The larger the particle, the
faster the sinking rate in proportion to the fluid resistance mass ratio.
As particles become smaller, settling times are increased and can be reduced
only by the application or force to them. In a cyclone, this force is
brought to bear in a centrifugal manner by introducing raw coal and water
tangentially into the upper part of the cyclone. In a typical cyclone the
centrifugal force acting on a particle in the inlet region is about
20 times greater than the gravitational force in a static bath. As the
feed descends in the conical section of the cyclone, the centrifugal
force is further increased and may approach over 200 times the force of
gravity at the apex. At this point, the cyclone has accomplished a size
classification of the particles with the larger particles at the perimeters
of the cyclone and the smaller particles near the center of the dense
medium cyclone. Figure 16 is a diagram of a dense media cyclone.
To achieve a gravimetric classification in a heavy media cyclone, the
water is made more dense by the addition of finely-ground magnetite with the
resultant effect that particles having a higher specific gravity are forced
to the perimeter of the cone and passed out through the apex as refuse,
while the particles of lesser specific gravity remain near the vortex finder
and pass out through the top of the cone as clean coal.
52
-------
FEED-t
WASHED COAL
1 FEED INLET
2 OVERFLOW CHAMBER
3 WASHED COAL OUTLET
4 CYLINDRICAL SECTION
5 CONICAL SECTION
6 REPLACEABLE UNDERFLOW ORIFICE
7 VORTEX FINDER
FIGURE 16 A DENSE MEDIUM CYCLONE
The general flow pattern of the medium in a cyclone consists of a de-
scending vortex that originates at the inlet and progresses through the
cyclone to the underflow outlet. As the descending vortex passes down the
cyclone, part of the fluid flows toward the center of the cyclone to form an
ascending vortex. This ascending vortex, in turn, surrounds a cylindrical
air core that encircles the entire longitudinal axis of the cyclone. An
additional factor that influences the separation is the progressive increase
in specific gravity of the medium as it descends toward the apex. This in-
crease occurs because the centrifugal force also tends to force the magnetite
particles in the medium towards the cyclone wall where they are preferentially
caught in the descending vortex, thus resulting in a progressively higher con-
centration of particles in the medium as the apex is approached. Therefore,
the specific gravity of the medium flowing through the underflow orifice is
higher than the specific gravity of the circulating medium; conversely, the
specific gravity of the medium passing through the overflow orifice is less.
The paths followed by the coal and impure particles in a cyclone
have been studied by observations in glass or clear plastic cyclones and are
still not fully understood. The refuse particles flow to the wall soon after
they enter the cyclone, are entrained in the descending vortex and are dis-
charged through the underflow orifice. The coal particles are also initially
entrained in the descending vortex in the upper part of the cyclone, and
53
-------
curiously, a large number of than descend well into the conical part of the
cyclone before they are trapped by the ascending vortex. This behavior has
been explained by postulating a barrier of high specific gravity that is due
to circulating medium particles in the lower part of the cyclone. When the
descending coal particles reach this zone, they migrate toward the central
air core where they are then caught in the ascending vortex and pass through
the overflow opening. The existence of a barrier, however, cannot entirely
explain the path of the coal particles because observation of the particles
in a glass cyclone when using an organic heavy liquid shows that they behave
similarly; that is, many coal particles descend well into the conical sec-
tion before they migrate to the ascending vortex. Clearly, a heavy organic
liquid is homogeneous and a barrier cannot be present, yet the separation is
very sharp. It is also interesting to note that the specific gravity of
separation is almost always higher than the specific gravity of the medium
when using either a heavy liquid medium or a magnetite dense medium.
Heavy Media Cyclone Performance
Table 5 provides operating results and test data on both washed coal
and refuse, along with the distribution curve or product distribution of wash-
ed coal and refuse. The separation for this particular coal was made at a
specific gravity of 1.35 and a 6% ash product was obtained. It is to be
noted that the near gravity material (+ or - 0.10%) was 77.3%. The washed
coal has no misplacanent of material above 1.45 specific gravity while the
refuse contains very little minus 1.30 specific gravity material.
Recent Technology
The swirl heavy media cyclone offers a new approach in that it is an
inverted cyclone which was patented by Dutch State Mines and developed fur-
ther in Japan by the Tagawa Machinery Works. The conical portion of the
cyclone is at the top and the cylindrical portion at the bottom, which results
in the refuse discharging from the top and the clean coal from the bottom.
This cyclone is said to be able to use a coarser magnetite which can result
in lower magnetite loss. The air column inside the conventional cyclone is
sometimes unstable; in the swirl cyclone this air column is stablized by an
air pipe which is open to the atmosphere. The design is such that the height
54
-------
(Jl
Ln
TABLE 5
HEAVY MEDIA CYCLONE OPERATING RESULTS FOR 31.75 x 0.75 Itm. RAW COAL FEED (1 1/4 x 1/32 in.)
Reconstituted Raw Coal
Gravity
Sink
1.25 x
1.30 x
1.35 x
1.40 x
1.45 x
1.50 x
1.55 x
1.60
Float
1.25
1.30
1.35
1.40
1.45
1.50
1.55
1.60
Direct
% Wt.
5.5
39.6
17.1
12.2
8.4
2.7
3.6
1.8
9.1
100.0
% Ash
2.32
5.02
9.24
13.70
18.74
21.60
24.70
31.00
50.80
13.59
Cumulative
% Wt.
5.5
45.1
62.2
74.4
82.8
85.5
89.1
90.0
100.0
% Ash
2.32
4.69
5.94
7.21
8.38
8.80
9.44
9.87
13.59
Hashed Coal
Feed
% Wt.
5.5
38.7
11.8
3.0
0.2
0.0
0.0
0.0
0.0
59.2
Distribution Curve
or
Product Distribution
Refuse Feed
% Ash %
2.32 0
5.03 0
8.95 5
12.18 9
16.10 8
2
3
Wt.
.0
.9
.3
.2
.2
.7
.6
1.8
9
5.96 40
.1
.8
% Ash
—
4.5
9.9
14.2
18.8
21.6
24.7
31.0
50.8
24.67
Washed
Coal
100.0
97.7
69.0
24.6
2.4
0.0
0.0
0.0
0.0
Refuse
0.0
2.3
31.0
75.4
97.6
100.0
100.0
100.0
100.0
The above performance at 1.35 specific gravity shows a washed ooal recovery of 59.2% when obtaining a 6%
ash product. These results are very good considering the wide size range of 31.75 x 0.75 mm. (1 1/4 x 1/32 in.)
being handled in the cyclone and the extreme amount of near gravity material (77.3%) at the separating gravity.
-------
of the air vent can be varied, as can the height of the vortex finder and the
diameter of the apex nozzle. The feed nozzle is also designed so that the
area of the opening can be adjusted by a regulating plate. This cyclone is
said to provide a sharpness of separation and efficiencies comparable to con-
ventional cyclones and nay offer possibilities in treating larger coal sizes.
THE OTISCA PROCESS
A discussion of the Otisca Industries, Ltd. process for coal benefica-
tion is based on those theories which control the process of sink-float
separation. The separation of coal (1.20 - 1.45 specific gravity) from the
undesirable mineral matter (ash of 2.0 - 2.5 specific gravity, and iron
pyrite of 4.9 - 5.0 specific gravity) can be accomplished by separating
particles of each in a liquid which has a specific gravity midway between
the components to be separated.
Stoke's Law states that the velocity of sink-float separation is in-
versely dependent on the viscosity of the separating fluid and directly pro-
portional to the square of the particle size, as well as directly proportion-
al to the difference in specific gravity between the coal and the separating
fluid. The Table below presents the viscosities of some common separating
liquids:
Viscosities of Common Separating Liquids
Viscosity at 20 °C in Centipoise
Carbon tetrachloride 0.969
Tetrachloroethane 1.844
Water 1.00
Bromoform 2.152
-325 mesh magnetite and water (fresh) 4.5
Otisca Media 1.4
An added effect is due to the surface tension of the separating fluid.
Surface tensions of common separating liquids are given below.
56
-------
Surface pensions of Cannon Separating Fluids
2S
Surface Tension in
Carbon Tetrachloride 21
Tetrachloroethane 36
Water 75
Bromoform 41.5
-325 mesh magnetite and water 75
Otisca Media Range of 20
If the coal is not completely wetted by the separating media, entrapped
air bubbles on the coal and on the ash matter tend to move both particles
toward a common density, irrespective of their inherent bulk densities. As
a consequence, separation becomes more difficult. Clearly, there is a major
problem in separations when the particle size of the coal decreases below
one millimeter. Liquids with properties which tend to permit the free sur-
faces (micro-cracks, etc.) of coal and rock to become wetted instantaneously
and completely will have a distinct advantage over the other liquids in that
they will tend to accentuate differences in specific gravity.
The loss in separation efficiency (often 10 to 30%) commonly experienced
during the transfer of "theoretical" washability data to the conventional
hydro-beheficiation systans is due, in part, to the fact that a water-magnetite
suspension has a higher viscosity than that of the heavy liquids used in lab-
oratory testing. Since the Otisca "media-only" Process uses the low-viscos-
ity fluid directly, little or no efficiency loss occurs between bench scale
testing and actual production.
An actual case involving settling-pond coal particles is illustrated
below. The size consist of the coal is given below.
Size Consist of Settling Pond Coal Fines
— Cumulative %
+0.25 mm. (+60 mesh) 0.6 0.6
+0.15 mm. (+100 mash) 2.0 2.6
+0.066 mm. (+230 mssh) 21.4 24.0
-0.066 mm. (-230 iresh) 76.0 100.0
Moisture 10 - 30%
Ash 20 - 30%
Sulfur* 2%
57
-------
The following results were observed after cleaning this coal with the
Otisca "media-only" Process:
Weight Yield 80% (wet coal)
Ash 8%
Sulfur Below 1.5%
These data demonstrate that the ultraf ine clay particles which contami-
nate the product coal have been separated.
The Otisca "media-only" beneficiation Process does not use any process
water and the resultant coal product contains about the same or less moisture
concentration as that which was charged into the process.
The Otisca Process makes the following claims:
• sharp separations even when processing fine coal;
• specific gravity of separation very near to that of
the clean coal itself; and
• separations unaffected by the size consist of the feed
coal.
Otisca Process Flow Sheet Description
The flow sheet presented in Figure 17 describes the main operations
involved in an Otisca "media-only" coal beneficiation plant.
Raw coal first enters a mechanical caminution stage, designed to optimize
the release of sulfur and ash. The caminution can optionally be carried out
in air or in a bath of media liquid, depending on the specific morphology of
the coal being cleaned. Next, the conminuted coal is conveyed to a static
bath of media liquid. Finally, the float-sink fractions are separately con-
veyed to respective media liquid stripping and drying operations, where
virtually all of the media is removed from the coal and refuse and delivered
to a media regeneration station. The regenerated media is then delivered to
the separation bath.
58
-------
/?AW COAL
MfCHAMCAL
BY
t/v SrAT/c
Meow BATH
FIGURE 17. Basic-Media-Only Coal Beneficiation Plant
59
-------
Performance
The Otisca Process has undergone pilot plant testing and the concept
utilizing an organic liquid of low viscosity has proven to be workable. The
major problem encountered to date has been the cost of media lost during the
process. Cost per ton of feed coal currently runs around $1.50.
CONE CONCENTRATORS
Cone concentrators were first developed in the early 1960's by Mineral
Deposits Ltd. of Australia, where they were used on mineral-bearing beach
sands. They are now being investigated for possible applications in coal pro-
cessing. The high capacity of 36 to 90 kkg/hr. (40 to 100 tph) at 55 to 70%
solids by weight, the low installation cost (due to modular desion, minimal
pumping distances and a minimal number of circuit stages), and the low operat-
ing costs (due to the absence of moving parts and low operator loading) com-
bine to make the cone concentrator a viable alternative to some of the equip-
ment currently being used in the United States, especially in the area of
secondary cleaning.
Cone Concentrator Operation
Cone concentrators use the principle of flowing-film concentration.
The basic separation element in the cone concentrator is a 1.90 meter
(75 in.) diameter, 17° inverted cone. Feed pulp is evenly distributed
around the top of the distribution cone. As the feed pulp flows down to
the periphery of the distribution cone, a rubber "finger" splitter evenly
divides the feed between two concentrating cones that operate in parallel
and are superimposed one above the other (the concentrating capacity in a
given floor area is therefore doubled). As the pulp flows toward the center
of the concentrating cone, the fine and heavy (pyritic sulfur) particles
separate to the bottom of the film and are removed by an annular slot in
the bottom of the concentrating cone. The efficiency of this process is
relatively low and is repeated several times within a single machine to
achieve an effective separation. The clean coal flows over the discharge
slot and passes into a central discharge pipe for repeated processing.
Cone concentrators accept feed particles in the ranges of 3.3 to
0.04 mm. (6 to 400 mesh); the optimal size being about 0.59 mm. (28 mesh).
60
-------
The best separation results are obtained when the feed tonnage, density
and character are not varied. Variations can be tolerated for short
periods of time, but only at the expense of recovery and wear rate.
VDR-SIV
The Vortex Dewatering Sieve, commonly known as a Vor-Siv, was developed
by the Polish Coal Industry to fit the need for a unit that would dewater
and classify slurries of high solids concentration. Today this device is
manufactured under license from the Polish government and, as of 1976, there
were 36 Vor-Sivs operating in the U.S.A., with another 20 in the process of
being installed or delivered. The Vor-Siv combines the operating characteris-
tics found in cyclones, sieve bends, cross flow screens, and vibrating
screens. It has no moving parts, and requires no energy to operate.
Vor-Siv Operation
Classifying of solids by the Vor-Siv is accomplished by the spiraling
or vortex flow of a slurry over a stationary, inverted cone-shaped, wire
screen. Material to be classified is gravity fed, or pumped, in slurry form
through the inlet nozzle. A covered trough directs the flow around a cir-
cular raceway and down onto a fine screen positioned at a 45° angle to the
raceway.
The vortex motion, created by centrifugal and gravitational forces,
causes the solids in the slurry to change from a radial flow to a downward
spiraling path on the surface of the inverted cone screen. A discharge out-
let is provided for the coarse material.
The undersize solids, with a topsize of less than 0.59 mm (28 mesh),
and the accompanied liquid circulate in an almost horizontal path and are
discharged through slits in the screen.
Wear is a major concern in all fine screening devices. While the
wedgebars in the Vor-Siv are much heavier than the wire used in fine screen
cloth, wear does occur and increase the gap between bars. Wear and blinding
(blocking of screen holes by near size particles) must be monitored to
assure efficient operation. The direction of flow can be changed by flip-
ping the inlet nozzle which presents the slurry with a new leading edge
61
-------
and therefore extends screen life. The basket can also be rotated either
counterclockwise or clockwise so that every area is corrpletely utilized.
Vor-Siv Performance
To date, the Vor-Siv has not been applied to the separation of sulfur
and mineral natter from raw coal. For classifying, the Vor-Siv has received
feeds to minus 8.3 mm. (5/16 in.) solids in slurries ranging from 10 to
30 percent solids and feed rates up to 200 I/sec (3,200 gpm). Dry solids
have been successfully treated at a rate of 135 kkg/hr. (150 ton) in a
single 2-meter-diameter (6.6 ft.) unit.
In general, a sieve bend or cross flow screen making a 0.59 mm. (28
mesh) separation can be expected to handle about 20 to 27 1/sec/m2 (30 to
40 gpm/ft2) of wire surface, whereas a Vor-Siv will handle from 35 to 48
1/sec/m2 (50 to 70 gpm/ft2) of screen area.
SPIRAL OMENTRATORS
Spiral concentrators are simple devices for separating pyritic sulfur
and asn from coal in accordance with their physical properties, primarily
specific gravity and particle size. Although not widely used in coal pre-
paration, tests on fine coal 4.7 to 0.4mm (4 to 35 mesh) suggest they should
be considered along with wet concentrating tables and hydrocyclones as a
"rougher" cleaning unit. They could be used to remove pyritic sulfur from
high-sulfur coals prior to flotation. Spiral concentrators have no moving
parts and require minimum energy.
Spiral Concentrator Operation
Gravimetric separation by spiral concentrators is accomplished by a
combination of centrifugal forces and gravitational forces. The spiral con-
centrator consists of a spiral conduit of modified semicircular cross-section.
Feed pulp is introduced at the top of the spiral and flows downward until
centrifugal forces cause the heavier particles (pyritic sulfur) to stratify
and to concentrate in a band along the inner side of the spiral. The heavier
particles are removed through adjustable portholes located on each turn of
the spiral at the lowest point in the cross section of the conduit. As the
spiral stream approaches the lower end of the spiral, an adjustable splitter
62
-------
divides the stream into two productions; a clean coal product and a
middlings product.
Spiral Concentrator Performance
Good results can be expected when spiraling coals of top sizes ranging
from 4.7 to 0.4 mm. (4 to 35 mesh) with 1.17 mm. (14 mesh) being the
optimum value. The reduction of sulfur rapidly diminishes for coal approach-
ing minus 0.15 mm. (minus 100 mesh). This is a normal characteristic of
specific gravity separators.
Capacities range up to 1.8 kkg/h (2 tph) for a single spiral and
banks of 80 or more spirals are common.
VORSYL
The Vorsyl was developed by the National Goal Board (NCB) of the United
Kingdom, where it is used extensively, to produce low-sulfur low-ash coking
coal and to recover anthracite from waste banks. The Vorsyl separator
incorporates the operating principles of centrifugal force developed by
vortex flow and dense medium cleaning, has no moving parts and requires no
energy for operation.
Vorsyl Operation
The Vorsyl separator is a completely cylindrical vessel usually install-
ed in the vertical postion. The vessel contains an axially-arranged open-
ended vortex finder for removal of clean coal through the base of the
chamber. The throat is concentric with the vortex finder and provides an
annular opening for discharge of the refuse and the remaining dense medium.
These are routed to the shale chamber and the vortextractor. The vortex-
tractor has a tangential inlet and axial outlet to induce an inward spiral
flow to the outlet so that pressure is dissipated.
Feed, Consisting of deslimed raw coal together with the dense medium,
is introduced tangentially at the top of the vessel. Material of specific
gravity considerably less than the separating gravity passes immediately
to the clean coal outlet via the vortex finder.
63
-------
The heavier refuse particles move to the wall of the vessel as a
result of the centrifugal force induced by the vortex motion of the feed
fluid. These particles move toward the base of the vessel where the
tangential velocity is reduced by drag forces. Zones of high centrifugal
force exist where the final precise separation takes place. Refuse is
discharged through the throat and near-gravity (±0.1 specific gravity
range) clean coal is carried up the outer wall of the vortex finder to
the product discharge.
Vorsyl Performance
Generally, Vorsyls show the same efficiency as conventional dense
medium cyclones except at the top end. This means that the Vorsyl will
have virtually no misplaced material that is susceptible to gravimetric
separation in the clean coal. The Vorsyl handles feed of 25 x 0.54 mm.
(1 in. x 28 mesh). The 76 cm. (30 in.) diameter unit has a capacity of
91 kkg/hr. (100 tph). The magnetite used should be less than 20% of the
medium (volumetric basis) and should contain minus 50 micron (270 mesh)
particles in excess of 90%.
The overall efficiency is not affected significantly by minor varia-
tions in design parameters such as feed pressure, vortextractor orifice
diameter or vortex finder length. This enables the Vorsyl to handle
variations in flow rates and consistencies without a loss in efficiency.
FLOTATION
Froth flotation is a physical-chemical process for the separation of
suspended solids, based upon selective adhesion to air of some solids and
the simultaneous adhesion to water of other solids. Separation of coal
from coal waste occurs as finely disseminated air bubbles are dispersed
throughout a coal-water slurry. The coal particles in the slurry
adhere to the air bubbles and are transported to the free surface of the
pulp. These bubbles and their attached coal particles, commonly referred
to as froth, are then removed; the ash and sulfur bearing material
remain in suspension and are passed through the cells.
64
-------
chemical process of bubble adhesion can be selectively modified by
the addition of certain reagents. These reagents either enhance the
floating characteristics of coals or the wetting characteristic (water
attraction) of the waste material. Reagents are also used to stabilize
the froth, thus allowing sufficient time for removal of the floated coal.
Coal flotation is influenced by several factors;
• particle size;
• oxidation and rank of coal;
• pulp density;
• chemistry of plant water - pH especially;
• flotation reagents; and
• flotation equipment
Particle Size:
Flotation processing is normally used in the cleaning of coal of a
top-size of 1.17 to 0.30 mm. (14 to 48 mesh). Flotation can be used on
coal as fine as 0.04 mm. (325 mesh) but product recovery and grade can be
affected. Flotation is a highly efficient method for reducing both sulfur
and ash in the 1.17 x 0.30 mm. (14 x 48 mesh) sizes while hydrocyclones
are more efficient for the reduction of pyrite in the minus 0.30 mm.
(48 mesh) coal.
Oxidation and Rank of Coal
The most important parameter in flotation chemistry is the surface
characteristics of the coal particle. Oxidized coals and high volatile
coals are not readily amenable to flotation.
Pulp Density;
The percentage of solids in the feed slurry has an overall effect on
cell productivity. Flotation cells are normally operated at pulp densities
of 3 to 20% solids. The size consist of the feed is the main factor in-
fluencing the percent solids at which maximum efficiency can be achieved.
Coal in the 1.17 x 0.30 mm. (14 x 8 mesh) range can be floated efficiently
65
-------
at around 15% solids, but ooal in the 0.59 x 0.07 ram. (28 x 200 mesh) range
must be floated at approximately 5 to 7% solids, if acceptable recovery is
to be maintained.
Chemistry of Plant Water;
The pH of the slurry water has an effect on the recovery and quality
of the floated coal. The relationship between pH and product recovery is
illustrated in Figure 18. The optimum pH for maximum recovery is between
6 and 7.5.
14
13
12
tu
-------
Flotation Iteagents;
The amounts arid kinds of reagents used are extremely critical to the
flotation process. There are three general classes of reagents; frothers,
collectors, and modifying agents. Frothers (frothing agents) are used
primarily to facilitate the production of a stable froth. Collectors or
promoter reagents are used to increase the contact between the coal particles
and the air bubbles by forming a thin coating over the particles, thus
rendering them water repellent. Modifying agents are a large group of
chemical reagents that are used to perform varying functions such as:
• depressing agents which inhibit flotation;
• activating agents which promote particle coating;
• pH regulators which control acidity or alkalinity of the
pulp;
• dispersing agents which are used to deflocculate troublesome"
minerals and control slime absorption; and
• protective colloids which have dispersive effects and can
be also used as depressants under some conditions.
Operation
Flotation equipment has evolved during approximately 60 years of develop-
ment. Flotation equipment, as used in coal processing plants, is based upon
well-tested designs. Flotation machines are of two basic types; mechanically-
agitated or pneumatic. Both types perform the following functions:
• pulp agitation;
• introduction and dissemination of fine air bubbles; and
• transportation and removal of the products and refuse.
These functions are, at times, in conflict; and the efficient operation
of a flotation cell requires that certain compromises be reached. Figure 19
is an illustration of the flow pattern? in a currently available cell.
Flotation cells are usually grouped in a bank of 4 or 5 cells. The
number of cells determines the particle retention time while the capacity of
each cell determines tonnages.
67
-------
FIGURE 19. FLOTATION CKT.T.
Performance
There are numerous complex variables that determine the performance of
coal flotation. In general, well-designed, properly maintained and efficient-
ly operated two-stage circuits can remove 50 to 60 percent of the liberated
pyrite while maintaining a product recovery of 80 to 85 percent.
Recent Technology—
Due to increasingly stringent air pollution standards, new technology is
being directed at improving sulfur rejection. Recent studies by the U.S.
Bureau of Mines and flotation manufacturers have resulted in improved coal
flotation processes, summarized as follows:
Rougher-Cleaner Flotation
• condition pulp at 10% solids with FeCl3.
• condition pulp at 10% solids with No. 2 fuel oil or kerosene.
• add Methyl isobutyl carbonyl (MIBC) sparingly to rougher
circuit, using stage addition of MIBC as needed, and produce
a rougher concentrate low in ash and pyritic sulfur. The
68
-------
sulfur and ash are rejected as tailings.
• add MIBC to rougher concentrate in cleaner circuit, dilute
to 10% solids, using stage addition of MIBC if needed, and
produce a final clean concentrate.
U.S. Bureau of Mines Two-Stage Flotation
• condition pulp at 10% solids with No. 4 fuel oil or
kerosene.
• add MIBC to first stage circuit, using stage addition
of MIBC to float the coal, and produce a first-stage coal
coal concentrate low in ash and high in pyritic sulfur.
• dilute the resulting first-stage coal concentrate to
10% solids, adjust pH to 6.5 with hydrochloric acid and
condition with Aero Depressant 633 followed by the addition
of a pyrite collector (potassium arnyl xanthate) to depress
the coal and float the pyrite.
• add MIBC to the pulp in the second stage circuit, using
stage addition of MIBC if needed, and produce a final con-
centrate from the second stage that is high in pyritic sulfur.
The resulting clean coal, low in ash and pyritic sulfur, is
the tailing product from the second stage of flotation.
In general, both of these methods are effective in removing sulfur.
Some coals are not amenable to either process.
69
-------
SECTION 4
CURRENT PROCESS TECHNOLOGY FOR CHEMICAL COAL CLEANING
Chemical coal cleaning processes are still under development. None of
these processes have been tested in units larger than eight metric tons per
day and only one process at that size. Consequently, performance and
cost comparisons are relatively uncertain at this time. The chemical coal
cleaning processes vary substantially in their approach, because of the large
number of possible reaction mechanisms and chemicals which can be used to
effect removal of sulfur and other reactive impurities in coal. Most chemical
processes remove over 90 percent of the pyritic sulfur and several are reported
to remove up to 40 percent of the organic sulfur as well.
Twenty-nine chemical coal cleaning processes were identified during a
technology overview study. Eleven U.S. processes are classified as major
processes. The other processes, which are considered minor, are listed in
tables later in this section. Further processes of importance may exist, but
were not identified in the search conducted in this study.
The eleven major chemical coal cleaning processes exhibit a great deal
of diversity with respect to such variables as:
• kinds and amounts of sulfur removed
• type of coal successfully desulfurized
• degree of coal crushing and grinding prior to chemical processing
• state of process development
• process chemistry
• major process steps
• prospects for technical and economic success
Table 6 is a listing of the major processes and summarizes some of the
above factors. The first four processes listed (Magnex™, Syracuse, TRW and
Ledgemont) will remove pyritic sulfur only; the remaining seven processes
70
-------
TABLE 6, PROCESS INFORMATION SUWRY OF MAJOR CHEMICAL COAL CLEANING PROCESSES
PROCESS a
SPONSOR
MAGNEX®,
HAZEN RESEARCH
INC., GOLDEN
COLORADO
"SYRACUSE"
SYRACUSE
RESEARCH CORP.,
SYRACUSE, N.Y.
"MEYERS", TRW,
INC. REDONDO
BEACH, CAL.
"LOtI' KENNECOTT
COPPER CO.
LEDGEMONT, MASS.
METHOD
DRY PULVERIZED COAL
TREATED WITH FE
(C0)5 CAUSES PYRITE
TO BECOME MAGNETIC.
MAGNETIC MATERIALS
REMOVED MAGNETICALLY
COAL IS COMMINUTED
BY EXPOSURE TO NH3
VAPOR j CONVENTIONAL
PHYSICAL CLEANING
SEPARATES COAL/ASH
OXIDATIVE LEACHING
USING FE2(S04)3 +
OXYGEN IN WATER
OXIDATIVE LEACHING
USING 0, AND WATER
9 MODERATE TEMP.
AND PRESSURE
TYPE SULFUR
REMOVED
UP TO 90Z
PYRITIC
50-70%
PYRITIC
90-95%
PYRITIC
90-95%
PYRITIC
STAGE OF
DEVELOPMENT
BENCH & 91 KG/DAY
(200 LB/DAY) PILOT
PLANT OPERATED
BENCH SCALE
8 METRIC TON/DAY
PDU FOR REACTION
SYSTEM. LAB OR
BENCH SCALE FOR
OTHER PROCESS
STEPS.
BENCH SCALE
PROBLEMS
DISPOSAL OF S-CONTAIN-
ING SOLID RESIDUES.
CONTINUOUS RECYCLE OF
CO TO PRODUCE FE
(CO)5 REQUIRES
DEMONSTRATION
DISPOSAL OF SULFUR
CONTAINING
RESIDUES.
DISPOSAL OF ACIDIC'
FES04 £ CAS04, SULFUR
EXTRACTION STEP
REQUIRES DEMONSTRA-
TION
DISPOSAL OF GYPSUM
SLUDGE, ACID
CORROSION OF
REACTORS
ANNUAL OPERATING
COST */TON CLEAN
COAL INCLUDING
COST OF COAL
38.7
39.14
45.3
RAW COAL COST IS INCLUDED AT $27.6/METRIC TON ($25/TON).
-------
TABLE 6. (CONTINUED)
to
PROCESS &
SPONSOR
"ERDA" (PERC)
BRUCETON, PA.
"GE" GENERAL
ELECTRIC CO.,
VALLEY FORGE,
PA.
"BATTELLE"
LABORATORIES
COLUMBUS, OHIO
"JPL" JET
PROPULSION
LABORATORY
PASADENA, CAL.
"iGT" INSTITUTE
OF GAS
TECHNOLOGY
CHICAGO, ILL.
METHOD
AIR OXIDATION &
WATER LEACHING 3
HIGH TEMPERATURE
AND PRESSURE
MICROWAVE TREATMENT
OF COAL PERMEATED
WITH NAOH SOLUTION
CONVERTS SULFUR
FORMS TO SOLUBLE
SULFIDES
MIXED ALKALI
LEACHING
CHLORINOLYSIS IN
ORGANIC SOLVENT
OX I DATIVE PRETREAT-
MENT FOLLOWED BY
HYDRODESULFUR I ZAT I ON
AT 8000C
TYPE SULFUR
REMOVED
%95% PYRITIC;
UP TO 40%
ORGANIC
V5% TOTAL S
^95% PYRITIC;
%25-50% ORGANIC
^90/b PYRITIC; UP
TO 70% ORGANIC
^95% PYRITIC; UP
TO 85% ORGANIC
STAGE OF
DEVELOPMENT
BENCH SCALE 11 KG/
DAY (25 LB/DAY)
CONTINUOUS UNIT
UNDER CONSTRUCTION
BENCH SCALE
9 KG/HR (20 LB/
HR) MINI PILOT
PLANT AND BENCH
SCALE
LAB SCALE BUT
PROCEEDING TO
BENCH AND MINI
PILOT PLANT
LAB AND BENCH
PROBLEMS
GYPSUM SLUDGE DISPOSAL
ACID CORRROS10N AT
HIGH TEMPERATURES
PROCESS CONDITIONS
NOT ESTABLISHED
CAUSTIC REGENERATION
PROCESS NOT
ESTABLISHED.
CLOSED LOOP REGENERA-
TION PROCESS UNPROVEN,
RESIDUAL SODIUM IN
COAL
ENVIRONMENTAL
PROBLEMS. CONVER-
SION OF HCL TO CL2
NOT ESTABLISHED
LOW BTU YIELD (<55%).
CHANGE OF COAL MATRIX
ANNUAL OPERATING
COST I/TON CLEAN
COAL INCLUDING
COST OF COAL
51.6
40.2
56.1
45.9
65.7
*RAW COAL COST IS INCLUDED AT $27.6/METRIC TON ($25/TON).
-------
TABLE 6. (CONTINUED)
PROCESS &
SPONSOR
"KVB" KVB, INC.
TUSTIN, CAL.
"ARCO"ATLANTIC
RICHFIELD
COMPANY
HARVEY, ILL.
METHOD
SULFUR IS OXIDIZED
IN N02-CONTAINING
ATMOSPHERE. SULFATES
ARE WASHED OUT.
TWO STAGE
CHEMICAL
OXIDATION
PROCEDURE
TYPE SULFUR
REMOVED
^95% PYRITICj TO
m ORGANIC
^95* PYRITIC;
SOME ORGANIC
STAGE OF
DEVELOPMENT
LABORATORY
CONTINUOUS 0,45
KG/HR (1 LB/HR)
BENCH SCALE UNIT
PROBLEMS
WASTE 8 POSSIBLY
HEAVY METALS DISPOSAL
POSSIBLE EXPLOSION
HAZARD VIA DRY OXIDA-
TION.
UNKNOWN
ANNUAL. OPERATING
COST $/TON CLEAN
COAL INCLUDING
COST IN COAL
^.90
'•6-58
(ESTIMATES)
CO
*RAW COAL COST IS INCLUDED AT $27.6/METRIC TON ($25/TON),
-------
(ERDA, GE, Battellef JPL, IGT, KVB and ARCO) claim to remove most of the
pyritic sulfur and varying amounts of organic sulfur. Also, the first two
processes are unique in that the coal is chemically pretreated, then sulfur
separation is subsequently achieved by mechanical means. Ihe remaining nine
processes are more typically chemical in that sulfur compounds in the coal
are chemically attacked and renewed. A capsule summary of each major pro-
cess follows.
MAGNEX^1 PROCESS
In this process, dry, pulverized (minus 14 mesh) coal is pretreated with
iron pentacarbonyl to render the mineral components of the coal magnetic.
Separation of pyrite and other mineral elements from the coal is then
accomplished magnetically. Figure 20 is a flow diagram of the Magnex*
process.l The cost of the Magnex™ process critically depends on the recycle
of the chemical reactant iron carbonyl. It is claimed that iron carbonyl can
be produced on-site from carbon monoxide released in the process, however, the
continuous recycle of carbon monoxide to produce low cost iron carbonyl
requires pilot plant demonstration. Approximately 40 coals, mostly of
Appalachian origin, have been evaluated on a laboratory scale. These coals
are rich in pyritic sulfur and are thus most applicable to this process.
For the most part, the process will produce coals which meet state regulations
for sulfur dioxide emissions of less than 4.3 kg SO2/WB kg cal (2.4 Ib S02/
106 BTU). This process does not remove organic sulfur from coal.
SYRACUSE PROCESS
Coal of about 3.8 on (IV) top size is chemically comminuted by exposure
to moist ammonia vapor at intermediate pressure.2 After removing the ammonia,
conventional physical coal cleaning then effects a separation of coal from
pyrite and ash. The Syracuse process flow diagram is shown in Figure 21.
This process is claimed to produce less fines than mechanical crushing, and
to achieve more effective liberation of pyritic sulfur. Generally, 50-70% of
pyritic sulfur has been removed from Appalachian and Eastern Interior coals,
producing coals which meet many state regulations for sulfur dioxide emission.
Construction of a 36 metric tons per day (40 tons per day) pilot plant is in
the planning stages.
74
-------
cn
ROM ,
COAL
CRUSH
AND
GRIND
CYCLONE
COLLECTOR
HEAT
TRANSFER
OIL
COAL HEATER
AND CONDITIONER
HOT OIL
HEATER
CARBONYL
BEACTOB AND
VAPORIZER -"
CARBONYL
VAPOR
RECYCLE CO
MAGNETIC SEPARATOR
CLEAN
COAL
COMPACTOR
FIGURE 20 MAGNEX PROCESS FLOW SHEET
-------
NH3 VAPOR
RAW
COAL
CRUSHING
ROM COAL
SIZED TO 1'/p"
X 0
;
1 t x-
-\.
DILUTE NH3 Lia
x
UN.
-------
MEYERS PROCESS
Ihis process is the most advanced of the chemical coal cleaning pro-
cesses, with a. 9 metric ton per day Reactor Test Unit (RTU) in operation. A
flow diagram of the process is shown in Figure 22. Ihe process removes 80-
99% of the pyritic sulfur from nominally 14 mesh top size coal. It uses an
aqueous solution of ferric sulfate and sulfuric acid to effect a chemical
leaching at moderate temperatures and pressures, but at rather long holding
periods (8-13 hours). Thirty-two different coals have been tested: twenty-
three from the Appalachian Basin; six from the Interior Basin; one from the
Vfestern Interior Basin and two western coals. The Meyers process is more
applicable to coals rich in pyritic sulfur, thus about one-third of Appala-
chian coal could be treated to sulfur contents of 0.6 to 0.9 percent to meet
the sulfur dioxide emission requirements of current EPA NSPS.3 Process by-
products are elemental sulfur, gypsum from wastewater treatment, and a mixture
of ferric and ferrous sulfate, with the latter presenting a disposal problem.
LEDGEMONT PROCESS
The Ledgemont oxygen leaching process is based on the aqueous oxida-
tion of pyritic sulfur in coal at moderately high temperatures and pressures.
Figure 23 represents the process flow scheme. The process has been shown to
remove more than 90 percent of the pyritic sulfur in coals of widely differing
ranks, including lignite, bituminous coals, and anthracite, in bench-scale
tests. However, little, if any, organic sulfur is removed by the process.
The process became inactive in 1975 during divestiture of Peabody Coal
Company by Kennecott Copper Co.1* Although not as well developed as the
Meyers process, the Ledgemont process is judged to be competitive in cost and
sulfur removal effectiveness. The principal engineering problem in this
process is the presence of corrosive dilute sulfuric acid, which may pose
difficulties in construction material selection and in equipment for pressure
letdown. The process also has a potential environmental problem associated
with the disposal of lime-gypsum-ferric hydroxide sludge which may contain
hazardous trace metals.
77
-------
IRONSULFATE
SULFUH
OEMOVAL
SECTION
CLEAN CO/Vl
FIGURE 22 PROCESS FLOWSHEET FOR THE (MEYERS') PROCESS
-------
ROM COAL '
MAKEUP
WATER "
OFFGAS
OXYGEN 1 i—»-(TO OXYGEN PLANT)
f J CO2. CO
BINDER'
DESULFURIZED
COAL
GYPSUM 4-
IRON HYDROXIDE
TO WASTE
DISPOSAL
NEUTRALIZER
FIGURE 23 LEDGEMONT OXYGEN LEACHING PROCESS FLOW SHEET
-------
EKDA PROCESS
A conceptualized flew sheet for the ERDA air and steam leaching process
is shown in Figure 24.5 The process is similar to the Ladgemont oxygen/
water process except that it employs higher temperature and pressure to effect
the removal of organic sulfur and uses air instead of oxygen. This process
can remove more than 90 percent of the pyritic sulfur and up to about 40 per-
cent of organic sulfur from finely crushed coals (minus 200 mesh). Coals
tested on a laboratory scale include Appalachian, Eastern Interior and
Western. The developer's claim is that using this process, an estimated
45 percent of the mines in the eastern United States could produce environ-
mentally acceptable boiler fuel in accordance with current EPA standards for
new utility installations. Effort to date is on a bench scale, but a mini-
pilot plant is expected to start up soon. The problems associated with this
process are engineering in nature. The major one is associated with the
selection of materials for the unit construction. Severe corrosion problems
can be expected as the process generates dilute sulfuric acid which is highly
corrosive at the operating temperatures and pressures.
G.E. PROCESS
A conceptualized flow sheet for the General Electric chemical coal
cleaning process is shown in Figure 25. Ground coal (40 to 100 mesh) is
wetted with sodium hydroxide solution and subjected to brief (-30 sec.)
irradiation with microwave energy in an inert atmosphere. After two such
treatments, as much as 75-90 percent of the total sulfur is converted to
sodium sulfide or polysulfide which can be removed by washing.6 No signifi-
cant coal degradation occurs. That portion of the process which recovers the
sulfur values and regenerates the NaOH is not proven. Work to date has been
with 100 gram coal samples, but scale-up to 1 kg quantities is presently in
progress. The process appears to attack both pyritic and organic sulfur,
possibly at about the same rate. Appalachian and Eastern Interior coals
having wide ranges of organic and pyritic sulfur contents have been tested
with about equivalent success.
80
-------
PULVERIZED COAL
00
OFFGAS
i
p
J
i
•" "•>
RE
ACTORS
F-
JU
t
jr~~*
| »
\/
FLASH
TANK
CLEAN COAL
GYPSUM
FILTER
FIGURE 24 ERDA PROCESS FLOW SHEET
-------
BINDER
HFCYC.I.EO
NriOtl SOLUTION
MICROWAVE
GENERATOR
AND
IRRADIATION
CHAMBER
CAUSTIC
GENERATOR
STEAM
A
EVAPORATOR
CONCENTRATED
NaOH SOLUTION
TO BLENDER
FILTER
FIGURE 25 GENERAL ELECTRIC MICROWAVE PROCESS FLOW SHEET
-------
BATTELLE PROCESS
Figure 26 is a flow diagram for the Battelle Hydro-thermal process. In
this process, 70 percent minus 200 mesh coal is treated with aqueous sodium
and calcium hydroxides at elevated temperatures and pressures, which removes
nearly all pyritic sulfur and 25-50 percent of organic sulfur.7 Test work
on a bench and pre-pilot level on Appalachian and Eastern Interior coals has
resulted in products which meet current EPA NSPS for sulfur dioxide emissions.
The conceptualized process, using lime-carbon dioxide regeneration of the
spent leachant, removes sulfur as hydrogen sulfides which is converted to
elemental sulfur using a Stretford process. In addition to being a costly
process, there are two major technical problems:
• The feasibility of the closed-loop caustic regeneration feature
in a continuous process is as yet undemonstrated; and
• The products may contain excessive sodium residues, causing low
melting slags and making the coal unusable in conventional dry-
bottom furnaces.
JPL PROCESS
Figure 27 shows a conceptualized process flow scheme for the JPL process.
This process uses chlorine gas as an oxidizing agent in a trichloroethane
solution to convert both pyritic and organic forms of sulfur in coal to
sulfuric acid. Since removal of sulfur can approach the 75 percent level,
without significant loss of coal or energy content, products should generally
meet current EPA NSPS for sulfur dioxide emissions.8 To date the process has
been tested on two Eastern Interior coals on a laboratory scale only, however,
the effort will progress to bench-scale and pre-pilot plant scale in the near
future. The project is currently supported by the Bureau of Mines.
There are some potential environmental problems with the process. The
trichloroethane solvent is listed by EPA as a priority pollutant in terms of
environmental effects. A major conceptual problem in the design of the process
is the need to recycle by-product hydrochloric acid for conversion to chlorine.
At a chlorine consumption rate of 250 kg per metric ton of coal, the
incorporation of a Kel-Chlor unit in the JPL system will add approximately
$10/mstric ton of coal. This may be a difficult economic problem for the
83
-------
MAKEUP WATER
CaOH
NaOH
CONDENSATE
BINDER
CRUSH
AND
GRIND
j
CO,
MAKEUP
00
RECIRCULATED
NaOH LEACHANT
OESULFURIZEO
COAL
SOLUTION
SULFIDE/CAUSTIC
SOLUTION
CONCENTRATOR
SULFIDE
STRIPPER
SODIUM
CARBONATE
SOLUTION
CARBON DIOXIDE
PURIFIER
o
CAUSTIC
REGENERATOR
CaCO3
SLURRY
/
FILTER
CaC03
STRETFORD
PROCESS
BLEED STREAM
RAW COAL
FUEL
CaO
CALCINER
LIME MAKEUP—•, r— H,O
T I T 2
CaO
LIME SLURRY
MIXER
-»- BLEED STREAM
-»- PRECIPITATED
COALORGANICS,
ASH AND TRACE
METALS
WASTE STREAM
FIGURE 26 BATTELLE HYDROTHERMAL PROCESS FLOW SHEET
-------
SOLVENT RECYCLE
WATER
ROM COAL '
00
en
MAKEUP
HCI
HCI RECOVERY UNIT
FIGURE 27 JPL PROCESS FLOW SHEET
-------
JPL process to surmount.
IGT PROCESS
The IGT conceptualized process is represented by the flow sheet shown
as Figure 28. This process employs essentially atmospheric pressure and
high temperatures [about 400°C (750°F) for pretreatment and 815°C (1,500°F)
for hydrodesulfurization] to accomplish desulfurization of coal. These high
temperatures cause considerable coal loss due to oxidation, hydrocarbon
volatilization and coal gasification, with subsequent loss of heating value.
Experimental results have indicated an average energy recovery potential of
60 percent for this process. The treated product is essentially a carbon char
with 80-90 percent of total sulfur removed. Most of the experimental work
to date has been accomplished with four selected bituminous coals with a size
of plus 40 mesh. Present effort is on a bench-scale level. The net energy
recovery potential of the system and the change in the coal matrix by the
process have been identified as possible severe problems for the IGT process.
The process must be developed to a stage vtfiere the process off-gas can be
satisfactorily utilized for its energy and hydrogen content. If this cannot
be technically and economically accomplished, the process will prove to be
inefficient and too costly for commercialization.
KVB PROCESS
A conceptualized process flow diagram for the KVB process is shown in
Figure 29. This process is based upon selective oxidation of the sulfur
constituents of the coal. Hh this process, dry coarsely ground coal (28
mesh) is heated to 120°C (250 °F) in the presence of nitrogen oxide gases
at a pressure of 2.4 atm. (35 psia) for the removal of a portion of the coal
sulfur as gaseous sulfur dioxide. Residence time is about 1/2 to 1 hour.
Any remaining reacted, non-gaseous sulfur compounds in the treated coal are
removed by water or caustic washing. The process has progressed through the
laboratory scale, but is currently not active for lack of support. Labora-
tory experiments with five different bituminous coals indicate that the
process has desulfurization potential of up to 63 percent of sulfur with
basic dry oxidation plus water washing treatment and up to 89 percent with
dry oxidation followed by caustic and water washing. The washing steps
also reduce the ash content of the coal.
86
-------
00
-J
SCRUBBER
LIQUOR
FLUIDIZED BED
PRETREATER
HEAT
EXCHANGER
Lr
r
OFF.GAS
SCRUBBER
PRETREATED
CRUSHED
C°AL
COMPRESSOR
-MAKEUPH,
RECOVERED
GAS SCRUBBER
SHIFT CONVERTER
CLEAN UP
ANDH,S/C02
RECOVERY
H2S
WASTE H2O WASTE SOLIDS
HEAT
EXCHANGER
AIR—4 STE/
I * HYDRODESULFUR1ZER
» >EXTRACTOR 1
MAKE-UP »•
CAUSTIC
LIME THEATER
RECYCLE CAUSTIC
ELEMENTAL
SULFUR
PLANT
ELEMENTAL
SULFUR
^v | f
FILTER >>|, > DRYER > CLEAN CHA
U U PRODUCT
FIGURE 28 IGT PROCESS FLOW SHEET
-------
02/N2/NO GASES
COMPRESSOR
SUCTION
DRUM
00
00
FIGURE 29 KVB PROCESS FLOW SHEET
-------
In cases where dry oxidation alone could remove sufficient sulfur to
meet the sulfur dioxide emission standards, this technology may provide a
very simple and inexpensive system. Potential problem areas for this
system are:
• oxygen concentration requirements in the treat gas exceed the
explosion limits for coal dust, and thus the application of
this process may be hazardous.
• nitrogen uptake by the coal structure may increase NOX emission.
SUMMARY CF MINOR AND MLSCELLRNEOUS PROCESSES
Tables 7 and 8 summarize process information on the minor and miscella-
neous chemical coal cleaning processes.
89
-------
7. PROCESS INPOFMATION SIM1ARY OF MINOR CHEMICAL COAL CLEANING PROCESSES
1$
Process
Colo. Sch. of
Mines Pes.
Inst.
Jolevil
U. of Houston
Olio State U.
NBC
(Canada)
W. 111. u.
Texaco
Method
Selective ferrofluid
wetting of pulverized
coal constituents
followed by magnetic
separation
Unknown
Hydrothermal alka-
line leaching
Microbiological
oxidation
Oil agglomeration
of very fine coal
particles leaving
rejected pyrites
in water slurry
Microbiological
oxidation of
pyrite particles
to increase hydro-
phobic properties
Hydrosulfuriza-
tlon in plasma arc
HzOj oxidation during
pipeline transport
Type S
Removed
Pyritic
Unknown
Pyritic &
some organic
Pyritic
Pyritic
Pyritic
Both
Pyritic
Stage of
Development
Starting lab work
Allegedly active
S being marketed
Lab scale
Lab scale
Active
Active on lab
scale
Active, at low
level
Inactive
Problems ,
Comments
EPRI funded
Unknown
Alleges improve-
ment on Battelle
process .
Internally funded
7+ day process.
Internally funded
May be especial-
ly applicable in
recovery of fines
May make oil
agglomeration
more efficient
Seeking funding
~
Economics
No data
Unknown
No data
No data
$2/ton applied
to fines recovery
No data
No data
-------
TABLE 7. (continued)
Process
U. of Fla.
Methonics, Inc.
Rare Earth
Industries
MIT
Gulf & Western
New South
Wales
Method
Gas oxidation/
reduction at very
high temperature
Wet nydrogenation
Rare earths recycled
as S-getters during
SRC liquefaction
Catalytic desulfuri-
zation of petroleum
fractions
Coal liquefaction
via graft polymeri-
zation
H2O2 oxidation
Type S
Removed
Both
Both
Both
Not Given
Not Given
Not Given
Stage of
Development
Inactive
Inactive
Inactive
Active
Inactive
Discontinued
Problems ,
Ccmnents
Poor yield?
no data since
1975
Company probably
no longer exists
Company probably
no longer exists
Not applicable
to coals
Changes coal
matrix. Prior
ERDA funding now
discontinued
Method is analyt-
ical, not meant
to be coal
cleaning
Economics
No data
No data
No data
No data
No data
No data
-------
TABLE 8. PROCESS INFORMATION SUMMARY OF MISCELLANEOUS CHEMICAL COAL CLEANING PROCESSES
Process
U.S. Steel
Chemioo
ERDA
(Laramie)
Rutgers
Dynatech
Kyoto Univ.
(Japan)
Kellogg
Method
Fused NaOH @ high
temperature
Wet oxidation using
air @ high terpera-
ture and pressure
Leaching using
H2SO,, or H2SOU+H202
Microbiological oxi-
dation of organic S
Microbiological
oxidation
C12/C>2 wet oxidation
High tetperature and
pressure
leaching in KOH solu-
tion w/Fe203 catalyst
Type S
Removed
Both
Both
Pyritic
Organic
Pyritic
Probably both
Both
Stage of
Development
Inactive
Inactive
Inactive-prev. on
lab basis only
Inactive
Inactive
Unknown
Discontinued
Problems,
Comments
Excess Na in
product; coal
matrix affected
Inactive since
1975
Recently dis-
continued;
negative results
Inactive
No answer to
our letter of
inquiry
Poor yield;
coal matrix
altered
Econcmics
No data
No data
No data
No data
-$4/ton (company
data)
No data
No data
-------
PROCESS AND COST COMPARISON
This section presents a comparison of technical and economic results
obtained from the assessment of major chemical coal cleaning processes as
described and discussed previously. The analysis and conclusions presented
herein are based on process claims made by individual developers, research
reports and published information.
Most processes described in this report are not at an adequate develop-
mental stage to permit the preparation of a precise engineering process flow
sheet for capital and operating cost evaluation. Thus, the process economics
presented for most processes are best engineering estimates, based upon the
information available.
Sulfur Removal and Energy Recovery Potential
A conparison of process performance and costs can best be acconplished by
looking at each process on a cannon coal feed. This basis allows the compari-
son of the following parameters process by process:
• Weight yield of clean coal product based upon a feed coal rate
(moisture free basis) of 7,110 metric tons (7,840 tons) per day
17,200 metric tons (8,000 tons) per day of 2 percent moisture coal];
• Weight percent sulfur in the clean coal product based upon the
sulfur removal efficiency of the process;
• Heating value yield of the process based upon a feed coal value of
6,800 kg cal/kg (12,300 BTU/lb) and net energy recovery in percent;
and
• Costs
- total capital costs for the process,
- total annual processing costs,
- annual costs per metric ton of clean coal, including coal costs
and excluding coal costs, and
- annual costs per heating value unit, including coal costs and
excluding coal costs.
This conparison data is shown in Table 9, arranged according to categories
of processes.
93
-------
The common coal feed selected is a bituminous coal from the Pittsburgh
seam, which cannot readily be cleaned by conventional washing techniques
to meet the current new source performance standards for sulfur dioxide
emission. This coal does have an organic sulfur content low enough (0.7
weight percent) so that complete removal of pyritic sulfur would result in a
product which will meet current NSPS for sulfur dioxide emission.
The percent removal of pyritic and organic sulfur assigned to each pro-
cess is based on data supplied by individual developers. The table indicates
a range of SO2 emission levels for the clean coal products of 1.5 to 3.8 kg/
106 kg cal (0.8 to 2.1 lb/106 BTU). The calculated S02 levels for pro-
cesses which remove both types of sulfur are lower than the 2.2 kg/106 kg cal
(1.2 lb/106 BTU) NSPS for sulfur dioxide emission. Of the four processes
which remove pyritic sulfur only, two (TRW and Ledgemont) will produce
slightly higher sulfur levels than that required to meet the current NSPS;
however, within the levels of accuracies involved they also might be con-
sidered to be in compliance. The remaining two processes [Magnex ^ and
Syracuse] would produce coal which would be in compliance only with a stand-
ard of 4.3 kg/106 kg cal (2.4 lb/106 BTU) for sulfur dioxide emission.
Processes which remove pyritic sulfur alone are primarily applicable to
coals rich in pyritic sulfur, so that efficient removal of pyritic sulfur
could bring these coals into compliance. Processes which remove both types
of sulfur are primarily applicable to coals which cannot be adequately
treated by pyritic removal processes. All chemical coal cleaning processes
are more selective and efficient than conventional coal cleaning techniques
and it is possible that each process may eventually find an area of
application.
As shown in Table 9, the energy recoveries estimated for these processes
are generally greater than 90 percent with a range from a low 57 percent
for the IGT process to a high of 96 percent for the GE process. All energy
recoveries listed in Table 9 reflect both the coal loss due to processing
and the coal used to provide in-process heating needs. However, with the
exception of the IGT process, the actual coal loss due to processing is
claimed to be small. For most processes, the major heating value loss is
due to the use of clean coal for in-process heating.
94
-------
TABLE 9.
PROCESS PERFOFMANCE AND COST COMPARISON FOR MAJOR CHEMICAL COAL d.KANING
PROCESSES
Not coal yield, metric tons
ler day (tens/day) *
Weiyht % sulfur in the product
llc-ating value, kg cal/kg
(mu/lb)
k
8.7
(7.9)
43.1
(39.1)
1.07
(0.27)
5.32
(1.34)
Processes which Remove Pyritic and Organic Sulfur
ERDA
6,400
(7,056)
0.65
7.100
(12,035)
1.6(0.9)
94
166.8
56.6
26.3
(23.8)
56.9
(51.6)
3.63
(0.92)
7.98
(2.00)
GE
6,826
(7,526)
0.50
6,800
(12,300!
1.5(0.8)
96
102.0
35.9
15.6
(14.2)
44.3
(40.2)
2.30
(0.50)
6.49
(1.63)
Battelle
6,755
(7,448)
0.65
6,300
(11,350)
2.2(1.2)
88
168.1
74.8
32.9
(29.0)
61.9
(56.1)
5.19
(1.31)
9.78
(2.46)
JPL
6,470
(7,135)
0.6
6,800
(12,300)
1.8(1.0)
91
103.0
44.2
20.3
18.4
50.6
(45.9)
2.97
(0.75)
7.40
(1.86)
IGT
4,270
(4,704)
0.55
6,500
(11,685)
1.6(0.9)
57
135
38.0
2G.5
(24.1)
72.5
(65.7)
4.09
(1-03)
11.2
(2.81)
KVB
6,070
(6,690)
0.61
7,300
'(13,120)
1.6(0.9)
91
65.5
39.6
19.4
(17. G)
51.7
(46.9)
2.66
(0.67)
7.09
(1.79)
AJ
-------
It is believed that the high yield estimated for the GE process may
not adequately reflect the heat requirements that may be needed to regener-
ate the caustic reagent employed in the process. This process is in its
early stage of development and as such, the energy requirements for the
process cannot be properly assessed at this time. It is possible, that in
the final analysis, the heating value recovery from this process will be
more in line with other chemical coal cleaning processes.
Cost Comparison for Major Chemical Coal Cleaning Processes
Estimates of capital and annual operating costs for each major chemical
coal cleaning process are also given in Table 9. These estimates are based
on an assured plant feed capacity of 7,200 metric tons (8,000 tons) per
day, equivalent to the coal need to fuel a 750 M.W. electric power plant.
The total annual operating costs for each process, including and excluding
cost of the raw coal, have been expressed also in terms of dollars per
metric ton and dollars per million kg cal heat content in the coal.
The capital cost estimate prepared by each process developer was used
as the basis of the cost estimates in this report. In some cases, these
costs were modified to allow the evaluation of the various processes on a
comparable basis. The estimated capital costs assume a grass roots opera-
tion including costs for coal crushing, grinding, product compacting and
feed and product handling. The capital costs also include land acquisition
and site development, off-site facilities, and engineering and design costs.
A contingency allowance of 20 percent has been included in all estimates,
with the exception of TRW's. A lower contingency allowance (10 percent)
was used for the TRW process since it is at a more advanced stage of devel-
opment and adequate process data is available to develop the economics of
this process with a greater degree of confidence.
Annual operating costs are based on a 24-hour workday, 90.4 percent
service factor (330 days per year) basis. The capital cost is amortized
over a period of 20 years at 10 percent interest per year. Where adequate
information was available, the utilities and chemical consumptions are based
upon actual process demand. The operating labor costs reflect wage rates
for the Pittsburgh, Pennsylvania, area. The estimates for the maintenance
96
-------
and supplies, general and administrative, taxes and insurance are taken as
5, 1.5, 2 and 1 percent on total installed plant capital cost (TPC),
respectively.
Capital Cost Comparisons—
In general, pyritic sulfur removal processes require the least amount
of capital investment. However, these processes have limited sulfur removal
efficiencies.
Among processes that remove both organic and pyritic sulfur, the KVB
process appears to have the lowest capital investment, since it is a
partially dry process requiring lower investment for the dry reaction
section. The high capital cost of the Battelle process is due to the pro-
cessing steps associated with reagent regeneration.
The high capital cost of the ERDA. process is due to costly equipment
associated with the handling of dilute sulfuric acid at elevated tempera-
tures and pressures. At the process operating conditions the dilute acid
is highly corrosive and it poses problems in terms of selection of construc-
tion material for equipment and devices which are exposed to the corrosive
atmosphere.
Very little is known about the ARCO process details and process
chemistry. Therefore, a capital cost estimate was not developed for that
process.
Operating Cost Comparisons—
Table 10 presents a summary of operating cost elements for each
process. The ranges of annual operating costs, including raw coal cost, in
terms of $/metric ton and $/106 kg cal are $43.10 to $72.50 and $5.32 to
$11.20, respectively. Pyritic sulfur removal processes using chemical
pretreatment are the least expensive of all processes listed in Table 10.
Operating cost for the Magnex process depends primarily on the cost of iron
pentacarbonyl manufacturing. In the estimate presented in Table 9, an
operating cost of $0.22/kg for the iron carbonyl manufacturing was used, as
projected by the developer. At a consumption rate of 10 kg/metric ton of coal,
97
-------
TAIII.K 10. QPiiwrjNn COST
IT)H mum CIIRMICAI, .G7
66,000 27.78
103,24.1 43.45
LF!X-.IM»rr
PIXCESS
QX7P S/'IOM
(51000) riiOLiucT
1,600 0.67
13,400 5.64
3,400 1.44
7,300 3.07
10,600 4.46
8,200 3.45
BOO 0.34
45,300 19.07
66,000 27.70
111,300 46.05
MW7JDX ^
nmnss
asr $/'n«
(Siooo) imnjuci'
7h5 0. ){,
4,274 2.04
J,002 O.r>2
i,ni9 o.n/
1,400 0.67
5,333 2.54
438 0.21
lr>,121 7.21
66,000 31.49
81.121 30.70
SYWVUUSF.
ruxiss
COST S/HJN
(51000) rujrxnT
620 0.31
5,9)9 2.80
1,512 0.72
1,335 1.58
1,040 0.49
4,220 1.99
-
16,646 7.09
66,000 31.25
02,646 39.14
aw
IWCESS
OUST 5/'K»4
(510CO) PliUXtT
3,255 1.36
19,600 8.25
5,004 2.11
8,340 3.51
13,224 5.57
6,932 2.92
240 0.10
56,595 23.82
66,000 27.78
122.595 51.60
G.E.
PROCESS
UIST 5/HJN
(51000) PRODUCT
1,830 0.72
11,900 4.73
3,790 1.49
5,310 2.10
7,170 2.82
5,860 2.31
-
35,900 14.17
66,000 2G.05
101.900 40.22
iwrnjij;
pi«xa-ss
COST 5/TON
(Siooo) pnaoucT
5,391 2.15
19,700 7.85
5,000 1.99
10,000 3.99
23,100 9.21
11,609 4.63
-
74,000 29.82
66,000 26.32
140,800 56.14
JPL
PROCESS
COST S/TON
($1000) PKOnUCT
3,700 1.54
6,900 2.87
1,310 0.55
2,300 0.96
1,400 0.58
28,600 11.90
-
44,200 10.40
66,000 27.47
110,200 45.07
ior
rntKfSS
COST S/TON
SIOOO) ITOHJCT
4,925 3.11
15,800 9.97
4,050 2.57
6,732 4.25
3,300 2.08
3,300 2.08
-
38,107 24.06
66,000 41.66
104,107 65.72
KVB
PHOCKSS
oonr S/'nxi
(51000) PPUXJCl
1,423 0.63
7,701 3. "2
1, 'Jf.fi 0.07
3,277 1.46
17,271 7.67
7,054 3.49
132 0.06
39,624 17.60
66,000 29.31
105,624 46.90
-------
each $0.20 cost increase per kilogram of iron carbonyl manufactured would
increase the annual operating cost of this process by about 27 percent.
Between the two processes which remove pyritic sulfur by leaching, the
TRW process appears to be slightly less costly. In the Ledgemont process
the fixed charges associated with the higher capital investment have an
adverse impact on the annual operating costs. Additionally, the TRW process
has a much higher probability of technical success since it is currently
active at a Process Development Unit stage. The Ledgemont process, tested
only at a mini-pilot plant level, is currently inactive.
The most expensive processes, in terms of energy output, are the ICT
process followed closely by the Battelle process. Laboratory data avail-
able at this time, indicate a very low BTU recovery for the ICT process.
The Battelle process is adversely impacted by the fixed charges associated
with the high capital investment and by the costs associated with chemicals
consumption and reagent regeneration operations.
The least expensive process capable of removing pyritic and organic
sulfur is the GE process followed closely by the JPL and KVB processes.
The GE estimate is based, however, on early laboratory data and it is
quite possible that the projected costs will prove somewhat inaccurate in
the long run. The basic process utilizes a caustic reagent in coal pretreat-
ment and the costs associated with caustic consumption and caustic regenera-
tion are questionable at this time. The JPL process estimates are also
preliminary since investigations on this process have been initiated
recently. More definitive cost information on this process will be avail-
able in 1978 when more process information and accurate material and heat
balance information becomes available. The annual costs reported for the KVB
process are also preliminary since the process is at its early stages of
development and accurate conceptualization of the process for purposes
of economic evaluation is not possible at this time. The main advantage
of the KVB process is the simplicity of the first stage dry oxidation
process. If the dry oxidation process can be successfully demonstrated
using coarse coals, this process would be an inexpensive technology for
99
-------
beneficiation of coals where partial removal of sulfur would substantially
upgrade the coal.
Among the processes capable of removing pyritic and organic sulfur
the ERDA process has one of the highest probabilities of technical success.
The process is currently active and most technologies employed in this
system have been already tested in other systems such as Ledgemont and TRW.
The process is attractive because it is claimed to remove both types of
sulfur and uses air as a major reagent. Furthermore, the sulfur by-product
from this process is claimed to be dilute sulfuric acid, rather than iron
sulfate, which greatly simplifies the coal washing operations. The process
is somewhat expensive due to high operating temperature and pressure require-
ments and the corrosive nature of dilute acid present in this system. The
dilute sulfuric acid at the operating conditions of the ERDA process will
require the use of expensive construction material and consequently a
higher capital investment cost.
Table 11 presents a cost effectiveness summary derived from information
presented in Table 9. Costs are presented in terms of dollars per percent
of sulfur removed from coal regardless of the quality of the treated
product. However, column 7 of the table shows whether the product would
comply with the current EPA's NSPS for SC>2 emissions. The processes are
then rated based upon the cost effectiveness of sulfur removal. The
subjective probability of success assigned to each process shown in column
8 of this table is based on integration of several factors such as:
• available experimental data;
• our understanding of the status; of the process;
• known product quality deficiencies;
• known process problems; and
• the degree and quality of effort assigned to the individual program.
In conclusion, all chemical coal cleaning processes discussed in this
section offer a possibility of converting coal into clean fuel. Each
process has an area of application. However, all of these processes are at
their early stages of development. Processes that remove both pyritic and
organic sulfur will have a greater impact in coal utilization. If chemical
100
-------
TABLE 11. COST EFFECTIVENESS AND OTHER COMPARISONS OF CHEMICAL COAL CLEANING PROCESSES*
Process
&i
Magnex'S-'
Syracuse &
Physical
Cleaning
TRW
LOL3
ERDA
GE
Battelle
JPL
IGT
KVB
moo
Type of
Sulfur
Removed
PA
P
P
P
(P&O)1
(P&O)
(P&O)
(P&O)
(P&O)
(P&O)
(P&O)
Percent
Sulfur
in *
Product
.97
i..^
.83
.83
.65
.50
.65
.60
.55
.68
.69
Percent
Sulfur
Removed
(%)*
0.96
0.43
1.10
1.10
1.28
1.43
1.23
1.33
1.38
1.25
1.24
Process
Cost($/
matric ton
incl. cost
of coal 4>
42.7
43.1
47,9
51.6
56.9
44.3
61.9
50.6
72.5
51.7
_A
Cost Effect-
iveness of
S removal,
($/% S)
Removed)
44.5
100
43.5
46.9
44.5
31.0
46.5
38.0
52.5
41.4
_A
Cost
Effect-
iveness
Rank-
ing
2
4
1
3
4
1
5
2
6
3
-
Meets
EPA
NSPS*
No
No
No
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Probability
of Success
(based on
available
info.)
85%
70%
90%
50%
70%
60%
35%
55%
20%
10%3
_A
Time Frame
for Commerci-
al Avail-
ability
(Years)":
2-3
2-3
5
5-3
5-8
5-8
4-6
5-8
5-8
5-8
_A
NOTES: Based on Pittsburgh seam coal from Pennsylvania which contains 1.22 weight percent pyritic,
0.01 percent sulfate and 0.70 percent organic sulfur.
, P = pyritic sulfur.
3. (P&O) = pyritic and organic sulfur.
. Tims frame assumas continuing effort or renewed effort starting iimediately.
^ Information available is insufficient to make educated guesses.
Processes not currently active, partially accounting for low probability of success.
^ 80 percent yield of product assumed in cleaning plant.
+ Difference between raw coal sulfur and clean coal sulfur
-------
coal cleaning is to be used as an approach for greater utilization of coal
as an environmentally acceptable fuel, the pyritic and organic sulfur
removal processes should be given the most emphasis and support.
102
-------
SECTION 4
REFERENCES
1. Kindig, J.K. and R. L. Turner (Hazen Research, Inc.). Dry Chemical
Process to Magnetize Pyrite and Ash for Removal from Coal. Preprint.
2. Catalytic, Inc. Chemical Comnninution, An lirproved Route to Clean Coal.
Philadelphia, Pennsylvania. 1977.
3. Hamersroa, J.W. and M. L. Kraft. Applicability of the Meyers' Process
for Chemical Desulfurization of Coal: Survey of Thirty-Five Coals.
4. Kennecott Chemical Coal Desulfurization Process. In-house report. 1977.
5. Ergun, S., R.R. Oder, L. Kulapaditharora and A. K. Lee (Bechtel Corpora-
tion) . An Analysis of Chemical Coal Cleaning Processes. Bureau of
Mines, U.S. Department of the Interior, Contract No. J0166191 (June
1977).
6. Zavitsanos, P. Coal Desulfurization by Microwave Energy. General
Electric Company. Re-entry and Environmental Systems Division.
Philadelphia, Pennsylvania 19101. Project References 1 through 9,
1976-1977.
7. Battelle in-house report, July 30, 1976.
8. Ganguli, P.S., G.C. Hsu, G.R. Gavalas, S. H. Kalfayan. Desulfurization
of Coal by Chlorinolysis. Vol. 21, No. 7, preprints of papers presented
at San Francisco, California. August 29-September 3, 1976.
9. Guth, E.D. and J. M. Robinson. KVB Coal Desulfurization Process. KVB
brochure. March 1977.
10. Personal Communications with Electric Power Research Institute.
103
-------
SECTION 5
CURRENT PROCESS TECHNOLOGY FOR FINE COAL
DEWATERING AND DRYING
INTRODUCTION
Current coal mining technology is based on the use of continuous under-
ground mining machines. These machines cut or rip the coal from the face and
produce an abundance of small particles. Those particles which will pass
through a 1.41 mm (14 mesh) opening are classified as fine coal. Approxi-
mately 33% of all coal produced by continuous mining methods is fine coal.
In surface mining the trend to large loading equipment minimizes size
degradation, but produces dirty coal due to the admixture of larger quantities
of non-coal mineral matter. With ever increasing prices for thermal units
and transportation, coal preparation becomes more attractive. Size consist
of surface mined coal has a lower content of fine coal and preparation equip-
ment is simpler and less costly. To remove the mineral sulfur (pyrite)
effectively, the coal must be crushed to a size sufficiently small to liberate
the pyrite crystals. These crystals range in size from 250 mm (10 inches) to
sub-micron sizes. The size reduction required to release pyrite may well
become the determining factor in establishing the cost effectiveness of
using a specific coal to meet a specified sulfur regulation. Beneficiation
of fine coal will require the use of coal-water slurries in the preparation
plant, with consequent requirement for dewatering and drying.
Vfeter sprays are used to suppress dust in underground mining, and rain
and snow add moisture to surface mined coal and coal preparation is almost
exclusively accomplished by wet processes. It has been estimated that for
each one percent of water in a metric ton of coal, approximately 6,300 kg
cal (25,000 BTU/ton) are required to evaporate that moisture. Ancillary
problems related to high moisture content are caking, freezing, and increased
transport costs.
104
-------
Moisture reduction also creates hazards, dry fine coal requires special
handling techniques to prevent dust pollution and explosions. Technology is
currently available for meeting the problems associated with handling dry
fine coal.
Mechanical dewatering devices can be divided into two categories:
• those which do not produce a final product—water only cyclones
(hydrocyclones) and static thickeners; and
• those which produce a final porduct—screens, centrifuges, spiral
classifiers and filters.
Coal driers are of two types:
• direct heat, in which the products of conbustion make direct
contact with the fine coal; and
• indirect heat, in which the fine coal does not come in contact
with the products of combustion.
The following table summarizes the product coal moisture ranges which
can be achieved by various dewatering and drying methods on coal which is
9.5 mm x 0 (3/8 in x 0).
Typical Moisture Content of Products by
Equipment or Process
Type of Equipment/Process Discharge Product
Dewatering screens 8 to 20% moisture
Centrifuges 10 to 20% moisture
Filters 20 to 50% moisture
Hydraulic cyclones 40 to 60% solids
Static thickeners 30 to 40% solids
Thermal driers 6 to 7^% moisture
Oil agglomeration processes 8 to 12% moisture
Table 12 shows the feed and product moistures for coal dewatering equip-
ment used on fine coal 0.6 mm x 0 (28 mesh x 0).
The operational and design variables that affect final coal moisture for
each of these devices are listed below. In all cases, an optimum solids feed
105
-------
o
CTl
TABLE 32, FINE COAL DENATURING AND HANDLING EQUIPMENT
28MxO
% SOLIDS
SCREENS
VIBRATING
LOAD REDUCERV
LOAD REDUCER + VIBRATING
CENTRIFUGES
HORIZONTAL- SOLID BOWL
VERTICAL + LOAD REDUCER
HYDRAULIC CYCLONES f
VACUUM FILTERS
STATIC THICKENER f
THERMAL DRYER
DIRECT*
INDIRECT
) 10 20 30 i
ill 1 1
1
1
|
10 50 60 70 80 90 100
1 . 1
\//////A
V//////A
\/////\
\/777//A
UZZZZ2
V////A
cm
_L
100 90
* 3/8" X 0
•V STATIONARY SCREEN, SIEVE BEND, VOR-SIV.
f PRIMARY DEWATERING
70
J_
60 50
% MOISTURE
40
30
20
4
-------
rate is assumed.
Screens - Particle size
Centrifuge - Initial density, centrifugal speed
Filters - Percent minus 44 microns (325 mesh),
Pressure differential, type of filter media
ffydraulic cyclones - Operating pressure, Diameter of cyclone
Static thickener - Area
Ihermal driers - Initial feed moisture
Oil agglomeration - Particle size, percent oil
processes
In summation, it may be said that the problems of fine coal processing,
dewatering, drying and handling, are not new and that current technology is
capable of meeting the needs created by the increasing volume of fine coal.
The most important element to be determined is what is the increased cost of
dewatering associated with a large fraction of fines or all fines necessary
for a high degree of sulfur removal.
MEOffiNICAL DEWATERING DEVICES
Screens
Vibrating screens, stationary screens and a number of specialized
screens are utilized primarily for sizing and media recovery in modern coal
cleaning plants. As sizing devices for coal 9.5 mm x 0 (3/8 in x 0), they
simultaneously dewater and discharge the oversize product with a moisture
content from 8 to 20 percent. However, their use in the dewatering of
minus 0.58 mm (28 mesh) coal is limited by the plugging of the screen aper-
tures with fine coal particles.
Vibrating Screens—
Ihe most common screens used in modern coal preparation plants are
perforated plate and woven wire screens that vibrate. Vibrating screens,
compared to fixed screens, have a large process capacity per unit of screen
area and floor space resulting in a lower cost of operation per ton of coal
screened. They can handle wet coal product down to 0.58 mm (28 mesh) top
size. Vibrating screens have also been utilized to sludge dewater material
down to 0.08 mm (200 mesh) but with an attendant increase in product moisture
107
-------
content.
The use of vibrating screens as dewatering devices for fine coal is
infrequent, but recent developments indicate that these devices may see
increased use in helping to recover ultra-fine coal from "black water" and
in the recovery of coal fines formerly sent to settling ponds. Hie following
table summarizes the typical feed rates, product moisture and percent
recovery rate of vibrating screens used in this particular application.
Recovery of Fine Coal from Slurry
Feed rate, kkg/h/m
2
Screen Arrangement
Vibrating screen
Vibrating screen with
sieve bend
Vibrating screen with
scalping deck
(tph)/ftz
15 (1.5)
80 (8.0)
80 (8.0)
moisture
28
25
25
recovery
15
49
37
Vibrating screen operation—Vibrating screens consist of a substantially
flat screening surface, usually stretched taut and inclined, which is caused
to vibrate with a small amplitude and a comparatively high frequency. High
frequency vibrations keep the screen openings clear of wedged particles aid
helps to stratify the particles. In general, the finer the size of the coal,
the smaller is the amplitude and the higher the vibration frequency. The
screening surface often has slotted openings and is usually constructed of
stainless stell to provide resistance to abrasion, corrosion and erosion.
The screening surface is set in a rectangular, circular or conical
frame having side walls suitable to confine the slurry flow. Vibratory
motions are either created by mechanical devices such as eccentric rotors,
cams or sprockets, or by electrical vibrators.
Particle movement on or over the screening surface is the result of
gravitational forces and the force exerted on the particles by the screen
surface. Most vibrating screens utilize gravity as the primary force. The
108
-------
screen surface is inclined to such an extent that a very slight impulse to
the screen is sufficient to cause the particles to progress down the
surface.
Vibrating screen performance—The performance of vibrating screens is
influenced by several factors, such as: depth of bed, screen fabric wear,
feed rate and, most important, percent of solids in the feed slurry.
Vibrating screens, combined with sieve bends or scalping decks, are efficient
devices in primary dewatering.
Recent technology; the Derrick screen—Derrick Manufacturing Company
claims to have developed a screen surface which is non-blinding. It consists
of a sandwich screen concept - the mounting together of two screen
surfaces - and utilizes a patented rubber screen surface to reduce
wear. The action of the two screens against each other, due to a special
high-frequency, low-amplitude vibrating motor, prevents near-size particles
from becoming lodged in the top screen. The openings in the bottom screen
are constructed slightly larger than those in the top screen to ensure that
particles do not become trapped between the screens. Ihe manufacturer claims
that this screen is capable of effecting a sharp separation down to 0.037 mm
(400 mesh) - The Derrick screen also incorporates a section consisting of a
slotted natural rubber surface fitted with a patented vacuum system to
assist in removing additional surface moisture.
Stationary Screens: The Sieve Bend—
die of the main types of stationary screens utilized in modem coal
cleaning plants is the sieve bend. Originally developed by the Dutch State
Mines in the early 1950's as a fine coal classifying device, the sieve bend
is now widely used for the primary dewatering of fine coal slurry ahead of
vibrating screens and centrifuges to reduce the water load on these latter
devices. The sieve bend has no vibrating or moving parts and utilizes only
gravitational forces. Because of these factors, the sieve bend has a low
operational cost.
109
-------
In essence, the sieve bend is a stationary wedge bar screen with the
bars oriented at right angles to and across the line of flow. The screen-
ing surface is comtonly curved in a 60 degree concave arc at a radius of
51 to 152 cm (20 to 60 inches) with 76 cm (30 inches) being the most coimion.
The feed end of the sieve is tangential to the vertical resulting in the dis-
charge end lying at 30 degrees to the horizontal for a 60 degree sieve bend.
The coal particles on the surface of the sieve bend start to slow down at the
discharge end, thus contributing to the dewatering function.
Sieve bend operation—The sieve bend is of greatest value when treating
slurries with high moisture content. Coal slurry at zero feed pressure is
introduced into the sieve bend feed box. A series of baffles in the feed box
spreads the slurry evenly across the entire width of the screen deck. The
slurry drops from the last baffle, passes through the feed spout, and is fed
tangentially onto the screen surface.
The full stream of coal slurry flowing over the sieve bend surface
decreases in depth in increments of about one-quarter of slot width each
time it passes a slot; the water passes through the slot openings while the
coal particles continue over the screen surface. The effluent from the sieve
bend flows into the effluent chamber and the particles of coal are discharged
over the lip of the sieve bend.
Sieve bend performance—A volume of 750 liters per minute (200 gallons
per minute) of 57 x 0 mm (2%f x 0 inch) coal slurry per 0.3 meter (foot) of
sieve bend width can be reduced from 85 percent moisture to 35 percent
moisture on a 76 on (30 inch) radius bend containing slot widths of 1/2- or
3/4-nm and 1-1/2-rrm wedge bars. Volumes as high as 3,735 liters/min/meter
(300 gpm per foot) of width can be processed by the sieve bend in preparation
for feed to tables or jigs where high moisture is not objectionable and under-
size can be tolerated in the overproduct.
110
-------
The Vor-Siv—
The Vortex Dewatering Sieve, commonly known as a Vor-Siv, was developed
by the Polish Coal Industry to fit the need for a unit that would dewater and
classify high volumes of coal particles in a water slurry. The Vor-Siv is
manufactured under license from the Polish government and, as of 1976, there
were 36 Vor-Sivs installed in the U.S.A., with another 29 in the process of
being installed or delivered. The Vor-Siv combines the operating character-
istics found in cyclones, sieve bends, cross-flow screens and vibrating
screens, has no moving parts and requires no energy for operation.
Vor-Siv operation—Dewatering of solids by the Vor-Siv is accomplished
by the spiraling or vortex flow of the slurry over a stationary inverted
cone-shaped wire screen. Material to be dewatered is gravity fed, or pumped
in slurry form through the inlet nozzle. A covered trough directs the
flow around a circular raceway and down onto a fine screen positioned at a
45 degree angle to the raceway.
The vortex motion, created by the centrifugal and gravitational forces,
causes the solid particles of the slurry to change from a radial flow to a
downward spiraling flow on the surface of the inverted cone screen. The
liquid with the undersize solids circulates in an almost horizontal path and
is discharged through slits of the screen into the effluent collecting
chamber. A discharge outlet is provided for the dewatered material.
Wear is a major concern in all fine screening devices. While the
wedge-bars in the Vor-Siv are much heavier than the wire used in fine screen
cloth, abrasive wear increases the gap between the bars. Wear and blinding
irust be monitored to assure efficient operation.
Vor-Siv performance—To date, the most popular applications of the Vor-
Siv in the U.S.A. has been as a dewatering device ahead of centrifuges, a
desliming device ahead of Deister Tables, and as a load reducer device ahead
of thickeners and flotation cells. The Vor-Siv has processed feeds of minus
9.53 mm (3/8 inch) particles in slurries ranging from 10 to 30 percent solids,
with flow rates up to 12,000 1/min (3,200 gpm) . A coal slurry containing
136 kkg/hr (150 tph) of dry solids has been successfully dewatered by a
single unit.
Ill
-------
In general, a sieve bend or cross-flow screen effecting a 0.58 irtn
(28 mssh) separation, can be expected to process about 114-151 I/tain (30 to
40 gpm) of slurry per 0.093 m2 (square foot) of wire surface, whereas a Vor-
Siv will process from 190 to 308 I/tain (50 to 55 gpm) up to 265 1/min (70 gpm)
of slurry per 0.093 m2 (square foot) of screen area. Sieve bends and cross-
flow screens providing a separation at 28 mesh will produce 35 to 50 percent
surface noisture in their plus 0.58 mm (28 mesh) material. The Vor-Siv can
produce a product with a moisture content of about 20%.
Rotating Wedge Wire Screens—
Another type of screen used for dewatering fine coal is the rotating
wedge-wire screen. This mechanism allows wastewater to pass vertically
downward from the outer surface to the inner surface of a drum by gravita-
tional forces. Ihe screened wastewater then passes out through the lower
half of the drum to a collection trough. Solids are retained on the outer
surface of the drum and are removed by a fixed scraper blade. A screen
spacing of 1.5 ran (0.06 inches) is reconmended for the dewatering of fine
coal.
The V-Screen—
Ihe V-Screen also can be used for dewatering fine coal. This screen
provides for wet coal to be fed into a vaned-disc distribution plate which
is located at the top of a drum and rotated in unison with the drum. Ihe
coal particles are dispersed outward to the top of the outer perineter of the
screening cloth which is attached to the drum. The combination of high speed
gyration and centrifugal forces throws the water through the screen. When a
feed product of 4.76 mm x 1.65 mm (3/16 inch x 10 mesh) containing 22% mois-
ture is processed by the V-Screen, the moisture content is reduced to 10%.
Centrifuges
Centrifuges - sometimes called centrifugal driers - are itechanical
devices which use centrifugal forces to dewater fine coal products received
from primary dewatering units. Ihe feed to a centrifuge usually has been
first processed by dewatering screens or primary dewatering devices. The
objective in the centrifuging of fine coal is to reitove as much of the
remaining free moisture as possible through the application of centrifugal
112
-------
force.
In general, the centrifuged product contains from 10 to 20 percent mois-
ture, depending upon several factors which are:
• amount of centrifugal force applied;
• amount of moisture in the feed;
• size consist of the feed;
• rate of flow to the centrifuge;
• type of screen cloth used;
• size of openings in the cloth; and
• chemical or heat additives used to reduce the surface tension.
Basic types of centrifuges in use in modern coal preparation plants are:
• screen bowl centrifuges;
• solid bowl centrifuges; and
• vibrating basket centrifuges.
Perforated Basket Centrifuges—
Perforated basket or screen bowl centrifuges commonly have baskets that
range frcm 30.5 to 137 cm (12 to 54 inches) in diameter. The types in
general use are those with positive transporting devices and those without
positive transporting devices.
Perforated basket centrifuge operations—Positive discharge perforated
basket centrifuges have two rotating conical elements: an outside conical
screen and an inside solid cone which carries spiral hindrance flights. A
single motor provides power for the gears which produce a differential speed
in the two rotating elements. The two elements rotate in the same direction,
but the screen element moves slightly faster than the cone carrying the
spirals. The outer drum is usually made of stainless steel wire with re-
placeable screens mounted on its inner surface.
The wet coal enters the machine at the top and falls on the apex of the
cone where the centrifugal force developed by the rotating cone throws the
coal slurry up against the screen. The water passes through the perforations
and is collected in an effluent chamber while the coal is gradually trans-
ported to the bottom of the screen cone as the flights spiral downward. The
113
-------
increasing diameter of the cone and screen subject the coal slurry to zones
of increasing centrifugal force.
Perforated basket centrifuges without transport are the simplest of all
centrifuges. Partially dewatered coal is introduced to the top of an
inverted, truncated, conical basket. The coal particles are accelerated by
the rotating basket and the water drains through the screen perforations.
The coal particles slide down the surface of the basket to the base of the
cone, where they are discharged. This type of centrifuge without transport
has found limited application.
Perforated basket centrifuge performance—The finished coal product has
a surface moisture content that generally is 20 percent lower than
similar coal dewatered by vacuum filters. On fine coal (20% minus 325 mesh),
the dewatered result is 12 to 14 percent surface moisture. Throughputs range
up to 91 kkg/hr (100 tph) and recovery almost always exceeds 90 percent.
Solid Bowl
The two principal elements of the solid bowl centrifuge are the contour-
ed, rotating bowl, which acts as a settling vessel, and the conveyor or
scroll for discharging the settled solids. The bowl has adjustable overflow
weirs at its larger end for the discharge of the effluent and fixed parts
at the opposite end for discharge of the solids.
Solid bowl centrifuge operation—As the bowl rotates, the centrifugal
force causes the slurry to form an angular pool whose depth is determined by
the adjustment of the effluent weirs. The solids discharge end of the bowl
is reduced in diameter so that it lies submerged in the pool thus forming a
drainage deck for dewatering the solids as they are conveyed across it by
the scroll. Feed enters through a stationary supply pipe and passes through
the conveyor into the bowl itself.
Solid bowl centrifuge performance—The major advantage of the solid
bowl centrifuge is that it can be used to dewater ultrafine coal slurries
with high clay content. Automatic addition of flocculating agents is also
available. A second advantage is that the solid bowl centrifuge v/ill
operate continuously without close attention from operators. In an attempt
to close the preparation plant water circuit, a recent modification of the
114
-------
solid bowl centrifuge has been introduced by a major manufacturer. By
increasing the depth of the pool and moving solids concurrently, a test model
has a proven solids recovery in excess of 99.9 percent.
A disadvantage of the solid bowl centrifuge is that the machine requires
additional power, because it must accelerate the water load as well as the
solids. Considerable power is also consumed because the scroll must push the
solids up to the discharge ports.
Centrifuge bowls are available in diameters of up to 152.4 cm (60 inches).
Capacities can reach 91 kkg/hr (100 tph).
Vibrating or Oscillating Basket Centrifuges—
The third type of centrifuge is the vibrating or oscillating basket
centrifuge, with either a horizontal or vertical axis design. Its primary
use is in the dewatering of a coal slurry containing particles coarser than
0.59 mm (28 mesh). Vibrating basket centrifuges are frequently installed
because of the lower power requirements and long basket life.
Vibrating basket centrifuge operation—The vibrating basket centrifuge
is distinguished from other basket types in that the rotating basket is
vibrated axially which causes the solids to move through the machine to the
larger diameter discharge end without the aid of a transport device. The
oscillations also keep the basket opening clear and constantly loosens the
cake while aiding dewatering. Two motors are generally employed, one for
basket rotation and one for vibratory action.
The feed enters at the smaller end of a truncated screen, where it is
subjected to a centrifugal force. Minimum degradation of the coal is in-
curred during the movement to the discharge end because of the lower rpm's
and consequently lower G forces (25 to 120). Lack of scrolls to crush the
coal also helps to keep degradation to a minimum. The lower centrifugal
force allows for a thicker coal bed for higher throughput. Wear on the
basket and horsepower requirements are also minimized because of the utiliza-
tion of a lower centrifugal force.
Horizontal axis vibrating basket centrifuges differ from vertical axis
units only in the orientation of the unit, feed mechanism and in the fact
that the horizontal unit requires less headroom for installation.
115
-------
Vibrating basket centrifuge performance—A disadvantage arising from
the lower centrifugal force is a higher coal moisture content than that
produced by perforated basket or solid bowl centrifuges treating similar
coal slurries. However, since the vibrating basket centrifuge primarily
treats the coarser, plus 0.59 mm (28 mesh) coal, the resultant moisture
content is acceptable. Solids recovery generally exceeds 97 percent, depend-
ing upon the friability of the feed coal. The current maximum capacity of
a vibrating basket centrifuge is 320 kkg/hr (350 tph).
Recent technology—A number of new developments are evident in various
centrifuges, although these are mostly in terms of capacity and improved
maintenance. Germany and U.S.S.R. have developed a double-basket type
centrifuge, which, it is claimed, does away with the need for dewatering
screens ahead of the centrifuge. One of these is the Soviet VGS-s, a
horizontally rotating and filtering centrifuge developed to dewater pulps of
low solids content produced during jigging operations. It vibrates horizon-
tally and features internal pushers to ensure a more effective contact
between the coal particles and the basket screen. The VGS-2 will accept a
top size feed of 19 mm x 0 (3/4 inch by 0). Depending upon the size consist
of the feed material, this centrifuge claims to provide a final product con-
taining from 5 to 10 percent moisture. A prototype machine is reportedly
under construction in the United States.
Spiral Classifiers—
Spiral classifiers are frequently used in coal preparation plants to
assist in the dewatering of coal slurries and coal refuse. However, their
most typical use is in pre-thickening of coal refuse suspended in the prepa-
ration plant water recycle circuit prior to a thickening or filtering opera-
tion.
The most common type of spiral classifier used in modern coal prepara-
tion plants is the horizontal current type in which the slurry is introduced
into the pool area and the settled solids are carried away by a spiral device.
Spiral classifier operation—Spiral classifiers consist of an inclined,
round-bottomed tank with one or two spirals mounted on a shaft parallel to
the bottom. The spiral struecure conveys the settled solids up the bottom
of the tank to a discharge lip. The slurry is fed into the classifier pool
116
-------
with a minimum head and at pool level to minimize undesirable agitation. The
pool level is maintained by adjusting the height of the overflow weirs. The
overflow drops into a collection pipe and is usually routed to a thickener.
The underflow may discharge directly to the refuse belt if sufficiently
dewatered or, if not, to some form of secondary dewatering device. The size
of separation is determined by the height of the overflow weir, feed rate of
the slurry and the rotational speed of the classifier spirals. In general,
the higher the flow rate or the increasing rotational speed of the spirals,
the coarser the size of separation.
Spiral classifier performance—One spiral classifier installation in
West Virginia processes 59 kkg/hr (65 tph) of coal refuse and 4,350 1/min
(1,150 gpm) of water. The split, at 0.30 mm (48 mesh), results in approxi-
mately 55 kkg (61 tons) of overflow material per hour. Elimination of much
of the coarse solids also reduces the load on the thickener.
Filters
Filters are used in dewatering of froth flotation concentrates and fine
coal slurries from thickeners. Filters process a slurry suspension contain-
ing a high percentage of solids and remove the water to produce a wet cake
of coal particles ranging from 20 to 30 percent moisture.
The two basic types of filters are disc-type and drum-type.
Disc Filters—
Disc filters offer several advantages over other types of filters.
large amounts of filter surface area , up to 400 square meters (4,300 square
feet) f can be contained in relatively small areas of preparation plant
floor space. The disc filter is ideal for use on relatively slow filtering
slurries requiring long drying periods with little or no cake washing.
The initial capital cost is about half the cost for the same surface area
contained in rotary drum-type filters. ¥he design of a disc composed of
many wedge-shaped sections facilitates ease of maintenance as each wedge
can be relieved and serviced separately, or it can be replaced with a spare
wedge while the original is being repaired.
117
-------
Disc filter operation—The disc filter consists of a number of round,
disc-shaped filter sections mounted on a rigid, hollow shaft. Each filter
section is composed of several wedge-shaped sectors.
The filter sections are connected tc the vacuum system by a series of
pipes that run through the hollow central shaft to a rotary valve mechanism
located at the end of the shaft- The wedge-shaped hollow sectors are
connected to the vacuum system through these pipes. The outside of each
sector is covered with a filter media, commonly constructed of stainless
steel wire mesh, saran, polyethylene or nylon.
The lower portion of each disc is submerged in the coal slurry
receiving tank or trough. In open tanks, an agitator is present to ensure
that settling of the coal particles does not occur. Newer filters employ
individual troughs for each filter disc so that additional agitation is
not required to prevent particle settling.
The entire disc assembly is rotated by a worm gear attached to the
hollow shaft. As each sector of the disc is submerged into the tank contain-
ing the coal slurry, it is connected to the vacuum outlet in the shaft by a
rotary valve. A minimum pressure differential of 0.67 to 0.80 atm. (20 to 24
inches of mercury) is required and remains in effect during the time the disc
sector is submerged and for a short time afterward to dry the cake further.
Filtrate passes through the filter media into a pipe, through the hollow
shaft and out the rotary valve, leaving a cake of coal particles on both sides
of the sector.
Removal of the filter cake is accomplished by a combination of scraping
and reverse blow-back. Prior to re-entering the slurry tank, the filter
sector is connected to the compressed air port causing air to be blown into
the sector. The cake is first loosened, then scraped off and finally con-
veyed away from the filter.
Cycle times range from one minute per revolution to three minutes per
revolution, depending upon the percentage of minus 0.07 mm (200 mesh) material
present, percentage of solids in solution and the amount of clay or slimes
contained in the finer fractions.
118
-------
Disc filter performance—At the usual vacuum of 0.53 to 0.60 atm. (16 to
18 inches of mercury) preparation plant filters processing a fine coal slurry
containing 30 percent solids produce a filter cake containing 18 to 28 per-
cent moisture at rates ranging from 40 to 60 kkg/hr (48 to 70 tph).
Recent technology—A new line of disc filters has been engineered by one
company. The features reportedly include a disc filter diameter of more than
5.27 m (17.3 feet), 30 sectors per disc [32 kg (70 pounds) per sector],
individual sector drain lines, operating speeds ranging up to 4+ KFM, a
narrow, individual filter trough for each disc, no need for an agitator,
deep sutmergence (55 percent), and precisely controlled blow-back.
Filtering efficiency is reportedly increased to discharge 95 percent
of a filter's caking per revolution. Another reported advantage is a 45
percent power saving attributed to a lower power requirement for the drive.
Drum Filters—
Drum filters range in size up to 3.15 m (10.33 feet) in diameter and
8.46 m (24.75 feet) in length. Longer versions are available if a compressed
air discharge is not used. Filter surface area can reach 93 square meters
(1,000 square feet).
The chief advantage of drum filters over disc filters is that drum
filters have a cleaner discharge because of the thinner filter take. Also
important is the drum filter's ability to operate under varying feed load
conditions.
Modern coal preparation plants utilize two types of drum filters: those
with inside filter media and those with outside filter media.
Drum filter operation-—In an outside filter media design, a drum covered
with filter media rotates in a tank containing the coal slurry. During the
time that a section of the filter drum is passing through the tank, and for a
period of time after the section emerges, a vacuum is applied to the inside
of the drum. This causes the liquids and solids to be drawn to the filter
media where the liquid and minus 0.07 mm (200 mesh) solids pass through the
media and out of the center of the drum, to the filtrate discharge portion of
the system. The solids in the slurry foun a filter cake which dries as the
119
-------
drum continues to rotate. The cake is discharged prior to re-entry into
the slurry tank.
The two types of discharge used in coal preparation are the belt dis-
charge and the blow-back discharge.
In the belt discharge, the filter belt is drawn away from the drum face.
The cake breaks away and falls into a receptacle as the belt turns at a sharp
angle over the discharge roller. In blow-back discharge, air pressure
"billows" the filter cloth gently against the scraper blade as the drum
rotates toward the discharge roller. Filter wear is minimized by the blow-
back discharge method.
On an inside filter media design, the division strips, drainage strip
and filter media are essentially the same as on the outside filter design.
One end of the drum is completely closed and the filter has no outer tank,
but rotates on rollers. A discharge chute is placed in the drum along the
axial center line.
Coal slurry is fed directly into the drum and forms a pool at the
bottcm. No agitator is necessary as the quick settling of the solids is a
desirable feature with filters of this type. As the filter cake emerges
from the slurry pool it is subjected to a spray wash. When a section
reaches the vertical centerline, a pressure blow-back causes the cake to
drop off into the discharge chute. In seme models, a conveyor belt is used
to bring the discharged cake out of the drum. This type of filter is not
used on slow filtering slurries because of the short time between emergence
fron the slurry pool and the cake discharge.
Drum filter performance—It is possible to reduce the iroisture content
of a flotation froth from 65 to about 18 percent, which greatly improves the
handling characteristics of the fine coal.
Other Filters—
Pressure filters have found limited applications in coal preparation
plants, primarily in foreign countries. New installations of these devices
are uncommon in the United States because of the high capital cost involved,
as much as 12 times the initial cost of a comparable capacity disc filter,
and because of the manpower required to operate and assist with the discharge
120
-------
of the filter cake. The addition of a fully automated discharge type
pressure filter reportedly has re-established American interest in these
units. A single operator may be able to operate up to 10 filters.
Recent Technology—
A recent development in filtration is the horizontal, belt-type vacuum
filter utilizing steam as a further drying aid. While this type of operation
is approximately 2-1/2 times as costly as conventional disc filtration at
ambient conditions, preliminary data indicates that product moisture can be
lowered to levels (6.9 to 9.0 percent) that would permit the elimination of
thermal driers in coal preparation plants.
Horizontal belt filter operation—Basically, the system consists of a
3.2 m (10.5 feet) long horizontal belt filter that conveys the coal slurry.
The filter system applies the negative pressure differential from below the
belt. The initial 1.9 meters (6 feet) of the belt are equipped with a hood,
the first 1.1 meters (3.5 feet) of which supplies steam to the filter cake
while the remaining 0.8 meters (2.5 feet) supplies hot, dry air. The last
1.2 meters (4 feet) of filter length is exposed to the atmosphere.
Horizontal belt filter performance—In laboratory tests, two major
manufacturers claimed the ability to dewater approximately 1 to 2 kkg of
feed per hour per square meter (50-70 pounds of feed per hour per square
foot) of active filter area to a final moisture content of 6.5 to 7.5 percent.
Hydraulic Cyclones
As they are used in fine coal dewatering, hydraulic cyclones, more
commonly referred to as hydro- or water-only cyclones, have similar applica-
tion to static thickeners: feed pulp percent solids is 8 to 15 percent and
is reduced to between 40 and 60 percent solids. However, the thickener will
more easily handle large volumes of slurry containing very fine coal.
In appearance/ the hydrocyclone differs from the conventional classify-
ing cyclone , the hydrocyclone has a short, stubby bottom, while the classi-
fying cyclone has a long, tapered conical bottom. The hydrocyclone also has
a longer and adjustable vortex finder and much larger orifice sizes for a
given diameter than the conventional cyclone. Numerous variations exist in
121
-------
available bottom design, with each manufacturer claiming the advantages of
his own particular design.
When processing fine coal, either all ceramic or rubber-lined cyclones
are standard, and in general, the smaller the particle size, the smaller the
cyclone diameter.
The hydrocyclone, as a primary dewatering or thickening device for fine
coal, has numerous advantages over other thickening devices. Among these are:
• a simple design incorporating no moving parts, and therefore,
little maintenance;
• once initial adjustments have been made, usually no further
adjustments are necessary;
• limited space requirements for operation; and
• pre-screening is not required.
Among the disadvantages are:
• the large quantity of water required for proper operation means
that more horsepower is necessary to operate the circuit; and
• fine coal is lost in the overflow.
Hydrocyclone Operation—
The cyclone body consists of a short cylindrical section which is
attached to an inverted truncated conical section. The centrifugal force
causes the particles in the feed to move to the inclined side walls, where
they slide down to the cone's apex and then out the underflow orifice in
the form of a thickened slurry. The liquid portion of the feed travels to
the center of the cone and is discharged to the overflow orifice.
The two most important factors which influence the application and
efficiency of the cyclone are its diameter and cone angle. For small feed
particles, a small diameter cyclone must be used for dewatering. Also, the
smaller the cone angle (greater than 10 °), the higher the recovery obtained
in the underflow product.
122
-------
Another factor affecting the operation of the cyclone is the orifice
size. A larger underflow orifice will increase the flow rate and percentage
of solids in the underflow, thereby decreasing the concentration of solids
in the overflow. Large overflow orifices will increase the total flow,
weight concentration and particle top size contained in the overflow while
the solids concentration in the underflow increases along with the percent-
age of coarse sizes contained in the underflow.
Hydrocyclone Performance—
Cyclone thickeners are available in many sizes. The size chosen for a
particular installation is directly dependent upon the size consist of the
feed. For example, 7.6 ram (3 inch) diameter cyclones are used to process
slurries containing particles having a size range of 2.4 mm x 0 (8 mesh by 0).
The units are normally arranged in banks containing 22 cyclones each with a
carman manifold in the single feed and overflow lines. One bank of cyclones
will handle a flow of approximately 950 1/min (250 gpm) of slurry at a feed
pressure of 3.72 atm. (40 psi).
The top size of feed to a 20.3 on (8 inch) diameter cyclone should be
less than 4.8 mm (3/16 inch). The standard 20.3 cm (8 inch) diameter cyclone
should be less than 4.8 mm (3/16 inch). The standard 20.3 cm (8 inch) diam-
eter cyclone will process approximately 415 1/min (110 gpm) of slurry at a
feed pressure of 3.72 atm. (40 psig). The 20.3 on (8 inch) diameter cyclones
are normally arranged in banks of two, three, or four units with common feed
and overflow manifolds.
The 35.6 cm (14 inch) diameter cyclone has a capacity of 1,230 1/nsin.
(325 gpm) at a feed pressure of 3.72 atm (40 psig) and is designed to handle
slurries with particles up to 6.4 mm (1/4 inch).
Recent Technology—
A new, small-diameter hydrocyclone that is reportedly inexpensive and
particularly suited to multiple assembly has been developed in the United
Kingdom.. A particular feature of the new 25 mm (1 inch) device is that it
can be easily cleaned either manually or automatically. Particle diameter
cut point is reportedly 2 to 20 microns. The new type of cyclonic separator
123
-------
will be manufactured under license from the U.K. National Research and
Development Corporation.
Static Thickeners
The term "thickener" is generally applied to a large-diameter sedimenta-
tion vessel. Its primary purpose is to provide maximum concentration of the
solids that are present in the feed. The solids in the feed pulp seldom
exceed 8-15 percent, while the discharge increases the concentration of
solids in suspension to 30 to 40 percent. The thickening step is usually
followed by additional dewatering. Three types of thickeners are considered,
conventional, Lamella, and Deep Cone.
Conventional Thickeners—
A conventional thickener, which may range in size from 1.8 to 99 meters
(6 to 325 feet) in diameter, consists of a shallow circular tank with a
gently sloping bottom and a slowly revolving arm which is equipped with
blades to rake the settled solids into a central discharge point. Size of
thickeners is determined by the influent flow rate to be handled, percent
solids in the pulp, size distribution of the solids, and the amount of solids
allowable in the overflow.
Conventional thickener operation—Influent pulp is introduced into the
center of the tank. The entering pulp displaces part of its volume as a
peripheral overflow of moderately clear water, while the solid particles fall
slowly downward. Material that is sufficiently coarse falls rapidly to the
raking zone, while the remainder of the solid particles slowly settle,
leaving a zone of clear water toward the upper periphery of the tank. A
transition stage occurs through which the pulp steadily increases in density
as it settles until the compression zone is reached where the individual
particles are squeezed together by the weight of the fluid above. The
thickener rakes slowly revolve through this compression zone gathering and
sweeping the settled particles ir.to the central discharge well, where they
are removed by a pump. The water overflowing from the periphery can be
processed further or recycled to the coal preparation plant.
124
-------
Conventional thickener performance—Raw coal, or plant feed having
generally raw coal characteristics , will require a unit area for thickener
operation of 3 to 10 sq meters/metric ton/day for solids that range between
25 to 85 percent minus 0.07 mm (200 mesh). Concentration of this type
material generally is between 30 and 40 percent solids in the thickener
underflow. Flotation tailings that are between 25 and 85 percent minus
0.07 mm (200 mesh) will require unit areas for thickener operation running
between 4.2 and 6.9 square meters/metric ton/day. The economic limit of
underflow concentration is usually about 35 percent solids by weight.
Flotation tailings that are finer than 85 percent minus 0.07 mm (200 mesh)
and have a very high ash content will need unit areas between 6.9 and 13.8
square meters/metric ton/day, depending upon the ash content, and can be
concentrated to about 30 percent solids in the underflow.
Lamella Thickeners—
The Lamella Gravity Settler, usually referred to as a Lamella Thickener,
is an inclined, shallow depth sedimentation device. It performs the same
function (the same settling area for a given flow rate) as a conventional
thickener, but occupies only a small fraction of the floor space. The unit
has no moving parts, consisting only of a tank with multiple inclined
plates onto which the solids settle and slide to the bottom for collection.
Lamella thickener operation—The Lamella thickener consists of a large
tank with inclined plates inside the tank. The feed enters through
a bottomless feed box vvhich distributes the feed over the Lamella plates.
The liquid flows upward over the plates while the solid particles settle on
the plates and slide downwards in counter-current flow to the clarified
liquid. From the feed point upwards there is a free settling clarification
area. Below the feed point is the hindered settling, thickening area.
The clarified liquid enters discharge boxes through throttling holes that
insure an even flow distribution between the plates. The particles slide
down the plates to a hopper where compression is increased by a low ampli-
tude vibrator. The thickened material is then removed by a pump.
125
-------
Lamella thickener performance—These thickeners generally can produce
overflows with less than 500 ppm solids. In laboratory tests, overflows of
less than 100 ppm and thickened-sludge underflows in excess of 20 percent
by weight were obtained. Feed loading rates [20.4 to 24.4 liters/min/n2
(0.5 to 0.6 gpm/ft2)] and flocculant dosages (about 2 ppm) are in line with
those of conventional thickeners. Lamellas will handle up to 5,600 liters/
min (1,500 gpm).
Deep Gone Thickeners-
Developed by the UK's National Coal Board, the evolution of the Deep
Cone Thickener followed conmercial introduction of high molecular weight
flocculants, usually called polyelectrolytes. The apparatus is essentially
a settling vessel of considerably less settling area than a conventional
thickener and is usually cylindro-conical in shape.
Deep cone thickener operation—Several cones are usually operated in
parallel to achieve the required throughput. Each cone is gravity fed
through a flow control valve via an open launder, the purpose of which is
to mix the feed (by the use cf baffles) thoroughly with the flocculant.
The launder discharges to a central feed well, where the solids begin to
settle and, with consolidation aided by gentle stirring, form a plug at the
apex of the cone. The clarified water overflows into a peripheral launder
cind can be reused within the plant. The pressure exerted by the suspended
solids is constantly measured by a pressure pad, which transmits a pneu-
matic signal to a valve controller. Upon reaching a certain set point, the
pneumatically controlled valve allows discharge of the thickened sludge
until tlie pressure drops below the certain set point, at which time the
valve closes.
Ceep cone thickener performance—Deep cones yield an underflow product
containing 35 to 45 percent moisture and perform well on flotation tailings
containing high percentages of minus 53 micron particles (-270 mesh).
126
-------
THERMAL DRYING
Thermal drying is a process of accelerated evaporation of surface
moisture. The wet coal is brought into contact with hot gases to produce
surface moisture in fine coal (6 to 12 percent range). The two types of
thermal driers in use in coal preparation plants are direct and indirect.
Direct Heat Driers
Direct heat driers bring the thermal transfer agent (hot gases) into
intimate contact with the fine coal. This must be done carefully so that
the coal neither catches fire/ loses volatile matter, or in the case of
coking coal preparation, is oxidized, which would destroy its coking
properties. At the same time, the exhaust gases, carrying moisture away
from the coal surfaces must be sufficiently hot to prevent condensation of
moisture before it is removed from the drier. A multitude of factors affect
the performance capability of a thermal coal drier: drying temperature,
fuel, inlet temperature, air volume and drier size. However, the greatest
single factor affecting performance is temperature. Temperature in the
drying zone should always be as high as safety will permit. When low temper-
atures are used, sensible heat losses in the exhaust gas are usually greatly
increased, because a high air flow is needed to deliver the required heat.
Moreover, lower temperature means lower thermal efficiency, higher fuel and
power requirements and increased amounts of dust carryout. There are six
basic types of direct heat thermal driers in use in coal preparation today.
Their usage is shown in the table on the following page.
127
-------
Thermal Drying of Bituminous Ccal and Lignite by-
Type of Equipment
Percentage
1974 1975
Drier type:
Fluidized-bed 68-3 72-5
Multilouver 8•3 5.5
Rotary 1-9 2.2
Screen 5.5 7.9
Suspension or flash 16.0 11.7
Vertical tray and cascade — 0*2
Ttotal 100.0 100.0
Based upon approximately 32.6 metric tons (36 million short tons) dried per
year.
SOURCE: Mineral Industry Surveys, Coal—Bituminous and Lignite in 1975,
U.S. Bureau of Mines
Fluidized Bed Driers—
Basically, the principle of fluid bed drying is uncomplicated: Air
heated by either a pulverized or stoker-fired coal furnace is pulled upward
through a constriction plate by a negative pressure suction fan. Heated air
suspends the coal particles in a buoyant effect and causes than to act like
a turbulent liquid. This "liquid" flows at a relatively even depth from the
feed end to the discharge end of the fluidized bed drier. Ihe fluidized
coal particles are completely surrounded by hot drying gases viiere intimate
contact is obtained between the gases and the coal particles for removal of
surface moisture. Ihe principle of fluidization as applied to the drying
process results in a thermally efficient method, for moisture removal from
coal solids.
Fluidized bed drier operation—Ihe fluidized bed drier utilizes a
constriction plate framed into a housing such that all drying air passes
uniformly through the plate. f!aterial to be dried is fed uniformly onto this
plate, while heated air is pulled through the constrictions under a negative
pressure by an induced draft fan. Eeated air passing through the plate
128
-------
creates extremely high velocity air currents which suspend the material
above the plate thus causing the mass to act like a liquid moving from feed
to discharge. Suction operating pressures assure that all the drying gases
pass through a dust collector, preventing the loss of ines through leakage.
Fluidized bed drier performance—1976 results of tests on one American
manufacturer's fluidized bed drier reported evaporation of 43.5 kkgph (48 tph)
of moisture for Western subbituminous ooal and 47.2 kkgph (52 tph) of
moisture for Eastern bituminous coal. Surface moisture in the Western coal
was reduced from 25 to 12 percent and in the Eastern coal from 11.5 to 3 per-
cent.
Multilouver Driers—
The Multilouver drier is adapted to handling large capacities and is
applicable to those materials requiring a comparatively short retention
time.
Maltilouver drier operation—The multilouver thermal drier consists of
a solid outer cylindrical shell containing full length louvers and an inner
shell of overlapping louvers which support the bed of material and increase
in diameter in the direction of flow. The material travels towards the dis-
charge end as the drum slowly revolves. In operation, the heat transfer
medium is introduced through the louver openings, permeates the bed and
intimately contacts every particle.
Multilouver drier performance—A 1976 test run on 4.5 metric tons
(5 tons) of Oolstrip coal resulted in a moisture reduction from 25 to 10
percent. Results indicated that Western subbituminous coal could be dried
at realistic feed rates without spontaneous combustion.
Rotary Driers—
The Rotary Drier, sometimes called a Drum Drier, usually consists of a
solid outer cylindrical shell and an inner shell of overlapping louvers
which support and cascade the coal particles as the drying gases pass
through the drier from the outer annulus to the inner tube.
129
-------
Rotary drier operation—The rotary-type drier essentially is a rotating
drum, pitched longitudinally about 2.9 ran per meter (3/8 inch per foot),
with the wet coal particles introduced at one end of the drum and the dried
coal discharged fron the other. Either heated air or hot ccmbustiori gases
may be used to dry the coal, and the flow of the air or gases may be either
with or counter to the flow of the coal. The air or gases may or may not
come in direct contact with the wet coal.
Screen-Type Driers—
Manufacturers of screen-type driers claim that this type of drier has
an additional advantage over other types: a considerable amount of excess
water is disposed of mechanically within the machine by applying gas pressure
alternately to each screen section. The gas pressure is strong enough to
clamp the coal to the screen deck during the cycle's suction phase and
excess water is literally squeezed from the coal.
Screen-type drier operation—As its name implies, this type drier pro-
cesses the wet coal on a screen while hot gases are passed either up or
down through the particles. The mechanical squeezing action has an addition-
al advantage: there is less tendency for the coal fines to be blown from the
bed. The coal is exposed to the hot gases for approximately 50 seconds.
Suspension or Flash Driers—
The term "flash" is derived from the fact that the wet coal is con-
tinuously introduced into a column of high temperature gases and moisture
removal is practically instantaneous. The system is based upon high inlet
gas temperatures that can reach 649°C (1,200°F) and outlet gas temperatures
of about 93°C (200°F). The high inlet temperature is possible because of
the nature of flash drying and does not result in any change in coal charac-
teristics.
130
-------
Flash drier operation—The Flash Drier introduces wet coal continuously
into a column of high temperature gases and the removal of any surface
moisture is practically instantaneous. This is accomplished by the applica-
tion of heat to the particles while they are transported in a turbulent
stream of hot air. The coal particles are in contact with the hottest gases
for approximately 0.5 seconds and in the gaseous stream:for approximately
5 seconds. The temperature for inlet gases is approximately 650°C (1,200°F).
Feed is introduced to the Flash Drier with a table, screw, or rotary star
feeders.
Flash drier performance—Flotation froth containing 25 to 30 percent
minus 0.07 mn (200 mesh) material can be reduced from 20 to 8 percent
moisture content.
Recent Technology—
The Turbo-Drier is of particular importance because it can contain the
inert nitrogen atmosphere that must be maintained during the drying process
to prevent the occurrence of explosions or fires and preclude any deteriora-
tion of the coal particles. Oxygen content in the unit is decreased to
three percent or less by totally sealing the unit in order for it to operate
successfully under the dusty conditions caused by the finely ground coal.
Turbo-Drier operation—Wet coal is fed through the top of the unit onto
a stack of rotating circular trays. The coal then moves continuously
through the drier by being transferred from each tray to the one below via
stationary wiper blades. The pile formed after each transfer is then
leveled to increase surface area, which enhances the drying rate.
Turbo-Drier performance—Minus 0.25 mm (60 mesh) coal is dried at a
rate of 114 kg/hr (250 Ib/hr). The moisture content is reduced from 30 to
5 percent.
Indirect Heat Driers—
In indirect heat drying, the coal being processed never comes in con-
tact with the heat transfer agent (usually hot oil, but sometimes steam or
hot water). Indirect heat exchangers are usually jacketed conveyors - some
have helix screws, while others are equipped vith fins, paddles or discs.
The thermal fluid circulates through both the jacket and the screws.
131
-------
Aircng the advantages of indirect heat exchanging are:
• exact temperature control;
• effective drying in the fine size range;
• gentle agitation of the coal;
• minimum particle degradation;
• continuous mass flow;
• elimination of fire and explosion hazards;
• automatic control and simplicity of operation;
• relatively dust-free operation;
• reduced space requirements;
• operation under controlled atmosphere;
• conservation of power and energy;
• heat recovery and recycle;
• iirproved working conditions, and
• ease of insulation.
The disadvantages of indirect heat transfer are:
• requirement of a secondary heat transfer agent handling system
(a steam boiler or hot oil heater); and
• relatively high initial cost.
Indirect heat drier operation—The continuous, hollow, helical screw
flights are welded to a stempipe to accomodate the flow of the hot oil.
The oil, preheated to about 900°C (1,650°F), first flows through a hollow
shaft and then through the hollow screw flights, which act as a screw
conveyor to dry as well as gently agitate, mix and advance the fine coal.
Indirect heat drier performance—A black water sludge filter cake was
fed to an indirect drier at the rate of 34 kkg per hour (37.5 tons per hour)
Moisture content of the feed was 20 percent; of the resultant discharge, 5
percent.
OIL AGGLOMERATION
The use of oil agglomeration in the dewatering of fine coal has
attracted a great deal of interest. It is, perhaps, the only large scale,
economic method presently available for recovering ultrafine particles
[minus 0.04 run (325 mesh)] from an aqueous slurry. This process, sometimes
132
-------
called pallatizing with oil, is a reapplication of the principles used by the
Trent bulk oil flotation process. The Trent process was tried at various
times in the past without commercial success, principally because of (at that
time) low market price of coal.
In general, oil agglomeration offers several possible benefits to
operators of modern coal cleaning plants. Among these are:
• reduction of product moisture of the fine fraction from 20 to 10
percent and a 1 to 2 percent decrease in overall product moisture;
• the formation of low moisture, easily drained pellets, which
eliminate dust problems in fine coal handling and transport;
• easy removal of deleterious clay and shale;
• a 2 to 3 percent increase , some estimate eight million tons
annually , in overall yield from coal preparation plants; and
• substantial removal of environmentally hazardous heavy metals,
such as mercury, arsenic and selenium, along with recovery of
potentially valuable vanadium, from ultrafine coal.
Cti the negative side, the processes currently in use require relatively
large amounts of oil, the supply and increasing cost of which makes economic
feasibility prediction at least risky if not extremely speculative.
Laboratory tests were conducted in 1973 on six different U.S. and
Canadian coals. The coals were wet ground as a slurry containing 40 to 50
percent solids and size distributions of 100 percent minus 0.07 mm (200 mesh)
and 40 to 50 percent minus 0.044 mm (325 mesh) particles. Agglomeration was
effected through actiition of a light petroleum distillate and agitation for
five minutes. The recovery in the agglomerates of the combustible content
of the feed exceeded 90 percent in every case.
At present, three basic processes are in use, Trent, Convertol and
Spherical.
The Trent Process
The bulk oil Trent Process was developed during World War I to remove
ash from powdered coal. Large quantities of oil , 30 to 50 percent of the
weight of the coal , are mixed with the coal/water/ash slurry. The oil and
133
-------
coal particles form an agglomerate that sinks to the bottom of the mixing
tank. The water and ash particles are drawn off, and the agglomerate is
screened to drain off the excess water, thus reducing moisture content to
8 to 12 percent. The pellet is then subjected to a carbonizing process to
produce easily handled, firm granules of coal.
The Convertol Process
More recently, the Convertol Process, which requires much less oil, was
developed in Germany. Five to eight percent by weight of oil is spread over
the coal surfaces. The Germans used a Palman-type mill to accomplish this,
while in the U.S. a high-speed conditioner is used. The agglomerate formed
was then dewatered either in a screen basket oscillating centrifuge (German)
or a solid bowl centrifuge (U.S. Steel).
Pilot plant tests in the U.S. were short-lived because of dust control
difficulties, the final product tended to become dusty as moisture evaporated
from the filter cake. Even with oil consutrption as low as five percent by
weight, the cost of the recoverable, low (10 to 12 percent) moisture coal is
high when compared to that of a usual cleaning plant froth flotation/vacuum
filter/thermal drier circuit.
Spherical Agglomeration
A family of techniques, labeled Spherical Agglomeration Processes, has
been developed by the National Research Council of Canada. These processes
go a step further than the Convertol and Trent methods - oil consumption is
reduced further to three percent by weight. The drained agglomerates are
rolled into uniform size pellets on a balling disc with as much as 25 per-
cent by weight additional oil being used as a binder. The pellets are then
thermally dried to effect further moisture reduction. The resultant pellets
contain only 8 to 12 percent moisture as conpared to 18 percent or more for
filter cake.
Full economic evaluation of this processing method has not yet been
completed.
134
-------
SECTION 6
COAL SLURRY SAMPLING
Fine coal cleaning involves the use of water as a washing and transport
medium. In a coal preparation plant the fine coal slurry is transferred
fron one unit of operation to the next by pipes, each unit receiving slurry
feed from one pipe and producing clean coal and refuse to be carried further
by other pipes. The characteristics of the flow and the slurry from each
of these pipes are vital to the optimum operational mode of each unit and
the entire system. Samples from the slurry streams must be taken to obtain
this information. If these samples are representative of the entire flow,
then the performance of each unit can be evaluated and optimized.
The objective of this section is to summarize a compilation of avail-
able information on slurry sampling. This information has been gathered
from literature references, vendor literature, types of systems in use, and
industry information obtained through telephone contacts. Two of the systems
have been selected by Pennsylvania Electric Co. to be evaluated and tested
for their applicability to the fine coal slurry circuit at the Homer City
Coal Preparation Plant.
SUMMARY CF SLURRY SAMPLING INFORMATION
The use of automatic samplers to obtain representative portions of a
full stream of slurry is a relatively new technological development. In
many cases where slurry pipelines have been used, a manual "grab" sample has
been sufficient to obtain general information about the characteristics of
the slurry.l Grab sampling is accomplished by simply opening a valve on the
pipe and collecting the slurry in a container, by collecting the slurry
from the end of the pipe as it empties into a reservoir, or by dipping
directly into the reservoir of slurry as it is being mixed.
135
-------
This inight supply sufficient information about the product delivered by
the pipe. However, when samples are needed to monitor process or unit opera-
tion performance it becomes necessary to sample the slurry in-line. This
can best be done by a continuous or time interval automatic sampler. Some
samplers continuously bleed slurry from the pumped stream, while others
obtain a sample at a preselected time interval. These samples can then be
analyzed on stream if auto-analyzers are available and practical or by
collection and submission to an analytical laboratory.
Slurries are used in numerous mineral industries. A description of some
slurry sampling and in-line sample analysis methods follows:
The Joy/Denver Sampler
The Joy/ttenver coal slurry sampler provides a sample cut across the
slurry stream diameter. The sampler mechanism is totally enclosed and has a
movable blade with an adjustable slit, enabling control of the amount of
slurry gathered. At preset variable time intervals the blade travels across
the slurry stream, perpendicular to the flow, gathering material from all
points across the diameter of the flow. Sampling time intervals can be set
in one minute increments from 2 to 60 minutes.2
The McGraw Sampler
The McGraw sampler was designed by Ray McGraw, Plant Manager of the
PENELEC Homer City Coal Preparation Plant. The McGraw sampler takes a
full-stream cut of the slurry. The cutter is a 13 on. (5 in.) diameter chute
which travels across a 10 cm. (4 in.) diameter stream. This effects full
slurry flow collection when the cutter is at the midpoint of travel and
partial flow collection as the cutter is entering and leaving the stream.
These two samplers will be installed in a fine coal circuit and will
be evaluated and tested for bias and applicability in the Homer City Coal
Preparation Plant.3
OTHER SAMPLERS AND SAMPLING SYSTEMS
The Miltronics, Ltd. Continuous In-line Sampler
The sampling device invented by Miltronics, Ltd. of Peterborough,
Ontario, differs markedly from other sampling concepts. Simple in design,
136
-------
the sampler unit consists essentially of a narrow slot, continuously rota-
ting about an axis parallel to the flow of the slurry- The kinetic energy
of the stream causes a portion of the stream to flow through the sampling
slot and be discharged through a hollow shaft. Total sample flow is a
function of such conditions as the slurry velocity, the relative areas of
the sampling slot and pipe, the back pressure developed at the discharge
spigot, and the pressure in the pipeline. For truly representative sampling,
the velocity of slurry through the sampling slot must be maintained equal to
the mean fluid velocity in the pipeline; this is easily accomplished by
installing a restricting valve in the discharge pipe to allow regulation of
the sanple flow.1*
The Mansanto Phosphate Rock Slurry Composite Sampler
The following is a description of an innovative use of a sampler to
obtain a composite sample at a phosphate rock slurry processing plant. A
small pipe with a 90° elbow is installed in the line. The sample flow pro-
duces a continuous slipstream, and the other end of the tube is attached by
flexible hose to the arm of a clarifier which turns in a circular motion.
There are several sample buckets placed in the path of the clarifier arm.
These buckets catch the sample as the stream intermittently passes over
them.5
The International Nickel Company of Canada; Qi-Line Sampling and Analysis
An automatic sampling and analysis system is now being used at the
Frood Stobie Mill.
Samples of flotation feed, concentrate and tailings are taken for
continuous on-line x-ray analyses and to provide an 8-hour shift composite
for analysis by wet chemical methods.
All pulp samples are obtained by either horizontal or vertical automatic
cutters, depending on whether the slurry stream is carried in a launder or a
pipe. The timing cycle on the cutters may vary from eight seconds to con-
tinuous to give a sample flow of approximately 30 liters/fain (8 gpm). 6
137
-------
The Outokumpu Courier 300 Qn-Stream Analysis System
The Courier 300 on-stream analysis system was developed by and is in
use at the Outokunpu Company in Finland. Ihe primary sanpler is a
continuously operating full stream cutter which diverts material to as
many as fourteen secondary samplers. Continuous full stream cuts from
the secondary sampler are automatically analyzed and then returned to the
process flow. The end of the return flow tube is piston controlled and
deposits timed samples into containers for shift and 24-hour analysis.7
BIAS TEST PLAN FOR TWD SLURRY SAMPLERS AT HOMER CITY COAL PREPARATION PLANT
The Joy/Denver and the McGraw automatic slurry samplers will be incor-
porated on the fine coal circuit of the GPU/PENELEC Hcmer City Coal Prepara-
tion Plant for mechanical bias testing.
Two operations will be conducted simultaneously for each sampler based
on equal number of cuts and equal volumes of samples obtained.
(1) Collecting and processing a primary bias cut (sampler sample); and
(2) Obtaining a comparison sample by taking and processing a full
width cut to obtain a total coal population sample.
The parameters used in determining bias will be established by
statistical comparison of those from the sampler cuts with the variability
characteristics of the full stream. Then size analysis and proximate
analysis on the bias cut samples will provide the basis for the statistical
development of the bias data for each sampler.
138
-------
SECTION 6
REFERENCES
1. Aude, T.C., and N.T. Cowper, T.L. Thompson, E.J. Wasp, "Slurry
Piping Systems: Trends, Design Methods, Guidelines", Chemical
Engineering, Vol. 78, No. 15, June 28, 1971.
2. Joy Manufacturing Company, Denver Equipment Division, "Samplers"
Specification Manual S1-B14B.
3. McGraw, R., Personal Coimtunication, PENELEC, Homer City Coal
Preparation Plant.
4. Osborne, B.F., "Continuous, In-Line Sampler for Slurry Systems
Developed in Canada", Engineering and Mining Journal, Sept. 1971,
pp 146-148.
5. Mullins, M., Personal Comunication, Mcnsanto-Columbia, Tennessee
Phosphate Rock Slurry Processing.
6. Fowler, H.B., and K.R. Kay, L. Kolpien, "Process Instrumentation and
Control at Frood-Stobie Mill", Canadian Mining Journal, Vol. 92,
No. 6, June 1971, pp 49-50, 52-54, 56.
7. Leppala, A., and J. Koskinen, T. Leskinen, P. Vanninen. "Courier
300 Cn-Stream Analysis System", Transactions of The American Institute
of Mining, Metallurgical, and Petroleum Engineers (Incorporated)
Vol. 250, Society of Mining Engineers of AIME 1971.
139
-------
SECTION 7
COAL PREPARATION REQUIREMENTS FOR SYNTHETIC FUEL CONVERSION PROCESSES
Preparation of feed ooal for synthetic fuel conversion plants uses
similar techniques for the various processes with various methods depending
on the process. All approaches use conventional mechanical equipment.
However, equipment selected differs somewhat from that used in conventional
coal preparation plants because the particle size distribution must be
closely controlled. Particle size control also requires close coordination
between mine and plant.
Coal preparation techniques are used to change the following character-
istics of coal to the required coal feed parameters for coal gasification
or liquefaction processes:
• The size distribution and mean particle size;
• The total moisture content;
• The ash content;
• The sulfur content; and
• The heating value.
Each of these parameters will have some bearing not only on the
economics of the coal conversion process, but also on the type of process
selected for the particular coal.
PARTICLE SIZE
Coal gasification processes utilize three types of reactor configura-
tions: fixed bed, fluid bed, and entrained bed systsns.
The pyrolytic liquefaction processes have the same constraints on
particle size as the various gasification processes, since they basically
utilize the same three types of reactors. The liquefaction processes that
require dissolution of the coal into a process derived solvent normally
140
-------
require a finely ground coal that can be readily dissolved
The fixed bed processes require the most stringent particle size re-
quirements, because the amount of fines going to the reactor must be mini-
mized. Generally, these processes require a coal particle size in the
range of 38 by 3 mm. (1-1/2 by 1/8 in.). The particle size distribution
determines the distribution of gases and steam through the coal bed. With
this type of reactor, particle size and distribution are critical factors,
since channeling gives incomplete gasification and lower thermal efficiencies.
With this type of reactor, particular attention must be given to the
design of the materials handling and coal preparation equipment so that
good size distribution is maintained.
Fines can be minimized or kept to acceptable levels by:
• The use of minimum degradation type coal storage facilities prior
to gasification, or the use of anti-degradation devices with more
conventional storage facilities;
• The careful design of conveying systems, particularly where there
are long hauls by the conveyor. The design must minimize the
degradation of the coal caused by transfer points from one conveyor
to another, and also the degradation caused by continuous flexing
of the belt;
• The proper selection of crushing equipment to maximize the product-
ion of crushed coal in the 38 by 3 mm (1-1/2 by 1/8 inch) size
range. This crusher selection will vary according to the character-
istics of the particular coal in question; and
• The proper selection and sizing of the screening equipment.
Because excessive fine material in the gasifier feed will lead to
channeling, the screening must be of the highest possible efficiency.
This might necessitate wet screening techniques, which would require
moisture reduction prior to gasification.
141
-------
Fine coal, minus 3 mm. (1/8 in.), could be briquetted to give a
particle size suitable for conversion. This procedure would involve
further grinding of the fine coal to give a size distribution suitable
for a briquetting process. This process could utilize some of the non-
marketable by-products from gasification as a binder to give a stable
briquet.
The fluid bed reactor requires a less stringent particle size require-
ment than the fixed bed type, but the fluid bed type reactor has other
constraints. While the fluid bed reactor is not as adverse to accepting
fine particles, it is mandatory that a uniform size distribution of the
coal feed be maintained at all times. If this is not achieved, the fluid
bed becomes unstable, affecting the gasification reactions. A large
percentage of any one particle size within the stated size range could have
an adverse effect on the fluid bed operation.
Coal preparation techniques will assist by:
• Maintaining the desired size range; and
• Maintaining the desired size distribution.
These are achieved by proper materials handling techniques combined
with the proper selection of the screening and crushing system. A stage
crushing system will screen out properly sized material before ach stage of
crushing and will minimize the production of unwanted fines.
The entrained bed reactor requires less stringent particle size
requirements than the other two types of reactors. Because the size
requirement for this system is normally very fine, such as 70 percent minus
0.074 mm. (200 mesh); closed circuit mineral crushing and grinding techni-
ques can be used to provide the material.
TOTAL MOISTURE CONTENT
The total moisture content of the coal can affect the overall thermal
efficiency of the gasification process. In some types of units water
is vaporized in the reactor, and this requires energy in the form of heat.
The heat is obtained from increased consumption of coal and increased oxygen
requirements to sustain the exothermic portion of the chemical reaction.
142
-------
Although water is required for the gasification reaction to give the
necessary source of hydrogen, steam from external sources is usually a more
economical means of meeting this requirement since it contributes to the
exothermal heat requirements and reduces process oxygen requirements. The
capital costs of an oxygen plant can be as high as 20 percent of the total
costs of the overall plant, therefore, reducing oxygen requirements can
have significant economic effects.
Some types of processes state a maximum feed coal total moisture con-
tent of 25 percent, but these units have operated on brown coals with
moisture as high as 40 percent. Certain other processes utilize high
pressure pumping of a thickened coal water slurry into the gasification
system.
Ihe moisture content of the coal, can affect screening efficiency,
particularly if the free moisture of the coal increases above an average
of five percent. The result is reduced screening efficiency with
excessive fines.
Coal preparation and materials handling techniques can be utilized for
the control of moisture, as follows:
• The use of conventional coal dewatering equipment such as screens,
cyclones, filters and centrifuges for moisture reduction;
• The use of conventional filtration equipment for fine coal,
ranging from vacuum to pressure filtration; and
• The use of thermal drying techniques ranging from indirect to
direct heat drying units, to reduce surface moisture.
Because any means of removing moisture from coal tends to be a relative-
ly costly operation, a trade-off study should be performed to determine the
optimum amount of moisture reduction with respect to cost, that should be
attained prior to gasification.
The characteristic high moisture content of some coals such as
lignites and sxfo-bituminous coals presents a problem to the coal preparation
engineer if moisture reduction is required. Because of the relatively high
143
-------
temperatures and residence tines required, a suitable method of drying this
type of coal still has to be demonstrated on a commercial scale.
ASH CONTENT
The ash content of a coal feed has a significant effect on the overall
thermal efficiency of the process. Ihe inert iraterial must be heated to the
reaction temperature, which requires corbustion of coal with the related
oxygen demand in the case of gasification processes. The total sensible
heat is not recovered fron the ash before it is discharged from the reactor;
therefore, there is a loss of thermal efficiency. Also, because of the
physical material handling problems of the ash within the reactor system, a
decrease in the ash content of the coal can affect either the sizing or
number of gasifier reactors required. With the fixed bed reactor, the
physical size of the reactor is decreased, whereas with the fluid bed reactor,
the number of gasif iers required is decreased, and lower volumes of circula-
ting gases are needed to maintain f luidization.
Several gasification processes utilize partial gasification to produce
a solid product in conjunction with the gas. This char is intended to be
burned for power generation in a conventional manner. Therefore, the ash
content of the char will have to meet certain specifications. Pesearch and
development presently being conducted to utilize low ash char by conversion
to hydrogen for process use will require a controlled feed coal of relatively
low ash content to produce a suitable char.
Coal preparation techniques will assist in ash control by such suitable
methods of beneficiation as:
• Wet jig type processes for coals down to a nominal 1 mm (8 mesh)
size;
• Heavy media separation processes for coals having a nominal top
size of 1/2 mm (20 mesh) size;
• Hydrocyclone separations for fine coals; and
• Froth flotation for fine coals.
Coal beneficiation can give the most significant upgrading benefit to
the gasification processes that require a finely ground coal, e.g., minus
100 mesh. With this fine grind, a high percentage of the finely disseminated
144
-------
ash or refuse material is liberated and is therefore in an ideal state for
separation. Capital and operating costs for fine coal beneficiation
are far in excess of coarse coal systems, and a detailed cost benefit
analysis must be performed for each case.
One other coal preparation technique that may prove beneficial to the
gasification process is the adjustment of the ash-fusion temperatures. This
factor could become very significant with the slagging processes and could
be achieved by adding such fluxes as lime.
SULFUR CONTENT
The sulfur content of the feed coal directly affects the capital and
operating costs of the synthetic fuel conversion process. As the sulfur
content of the coal increases, the sulfur content in the resulting
synthetic fuel also increases. The sulfur removal processing units
inmediately following the reactors depend on the gas flow, sulfur concentra-
tion- and the type of sulfur compound. The conversion of the sulfur compounds
to elemental sulfur depends on sulfur content of the stream. Also by-
product credit for elemental sulfur becomes questionable because of a
traditionally unstable world market for sulfur.
The sulfur content of the feed coal also influences the overall effi-
ciency of the gasification process. This relation is due to the reaction
between the sulfur and process hydrogen to form hydrogen sulfide in the
gasifier. The process hydrogen can be used more efficiently in the overall
process for methanation.
In the char producing processes, sulfur content becomes very significant
since the char is burned in a conventional manner for steam raising purposes.
Sulfur removal from a flue gas is a costly process, therefore, more emphasis
should be placed upon sulfur removal from coals prior to combustion.
HEATING VALUE
As the heating value of the coal increases, potential yield of
gas per unit weight of the coal also increases. Therefore, for a gasifica-
tion plant constructed to produce a fixed volume of specified gas, the
reactor requirements will change with changes in the heating value
145
-------
of the feed coal. As the heating value increases, the reactor volume
required is reduced, and the associated plant capital and operating costs
are reduced.
•There is an inverse relationship, although not proportional, between
the ash and heating value of a coal. If a bituminous coal is
being considered, the heating value of the coal increases as the ash
content of the coal decreases. If the ash content of the coal is reduced,
this will bring about an increase in the heating value of the coal.
Certain gasification processes have quite stringent tolerances for
the day-to-day variation of the feed coal. For example, the Lurgi process
specifies a coal with a maximum variation of ± 63 kg cal (250 BTU). As the
heating value varies, the oxygen consumption also varies, and to
accotnodate the heating value variation, the coal tonnage feed rate must be
varied. large deviations in coal feed rate causes excessive temperature
fluctuations leading to clinker formation and other problems. Other types
of reactors may require even more stringent tolerances than those already
mentioned. The reaction in a fluid bed reactor becomes very unstable as the
heating value of the feed coal varies.
Because the heating value of any coal is likely to exceed the previous-
ly mentioned tolerance within the same coal seam, or between different seams,
coal preparation must be used to stabilize these tolerances. The beneficia-
tion process alone may be sufficient to give the required control. If not,
then some form of blending of the various types of coal is used.
EFFECT OF COAL BENEFICIATTCN ON COAL CONVERSION AND UTILIZATION
An experimental program entitled "Effects of Coal Beneficiation on Coal
Conversion and Utilization" was initiated at the Department of Energy
Pittsburgh Energy Research Center in FY 1977. This is an initial step in a
broad ranging program to evaluate the impact of coal benef iciation, both
physical and chemical, upon technologies for gasification, liquefaction,
and fluidized-bed combustion of coal.
The objective of the program is to determine whether coal benef iciation,
by removing sulfur or other mineral matter, has, beneficial or adverse effects
146
-------
on coal conversion to liquids arid gases and on fluidized-bed conbustion;
that is, whether a coal conversion or utilization facility designed to use
a particular coal should include cleaning of the coal as part of the
optimum design.
The program is planned will evaluate the overall impacts of the
combination of coal beneficiation with conversion and utilization of coal.
In coal liquefaction, for example, mineral matter present in a specific
coal could affect the liquefaction process in several ways. Some evidence
indicates that coal pyrite catalyzes coal liquefaction. Workers at several
laboratories including the Pittsburgh Energy Research Center, are studying
this hypothesis. Some trace elements may also act as catalysts. On the
other hand, mineral matter in coal (1) decreases the throughput of organic
matter in the conversion reactor by increasing the total amount of material
processed for a given amount of product, (2) increases the process heat
requirements, (3) complicates downstream purification and solids separation
or increases the ash content of the product, and (4) may poison conventional
hydro-desulfurization and shift conversion catalysts. The trade-offs of
identified advantages and disadvantages of coal beneficiation is yet to
be established.
SUMMARY OF COAL PREPARATION REQUIREMENTS FOR COMMERCIAL AND FEDERAL
GOVERNMENT SPONSORED SYNTHETIC FUEL PROCESSES.
Tables 13 through 15 summarize the current status of coal preparation
requirements for major synthetic fuel processes.
147
-------
TABIi: 13. SUMMARY STMUS OF COM. PREPARATION TOR HIGH BflJ GASIFICATION PKXESSES '
PKtXTlBS
COa Acceptor
BI-GAS
HYGAS
Agglomerating
Burner
Syn thane
COGAS
cotiTRACTOR(s)/
D1.VELGPERS
Conoco Coal
Development Co. ,
Library / Pa .
Bituminous Coal
Research, Inc.,
Monroeville, Pa.
Institute of Gas
'Ifechnology,
Chicago, 111.
Battelle Memorial
Institute,
Colon-bus Labora-
tories, Columbus,
Ohio
Tile Lutmius Co. ,
Bloomfield, N.J.
COGAS Develop-
ment Co. ,
Princeton, N.J.
PI ANT TYPE
AND STATUS
Pilot Plant
(operation)
Pilot Plant
(operation)
Pilot Plant
(operation)
Process
Development
Unit
(last pliase)
Pilot Plant
Demonstration
Plant
(operation)
PLANT SIZE
Ktetrir
tons/day
(tons/dav)
35
(40)
110
(120)
68
(75)
23
(25)
68
(75)
45
(50)
PIANT
ICCATION
Rapid City,
South Dakota
lloiner City,
Pennsylvania
Chicago, 111.
West Jefferson,
Ohio
Allegheny County,
Pennsylvania
leatherhead.
England
CCAL TYPE(S)/
SOUHCF.(S)
IV/odak Sulibituminous,
Glenharold lignite.
Velvo lignite, Husky
char (lignite, sub-
bituminous)
All, for greatest
yield - N.D. lignite,
I'/yoming subbituniinous
C coal. Pa. high vola-
tile A bituminous coal,
Wsstern Ky. coal No. 11
All, Illinois No. 6
bituminous coal to be
tested, Montana Rosebuc
seam, subbitminous
All, Pittsburgh No. 8,
subbiturainous from
Lake Deatet, Wyoming,
from Ohio Clarion
All, Rosebud Coal
All
PREPARATION -
(FINAL SIZE/
MOIS1URE CONTEWT)
Crushed to 2.4 x 0.15 nm
(8 x 100 itesh) ,in a hot-gas-
swept iitpact mill and
flash dried to 3-5 wt. %
IhO
Crushed to minus 38 nm
(1 1/2 in) washed, screened
& pulverized to 70% minus
0.074 nm (200 nesh) . Dried
bo 1.3% moisture
Crushed to minus 2.4 mm
(8 mesh)
Crushed/separated 2.4 x
-0.15 nm (8 + 100 mesh) and
0.15 mm x 0 (100 mesh x 0)
Crushed to minus 0.84 mm
(20 mesh) , dried.
Crushed to minus 0.074 nm
(200 mash)
CO
-------
TABLE 14. SUMMARY STA1US OF COAL PREPARATION FOR IjOW BTU GASTKICATION PROCESSES >
PROCESS
WBllman-
Galusha
Winkler
Chapman
(Wilputte)
Gasifier
•• "i
Riley Morgan
Koppers-
Totzek
GFERC
Slagging
Gasifier
Texaco
Gasifier
CCNTRACTOR(S)/
DEVELOPER(S)
McDowell
Wellman Engineer-
ing Conpany,
Cleveland, Ohio
Davey Powergas
lakeland, Florida
Wilputte Corp.
Murray Hill, N.J.
Riley Stoker Corp.
Worcester, Mass.
Kbppers Co. , Inc.
Pittsburgh, Pa.
GFERC
Grand Forks,
North Dakota
Texaco Develop-
ment Corp.
New York, N.Y.
PLANT TYPE
AND STATUS
Commercial Unit
Commercial Unit
Conmercial Unit
Pilot Plant
(operation)
Commercial Unit
Pilot Plant
(operation).
Pilot Plant
(operation)
PLANT 'SITE
METRIC
tons/day
(tons/day)
*'•*•
9 metric tons
tons/hour
(1 ton/hour)
5,800
(6,400)
*
A
Up to 770
(850)
*
*
PIANT
LOCATION
N.C.
Largest in Madras,
India; none in U.S.
Kingsport, Term.
Worcester, Mass.
Largest unit in
Johannesburg!! ,
South Africa
Grand Forks, N.D.
"S
Montebello,
California
COAL TYPE(S)/
SOURCE (S)
Bituminous, Anthra-
cite, Charcoal and
Coke
Lignite, subbitu-
minous, bituminous
All; lignite
Anthracite, bitu-
minous, caking
bituminous
Bituminous , all
Bituminous char,
lignite char and
lignite
Lignite, bituminous
PREPARATION -
(FINAL SIZE/
MOISTURE CONTENT)
Crushed and sized to
14.3 x 7.9 mn (9/16 x 5/16 in.)
for anthracite, 51 x 26 mm
(2 x 1 in.) for bituminous
Crushed and sized to minus
8 mm (5/16 in. ) dried
Crushed and sized to minus
100 mn (4 in.)
Crushed and sized 51 x 3.2 mn
(2 x 1/8 in.)
Crushed and sized to 70%
minus 0.074 mm (200 mesh) ,
dried to 1-8% moisture
Crushed and sized to 19 x 6.4
mm (3/4 x 1/4 in.) dried to
less than 35% moisture
:rushed and sized to
70% minus 0.074 mm
(200 mesh) ; slurried
K
(CONTINUED)
-------
TAME 14. (CONTINUED)
rut'KSs
Q xi lex
Mol ten Salt
Advanced Coal
Gasj fication
for Electric
Power Genera-
tion CiVcr-
Staqe Pressur-
ized Fluid Bed)
low B'KI Gasifi-
cation for
Electrici ty
Generation
(CE Atmo-
spheric
Entrained
ted Gasifier)
Woodall-
Duckhcim
Lurgi
Pressure
Gasifier
CUNTI
-------
TAME 15. SUMMARY STATUS OF COAL PREPARATION FOR LIQUEFACTION PROCESSES'*
PROCESS
Solvent
Refined
Coal
Il-Coal
Clean Coke
Zinc Chloride
Catalyst
Donor
Solvent
eOOTRACTOR(S)/
DEVELOPERS
Pittsburgh &
Midway Coal Mining
Conipany, Merriani,
Kansas
Hydrocarbon
Research , Inc . ,
Trenton, N.J.
U.S. Steel
Pittsburgh, Pa.
Conoco Coal
Development Co. ,
Liberty , Pa .
Exxon Research
and Engineering
Co., Baytown,
Texas
PLANT TYPE
AND STA-1US
Pilot Plant
(operation)
(converting to
SRC II and back
to SRC I)
Pilot Plant
(construction)
Process Devel-
opment Unit
(operation)
Process Devel-
opment t)nit
(early start-
up)
Pilot Plant
(design)
T-LANT'SIZE" -
Hetric
tons/day
(tons/day)
45
(50)
540
(600)
0.45
(0.5)
.9
(1)
230
(250)
PLANT
LOCATION
Tacoma, Washington
Catlettsburgh,
Kentucky
Pittsburgh,
Pennsylvania
Liberty,
Pennsylvania
Baytown , Texas
COM, TYPE(S)/
SOURCE (S)
All
All
Bituminous ; Vfestem
Kentucky and Pitts-
burgh seam coals
All ; Col strip sub-
bituminous; or SRC
All; lignite, Wyoming
subbitimiinous, Illinois
No. 6 bituminous,
North Dakota lignite,
Wycdak
PREPARATION -
(FINAL SIZE/
MOISTURE CONTENT)
Crushed to 0.074 mm (200
mesh) dried to 1-3%
moisture
Crushed to minus 18 mn
(3/4 in.) dried to 4%
moisture. Crushed to minus
0.25 nm (60 mesh)
R.O.M. coal is dried and
crushed.
Dried, crushed and slurried
Dried, crushed and sized to
minus 0.5 nm (30 mesh)
then slurried
01
-------
SECTION 7
REFERENCES
1. Energy Research and Development Administration Coal Gasification,
Quarterly Reports 1975-1977.
2. Bloom, Ralph, Jr. The Illinois Coal Gasification Group Project
Incorporating the COGAS Process. Presented at the Eight Synthetic
Pipeline Gas Synposium. COGAS Development Co. Princeton, New Jersey,
1976.
3. Cavanaugh E.G. et.al. Technology Status Report: Low/Medium BTU Coal
Gasification and Related Environnental Controls. Volume II, Radian
Corporation, Austin, Texas, June 1977.
4. Energy Research and Development Administration Coal Liquefaction,
Quarterly Reports 1975-1977.
152
-------
TABLE 16
METRIC UNITS
CONVERSION TABLE
GO
Multiply (English Units)
ENGLISH UNIT
acre
acre - feet
British Thermal Unit
British Thermal Unit/pound
cubic feet/niinute
cubic feet/second
cubic feet
cubic feet
cubic inches
degree Fahrenheit
feet
gallon
gallon/minute
horsepower
inches
inches of mercury
pounds
million gallons/day
mile
pound/square inch (gauge)
square feet
square inches
tons (short)
yard
ABBREVIATION
ac
ac ft
BTU
BTU/lb
cfm
cfs
cu ft
cu ft
cu in
F
ft
gal
gpm
hp
in
in llg
Ib
ingd
mi
psig
sq ft
sq in
t
y
by
CONVERSION
0/tnc
. 1U 3
1 OTi C
1/JJ.D
0.252
0.555
0.028
1.7
0.028
28.32
16.39
0.555 (°F-32)*
0.3048
3TQC
. /o_>
0.0631
0-]A C -|
. /43 /
2C A
. j3
0.03342
0.454
3,785
1AAQ
* \)\jy
(0.06805 paig +1)*
0.0929
6.452
0.907
0.9144
To obtain (Metric
ABBREVIATION
ha
cu in
kg cal
kg cal/kg
cu m/min
cu m/min
cu m
1
X
cu an
°C
m
1
I/sec
kw
cm
atm
kg
cu m/day
km
atm
sq m
sq cm
kkg
m
Units)
METRIC UNIT
hectares
cubic meters
kilogram - calories
kilogram caloriesAilogram
cubic meters/minute
cubic meters/minute
cubic meters
liters
cubic centimeters
degree Centigrade
meters
liters
liters/second
kilowatts
centimeters
atmospheres
kilograms
cubic meters/day
kilometer
atmospheres (absolute)
square meters
square centimeters
metric tons (1000 kilograms)
meters
*Actual conversion, not a multiplier
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1 REPORT NO
EPA-600/7-78-150
2.
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Assessment of Coal Cleaning Technology: First
Annual Report
5. REPORT DATE
July 1978
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Lee C. McCandless and Robert B. Shaver
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Versar, Inc.
6621 Electronic Drive
Springfield, Virginia 22151
10. PROGRAM ELEMENT NO.
EHE624A
11. CONTRACT/GRANT NO.
68-02-2199
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Annual; 1/77-1/78
14. SPONSORING AGENCY CODE
EPA/600/13
15. SUPPLEMENTARY NOTES JERL-RTP project officer is James D. Kilgroe, Mail Drop 61,
919/541-2851.
16. ABSTRACT
The report gives results to date of a continuing assessment of coal cleaning
technology. It discusses: (1) washability characteristics of coal, with emphasis on
the correlations of various washability parameters; (2) current technology on coal
comminution and gravity separation processes; (3) 11 major chemical coal cleaning
processes, including evaluations and comparisons; (4) current technology on mechan-
ical and thermal drying and oil agglomeration; (5) slurry sampling techniques,
including a summary of a Bias Test Plan for two slurry samplers; and (6) coal pre-
paration requirements for synthetic fuel conversion processes, in terms of particle
size, moisture, ash, sulfur, and heating value. In addition, coal preparation
requirements for 6 high-Btu gasification processes, 13 low-Btu gasification proces-
ses, and 5 liquefaction processes are summarized.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Pollution
oal
oal Preparation
Assessments
omminution
Separation
Dewatering
Oils
Agglomeration
Slurries
Sampling
Fuels
Gasification
Liquefaction
Pollution Control
Stationary Sources
Coal Cleaning
Washability
Gravity Separation
Synthetic Fuels
13 B HH
2 ID
081 HG
14 B
13H,07A
07D
Unlimited
19. SECURITY CLASS (This Report)
Unclassified
20. SECURITY CLASS (This page)
Unclassified
21. NO. OF PAGES
161
22. PRICE
EPA Form 2220-1 (9-73)
154
------- |