-------
slurry was adjusted to a pH greater than 7 with NaOH to precipitate iron
hydroxide and then was treated with CaCl^ solution to a pH of about 7 to
precipitate calcium sulfate. Recovery experiments with these treated
slurries using kerosene and fuel oil-kerosene mixtures indicated that the
coal particles had agglomerated well. However the separation of the
agglomerated coal from the residue suspension was extremely difficult due
to plugging of the sieve openings by the precipitated residue. In order to
control this effect, several dispersing agents were tried. The results of
their use on the agglomeration of coal slurries neutralized with NaOH (no
CaCl~ addition) are given in Table 20. The dispersing agents improved the
separation of the agglomerated coal from the residue suspension, but the
coal recovery and quality were not satisfactory. Sodium metaphosphate
appeared to be most promising.
The data given in Table 21 are the results of experiments in which
a sequence of treatments were tried. The treatment consisted of a combina-
tion of a single stage washing with water followed by adjustment of the pH
either to about 6 or 5 with sodium hydroxide which was followed by addition
of sufficient sodium silicate solution to raise the pH to either about 9
or 8. The results indicate coal recovery is greater when coal is washed
and sodium silicate is used. Washing the slurry before addition of sodium
hydroxide and sodium silicate reduces the ash in the recovered coal. The
lowest ash in the recovered coal occurred when the pH was adjusted to
about 5 with sodium hydroxide alone.
This behavior was attributed to differences in the
composition of the partially dried old slurry pond sediment that became
segregated during use of the material. The major portion of the sediment
was a loose material which had mixed with it firm lumps ranging in various
sizes up to 3 inches in diameter. Towards the end of these studies, the
larger lumps were used as part of the feed material.
The behavior of these two types of materials during slurry
preparation and agglomeration were compared. As shown in Table 22,
the aqueous slurry made from the aggregated sediment had a pH of about 6.5
while the slurry made from the loose sediment had a pH of 2.9. When the pH
2.9-slurry was adjusted to a pH of 6.5 to match that of the aggregate
slurry and then agglomerated using a 10 percent concentration of No. 2 fuel
-------
TABLE 21. EFFECT OF SINGLE WASH OF SEDIMENTS, pH ADJUSTMENT AND LOW LEVEL
SODIUM SILICATE ADDITIONS BEFORE AGGLOMERATION ON COAL
RECOVERY FROM PARTIALLY DRIED OLD SEDIMENTS(a)
Slurry pH
Initial
Unwashed
Washed
Washed
Unwashed
Washed
3.
3.
3.
3.
3.
2
3
3
1
3
Adjusted
NaOH
5
5
5
5
5
.8
.8
.0
.1
.3
After Sodium^ Product,
Silicate Of Feed
9.2 64.
9.6 66.
8.2 66.
— (°) 62.
65.
8
7
7
3
0
percent
MF
7
6
6
5
5
Ash
.43
.13
.64
.40
.36
Coal
Recovered
percent
88
92
92
87
91
(a) Slurry concentration, 10 percent; No. 2 fuel oil concentration, 10 percent,
(b) pH values after adding sodium silicate solution (8.9% Na~0 + 24% Si00) .
(c) No sodium silicate added.
45
-------
-P-
O\
TABLE 22. COMPARISON OF AGGLOMERATION BEHAVIOR OF SEGREGATED LUMPS
OF LOOSE MATERIAL FOUND IN PARTIALLY DRIED OLD COAL SEDIMENT
Loose Sediments
Aggregated Sediment
Original
Slurry
2.6
6.4
Slurry pH
Diluted for
Agglomeration
2.9
6.5
At
Agglomera tion (a )
6,
6,
,5(b)
.5(C)
Produc t ,
Of Feed
8.47
60.4
percent
MF Ash
8.19
5.02
(a) Slurry concentration, 10 percent; No. 2 fuel oil concentration, 10 percent.
(b) pll adjusted with NaOH.
(c) No adjustment required.
-------
oil, only 8.5 percent of the material was separated. In contrast,
agglomeration of the slurry of the aggregate yielded 60 percent of the feed
and the recovered coal had lower ash content. This may have been due to the
reduced exposure to weathering that these aggregates had compared to the loose
materials. It could also be due to a higher coal concentration in the aggregate.
RESIDUE CHARACTERIZATION
Once coal values have been removed, the residue material is
relatively free of combustible material. For example, the solids recovered
from the aqueous residue suspension after agglomerating with a mixture of
No. 6 fuel oil and kerosene contain about 90 percent moisture free ash
while the residue solids from the aged slurry pond sediments contained
about 89 percent moisture free ash. Part of the weight loss during ash
determination could have been due to the loss of water so in effect the
quantity of coal present is very small. As reported by Capes et al. many
of the trace elements in coal considered to be hazardous are removed with the
mineral matter during agglomeration.
Any process designed to use the agglomeration technique for
recovery of fine coal from coal slurry waste streams or to reduce the
environmental threat of slurry ponds will be concerned with the disposal
of the residues from the operation. To this end studies were initiated to
measure the settling characteristics of the residue suspensions and the
concentration of the agglomerating oil they may retain.
Settling Rates
The settling rates of the original coal slurry and of the residue
suspension were obtained by preparing sufficient 10 percent slurry to make
up 1700 ml and shaking the suspension thoroughly in a 2000 ml graduated
cylinder for several minutes. The settling rate was measured by following
the level of the clear interface with time. After completion of this cycle,
the slurry was agglomerated using a 20 percent kerosene concentration in a
one-gallon blender. After the usual coal recovery technique (including
washing on the screen), the residue suspension was now contained in
2000 ml. The settling rate was measured again.
47
-------
The settling rates for the aged slurry pond sediment and the
resulting residue suspension are given in Figure 6. Settling rates for the
fresh black water sediments and the resulting residue suspension are shown
in Figure 7. The settling rates of both residue suspensions are faster
than the parent sediment slurries containing the coal. Removal of the
coal from the residues which appear to be mostly fine clays probably
increases settling in part due to the decreased slurry concentration. This
is a very desirable result. In addition, specific agents.known to improve
clay flocculation can now be applied more specifically.
The pulp density of the residues (1.09 and 1.14 g/ml) at the end
of the 3-day settling period suggest that they remain pumpable. The
residues recovered after settling were dried at 110 C and then analyzed (see
Table 23). As expected the results for the two materials from the same
mine are similar and the composition is characteristic of clays with SiCL
to A^O., ratio of about 4 (perhaps illites since iron, calcium and
potassium are present).
Oil in Residue
An estimate of the amount of agglomerating oil that is lost to
the residue suspension after agglomeration was made. The procedures used
for determining the oil content in the residue slurry were as follows:
1. The residue water suspension from the oil agglomeration
(mostly mineral matter) was mixed well with 50 ml of
carbon tetrachloride in a 1000 ml separating funnel.
2. The mixture was allowed to separate into two phases
and the heavier CC1/ phase was drained off.
3. These extraction steps were repeated five times.
4. A 24 ml aliquot of the CC1, extract (250 ml) was placed
in a weighed dish and evaporated in a constant tempera-
ture bath at 80 C for 20 min.
5. The weight of remaining oil was measured and multiplied
by 10 to calculate the total amount of oil in the
residue slurry.
48
-------
2000k-
1800
1600
1400
1200
•o
a
1000
800
600
Residue
Suspension
400
200
Final Pulp Density of
Residue = 1.09 g/ml
30
60
90 120
Time, min
150
180
210
FIGURE 6. SETTLING RATES FOR 10 PERCENT SLURRY-OF AGED SEDIMENTS AND
THE RESIDUE AFTER REMOVAL OF COAL VALUES BY AGGLOMERATION
(20 PERCENT KEROSENE CONCENTRATION)
49
-------
2000
1800
1600
1400
1200
1000
c
0)
800
,•'1600
400
Residue
Suspension
460 at
1125 min
390 at
1335 min
200
Final Pulp Density of
Residue =1.14 g/nl
J \ 1 L
_I__J II II
30
60
90
120
Time, min
150
.180
210
FIGURE 7. SETTLING RATES FOR 10 PERCENT SLURRY OF FRESH BLACK WATER
SEDIMENT AND THE RESIDUE AFTER REMOVAL OF COAL VALUES BY
AGGLOMERATION (20 PERCENT KEROSENE CONCENTRATION)
BULK DENSITY AFTER 3 day = 1.14 g/cc
50
-------
TABLE 23. ANALYSIS OF RESIDUES FROM AGGLOMERATION
Source
of Residue SiO« ^-9^7
Ag'ed Sediment 54.4 21.5
Fresh Sediment 55.6 24.8
Composition, percent
Na20
0.77
0.71
CaO MgO
6.07 0.32
4.05 0.33'
K20 Fe203
].31 3.49
2.55 3.30
The above method, however, suffers from the following uncertainties:
• The CC1, does not appear to extract all of the oil
present in the residue slurry.
9 Some of the.oil evaporates together with CC1, .
o The original sediments may contain oily substances
which can be extracted by CC1,.
Base-line experiments were conducted to determine the amount of
No. 2 fuel oil evaporated with CC1, at several known oil concentrations.
The results indicated that the amount of oil lost during evaporation varied
with the oil content. Figure 8 is the calibration curve which shows the
relationship between the actual oil content and that measured by the
evaporation technique." In order to find the amount of oily substances in
the original samples, the CC1,-extraction was conducted with the aged
4 i 5
slurry sediment. It was found that essentially no oil could be found in
the sediment with this technique. Another base-line experiment was carried
out to determine the amount of oil which could not be extracted by the
CCl^-extraction. The extraction experiments on the aged slurry sediment
mixed with the known amount of No. 2 fuel oil indicated that about 10 percent
of the oil could not be extracted by this technique.
* More precise measurements using infrared spectral analysis of the CC1/
oil solutions are available, but were not applied to these preliminary
evaluations.
51
-------
3.0
0.5 1.0 1.5 ' 2.0 2.5, r 3.0
Oil Content (Measured), g of No. 2 fuel oil/100 ml of CC14
FIGURE 8. CALIBRATION CURVE FOR DETERMINATION OF
OIL CONTENT BY CCl^-EXlRACTION•
52
-------
This modified CC1,-extraction technique was applied to determine
the amount of oil loss to the residue suspension during agglomeration of the
aged slurry sediment with No. 2 fuel oil. The results are summarized in
Table 24. It is shown that the amount of oil loss in the residue varied
depending upon the amount of oil used. The smallest oil loss was 2.24 per-
cent at 10 percent oil concentration. The 20 percent oil concentration
resulted in larger oil loss due to the "free" oil unattached to coal
particles. At the 5 percent oil concentration, the sizes of agglomerated
coal were so small that some of the coal was lost into the residue and oil
was carried with it.
Staged Washing of Agglomerated Coal
(3)
Min suggested that resuspension of the recovered agglomerated
coals in water can be effective in further reducing the ash content of
the recovered coal as well as producing large size, strong agglomerates of
the coal. The washing technique consisted of performing an agglomera-
tion of a 10 percent slurry using a 10 percent oil concentration and
then placing the entire mixture into an air-float cell (60 mm ID x
300 mm tube fitted with a fritted glass disc at the bottom). A very
slow flow of air bubbles lifted the agglomerated coal to the top of
the cell where it was removed with a screen. Once all of the coal
fraction had been collected it was either dried and analyzed or it was
resuspended in clean water and air floated a second time and
recovered.
The results of these air float washings are given in Table 25.
The single stage washing gave only a minor improvement in the ash content
(8.90 vs. 9.11 percent) of the coal recovered from the aged sediment and
the 85 percent coal recovery was significantly lower than the 97 percent
recovery obtained by the usual procedure. A significant improvement
in ash in the recovered coal was obtained in the two stage washing (5.66
vs. 8.90 percent) at only a slightly lower coal recovery than in the single
stage washing (80 vs. 85 percent). (No. 2 fuel oil and kerosene containing
1 percent palmitic acid did not offer any improvement in the two stage
washings.) The pyritic sulfur in the recovered coal after the second
53
-------
-P-
TABLE 24. AMOUNT OF OIL LOST TO THE RESIDUE SLURRY DURING AGGLOMERATION
OF THE AGED SLURRY SEDIMENT
(a)
Oil
g
1.25
2.5
5.0
Used
Cone. ,
%
5
10
20
I
Oil Content
(experimental) ,
g Oil/100 ml C04
0.0332
0.0156
0.0590
Oil Content^
(corrected) ,
g Oil/ 100 ml C04
0.05
0.02
0.10
Oil Extracted,
8
0.125
0.05
0.25
Total Oil(c)
in Residue,
g
0.139
0.056
0.278
Oil Loss
in Residue,
%
11.12
2.24
5.56
(a) 250 g of-'Slurry (10% slurry concentration) was agglomerated with No. 2 fuel oil
(b) Oil content is corrected according to the calibration curve (Figure 9).
(c) Total amount of oil in the residue is estimated as
oil extracted (g)
0.9
-------
TABLE 25. EFFECT OF SECONDARY WASHING OF COAL RECOVERED BY AGGLOMERATION
(a)
Ui
Oi
Single Stage Wash
, . Product, percent Coal Recovery,
Oil Used^d' Of Feed MF Ash percent
Aged Sediment
Kerosene 50.8 8.90 85
No. 2 fuel oil^
„ Kerosene — — —
Black Water Sediment
Kerosene 50. 4 8.15 94
Product
Of Feed
46.2
41.0
45.8
48.1-
Two Stage
, percent
MF Ash
5.66
5.22
5.40(C)
5.78
Wash
Coal Recovery,
percent
80
71
80
92
(a) Slurry concen tra t.i.on, 10 percent; oil concentration, 10 percent.
(b) Oils contain 1 percent palmitic acid.
(c) For coal recovered $t , = 3.38 percent, S = 1.05 percent (MAF).
-------
washing was about one-half of that ob.tained in coal recovered by the conven-
tional method (1.05 vs. 1.98 percent).
The results from the experiment using fresh black water sediments
were better. Coal recovery after two stage air float washing was 92 percent
and the ash content of the recovered coal was 5.78 percent which is
significantly lower than the 8.16 percent ash for a coal recovery of 94 per-
cent using the usual agglomeration technique.
ENHANCEMENT OF PYRITE REMOVAL
Because of similarity in the surface characteristics of coal and
pyrite (i.e., both are hydrophobic), oil agglomeration faces difficulty
in removing pyritic sulfur from coal. The earlier study by Sun and
(16)
McMorris showed that the oil agglomeration process was very effective for
recovery of low ash coal but the reduction in sulfur content, was negligible.
The Bureau of Mines recent study also confirmed .that the agglomeration
process provided only limited sulfur reduction. Thus, in order to enhance
pyrite removal during oil agglomeration, it is necessary to modify the sur-
face characteristics of pyrite to become hydrophilic.
There are many chemicals which are considered to be good pyrite
depressants in the flotation practice of sulfide minerals. They include lime,
cyanides, permanganates, sodium sulfite or sulfide, etc. The effectiveness
of these chemicals during agglomeration of coal had .been tested by Cape
et al. and Min. The results showed that these chemicals provided only
limited success on depressing pyrite. The preliminary work in earlier phases
(2)
of this program also confirmed that the pyrite depressants tried did not
improve the separation of pyrite from coal significantly. This ineffective-
ness may be due to the highly porous structure of coal. The coal particles
may adsorb the 'depressing agents rapidly before they can alter the surface
property of pyrite particles. Hence, an excessive amount of depressing
agents may be needed to affect the pyrite. In general, however, the use of
excessive agents deteriorates coal recovery.
Instead of using the conventional pyrite depressants, other
approaches have been studied to alter the surfacial characteristics of the
pyrite. Capes, et al. showed that the surface of the pyrite could become
56
-------
hydrophilic during grinding in a slurry containing iron-oxidizing bacteria,
Ferrobacillus-Thiobacillus group. More than 90 percent of the pyritic sulfur
could be removed after such a treatment. It was postulated that the surface
of the pyrite particles were oxidized by bacterial action rendering hydro-
philic surfaces. On the other hand, Min reported that the surface of the
pyrite could also be altered by mild chemical treatments in an aqueous
alkaline solution with air at elevated temperature. Presumably the surface
oxidation of pyrite in water led to the formation of a film of hydrated
ferric oxide which reduced the floatability of the pyrite.
During this screening study, the following alternative approaches
were tried to find an effective method which could selectively alter the
pyrite surface to become hydrophilic:
• Aqueous oxidation treatment
« Electrolytic treatment
• Microwave treatment
9 Wetting-agent treatment
The approaches and results are described in the subsequent sections.
Aqueous Oxidation Treatment
Unheated Treatments—
Pretreatment of the coal with a pyrite depressant at 25 C was done
before any attempt at agglomeration. Known pyrite depressants were used at
-4
concentrations of 5 x 10 g per g of coal (or one Ib per ton of coal) or
less. An attempt was also made to pretreat coal with ferric sulfate solution
under conditions more mild than the TRW-Meyers process which is known to
react with pyrites. When this was done, the coal was separated by
filtration before agglomeration was attempted. The slurry pH was adjusted to
a specied value before agglomeration with dilute sodium hydroxide or sulfuric
acid solutions. Studies on the use of depressants to improve pyrite removal
were done using kerosene with a loading of 20 percent of the dry coal weight
and a blending time of 2.5 minutes to accomplish the agglomeration. The
results of the experiments on the effect of chemical treatment at 25 C on
pyrite removal during agglomeration are given in Table 26.
In Table 27, the reduction of pyritic sulfur of untreated and
57
-------
TABLE 26. RESULTS OF CHEMICAL PRETREATMENT AT 25 C 'ON PYRITE SEPARATION FROM ILLINOIS
NO. 6 COAL DURING AGGLOMERATION (25 Grams Coal in 225 Grams Water)
Pretreatment of Coal
Experiment
No.
Feed Coal
60
62
65
66
67 .
70
71
72
U1
CO 75
76
77
78
79
80
1
Feed Coal
2
3
4
5
6
8
Mesh
Size
-14
-14
-14
-14
-14
-14
-14
-14
-14
-14
-14
-14
-14
-14
-14
-14
-14
-48
-48
-48
-48
-48
-48
-48
Amount Time,
Agent ml/25 g Coal mln
NaOCl Solution
NaOCl Solution
2N Fe2(S04>3
2N Fe2(S04)3
2N Fe2(S04)3
.IN Fe2(S04>3
.IN Fe2 (804)5
.IN Fe2(S04>3
.IN Fe2(S04^3
.IN Fe2(S04>3
.IN Fe2(S04)3
.IN Fe2(S04>3
NaOII
0.1M K4Fe(CN)6
0.1M K3Fe(CN)6
.IN Fe2(S04>3
.IN Fe2(S04)3
.IN Fe2(S04>3
0.1M l<4Fe(CN)6
0.1M K3Fe(CN)6
0.1M K4Fe(CN)0
NaOII
4.0
0.4
150
150
150^
2.0
2.0
150
2.0
2.0
2.0
2.0
150
3.0
4.0
150
150
2
3
4
3
150
0.5
10.7
15
60
15
0.5
2.5
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
Temp. ,
C
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
Agglomeration Conditions
Type(a)
A
A
A
A -
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
C
Oil
Charge, f
7.5
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
Time,
5 mln
0.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5"
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
PH
10.8
10.7
2.3
• 2.3
11.0
4.0
4.0
2.3
4.5
10.9
7.4
4.4 to 8.0
8.0 to 6.2
10.7
4.6
4.4
2.1
2.3
5.2
5.3
5.2
5.2
5.4
Product, percent
Feed
10.8
58.4
2.0
1.6
3.2
63.6
56.4
24.8
44.0
0.4
23.6
9.2
72.0
68.8
61.6
2.0
0.4
85.6
83.5
94.0
94.8
86.4
MF
Ash(b)
14.0
9.47
7.92
6.51
)
8.34
8.82
8:15
7.42
8.06
--
7.59
--
8.24
7.93
8.27
--
12.9
--
7.47
8.08
7.81
7.81
8.62
Residue, percent
MAF(c)
STOT
5.93
4.64
4.38
3.75
--
4.10
4.23
4.62
4.68
4.21
4.52.
--
3.78
--
4.51
4.39
4.37
--
5.75
--
4.59
4.54
4.43
4.52
5.01
SPYR
3.12
2.33
1.67
2.02
--
1.51
1.50
1.80
1.83
1.62
1.79
--
1.44
--
1.80
1.52
1.65
--
2.89
--
1.99
1.89
1.81
1.87
2.11
Feed
84.8
39.2
96.0
93.2
52.0
32.4
39.2
70.4
53.2
101.6
72.8
87.6
26.8
29.2
36.8
97.6
94.8
9.6
12.8
7.6
7.6
7: 6
MF
Ash(b)
12.7
15.0
11.7
--
14.0
16.6
17.4
12.6
14.7
--
13.1
--
20.96
19.2
16.6
--
--
40.5
33^9
53.6
53.7
41.3
MAF(c)
STOT
5.20
5.45
4.84
--
5.09
5.60
6.00
5.07
5.48
--
4.76
--
6.91
6.24
6.09
--
--
11.06
9.64
20.3
19.9
8.93
SPYR
2.41
2.73
2.32
--
2.42
2.82
3.19
2.49
2.81
--
2.60
--
3.94
3.44
3.01
--
--
8.53
7.38
17.0
17.2
6.39
(a) Oil types: A = Kerosene; B - No. 2 Fuel Oil; C = Tetralln
(b) MF = Moisture-Free basis
(c) MAF - Moisture- and Ash-Free basis
(d) Pretreatment of coal In separate vessel and filtered before: use In,agglomeration
(e) Separation on 28 meah sieve
(f) Separation on 48 mesh sieve.
-------
TABLE 27. EFFECTIVENESS OF DEPRESSANTS FOR REMOVAL OF PYRITE FROM ILLINOIS NO. 6 COAL
DURING AGGLOMERATION COMPARED TO RESULTS FROM FLOAT-SINK ANALYSIS
Product (Float)
Feed Coal
Agglom.
Agglom.
Agglom.
Agglom.
41(c)
61
70
79
Float-Sink
Analysis
Feed Coal
Agglom.
Agglom.
Agglom.
50
3
6
Float-Sink
Analysis
Mesh
Size
-14
-14
-14
-14
-14
-14
-48
-48
-48
-48
-48
Residue (Sink)
Pretreatment Z Re- MF(«) Ash, MAF,(b) 70 % Re_ MF Ash, MAF, %
of Coal covered
None 43.6
pH 10.7 64.8
2 ml, 0.1N
Fe2(S04>3 63.6
30 seconds
3 ml, 0.1M
K4Fe(CN)6 68.8
15 rain
1.9 specific 94.1
gravity
None 84.4
2 ml, 0.1N
Fe2(S04)3 85.6
15 min
3 ml, 0.1M 94.8
KAFe(CN)6 15 min
1.9 specific 93.3
gravity
14
9
7
8
7
9
12
7
7
7
9
%
.0
.18
.97
.82
.93
.77
.9
.83
.47
.81
.32
S
5
4
4
4
4
3
5
4
4
4
3
TOT
.93
.43
.18
.62
.39
.89
.75
.15
.59
.53
.87
Pyritic
Sulfur
J>PYR covered % ^fQf ^PYR Reduction, %
3.12
1.56 49.2 13.8 5.52 2.70
1.62 32.8 21.1 6.48 3.56
1.80 32.4 16.6 5.60 2.82
1.52 29.2 19.2 6.24 3.44
1.28 5.9 51.3 37.5 25.9
2.89
1.53 10.0 42.3 17.4 14.4
1.99 9.6 40.5 11.1 8.53
1.87 7.6 53.7 19.9 17.2
1.09 6.7 49.7 40.10 34.5
50
48
42
51
59
47
31
35
62
(a) MF = Moisture Free basis
(b) MAF = Moisture and Ash Free basis
(c) All agglomerations were performed with kerosene using 20 percent of the weight of the coal
feed.
-------
pretreated coal are compared to the reduction attainable from float sink
analysis for the two coal sizes studied. These data suggest that of the
reagents tried with the coarser coal, only potassium ferrocyanide seemed to
give good coal recovery and pyrite separation. However, it does not attain
the performance of float-sink separation. The next best system appears to be
the adjustment of the pH of 10 to 11. Potassium ferrocyanide exhibited
similar improvements in coal recovery and pyrite removal for the minus
48 mesh coal. Although these improvements with the finer coal are signifi-
cant, they do not match the separation possible at a specific gravity of 1.9.
This is especially true for pyrite in the residue. It should be noted that
ash remaining with coal after agglomeration treatment is less than that
obtained by the float-sink separation. In all but the experiment in which
potassium ferrocyanide was used, the amount of misplaced coal may account
for this effect.
Solutions with concentrations typical of that used by the TRW-
Meyers process (i.e., 2 Normal) were used to treat the coal for period of
15 and 60 minutes but with no heating. The intent of these experiments was
to attack the surface of the pyrites through the same reaction that consumes
them during the TRW-Meyers process. It was found that the hydrophobic
character of the coal was destroyed, essentially no agglomeration occured
under conditions known to be successful, and more than 90 percent of the coal
passed through the screen (Experiments 65, 66, and 67, Table 26). Even
solutions of 0.IN Fe9(SO,)» caused the coal to be wetted by water. By
decreasing the dosage to only 2 ml of 0.1N Fe^SO per 25 g of coal (0.6 g
Fe«(SO,)~/100 g) only slight improvements were noted. Adjustment of the pH
to 10.9 compounded the hydrophylic effect probably because the absorbed
ferric ion on the coal surface formed the hydroxide. (The effect of acid was
not investigated.)
!
In those experiments where agglomeration gave 80 percent coal
recovery or better, the water occluded in the coal-kerosene agglomerate was
about 46 percent of the wet cake. This amount was reduced to 24 by only
mechanical action (kneading on the screen).
Heated Treatments—
The jib-washed coal samples from Illinois No. 6 described earlier
60
-------
were ground to minus 48 mesh and pretreated in an aqueous slurry before oil
agglomeration. Figure 9 shows the apparatus used for this treatment step.
One hundred grams of ground coal was mixed with 400 ml of distilled water in
a 500 ml gas washing bottle and a small amount of chemical was added. The
slurry was agitated with a magnetic stirrer and heated by an electrical heat-
ing tape. After the temperature reached above 90 C, air bubbles were intro-
duced through a porous glass plate at about 400 cc/min for 30 minutes. The
reaction temperature was maintained about 90 C and a condenser was mounted
above the reactor to prevent loss of water by evaporation.
After the pretreatment, the slurry was diluted to 10 percent
slurry concentration by adding distilled water and cooled to room tempera-
ture. One-quarter of the slurry was taken and agglomerated with 5 g of
kerosene. With each pretreated slurry, three agglomeration runs were made
at acid, neutral and basic pH values by adjusting with either hydrochloric
acid or sodium hydroxide solution. The experimental results are summarized
in Table 28. The data.show that the aqueous oxidation treatments signifi-
cantly improved pyrite removal during agglomeration as compared with the
untreated coal samples (Run #24). However, the observed differences in
sulfur reduction among the different chemical conditions are too small to
choose which condition is the best for treating the pyrite. It appears that
the adjustment of pH after the chemical treatment did not have any marked
effect on the agglomeration results.
Another series of aqueous oxidation treatments were carried out
with the jig-washed coal sample from Ohio Pittsburgh Seam No. 8 coal to
examine the effects of wider variety of chemical conditions. The apparatus
and experimental procedures were the same as the previous runs except that
the treated slurry was agglomerated with 2'. 5 g of No. 2 fuel oil. Table 29
summarizes the pretreatment conditions and the agglomeration results. For
the sake of comparison, a float-sink analysis of the same coal (minus
48 mesh) at 1.59 specific gravity is also shown in the table. Since the
specific gravity of pyrite is about 5.0, the float fraction at the specific
gravity 1.59 is supposed to be free from liberated pyrite and the remaining
pyrite represents that locked in the coal matrix. Hence, the pyritic
sulfur reduction obtained by the 1.59 specific gravity float was set as the
attainable goal by the oil agglomeration technique.
61
-------
Thermometer
Rota-
Meter
Air
Reflux Condenser
«-Cooling Water
Electrical Heating Tape
Porous Glass
Magnetic
Stirrer
FIGURE 9. EQUIPMENT FOR CHEMICAL PRETREATMENT OF
A COAL SLURRY
62
-------
TABLE 28. EFFECT OF AQUEOUS OXIDATION PRETREATMENT ON PYRITE REMOVAL DURING
OIL AGGLOMERATION OF ILLINOIS NO. 6 COAL (FRESH GROUND)
ON
U)
Run No.
Chemical Agent
0.5 g/100 g coal
PH
Of
Feed
Feed coal
24-1
-2
-3
7-1
-2
-3
12-1
-2
-3
14-1
-2
-3
16-1
-2
-3
22-1
-2
-3
27-1
-2
-3
None
None
None
Air
Air
Air
Na -S+Air
2
Na S+Air
Na 2 S+Air
Fe7(SO K+Air
Fe"(SO K+Air
Fe0(SO,K+Air
z 43
NaHCO +Air
NaHCO^+Air
3
NaHCO +Air
(NH ) CO +Air
(NlO^COil+Air
(Nil ) CO +Air
Na 02+Air
Na 02+Air
Na 0 +Air
4.3
7.4
9.0
5.3
7.2
8.9
7.2
5.0
8.9
3.1
5.1
6.9
7.0
5.1
9.1
7.8
5.2
8.8
8.3
5.1
6.9
98.0
94.0
97.7
94.0
90.9
87.8
92.1
91.6
83.9
93.7
91.3
94.3
87.1
93.1
92.4
95.4
86.6
92.6
90.5
92.3
91.2
Product ,
percent
MAP
MF Ash
11.3
8.70
9.02
9.15
7.64
7.78
7.82
7.50
7.30
7.32
8.05
8.20
8.41
7.63
7.70
8.04
8.03
7.62
7.78
7.72
7.77
7.78
STot
4.07
3.79
—
3.53
3.56
3.61
3.74
—
—
3.58
—
—
3.51
—
—
3.76
—
—
,3.70
--
—
SPyr
1.36
1.27
--
1.00
1.00
1.15
1.01
--
—
1 . 08
—
—
1.03
—
—
1.06
—
—
0.98
—
—
Reduction, percent
Ash
23.0
—
32.4
31.2
30.8
33.6
35.4
35.2
28.8
27.4
25.6
32.5
31.9
28.8
28.9
32.6
31.2
31.7
31 . 2
31.2
STot
6.9
--
13.3
12.5
11.3
8.1
—
—
12.0
—
—
13.8
—
—
7.6
—
—
9.1
—
—
SPyr
6.6
—
26.5
26.5
15.4
25.7
—
—
20.6
—
—
24.3
—
—
22.1
—
—
27.9
—
—
-------
TABLE 29. EFFECT OF AQUEOUS OXIDATION PRETREATMENT ON PYRITE REMOVAL
DURING OIL AGGLOMERATION OF OHIO PITTSBURGH SEAM NO. 8 COAL
Product ,
Run No .
Feed Coal
Float at
1.59 Sp.
44-1
46-1
41-1
89-1
91-1
97-1
7-1
52-1
48-1
93-1
99-1
9-1
95-1
65-1
74-1
Chemical ,
g/100 g coal
Cr.
None
Air
Air
°2
°2
Na S (0.1)
Na^S (0.1)
Air +'"Na2S (1.0)
Air + Na?S (1.0)
07 + Na2S (0.1)
Na2CO3 (0.1)
Na2C03 (0.1)
02 + Na2C03 (0.1)
Sediment wash
water (a)
Sediment wash
water 'a/'
Na202 (1.0) + Air
Temp . ,
C
RT
96
RT
92
RT
95
RT
96
RT
RT
99
RT
RT
RT
96
Time,
min
900
30
30
30
30
30
900
30
30
30
30
30
240
5280
30
PH
6.7
7.2
8.3
3.7
2.9
4.7
4.0
9.7
10.9
4.7
4.5
3.7
4.5
3.8
2.7
11.5
Of
Feed
95.4
97.2
96.5
95.1
96.4
96.5
96.8
95.8
95.6
94.1
96.0
96.0
95.6
96.4
96.0
90.5
79.6
percent
MAF
MF Ash
10.50
8.69
7.44
6.50
6.16
6.98
6.91
6.68
6.62
6.03
5.84
6.67
7.13
6.53
6.86
6.96
6.54
6.79
STot
5.25
4.22
4.89
4.19
4.18
4.20
4.19
4.22
4.15
4.17
4.09
4.12
3.95
4.04
4.2,4
4.17
4.18
SPyr
2.24
1.30
2.32
1.65
1.57
1.63
1.54
1.58
1.51
1.49
1.48
1.93
1.39
1.35
1.82
1.57
1.65
Ash
17.2
29.1
38.1
41.0
33.5
34.2
36.3
36.9
42.5
44.3
36.4
32.1
37.8
34.6
33.7
37.7
26.5
Reduction ,
percent
STot
19.6
6.9
20.2
20.4
20.0
20.2
19.6
21.0
20.6
22.1
21.5
24.8
23.0
19.2
20.6
20.6
sPyr
42.0
(-3.6)
26.3
29.9
27.2
31.3
29.5
32.6
33.5
33.9
• 13.8
37.9
39.7
18.8
29.9
29.9
(a) The sediment wash water was filtrate of the slurry containing 100 g partially dried old coal
sediment and 500 ml distilled water.
-------
With reference to Table 29, the following qualitative observations
can be made:
1. As compared with the float-sink values, the oil
agglomeration products were lower in ash content but
higher in pyritic sulfur content. However, the total
sulfur contents of the pretreated samples were as low
as or lower than that of the float-sink value.
2. All of the pretreatments provided a cleaner product
than could be obtained by direct oil agglomeration
(Run 44-1). '
3. In general, the pyritic sulfur reductions were higher
with oxygen treatments than those with air treatments,
while the ash reductions were reversed.
4. A comparison of Runs 89-1 and 91-1 indicates that
the low temperature provided a higher pyrite reduction
than the high temperature. This may be due to the
lower solubility of oxygen at higher temperature.
5. In the sodium sulfide treatments (Runs 97-1 and 7-1),
a higher pyritic sulfur reduction was obtained at
lower temperature. This may indicate that the
depressing action of sodium sulfide was caused by
adsorption rather than by chemical reaction.
6. The results of Runs 89-1, 97-1 and 93-1 show that the
treatment with sodium sulfide and oxygen together pro-
vided a higher pyritic sulfur reduction than could be
obtained by either one alone. The results of Runs
89-1, 99-1 and 95-1 also show the similar synergetic
effect.
7. In the sodium carbonate treatments (Runs 99-1 and 9-1),
a higher temperature provided significantly cleaner
product. This suggests that the depressing effect of
sodium carbonate was apparently caused by chemical
reaction.
65
-------
8. The treatments with sediment wash water, which may
contain some iron-oxidizing bacteria, show that pro-
longed reaction was needed to depress the pyrite.
Electrolytic Treatment
Another approach was the electrolytic pretreatment of a coal slurry
prior to oil agglomeration. It was hoped that electrolytic reaction might
offer a selective means of modifying the surface characteristics of pyrite
particles. The apparatus used for the electrolysis is shown in Figure 10.
A stainless steel rod (3/4 in. dia. x 4 in.) was used as a rotating '
cathode and a sheet of brass or steel (4 x 8-1/2 in.) rolled into a cylinder
was used as an anode. The cathode chamber was kept separated from the anode
chamber by a porous alundum thimble (1-3/4 in. dia. x 5 in.). A 400 ml
beaker was used in all experiments. Coal slurry containing 25 g of the
ground Ohio Pittsburgh Seam No. 8 .(minus 48 mesh) was electrolyzed for
30 minutes at 10 volt potential and 0.5 ampere limiting current. Elec-
trolysis was carried out in two different configurations: one with the
porous thimble in place and the other without the thimble. The treated
, coal slurry was then agglomerated with 2.5 g of No. 2 fuel oil at 10 percent
slurry concentration. Tables 30 and 31 summarize the experimental conditions
and agglomeration results. In general, the electrolytic treatments were hot
as effective as the aqueous oxidation treatments on the pyrite removal
during agglomeration. This might be due to the poor reactor configuration
in which the oxygen bubbles generated from the . anode surface failed to make
good contact with the pyrite particles. The best pyrite removal was
obtained when sodium carbonate was used as an electrolyte (Run 42-1),
but the amount of coal recovery was very low.
Microwave Treatment
Another approach employed was the use of 2450 MHz microwave
energy. It was hoped that pyrite particles might be preferentially heated
by the microwave energy, thereby modifying their surface characteristics.
The ground Ohio Pittsburgh Seam No. 8 coal (minus, 48 mesh)' in either dry
or wet state was exposed to microwave energy under a nitrogen- atmosphere in a
66
-------
DC Power ©
Source ©
Ammeter
Variable Speed
Electric Motor
c
Magnetic Stirrer
— Rotating Cathode.
•Stationary Anode
""Porous Alumdum Thimble
FIGURE 10. EQUIPMENT FOR ELECTROLYTIC PRETREATMENT OF COAL SLURRY
-------
TABLE 30. EFFECT OF ELECTROLYSIS ON THE PYRITE REMOVAL DURING OIL AGGLOMERATION
OF OHIO PITTSBURGH SEAM NO. 8 COAL (WITHOUT THIMBLE)
00
Eroduct
PH
Run No .
Anode Electrolyte
Before
After
, percent
Of
Feed
'Feed Coal
Float at
1.59 Sp
44-1
19-1
19-3
19-4
34-1
34-2
34-3
42-r
. Gr.
No Pretreatment
Brass None
Brass .NaOH'- '
Brass NaOH^a^
Steel 0.5% NH^OH
Steel 0.1% Fe2(SO^)3
Steel 0.1% Na2"S
Steel 0.03% Na2C03
6.
3.
10.
7.
10.
2.
8.
5.
7
5
0
0
8
7
4
4
6.
5.
10.
10.
10.
2.
8.
6.
7
3
3
0
5 .
6
2
5
95
97
92
95
96
89
9.3
94
84
.4
.2
.3"
.6
.6
.2
.7
.6
.3
MF Ash
10
8
7
7
10
10
7
7
8
7
.50
.69
.44
.86
.21
.39
.73
.78
.66
.99
5
4
4
4
4
4
4
4
4
MAF
STot
.25
.22
.89
.31
—
.62
.38
.73
.92
.25
S
2.
1.
2.
1.
-
2.
'I.
2.
2.
1;
Reduction,
Pyr
24
30
32
90
-
33
86
13
17
48
Ash
17,2
29.1
25.1
2.8
1.0
26.4
25.9
17 .5
23.9
STot
19
6
17
-
12
16
9
6
19
.6
.9
.9
-
.0
.6
.9
.3
.0
percent
SPyr
42.0
(-3.6)
15.2
—
(-4.0)
17.0
4.9
3.1
33.9
(a) One molar sodium hydroxide solution w.as added to adjust the pH.
-------
TABLE 31. EFFECT OF ELECTROLYSIS ON PYRITE REMOVAL DURING OIL AGGLOMERATION
OF OHIO PITTSBURGH SEAM NO. 8 COAL (WITH THIMBLE)
PH
Run No.
Feed Coal
Float at
1.59 Sp.
44-1
19-11
19-10
32-1
32-2
19-7
19-5
Cr.
Anode (brass)
Cathode
Anode (brass)
Cathode
Anode (steel)
Cathode
Anode (steel)
Cathode
Anode (brass)
Cathode
Anode (brass)
Cathode
Coal
1 wt
1 wt
Coal
Coal
1 wt
Coal
1 wt
Coal
Old
Old
Coal
Content
slurry
. % CaCl2
. % CaCl2
slurry
slurry
. % NaOll ..
slurry
• % Na2S
slurry
coal sediment
coal sediment
slurry
Before
6
3
7
7
3
3
11
3
11
3
2
2
3
.7
.5
.7
.7
.5
.3
.9
.3
.8
.3
.5
.5
.5
After
6.7
5.2
12.0
5.9
11.8
2.6
11.8
2.3
11.8
4.0
3.3
3.8
10.8
Of
Feed
95.
97.
95.
87.
91.
87.
93.
53.
60.
91.
4
2
3
9
7
3
8
5
8
9
Product, percent
MAF Reduction, percent
MF
10
8.
7.
7.
8.
7.
7.
7.
27.
20.
9.
Ash STot Spyr Ash STot
.5 5.25 2.24
69 4.22 1.30 17.2 19.6
44 4.89 2.32 29.1 6.9
57 4.53 2.08 27.9 13.7
48 — — 19.2
44 4.26 1.94 29.1 18.9
86 4.42 2.01 25.1 15.8
14 4.16 1.92 32.0 20.8
67
32
23 -- — 12.1
sPyr
42.0
(-3.6)
7.1
—
13.4
30.27
14.3
—
-------
Varian Industrial Microwave Systems Model EW3-DPM35. After treatment, 25 g
of coal was taken and agglomerated with 2.5 g of No. 2 fuel oil at 10 per^
cent slurry concentration. Table 32 summarizes the treatment conditions and
agglomeration results.
Improvement in pyrite reduction could be noticed when the wet
coal impregnated with sodium hydroxide (Run 29-3) was heated by the micro-
wave. One advantage of the microwave treatment is short reaction time, i.e.,
seconds vs. minutes, and hence this treatment may be suited for continuous
pretreatment^before oil agglomeration
Wetting-Agent Treatment
Another approach was addition of chemical reagent from the known
classes of either wetting or dispersing agents. In the; pretreatment step,
100 g of ground coal (minus 48 me'sh) was mixed with 400 ml of distilled water
and a small amount of reagent. The slurry was agitated with a magnetic
stirrer at the room temperature for a specified time. After the pretreatment,
one quarter of the slurry was taken and agglomerated with 2.5 g of No. 2
fuel oil at 10 percent slurry concentration.
The specific- chemical reagents were chosen for the following
reasons:
a Sodium-di-octyl sulfosuccinate and sodium dodecyl sulfate
are commonly known as wetting agents. In a U.S. patent
/ 1 Q \
by Chia, these agents are. claimed to be effective f:or
desulfurizing high sulfur low grade coal.
9 Phytic acid and myo-inositol-2-monophosphate are chosen
because of their structural similarity with phosph'atidal-
inositol which is identified as the. wetting agent
secreted by bacteria during their oxidation of sulfide
minerals.
« Sodium metaphosphate, sodium carboxy methyl cellulose
and sodium tripolyphosphate are dispersing agents. A
(20)
U.S. Bureau of Mines study shows that they are
effective in dispersing mineral particles during selec-
tive flocculation of coal slimes.
70
-------
TABLE 32. EFFECT OF MICROWAVE TREATMENT ON PYR1TE REMOVAL DURING OIL
AGGLOMERATION OF OHIO PITTSBURGH SEAM NO. 8 COAL
Product,
Microwave
Run No. Coal Sample Energy, watt
Feed Coal
Float at
1.59 Sp. Cr.
16-1 Dry
29-1 Coal
powder 500
1000
+ 1% NaOH 500
Overdrled
29-3 Coal
Wet
+ 1% NaOH 500
paste
Exposure Temp., Of
Time, mln C pll Feed
95
3
2 238 4.7 96
3 120 10.4 96
5.9(a) 96
3 98 H-4, , 8*
6.9CdJ 92
.4
.6
.8
.9
.5
.5
percent
MAP Reduction, percent
MF
10.
8.
9.
8.
8.
7.
8.
Ash
50
69
10
18
12
31
21
STot SPyr Ash
5.25 2.24
4.22 1.30 17
4.77 2.22
13
4.27 2.14 22
22
4.14 1.69 30
21
.2
.3
.2
.7
.4
.8
STot SPyr
19.6 42.0
9.1 0.9
18.7 4.5
—
21.1 24.6
(a) pH was adjusted with HC1 solution before agglomeration.
-------
Table 33 summarizes the experimental results. The data show that
the reagents generally provided cleaner product than could be obtained
without the reagent. In particular, the results obtained with sodium
metaphosphate are surprisingly good and the product was cleaner than that
of 1.59 Sp. Gr. float. Although more work is needed to verify this
exceptionally good result, it appears that sodium metaphosphate disperses
the liberated, pyritic particles to prevent them from being -entrapped in the
agglomerated coal.
Combined Treatment
An exploratory experiment was conducted to examine the added
effectiveness of successive pretreatment steps. One hundred grams of the
ground Ohio Pittsburgh Seam No. 8 coal was mixed with 0.1 g sodium
carbonate in 300 ml distilled water and electrolyzed without the thimble as
described previously. After the electrolysis,' the treated slurry was
''divided into four parts and each part was agglomerated with 2.5 g of No. 2
fuel oil at the following conditions:
1. Agglomeration without further treatment.
2. Agglomeration after mixing with 0.25 ml of 5 percent
sodium metaphosphate solution.
3. Same as (-2) except after agglomeration, 0.25 ml of
0.1 percent Percol 728 cationic flocculant was added
before screening.
4. Same as (3) except sodium metaphosphate solution
was added.
The experimental results are shown in Table 34. Except Run 42-4,
the data indicate that the agglomerated product became cleaner as more
steps were added. The product of Run 42-3 was much cleaner than that of
1.59 Sp. Gr. float.
72
-------
TABLE 33. EFFECT OF CHEMICAL REAGENT ON PYRITE REMOVAL DURING OIL
AGGLOMERATION OF OHIO PITTSBURGH SEAM NO. 8 COAL
Product, percent " Reduction,
Reagent, Time, Of MAF percent
Run No .
Feed Coal
Float
1.59 Sp.
44-1
56-1
58-1
62-1
60-1
1-1
3-1
g/100 g coal min pH Feed MF Ash STot Spyr Ash STot
10.50 5.25 2.24
Cr. 95.41 8.69 4.22 1.30 17.2 19.6
None 6.7 97.2 7.44 4.89 2.32 19.5 6.9
Sodium-dioc tyl
sulfosuccinate (0.2) 100 7.3 96.6 6.98 4.20 1.61 24.5 20.0
Sodium dodecyl
sulfate (0.2) 100 7.3 96.0 8T05 4.37 1.77 12.9 16.8
Phytic acid
calcium salt (0.2) 100 6.8 96.2 6.37 4.37 1.73 31.1 16.8
Myo-inositol-2-
monophosphate (0.0004) 3900 4.8 92.5 6.61 4.44 1.77 28.5 15.4
Sodium metaphosphate (0.5) 30 4.2 94.0 6.39 4.10 1.29 30.8 21.9
Sodium carboxy methyl
SPyr
42
(-3
28
22
22
22
42
.0
.6)
.1
.0
.8
.0
.4
5-1
cellulose (0.1)
Sodium tripolyphosphate
(0.5)
30 4.1 95.2 7.85 4.23 1.60 15.0 19.4 28.6
30 5.1 97 6.96 4.31 1.68 24.7 17.9 25.0
-------
TABLE 34. EFFECT OF COMBINED TREATMENT ON OIL AGGLOMERATION
OF OHIO PITTSBURGH SEAM NO. 8 COAL
Run No.
Product, percent
Of MAP Reduction, percent
Treatment Feed MF Ash Sfot spyr Ash STot SPyr
Feed . 10.50 5.25 2.24
Float at
1.59 Sp. Gr. 95.4 8.69 4.22 1.30 17.2 19.6 42.0
42-1 Electrolysis 84.3 7.99 4.25 1.48 23.9 19.0 33.9
42-2 Electrolysis + sodium metaphosphate 86.1 7.21 4.12 1.38 31.3 21.5 38.4
42-3 Electrolysis + sodium metaphosphate
+ Percol 728 85.0 6.55 3.99 1.26 37.6 24.0 43.8
42-4 Electrolysis + Percol 728 86.5 10.19 — — 3.0
-------
SECTION 6
DISCUSSION
COAL RECOVERY FROM WASTE STREAMS
The coal mining industry is under constant pressure to limit the
effluents from coal mining and preparation operations and coal refuse
handling and water management are two areas of immediate concern for the
industry. Laws regulating pollution control and tailing water runoff exist.
The ultimate goal is to produce clear water for reuse in the preparation
plant, while the undesirable solids end up as a dry manageable solid that
can be disposed of as a landfill or as mine backfill to reduce acid mine
drainage. The less coal present in this solid, the more attractive it is
with respect to minimizing environmental impact and land requirements for
disposal while maximizing the recovery of a valuable fuel from what usually
is considered a waste.
As mentioned at the beginning of this report the present strategy
is to impound fine coal cleaning wastes in slurry ponds to permit sufficient
residence time for settling to occur before the water is discharged to
natural waterways or reused. This approach has resulted the accumulation of
billions of tons of refuse in active and abandoned waste piles and impound-
ments. A greater accumulation is anticipated as mining and coal cleaning
activity increase to meet the Nation's increasing dependence on coal.
The results from this study on the recovery of coal from fine
(minus 28 to 0 mesh) coal cleaning wastes suggest that good coal recovery
from sediments by the oil agglomeration technique is technically feasible.
The technique can be applied to black water and aged slurry pond sediments
directly, but partially dried and weathered sediments excavated from slurry
ponds require modest treatment before coal recovery is attempted. This
study has identified what appears to be the conditions to recover 90 percent
75
-------
or more of the coal value from wastes containing about 50 percent coal for
one small region in southeastern Ohio. Even within this region it was found
that conditions for recovery of coal values from black water sediments are
slightly different from those used for recovery of aged sediments. This was
due in part to recent changes in the coal cleaning plant unit operations
which produced a much finer suspension making the waste stream more responsive
to coal recovery by the oil agglomeration technique. What is even more
significant is that the technique can be modified slightly and applied to aged
sediments as well for recovery of a quality fuel without the associated mining
costs.
The quality of the coal recovered* from aged slurry pond
sediments and black .water sediments compared to that presently being
shipped from the mine (after jig washing), is given in Table 35.
The quality of the dryer product is also included to give .a com-
parison to fine coal. It can be seen that the recovered coals are lower in
ash and sulfur and generally of better quality than the coal shipped from
the mine. In addition the recovered coal can be dewatered easily. The
residues from the operations retain only about 2 to 5 percent of the
agglomerating oil used and consist of about 92 percent ash. Their mineral
composition probably consists of the unaltered major minerals normally
associated with coal (i.e., illites, kaolinite, montmorillonite, quartz,
gypsum, .pyrite, calcite, dolomite, siderite) and possibly many of the minor
minerals (and the trace elements commonly associated with them). Indica-
tions are that some of the pyrites liberated from the coal in the wastes
are removed in the process. Chemical treatment of the slurry by techniques
i
developed for reducing the pyrites in the coal (discussed later) could be
used to further reduce the pyritic sulfur in the recovered coal.
Since the residues from the coal recovery operation consist mostly
of unaltered mineral matter taken from the earth, they would be compatible
with any landfill disposal method. The residue slurry settles rapidly and
can be concentrated by settling. Since it now consists mostly of clays,
technology developed to improve settling of such materials could be applied
Using agglomeration technique without chemical treatment for enhanced
pyrite removal.
76
-------
TABLE 35. COMPARISON OF QUALITY OF COAL RECOVERED
FROM SEDIMENTS TO CLEANED COAL
Shipped
• Coal
Product
Clean Coal(a)
Dryer Product
MF
9
9
Ash,
%
.24
.84
MAF
Pyr
2.04
1.71
Sulfur, %
Total
4.88
4.97
Recovered from Sediment, .
from Slurry Pond (Avg)(C) 8.89 1.70 3.40
Recovered from Sediment/ v
from Black Water (Avg)U; 6.13 1.41 3.66
(a) Jig cleaned. '
(b) Fines before mixing for shipping.
(c) Average of values for high coal recovery-high ash, high recovery-
low ash, and low recovery-low ash.
77
-------
more specifically through the use of flocculants and/or coagulants. The
material also may be used as raw material directly (e.g., for ceramic
products) or as a source of mineral for metal recovery (e.g., alumina).
In addition to the coal recovery from the slurry wastes, the oil
agglomeration technique can be utilized for cleaning and dewatering of fine
coal. For example, oil agglomeration can provide more effective means
of cleaning ultrafine coal than the froth flotation. It was demonstrated
in Germany that the use of the oil agglomeration process for the minus 250
mesh coal improved the capacity and selectivity of fine coal cleaning cir-
(9)
cuit, and the product was easy to filter, rendering low moisture content.
If the coal cleaning industry adopts new desulfurization techniques such as
high-gradient magnetic separation, chemical treatments and bacterial treat-
ments in which the coal sizes are required to be very small for effective
treatment, the oil agglomeration technique will be a viable means of
isolating and dewatering of the desulfurized product. Furthermore, if the
use of coal-oil mixture becomes widely -accepted in commercial boilers, the
oil agglomeration technique will be an ideal method of coal cleaning
because the oil used for agglomeration will not represent any additional
cost and the agglomerated product can be readily dispersed in the oil.
PREPARATION OF COAL FEED TO LIQUEFACTION PLANTS
Reports about facilities proposed to prepare clean fuels by
liquefaction of coal often skim over the critical aspects of the coal
preparation plant operation that prepares the fine coal to be fed into the
oil slurry preparation prior to entering the liquefaction process. The
coal must be ground to a fine size usually 70 percent minus 200 mesh
(0.05 mm) and then mixed with recycle oil to prepare a slurry. Grinding
the coal to this size is an extremely dusty operation when done dry but
wet grinding is usually ruled out because of the energy required for drying
before the coal can be utilized. However, wet grinding would be advan-
tageous for environmental and health reasons.
This study demonstrated that coal can be recovered from fine coal
slurries with coal liquid distillates as well as with hydrocarbons derived
from petroleum. These agglomerates could be dewatered to levels of
73
-------
^ 15 percent. In addition when a coal is ground to minus 200 mesh, signif-
icant amount of mineral matter is liberated and can be removed during
agglomeration. Thus a process is envisioned where wet grinding can be
utilized to remove significant ash from the coal feed before entering the
liquefaction step. The feed would be agglomerated with an oil used to
prepare the oil slurry feed to the liquefaction process preheater or
dissolver. The reduced ash loading would significantly reduce filtration
and/or catalyst life problems and reduce trace element emissions in the
subsequent gasification of the char or liquefaction residues.
ENHANCEMENT OF PYRITE REMOVAL
These studies have shown that chemical treatment specifically
directed at making the surface of pyrite hydrophylic have improved the
removal of pyrite liberated from the coal. Pyrite removal of 42 percent
has been reached and is equal to that obtained by float-sink separation at
a specific gravity.of 1.59. This is especially significant for the fine
size of the fresh coal evaluated (minus 48 mesh Tyler). The two chemical
treatments capable of producing this level of removal are (1) the treatment
with oxygen at room temperature of a slurry of coal containing 0.1 g of
N32COT/100 g of coal and (2) the treatment of a coal slurry with 0.5 g of
sodium metaphosphate/100 g of coal. In Treatment 1, it is believed that
the pyrite surface was oxidized to produce a hydrated ferric oxide layer
which renders the surface hydrophylic. When Treatment 2 is used, it is
believed that the pyrite surface adsorbed phosphate ion to make them
wettable. When the treatment with sodium metaphosphate was preceeded by
electrolysis and then followed by the addition of a flocculant, further
reductions in pyritic sulfur were realized but in general did not exceed
that attainable in laboratory float-sink separations.
ESTIMATION OF PRODUCT COST
The experimental data have shown that the oil agglomeration
technique can recover low ash coal from the coal preparation plant wastes.
Furthermore, the sulfur content of the agglomerated product can be
79
-------
significantly reduced through suitable treatments. For the purpose of
preliminary estimation of product cost, a conceptual design of an oil
agglomeration plant was made. Figure 11 shows a hypothetical flowsheet of
the oil agglomeration unit which can be either an add-on unit to an existing
coal cleaning plant for the purpose of recovering coal fines from the
thickener underflow or a portable plant for reclaiming coal values from the
coal waste .pond sediments.
The thickener underflow or waste pond sediment containing 35 wt.
percent solids is fed to the pretreatment tank where sodium carbonate and
air bubbles are introduced to oxidize the slurry. The pretreated slurry is
then sent to the conditioning tank where the slurry is diluted to 10 wt.
percent solid and sodium metaphosphate is added to provide good dispersion.
The conditioned slurry is pumped to the oil agglomeration tank
where oil is added and agglomeration is carried out with a high-shear
agitation. After agglomeration is completed, the slurry is transferred to
the flqcculation tank where flocculating agent is added to facilitate
settling of reject materials. The slurry is then sent to a vibrating screen
where agglomerated coal is separated from residue slurry. The recovered
coal is fed into a dewatering centrifuge to reduce the moisture content.
The residue slurry is pumped to a static thickener where reject materials'
are settled, and water and some oil are recovered for recycling.
The material balances shown in Figure 11 are chosen based on the
information obtained from this experimental work. From an economic point
of view, one of the most important variables in the process is the amount
of oil used. The experimental data have shown that a 5 percent oil concen-
tration will give adequate recovery of coal when the No. 6 fuel oil-kerosene
mixture.'was used (Figure 5). With these values, material costs for the
agglomeration process are estimated and summarized in Table 3.6.
The results show that the total material cost is $11.48/ton of
product, of,which about 97 percent represents the cost of oil. In addition,
cost of power requirement will range from 20 to 30c/ton of .product (@ 2.5c/
kWh). The total capital and fixed costs are approximately $2.00/ton of
(21)
product. With these economic estimates, the product cost will be about
$14/ton. Considering the current market price of coal at $20/ton, it
appears that the oil agglomeration process is economically viable in the
80
-------
Thickener Underflow
or
Waste Pond Sediment
35 TPHS
265 GPMW
Soda Ash
Air
70 Ib/hr
Pretreatment
Tank
Sodium
Metaphosphate
7 Ib/hr
Conditioning
Tank
Oil
505 gal/hr
Oil Agglomeration
Tank
Flocculant
0.35 Ib/hr
Flocculation
Tank
Recovery Screen
995 GPMW Water
Recycle
Water
Refuse
Coal
18 TPHS
72 GPMW
17 TPHS'
1188
GPMW
Dewatering
Centrifuge
Static Thickener
0.5 TPHS
60 GPMW
Product
17.5 TPHS
12 GPMW (15% Moisture)
FIGURE 11. FLOW DIAGRAM OF OIL AGGLOMERATION PROCESS
81
Refuse
17.5 TPHS
253 GPMW
-------
TABLE 36. MATERIAL COSTS FOR OIL AGGLOMERATION PROCESS
Item
Soda ash
Sodium metaphosphate
Flocculant
Kerosene
No. 6 fuel oil
Total
Feed Rate
70 Ib/hr
7 Ib/hr
. 0.35 Ib/hr
444 gal/hr
61 gal/hr
Unit Pric.e
$91/ton
$33/100 Ib
$1.77/lb
$0.40/gal
$0.28/gal
Costs
$/Tdn of Clean Coal
0.18 '
0.13
0.04
10.15
0.98
11.48
82
-------
case of coal recovery from wastes that have essentially no value.
As discussed above, the major cost determinant of the oil
agglomeration product is the cost of the oil used. The oil used, however,
will contribute to the overall heating value of. the product. The heating
value of agglomerated product using 5 percent oil can be estimated as:
Heating value of marketed coal - 11,000 Btu/lb
Heating value of oil - 140,000 Btu/gal
Heating value of agglomerated coal - 11,484 Btu/lb
Hence, the product cost based on .the heating value will be 61C/MM Btu
compared to the market price of coal at 91C/MM Btu.
In addition to the profitability, oil agglomeration provides a
number of additional possible advantages. In particular, the oil agglomer-
ation process will significantly reduce the waste volume which will not
only save handling and disposal costs but also reduce environmental
problems from its disposal.
83
-------
REFERENCES
1. Mezey, E. J., S. Singh and D. W. Hissong. Fuel Contaminants: Volume 2,
Removal Technology Evaluation, EPA 600/2-76-177b, U.S. Environmental
Protection Agency, Research Triagnle Park, N.C., 1976, 316 pp.
2. Mezey, E. J. and S. Singh. Fuel Contaminants: Volume 3, Removal
Feasibility Studies, Draft Report prepared by Battelle's Columbus
Laboratories for EPA, Industrial Environmental Research Laboratories
at Research Triangle Park, N.C.. , 1977.
3. Min, S. Physical Desulfurization of Iowa Coal, Energy and Mineral
Resources Inst. Publication IS-1CP-35, Iowa State University, Ames,
Iowa, March 1977.
4. Capes, E. C., A. E. Mcllkinney and A. F. Sirianni. Agglomeration from
Liquid Suspensions—Research and Applications. In: Agglomeration 77
(Proc. of the 2nd Int. Symp. on Agglomeration) Vol. 2, Am. Inst.
Min. Met. and Pet. Eng., New York, N.Y.,'l977.
5. " Bhattachayya, R. N., A. K. Moza and G. E. Sarkar. Role of Operating
Variables in Oil Agglomeration of Coal, Ibid.
6. Swanson, A. R.., C. Bensley and S. K. Nical. Some Fundamental Aspects
of the Selective Agglomeration of Fine Coal, Ibid.
7. Hoffman, L., J. B. Truett, and S. J. Aresco. An Interpretative
Compilation of EPA Studies Related to Coal Quality and Cleanability,
MITRE Technical Report, MTR-6648, March 1974, pp. 107-112.
8. Deurbrouck, A. W. Personal Communication, October' 1977.
9. Bogenschneider, B. and H. Kubitza. The Olifloc Process for the
Dewatering and ^Cleaning of Ultras-Fine Coal Slurries. Paper presented
a;t the 15th Biennial Conference of the Institute for Briquetting and
Agglomeration, August 22-25, 1977, Montreal, Quebec, Canada.
10. Fisher, R. W. and T. D. Wheelock. Advanced Development of Fine Coal
Desulfurization and Recovery Technology. Fossil Energy Division
Quarterly Techincal Progress Report, Oct. 1, 1976-Dec. 31, 1976.
Ames Laboratory, ERDA, Ames, Iowa.
11. Busch, R. A., R. -R. Backer and L. A. Atkins. Physical Property Data on
Coal Waste Embankment Material. U.S. Bur. of Mines- RI-7964, 1974.
84
-------
12. Busch, R. A., R. R. Backer, L. A. Atkins. Physical Property Data on
Fine Coal. Refuse. U.S. Bur. of Mines RI-8062, 1975.
13. Browning, J. S. Recovering Fine-Size Coal from Alabama Surface Mine
Washer Wastes Using the Humphreys Spriral. MRI Technical Series
IR No. 2, Mine Resources Institute and Experiment Station, University
of Alabama, 1977.
14. Anon. Underground Disposal of Coal Mine Wastes. Report to the
National Science Foundation by the Study Committee to Assess the
Feasibility of Returning Underground Coal Mine Wastes to Mined-Out
Areas, Environmental Studies Board, Board on Energy Studies, National
Academy of Science, Washington, D.C., 1975.
15. Capes, C. E. et al. Rejection of Trace Metals from Coal During
Beneficiation by Agglomeration. Env. Sci. and Tech. 8(1), 35, 1974.
16. Sun, S. C. and W. L. McMorris, Factors Affecting the Cleaning of Fine
Coals by the Convertol Process. Mining Eng. 11(11), 1151 (1959).
17. Capes, C. E., A. E. Mcllhinney, A. F. Sirianni, and I. E. Puddington.
Bacterial Oxidation in Upgrading Pyritic Coals. Canadian Mining and
Metallurgical (CIM) Bulletin 66, 88 (1973).
18. Chia, T. Y. Method of Desulfurizing Coal, U.S. Patent 3,988,120,
Oct. 26, 1976.
19. Zabic, J. E. Microbial Biogeochemistry, Academic Press, New York
(1969), p. 133.
20. Hucko, R. E. Beneficiation of Coal by Selective Flocculation - A
Laboratory Study. U.S. Bureau of Mines, RI-8234 (1977).
21. Capes, E. C., A. E. Smith and I. E. Puddington. Economic Assessment
of the Application of Oil Agglomeration to Coal Preparation.
Canadian Mining and Metallurgical (CIM) Bulletin 67, 115 (1974).
85
-------
TECHNICAL REPORT DATA
(Please read faarucrions on the reverse before completing]
1. REPORT NO.
EPA-600/7-79-025b
2.
3. RECIPIENT'S ACCESSION*-NO.
4. TITLE AND SUBTITLE
Fuel Contaminants: Volume 4. Application of Oil
Agglomeration to Coal Wastes
5. REPORT DATE
January 1979
6. PERFORMING ORGANIZATION CODE
7. AUTHOH(S)
E.J.Mezey, Seongwoo Min, and Dale Folsom
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Batte lie-Columbus Laboratories
505 King Avenue
Columbus , Ohio 43201
10. PROGRAM ELEMENT NO.
EHE623
11. CONTRACT/GRANT NO.
68-02-2112
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
Final: 8/77 - 4/78
14. SPONSORING AGENCY CODE
EPA/600/13
is. SUPPLEMENTARY NOTES ffiRL-RTP project officer is Lewis D. Tamny, Mail Drop 61, 919/
541-2709. EPA-600/2-76-177a and -177b are earlier reports in this series.
is. ABSTRACT
rep0rt gives results of a study of the application of oil agglomeration
to coal wastes. There are an estimated 3000-5000 sizeable active and abandoned
coal waste piles and impoundments in the eastern U.S. coal fields alone, containing
3 billion tons of refuse, part of which are slurry ponds. The impoundments, contain-
ing coal fines from coal preparation/cleaning plants, are a ready reserve of mined
fuel for use in times of shortages. It appears that oil agglomeration could contribute
significantly to the removal of contaminants before the conversion process is under-
taken. The ability of agglomeration to dewater finely ground wet coal also suggests
further incorporation of the process in any environmentally sound preparation plant
supplying these conversion plants. Early studies indicated that, although agglomer-""
ation can effectively remove much of the ash forming minerals . it was unable to
separate the liberated pyrite from coal. This program was undertaken to investigate
several approaches, identified during the first phase, to enhance pyrite removal du-
ring agglomeration and to demonstrate the utility of the technology to reduce the
environmental impact of increased quantities of coal cleaning refuse. Study results
show that the coal recovered is of better quality than the coal now being shipped from
the mine, in that sulfur and ash values are lower. Coal value recoveries were >90%.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Croup
Pollution
Coal
Wastes
Agglomeration
Dewatering
Coal Preparation
Liquefaction
Tailings
Slime
Pyrite
Desulfurization
Pollution Control.
Stationary Sources
Coal Waste Piles
Oil Agglomeration
13B 07D
21D,08G
08H
13H,07A
081
13. (DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
95
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
36
-------