United States
Environmental Protection
Agency
Air and Energy Engineering
Research Laboratory
Research Triangle Park NC 2771
Research and Development
EPA/600/S7-86/037 Feb. 1987
SEPA Project Summary
Assessment of Coal Cleaning
Technology: Final Report
Lee C. McCandless, A. Bekir Onursal, and Jean M. Moore
Tests at seven coal preparation
plants evaluated the performance of
froth flotation cells and dense-medium
cyclones in removing ash and sulfur (S)
from fine coal (minus 28 mesh). Flota-
tion circuits tested at four plants
showed substantial reductions in coal
ash content (64-88%), pyritic S content
(48-65%), and sulfur dioxide (S02) emis-
sion (expressed as ng SO2/J or Ib SO2/
106 Btu; 15-87%) at mean weight recov-
eries of 11-54%. Dense-medium
cyclones tested at three plants showed
reductions in coal ash content (43-75%),
pyritic S content (29-67%), and SO2
emission (16-40%) at mean weight re-
coveries of 63-83%. Data from other
coal preparation plants demonstrated
that physical coal cleaning (PCC) re-
duces the variability as well as the
mean value of the coal ash and S con-
tents. Raw and clean coal data sets
were found to exhibit statistical proper-
ities which can be characterized by time
series models. The use of low S coal,
PCC, or chemical coal cleaning (CCC)
was evaluated for compliance with po-
tential SO2 emission limits for indus-
trial boilers. PCC can achieve moderate
S reductions in (high S) Northern Ap-
palachian and Midwestern coals, but
few of these coals can be cleaned to
meet a 516 ng S02/106 Btu standard.
Many Southern Appalachian, Alabama,
or Western coals are capable of meet-
ing this standard as mined or after
cleaning. Many CCC processes can be
used to desulfurize high S coals for
compliance with this standard. Eleven
major CCC processes were evaluated
for their performance potential. Some
processes can remove as much as 90-
95% of the pyritic S and up to 40% of
the organic S from raw coal.
This Project Summary was devel-
oped by EPA's Air and Energy Engineer-
ing Research Laboratory, Research Tri-
angle Park, NC, to announce key
findings of the research project that is
fully documented in a separate report
of the same title (see Project Report
ordering information at back).
Introduction
Although approximately 60% of un-
derground coal and 20% of surface
mined coal is cleaned in some way,
physical coal cleaning (PCC) has not
been fully exploited. Chemical coal
cleaning (CCC) has not been used com-
mercially. Many facets of these tech-
nologies were explored under EPA's
sponsorship, and this Coal Cleaning
Technology Assessment project was
part of that effort. This project included:
A comprehensive evaluation of ex-
isting performance data and costs
of PCC equipment with respect to S
removal.
Development of new data neces-
sary to complete the evaluation of
the performance of coal cleaning
equipment and processes.
An evaluation of fine coal dewater-
ing and handling technology, in-
cluding costs.
An evaluation of coal preparation
requirements for synthetic fuel con-
version processes.
An engineering and economic eval-
uation of CCC processes.
An assessment of coal cleaning as a
pollution control technology for in-
dustrial boilers.
An evaluation of the reduction in S
variability of coal by commercially
operating coal cleaning plants.
-------
The report is based on information
gathered, data generated, and engi-
neering analyses performed by Versar
during the period 1978 through 1980.
Equipment Performance
Studies
Domestic and foreign equipment
manufacturers were contacted and their
equipment data were compiled and
evaluated as part of this effort. Also, ac-
tual in-plant performance of froth flota-
tion cells and dense-medium cyclones
was evaluated for S and ash removal. A
mobile laboratory was outfitted and de-
ployed to support the sampling and an-
alytical work at seven coal preparation
plants. The equipment types, plant loca-
tions, and coal sources tested are
shown in Table 1.
As shown in Table 1, flotation circuits
at four selected coal preparation plants
were sampled. In addition, samples of
the plant raw coal feed, product coal,
and refuse were characterized in an ef-
fort to gain information on the overall
performance of the plants. Each plant
contains one flotation circuit except
plant 1-D, which has a fine coal and a
coarse coal flotation circuit; however,
only the coarse coal flotation circuit was
evaluated. In all but one plant, five sets
of coal samples were collected (each set
on a different day) to determine the vari-
ability of the measured parameters
(e.g., pyritic S content) with time.
Ash, pyritic S content, and SO2 emis-
sion (ng SO2/J or Ib SO./106 Btu) of the
product streams for all circuits were
found to be lower than those for the cor-
responding feed streams at mean
weight recoveries of 11-54%. Mean ash
reductions for all tested circuits were
64-88%. The range for pyritic S reduc-
tion was 48-65%. Mean reduction in
SO2 emission was 15-87%.
All of the tested flotation circuits, ex-
cept the one in plant 1-C, show that the
flotation products contain less weight
percent S than the feed. The result for
plant 1-C was verified by repeating the
total S tests at Versar's analytical labo-
ratory: the same conclusion was
reached from the analysis of these test
results. The increase in total S concen-
tration can be explained by the constant
organic S concentration in the pure coal
portion (as opposed to ash) of the feed
and product streams. Based on the lab-
oratory analyses, the organic S content
in the feed coal is about 30% of the total
S. As most of the ash in the feed coal is
removed by the flotation process, the
pure coal content, and therefore the
Table 1. Coal Types and Circuits Sampled for Equipment Performance Testing
Plant Circuit Tested Plant Location Coal Type
1-A Froth flotation
7-8 Froth flotation
1-C
1-D
Froth flotation
Froth flotation
(Shakedown tests)
2-A Dense-medium cyclone
2-B Dense-medium cyclone
2-C Dense-medium cyclone
Franklin County, IL
Indiana County, PA
Raleigh County, WV
Co/fax County, NM
Raleigh County, WV
Wyoming County, WV
Wise County, VA
Illinois No. 6, Franklin County
Upper Freeport Coal, Indiana
County
Peerless Seam, Raleigh County
Colfax County
Pocahontas No. 3 Seam
Williamson Seam No. 2
Blend of Norton, Dorchester,
Lyons, Clintwood, and
Elkhorn Rider Seams
weight concentration of S, in the
product stream increases.
Figure 1 shows the reduction in per-
cent S02 emission as a function of per-
cent weight recovery from the froth flo-
tation circuit for each plant tested. For
plants 1-A (Bank 1 and 2) and 1-C, the
reduction in SO2 emission increases for
lower percent weight recoveries. For
plant 1-B, the data are scattered.
Three coal preparation plants (plants
2-A, 2-B, and 2-C) shown in Table 1 were
selected for testing of dense-medium
cyclones. Each of the three plants uses
61 cm (24 in.) diameter dense-medium
cyclones as the only coal cleaning de-
vice in the process. Feed size of the coal
to the cyclones is 38.1 mm x 0 (1-1/2
in. x 0) for plants 2-A and 2-C, and 9.4
mm x 0 (3/8 in. x 0) for plant 2-B.
Feed, product, and refuse streams as-
sociated with the dense-medium cy-
clones were sampled for 5 consecutive
days, and analyzed for ash, pyritic and
total S, and heating value. The test re-
sults show that cleaning of coal in
dense-medium cyclones resulted in a
product containing less ash, less total
and pyritic S, and higher heating value.
S02 emission decreased as a result of
this cleaning process. Mean reductions
from the three dense-medium cyclone
circuits were 43-75% for ash, 39-67%
for pyritic S, 2-18% for total S, and 16-
40% for S02 emission at mean recover-
ies of 63-83%.
The primary interest of this study is
the performance of a dense-medium cy-
clone in reducing SO2 emission. Data
from 5-day samples were used to plot
the weight recovery of the cleaned coal
as a function of S02 emission reduction
(Figure 2). The results show that the
quality of the product, in terms of S02
emission, becomes poorer as more ma-
terial is recovered from the dense-
medium cyclones.
Fine Coal Dewatering
An engineering study evaluated some
alternatives for fine coal dewatering
and drying. Costs to a preparation plant
operator for alternative dewatering and
drying schemes were compared to the
economic benefits achieved by ship-
ping drier coal to a 580 MW electric utjf
ity. A base case with no dewatering was
also included in the study.
Seven alternative schemes for coal
dewatering and drying were evaluatec
in this study:
Case O -Base case using no de-
watering.
Case A -9.5 x 0.6 mm fraction cen-
trifuged and 0.6 mm x 0
fraction filtered.
Case B -Same as A, but the
0.6 mm x 0 filter cake is
dried with a heat ex-
changer.
Case C -Same as A, but the
0.6 mm x 0 filter cake is
dried in a direct heat ther-
mal dryer.
Case D -9.5 x 0.6 mm centrate is
processed in a hydro-
cyclone for slimes remova
Case E - Same as D but slimes are rt
moved by flotation cells.
Case F -9.5 x 0.6 mm is not cen
trifuged, but combined wit
0.6 m x 0 filter cake fror
vacuum filter and dried in
direct heat thermal dryer.
The coal user (e.g., an electric utilit
was assumed to contract for net heatin
value. For the purpose of analyzing qj
watering and drying operations only,
-------
700 1
30.
o
g 60
s
I 5ฐ
I
CM
to 40-
30-
20
10'
A-5=Test Sample No. 5 at Plant 1-A. Bank 1
A'-5=Test Sample No. 6 at Plant 1-A, Bank 2
Plant 1-D. Coarse Coal Flotation Circuit
A-5
Plant A-Bank 1
c_5 . A'-J Plant A -Bank 2
Plant B
10
20
60
70
Figure 1.
30 40 SO
% Weight Recovery
Plot of percent of SOz emission reduction vs. percent weight recovery lor the
flotation units tested at four plants.
constant heating value (31,751 J/g or
13,650 Btu/lb) of dry coal was assumed,
since no appreciable change in coal
composition results from these opera-
tions. However, any associated mois-
ture in the coal received was penalized,
for the purpose of this study, by the re-
quirement for sufficient additional coal
to vaporize this moisture. This addi-
tional coal penalty is the total cost of
such coal through mining and the entire
cleaning plant beneficiation process in-
cluding separation, dewatering and dry-
ing, and refuse disposal; and was as-
sumed to be $22/Mg ($20/ton) on a dry
basis. In addition, a power plant pulver-
ization cost of 60 cents per wet ton was
assessed to the additional coal require-
ment.
Table 2 shows that the fine coal de-
Catering and drying alternatives have
Pgnificant benefits compared to the
baseline case of no dewatering. It is in-
structive to compare the net benefits to
those of Case A $3.41/Mg ($3.10/ton),
which is limited to mechanical dewater-
ing processes. Case B, in which the filter
cake is dried in an indirect heat ex-
changer, is only marginally more attrac-
tive. Case C, where a direct thermal
dryer is used, is significantly less attrac-
tive than Case A. In Cases D and E, the
recovery of solids from the centrate ap-
pears attractive, reflecting lower refuse
disposal costs as well as recovered
product values. The use of a thermal
dryer in Case F to avoid centrate solids
losses is apparently competitive with
Cases D and E.
Pollution From Coal Cleaning
Processes
Coal cleaning can significantly reduce
SO2 emissions, scrubber sludge from
air pollution control equipment, and ash
from coal-fired boilers. However, the
coal cleaning process itself generates
emissions to air, water, and land.
Samples obtained from 11 coal
preparation plants and auxiliary areas
(e.g., refuse piles) were analyzed for the
65 classes of pollutants identified under
the court-approved Consent Decree of
July 7, 1976. Among the non-organic
priority pollutants detected in untreated
coal preparation plant wastewaters
were Sb, As, asbestos, Be, Cd, Cr, Cu,
cyanide (CN), Pb, Hg, Ni, Ag, Se, Tl, and
Zn compounds. Settling appeared to be
effective in removing all these elements
except Cd, Pb, Hg, Ag, Se, and Tl. In
general, analytical data showed signifi-
cant amounts of dissolved metallic ele-
ments in the process waters. This result
agrees with the fact that the coal proc-
essing medium remains slightly alka-
line. Such a medium is not likely to dis-
solve metallic minerals present in the
coal. Some organic compounds were
detected, but these were found to be the
results of laboratory contamination or
processes other than the mining or
cleaning of coal. Suspended solids were
found to be the principal pollutant in
coal preparation plant wastewaters.
Analysis of data for leachate and
runoff from coal storage, refuse piles,
and coal preparation plant ancillary
areas showed that waste loadings and
resulting effluent qualities were very
similar and appeared to be independent
of the processing methods that were
used in the respective plants. The princi-
pal pollution control measure associ-
ated with coarse waste disposal is com-
paction and coverage with soil to
minimize the chances for oxidation and
percolation. This also reduces the possi-
bility of fire, another major environmen-
tal problem with refuse piles.
Sulfur Reduction and
Variability
The ability of boiler operators to com-
ply with emission regulations and the
costs associated with such compliance
also depend on the variabilities of coal S
content and heating value. When the
emission regulation is expressed in
terms of maximum SC>2 emission in ng
S02/J db SO2/106 Btu), the mean SO2
emission of a coal burned in boilers
must be lower than this maximum
value. The reason for using a coal with
a lower S02 emission is to prevent non-
compliance (exceedance) during posi-
tive excursions around the mean. Two
factors determine how much lower the
-------
50
40 .
.i
.8
I
ง30
I
i
I20
10
Notation:
A-5=Test Sample No. 5 at Plant 2-A
A-3
B-S
B-2
B-1
\ Plant B
SO
Figure 2.
60 70 80
Percent (%) Weight Recovery in Clean Coal
90
Plots of percent weight recovery vs. percent reduction of SOi emission for the
dense-medium cyclones tested at three coal preparation plants.
Table 2. Costs and Benefits Per Dry Ton of Fine Coal Product3*
Case Fine Coal Dewatering and Drying Operations
Cost,
$/Mg
Benefit,
$/Mg
Net Benefit,
$/Mg
0 None 0.00 0.00 0.00
A Centrifugation, Filtration 1.02 4.43 3.41
B Centrifugation, Filtration, Indirect Heat Ex-
change 1.35 4.79 3.44
C Centrifugation, Filtration, Direct Thermal 2.16 4.91 2.75
Dryer
D Centrifugation, Filtration, Hydrocyclone, Filtra-
tion 0.97 4.69 3.72
E Centrifugation, Filtration, Flotation, Filtration 0.88 4.87 3.99
F Filtration, Direct Thermal Dryer 7.85 5.57 3.72
a 1977 dollars.
bOperations listed were performed on partial streams.
mean SO2 emission must be than th
emission limit: (1) the fractional time'
that the regulations permit a boiler to
exceed the nominal limit (confidence
level), and (2) the characteristic variabil-
ity in the coal feed (mean value, stand-
ard deviation, and autocorrelation
structure). Quantification of the second
factor (i.e., the characteristic variability
of heat-specific S content in coal) was a
prime objective of two studies: (1) one
that discusses the effect of PCC on the S
content and S variability in coal, and
(2) an evaluation of the effect of PCC on
attenuating coal S variability.
Effect of PCC on the Content
and Variability of Sulfur
Using existing PCC plant data as a
basis, the first study sought to achieve
two primary objectives: (1) documenta-
tion of the performance of commercial
coal cleaning facilities in removing S
from coal, and (2) quantification of vari-
ability of the coal's S02 emission.
The database used in this study con-
sits of 53 data sets, with a total of 3,204
data points. Each data set represents an
identifiable and unique coal stream,
either raw coal or cleaned coal, from a
particular cleaning plant or loading fa-
cility, with the source of the coal (searrt
and county) and cleaning level specfl
fied. Eight coal preparation plants pro-
vided data sets for both feed and
product coal; and approximately 40
others submitted only single values for
feed and product measurements. The
remaining plants provided product data
without the corresponding feed values.
The analysis from the matched pairs
of data sets for S, Btu, and SC>2 emission
supported the following conclusions:
In each of the eight plants for which
matched pairs of feed and product
data were available, both the abso-
lute standard deviation and the re I
ative standard deviation (RSD) foi
all three coal characteristics were
reduced by the coal preparatior
process.
The raw and clean coal RSDs van
from plant to plant, and no typica
values are universally valid.
This study revealed a need for furthe
investigation of coal S variability fo
several reasons. One of the reasons wa
that the effect of cleaning on S variabil
ity could not be well quantified from thi
study because of the insufficient paire<
data for raw and cleaned coal. Also, coi
relation of S content in coal sample
could not be quantified reliably becaus
sampling and analysis procedures us4
-------
' the various coal companies who pro-
ftded the data were not uniform.
Effect of PCC on Attenuating
Coal Sulfur Variability
The second study was conducted as a
controlled experimental investigation to
accurately collect and analyze represen-
tative samples of raw and clean coal.
Raw and clean coal samples were col-
lected from two coal preparation plants
so that hour-to-hour and day-to-day
changes in the coal characteristics
could be monitored. These samples
were collected and analyzed consis-
tently during the study period using
standard techniques so that the vari-
ance associated with sampling and
analysis would not mask changes in the
coal characteristics.
Sampling both feed and product
coals from each of two coal preparation
plants, under carefully controlled condi-
tions, confirmed the results of prior
studies. Both the mean total S content
and the mean S02 emission are signifi-
cantly reduced by the cleaning process
as shown in Table 3.
The extent of the reduction is quite
different for the two plants. In fact, the
3.4% SO2 emission reduction at Plant
o. 2 is uncharacteristically high for
many existing coal preparation plants.
However, a wide range of reductions
among preparation plants is a result
consistent with prior findings.
Prior to analyzing the variability in
coal data, the measurement uncertainty
in the data was independently deter-
mined. This uncertainty, attributable to
the process of sampling, compositing,
sample preparation, and laboratory
analysis, provides a quantitative limita-
tion to subsequent explanations of coal
variability. All values for aggregate
measurement uncertainty were signifi-
cantly less than the total variations. Real
variability in coal characteristics there-
fore was observed, over and above the
measurement noise level.
For much of the data acquired in this
study, strong autocorrelation was indi-
cated. The 30-minute increment data
from Plant No. 1 were more highly auto-
correlated than composite data over
longer time intervals. The data from the
Plant No. 2 exhibited weaker autocorre-
lation than the Plant No. 1 data. How-
ever, the results from both plants con-
firm that serial correlation of coal data
does exist over short time intervals.
Two analytical techniques were uti-
to quantify the correlated and ran-
Tablo 3. Effect of PCC on Coal Sulfur
Parameter
Plant No. 1
30-Min
Increments
Plant No. 2
1-Hour
Increments
Total S, %
lbSO2
106 Btu
Raw Coal
Cleaned Coal
Reduction
Raw Coal
Cleaned Coal
Reduction
3.076
2.612
15.1%
5.476
4.237
22.6%
2.576
1.309
49.2%
5.117
1.875
63.4%
dom components of the variability in
coal data: geostatistics and time-series
analysis. Time-series analysis proved to
be the more useful.
Time-series models can be used in a
predictive way, to generate data sets
much longer than the empirical (meas-
ured) data set. The random component
in the predictive model is obtained from
a random number generator. Since this
model is probabilistic, many different
time series, equally likely, may be gen-
erated, all based on the same mean,
same variance, and same correlation
structure. From a large number of time
series based on the model for any single
data set, the average expected number
of emission violations by a power plant
burning this coal (either raw or cleaned)
can be estimated.
The time-series predictive model was
also used to develop the effect of lot size
on variability. The data generated by the
time series were mathematically com-
posited into successively longer time in-
tervals (corresponding to successively
larger quantities of coal in each inter-
val). The sample mean variance de-
creases with increasing lot size, but at a
smaller rate than would be expected
from serially independent data. This re-
lationship was more pronounced for
clean coal than ROM coal at Plant No. 1.
Results of the second study showed
that serial dependence (also called auto
correlation) of coal characteristics must
be incorporated into any analysis of the
ability of coal to comply with SO2 emis-
sion regulations. The misapplication of
Gaussian statistics, which assumes se-
rial independence of coal data, leads to
a gross underestimation of the fre-
quency of short-term emission viola-
tions. Time series analysis, which com-
bines serial dependence with a
stochastic component to construct a
predictive model, provides an alterna-
tive to Gaussian statistics. The tech-
niques and computer programs for ap-
plying time-series analysis are
generally available for use.
Although the two diverse coals stud-
ied in detail both exhibited autocorrela-
tion the magnitude of the autocorrela-
tion component of the total variance
differed from one coal to another and
from raw to cleaned coal. Therefore,
each coal's ability to meet short-term
emission regulations must be deter-
mined separately until the number of
different coals characterized is sufficient
to generalize the variability of coal char-
acteristics.
Evaluation of PCC as a Sulfur
Control Technology for Indus-
trial Boilers
A study was performed to support the
EPA Office of Air Quality Planning and
Standards in developing New Source
Performance Standards (NSPS) for in-
dustrial boilers. The results were com-
piled in one of eight technology assess-
ment reports for industrial boiler
applications. This study was performed
to determine the Best System of Emis-
sion Reduction (BSER) for industrial
boilers. BSERs were defined as control
technologies that could comply with a
specified emission control level at mini-
mum cost, energy, and environmental
impact. The major pollutant considered
for control was SOj, although particu-
lates, nitrogen oxides (NOX), and other
pollutants were included in relation to
energy and environmental impacts of
the chosen technologies.
Major decision variables considered
in this study included the coal type and
S control options, boiler types, and SO2
emission control levels.
The sulfur control options are:
Use of naturally occurring low S
coal: considered to be coal with a S
content of approximately 1% or
less.
-------
Beneficiation of raw coal by PCC to
remove ash and pyritic S.
Beneficiation of raw coal by CCC to
remove pyritic and/or organic S.
Multiple options were available within
each type of control, and some prelimi-
nary evaluation and screening was re-
quired before detailed evaluations were
performed.
The following items were considered
in relation to the selection of air pollu-
tion control technologies for new
source standards development:
Performance and operating data.
Reliability of control systems.
Compatibility with other systems.
Applicability of control systems to
different boiler sizes and types.
Estimated capital and operating
cost of the control systems.
Control system cost as a function of
removal efficiency.
Status of development.
Commercial availability.
Energy requirements of the control
system.
The S control options evaluated in
this study included PCC and CCC. For
PCC, the following types of coals and
levels of cleaning were considered.
Coal Type
Control Technology
High S eastern
Medium S
eastern
Low S eastern
Low S western
PCC Level 5-Deep
cleaned middlings
PCC Level 3
PCC Level 4
PCC Level 2
For the evaluation of CCC, the Meyers,
Gravichem, and ERDA processes were
selected.
Although the control technologies
could be used in combination, they
were considered separately for com-
parison in this study. Final evaluations
were based on projected emissions
from a set of five reference boilers using
four reference coals. The coal-fired boil-
ers chosen for this study and their re-
spective heat input are:
Thermal input.
Boiler Type MW (106 Btu/h)
Package, watertube
underfeed
Field-erected, water-
tube, chain grate
Field-erected, water-
tube, spreader
8.8 (30)
22.9 (75)
44.0 (150)
Field-erected, water-
tube, pulverized
coal
Field-erected, water-
tube, pulverized
coal
58.6 (200)
118.0(400)
For this study, five emission levels were
chosen:
Stringent-516 ng SO2/J (1.2 Ib SO2/
106 Btu).
lntermediate-645 ng S02/J (1.5 Ib
S02/106 Btu).
"Optional" moderate-860 ng S02/J
(2.0 Ib S02/106 Btu).
Moderate-1,290 ng S02/J (3.0 Ib
SO2/106Btu).
A State Implementation Plan (SIP)
level of 1,075 ng S02/J (2.5 Ib S02/
106 Btu).
Equipment and process data com-
piled previously were used to project
the results of applying certain cleaning
processes to the reference coals.
Based on performance, cost, energy
requirements, and environmental im-
pacts, five best systems of emission re-
duction were chosen from the original
approximately 17 options. The final
choices are summarized in Table 4.
Chemical Coal Cleaning
Recognizing the importance of CCC
as a potential S02 pollutant control op-
tion, EPA directed a study in 1977 to in-
vestigate the technical and economic
feasibility of developing CCC.
The objective of the study was to sur-
vey the field of CCC, to identify active
and inactive processes, and to perform
a critical evaluation of competing proc-
esses. The purposes of this evaluation
were fourfold:
To provide updated information on
technical and economic viability of
these processes and to identify
their developmental stage.
To examine their performance
characteristics and environmental
aspects.
To develop quantifiable technical
and economic parameters for pur-
poses of process comparison.
To identify specific research and
development needs for processes
showing a potential for substantial
reduction of S in coals.
Twenty-nine CCC processes were
identified for study. Eleven U.S.,
Japanese, and Australian processes
were judged to deserve no further con-
sideration, because they were inactive
or proved to be inapplicable to most
U.S. coals. Seven U.S. and Canadi
processes were considered to be d1
minor relevance, because of their early
stage of development or inactive status.
Eleven U.S. processes were considered
to be of major relevance, and these
were evaluated in detail with respect to:
description; developmental status;
technical evaluation, including S re-
moval potential, S by-products, benefits
analysis, environmental aspects, and
research and developmental efforts and
needs; and economics.
Five basic reactions were involved in
desulfurization by major CCC proc-
esses: oxidative leaching, hydrogen
leaching, alkali leaching, chlorine sub-
stitution, and iron adsorption. One addi-
tional technique was a chemical fractur-
ing step that prepared the coal for
desulfurization by conventional PCC.
Detailed comparisons were made on
the basis of a common coal feed of Pitts-
burgh seam bituminous coal. In addi-
tion to costs, the following parameters
were evaluated:
Weight yield of cleaned coal based
on a common feed coal rate.
Weight percent S in the cleaned
coal product based on the S re-
moval efficiency of the process.
Heating value yield of the proceซ
based on feed coal heating value
and the net energy recovery.
SO2 emission levels were calculated for
the cleaned coal products. Emission lev-
els for processes which removed both
types of S were below 520 ng SO2/J (1.2
Ib SO2/106 Btu). Of the four processes
which removed only pyritic S, the two
that used chemical removal methods
(Meyers and LOL) were very close to
compliance with 520 ng S02/J, but the
two processes that used mechanical re
moval (Syracuse and Magnexฎ) coulc
only reach an emission limit of 1040 nc
SO2/J (2.4 Ib SO2/106).
Estimated energy recoveries were
generally greater than 90% except foi
the IGT process which was low witr
57% recovery. All energy recoveries re
fleeted both the coal loss from process
ing and the coal used to provide in
process heating needs. However
except for the IGT process, the actua
coal loss from processing was claimee
to be small. For most processes, th<
major heating value loss was due to th<
use of cleaned coal for in-process heat
ing.
CCC processes were still under devel
opment at the time of this study; there
fore, the economic evaluations weq
-------
'able 4. Best Systems of Emission Reduction for Four Candidate Coals and Five SO2 Emis-
sion Control Levels
Coal
High S
Eastern
Moderate
1,290"
(3.0)
PCC Level 5
Middlings
SIP"
1,075
(2.S)
PCC Level 5
Middlings
"Optional"
Moderate
860 (2.0)
PCC Level 5
"Deep
Cleaned"
Inter-
mediate
645 (1.51
PCC Level 5
"Deep
Cleaned"
Stringent
516
<1.2)
CCC-EFtDA
Medium S
Eastern
Low S
Eastern
Low S
Western
Raw Coal
Raw Coal
Raw Coal
PCC Level 3
Raw Coal
Raw Coal
PCC Level 3
Raw Coal
Raw Coal
CCC-ERDA
PCC Level 4
Raw Coal
CCC-ERDA
PCC Level 4
Raw Coal
ang SOJJ (Ib SO^W6 Btu).
bState Implementation Plan.
best engineering estimates based on
the information available. Capital and
annual operating costs for each major
CCC process were estimated. These
were based on an assumed plant feed
capacity of 7,200 metric tons (8,000
tons) per day, equivalent to the coal
needed to fuel a 750 MW electric power
plant. The total annual operating costs
jor each process, including and exclud-
ig cost of the raw coal, were also ex-
pressed in terms of dollars per metric
ton and dollars per 109 calories heat
content in the coal. Annual operating
costs in 1978 $, including raw coal cost,
ranged from $43.10 to $72.50 per metric
ton ($39.10 to $65.80 per ton) or $5.32 to
$11.20 per 109 cal ($1.34-$2.82 per mil-
lion Btu).
In general, pyritic S removal proc-
esses required the least amount of capi-
tal and had the lowest operating costs,
but they had limited S removal efficien-
cies.
-------
L C. McCandless. A. B. Onursal, and J. M. Moore are with Versar, Inc..
Springfield. VA 22151.
Julian W. Jones is the EPA Project Officer (see below/.
The complete report, entitled Assessment of Coal Cleaning Technology: Final
Report," (Order No. PB 87-120 51S/AS; Cost: $24.95, subject to change)
will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Air and Energy Engineering Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
EPA/600/S7-86/037
0000329
60604
------- |