REPORT
EPA-R2-73-262
May 1973
Evaluation of Dewatering of
Limestone Wet Scrubbing Process Sludges
Environmental Protection Agency
Research Triangle Park, N. C. 27711
COAL RESEARCH BUREAU
MINERAL INDUSTRIES BUILDING
WEST VIRGINIA UNIVERSITY
Morgantown, West Virginia
-------
EPA-R2-73-262
EVALUATION OF DEWATERING OF
LIMESTONE WET SCRUBBING
PROCESS SLUDGES
By
Coal Research Bureau
West Virginia University
Morgantown, West Virginia 26505
Contract No. EHSD 71-11
Program Element No. 1A2013
EPA Project Officer: L.H. Garcia
Control Systems Laboratory
• National Environmental Research Center
Research Triangle Park, North Carolina 27711
Prepared for
OFFICE OF RESEARCH AND MONITORING
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
May 1973
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Monitoring.
Environmental Protection Agency, have been grouped into five
series. These five broad categories were established to
facilitate further development and application of environmental
technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface
in related fields. The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL
PROTECTION TECHNOLOGY series. This series describes
research performed to develop and demonstrate instrumentation,
equipment, and methodology to repair or prevent environmental
degradation from point and non-point sources of pollution. This
work provides the new or improved technology required for the
control and treatment of pollution sources to meet environmental
quality standards.
EPA REVIEW NOTICE
This report has been reviewed by the Office of Research and
Monitoring, EPA, and approved for publication. Approval does
not signify that the contents necessarily reflect the views and
policies of the Environmental Protection Agency, nor does
mention of trade names or commercial products constitute
endorsement or recommendation for use.
-------
ABSTRACT
Several methods of dewatering solid materials were applied on a bench
scale to wet-collected, limestone-modified flyash from a coal-fired electric
power plant using limestone-injection and wet-scrubbing methods to control
the emission of gaseous sulfur oxides. Porous-bed sand filtration, lagoon-
ing and possibly pressure filtration appear to hold the most promise.
Aluminum extraction tests using sodium hydroxide, sodium carbonate and
combinations of the two yielded less than 50 percent of the aluminum avail-
able in the leach liquor while structural materials testing indicated that
there is insufficient free lime available in the modified ash to act as
a suitable binding agent.
ii
-------
CONTENTS
Abstract
List of Figures iv
List of Tables
Introduction ,
Dewatering Studies j
A. Literature Studies 2
B. Pressure Filtration 2
C. Coagulation g
D. Flocculation and Vacuum Filtration fi
E. Sand Filtration 14
F. Impoundment ^
G. Conclusions lg
Structural Products 18
Conclusions 23
Aluminum Extraction 23
A. Literature Studies 24
B. Experimental Methods and Results 25
C. Conclusions 01
Appendix A. Dewatering Methods 33
Introduction 34
Dewatering Methods 34
Dewatering Aids 38
Footnotes 42
Appendix B. Flocculant Filterability Testing 45
A. Buchner Funnel Test 45
B. Filter Break Test 47
C. Filter Leaf Test 47
Appendix C. Alumina Leaching 49
Sintering Processes 50
Nitric and Hydrochloric Acid Processes 52
Sulfuric Acid Processes 55
Sullurous Acid-Caustic Process 57
Ammonium Alum Process 53
Potassium Alum Process 59
References cn
bO
iii
-------
FIGURES
No. Page
1 The Effect of Pressure Level Upon the Dewatering
Capability of Sperry Pressure Filter Using KPL
Slurry ' 5
2 Sand Filtration Tests on KPL Slurry 16
3 Settling Test of KPL Slurry 17
4 Flow Sheet: Optimum Conditions for the Production
of Calcium-Silicate Brick From Wet-Collected
Modified Ash 21
5 Final Flow Chart for the Production of Structural
Products From 100% Wet-Collected Limestone-
Modified Flyash 22
6 The Effect of Autoclaving Upon Aluminum
Extraction Efficiency 26
7 The Effect of Leaching Time Upon Aluminum
Extraction Efficiency 27
8 The Effect of Sodium Ion Concentration Upon
- Aluminum Extraction Efficiency 29
9 The Effect of Slurry Concentration Upon
Aluminum Extraction Efficiency 30
iv
-------
TABLES
No. Page
1 Pressure Filt ration Characteristics of KPL Slurry 4
2 Variation of Z'eta Potential Values as a Function of
Inorganic Coagulant Dosage on Wet-Collected
Limestone-Modified Flyash from Kansas Power
and Light 8
3 Vacuum Filtration Test Results 9
4 Description of Filter Media Examined 10
5 Filter Leaf Test Results Using Concentrated
(10. 5 g/100 ml) Slurry and 15" Hg Vacuum,
Filtration Time 1-1/2 Minutes 11
6 Comparison of % Moisture in Raw, 835A-
Flocculated and Air-Dried Sludges 12
7 (Effect of Increasing Dip Time) 13
8 A Comparison of Filtrate Recovery and Cake %
Solids for Filtration of an Unflocculated and
Flocculated Slurry 14
9 (Results of High-Pressure Alumina Leaching) 20
-------
INTRODUCTION
The purpose of the subject contract has been to investigate the potential
utilization of wet-collected, limestone or dolomite-modified flyashes resulting
from sulfur dioxide control methods. Specific areas under investigation have
included dewatering, alumina leaching, and production of structural products by
employing potential cementitious properties of the modified flyashes. In
addition, literature surveys have been made to aid in the determination of
the state of the art and how it may be applied for each particular area.
The studies conducted under this contract (EHS-D-71-11) are complimentary
to three previous studies under Contracts CPA 70-66, PH 86-67-122 and
FH 22-68-18. All contracts have been concerned with various aspects of
characterization and utilization of wet- or dry-collected modified flyash.
However, owing to a scarcity of wet collected modified flyash samples, the
data reported in this study have been obtained using material generated
by the Combustion Engineering limestone-injection, wet-collection process
at the Lawrence Station of the Kansas Power and Light Co. (KPL).
DEWATERING STUDIES
Potential utilization of wet-collected modified flyash will be
dependent in part upon the successful dewatering of this material. Also,
disposal will be facilitated if the water can be removed from the ash
materials and returned to the power station for recycling.
-------
The modified flyash used in this program is a wet-collected limestone
material and contains approximately 1/2 to 2 percent solid materials, 85 percent
of which is smaller than 50 microns. With time, natural sedimentation processes
yield a sludge which may contain 50 to 60 percent solids. This figure is mis-
leading; however, as layering occurs and it appears that chemical reaction,
eg, hydration, etc may account for the removal of a portion of the interstitial
and intralayer waters. Disturbance of the sediments, such as would occur
during dredging operations, allows any water remaining in the pond area
to remix with the flyash and produce a slurry with 20-45 percent solids.
A. Literature Studies. In order to properly define specific areas to
be investigated in the potential dewatering of modified flyash slurries, a
literature study (see Appendix A) was undertaken. Based upon information
obtained in this study, the following dewatering processes and aids were
selected for bench-scale laboratory examination:
a) filtration
b) coagulation, and
c) flocculation.
Filtration was chosen as it is a commonly used dewatering method for
fine particles. The literature study indicated that vacuum and pressure
filtration methods were most feasible. Further, it was decided that dewatering
aids such as flocculants and coagulants should also be examined to provide
basic information for other dewatering processes. In order to differentiate
between the two sometimes interchangeably used terms, coagulants were
considered as salts which alter the surface charge of individual colloidal size
particles to lower their repulsive effects while flocculants were considered
to be long-chain polymers which actually form "bridges" between particles.
In either case, a grouping together of particles to aid in settling or
-------
filtration was the goal.
B. Pressure Filtration was selected for laboratory study because preliminary
work had indicated that modified flyash slurries might prove amenable to such
a dewatering method. A 3 5/8 inch square laboratory scale pressure filter
was obtained through D. R. Sperry and Company* and tests were performed to
determine the effects of such parameters as slurry concentration, gas
pressure, and cycle duration between filter cake removal upon the solids
content of the final filter cake.
Initial tests indicated that varying the solids content of the slurry
had little effect upon the final filter cake as long as the slurry content
was below 10-15 percent solids. Accordingly, the slurry concentration was
maintained between 2 1/2 and 3 percent solids as this was estimated to be
the highest obtainable during normal operation of the scrubbing system at
Kansas Power and Light Co. The filtration pressure levels examined were
60, 80 and 100 pslg held for two-hour intervals for each test run. At the
end of each test, the filter cake was examined and the volume of filtrate
recovered was measured. The two-hour test duration period was selected as
it represents a typical operating cycle between filter change and service for
a commercial pressure filter operation. Compressed air was used to provide
the necessary pressure in all of the tests.
Because the capacity of the laboratory apparatus was insufficient to
allow a two-hour test run, it had to be opened and refilled several times
during each run. Extreme care was taken to ensure that the existing cake
was not disturbed or allowed to dry as cracking of the filter cake would
effectively terminate the test run. The volume of filtrate recovered was plotted
*The use of commercial firm or trade names does not imply endorsement
by the Coal Research Bureau, but is simply intended for clarification
or description of the equipment used.
, 3
-------
as a function of time for each test performed. Two representative curves are
included as Figure 1. The even filtering rate with respect to time as shown
by the curves indicate that the filter was not blinding* and also that there
was very little difference when the pressure was increased from SO to 100
psig. Representative data from the tests performed is included below:
TABLE 1
PRESSURE FILTRATION CHARACTERISTICS OF KFL SLURRY
Pressure
(psig)
60
80
100
80
100
100
Percent
Solids
(Initial)
2.60
2.60
2.50
2.75
2.75
2.55
Slurry
Volume
(Gallons)
1.32
1.39
1.24
1.26
1.29
1.88
Filtrate
Volume
(Gallons)
1.18
1.26
1.10
1.13
1.15
1.73
Filtrate
Slurry
Ratio
0.89
0.90
0.88
0.89
0.89
0.92
Filter
Cake
% Solids
34.5
33.0
34.0
35.0
36.0
39.0
Hours
2
2
2
2
2
4
The results of the pressure filtration tests indicate that modified
flyash slurries could be dewatered to produce a filter cake containing 35 to
40 percent solids. The water volume recovered by such a method is approximately
90 percent of the original slurry volume. It would be satisfactory for
recycling as it is very clear and the addition of reagents such as flocculants
should not be necessary for further clarification. Other tests performed
indicate that blinding of the filter cake does not occur within 10 hours of
operation and that the length of the filtering cycle would be determined by
the capacity of the filter apparatus. Economic data was not obtained as it
would be dependent upon the particular power plant scrubbing slurry concentration
and type of alkaline material used for injection. It also appears probable that
the underflow from a classifier-thickener would be amenable for further dewatering
by this method. However, additional research is needed in this area.
*Blinding refers to the physical blocking of pore spaces in the filter
media by small, solid particles of the material being filtered.
-------
Figure I
THE EFFECT OF PRESSURE LEVEL UPON THE DEWATERING
CAPABILITY OF SPERRY PRESSURE FILTER USING KPL SLURRY
C/J
5000
CO
Z 4000
o
111
oc
III
o
oc 3000
hi
2000
<
P
1000
100 PSIG (Initial Solids 2.75g./IOO ml)
-80 PSIG (Initial Solids 2.75g/IOOml.)
I.O 2.0
TIME-HOURS
3.0
-------
C. Coagulation was examined as a means of increasing the settling rates
of the smaller modified ash particles. Although the majority of the ash
would settle within 5 to 10 minutes, it was noted that the finer particles
remained in suspension, leaving the water cloudy or turbid, for several
hours. Zeta potential tests performed upon the modified flyashes
indicated that they were already within the range, + 15 millivolts, in which
coagulation should occur. This was verified in the laboratory by adding -
various concentrations of trivalent, divalent and monovalent inorganic
coagulants to the slurry. The resultant data, presented as Table 2, indicated
that there were no significant changes affected in the zeta potential of the
modified ash particles due to the presence of the coagulants. It was there-
fore concluded that coagulation of the fine ash particles did not appear
promising and the study was terminated.
D. Flocculation and Vacuum Filtration were investigated as a combination
method for dewatering modified fly ash slurries. Rather than acting primarily
to change particle surface charges, flocculants act as physical bridges to
bind particles together and their effectiveness is not dependent upon the
surface charge of the particles as are coagulants. Flocculants, generally
long-chain polymers, are commonly used in conjunction with mechanical dewatering
processes and a general testing method has been devised to evaluate the
effectiveness of flocculants when used with vacuum filtration. This commonly
used test series includes the Buchner funnel test, the vacuum break test and
the filter leaf test. The detailed procedures involved in these tests are
included in Appendix B.
Buchner Funnel Test
The Buchner funnel test is designed to aid in the selection of a particular
flocculant and concentration to provide optimum dewatering of a slurry with the
-------
aid of vacuum filtration. In the subject testing program, a total of thirty-
i
five series of tests were performed with a variety of anionic, non-ionic,
and cationic polymers. The average optimum dosage, ie, the dosage
which gave the largest volume of filtrate for all thirty-five flocculants, was
3.19 ppm and the average percentage recovery of filtrate was 30.4.
The flocculants studied, their optimum concentrations, and the percentage
recovery are presented in Table 3. The primary conclusion to be drawn from
the presented data is that the better flocculants were strongly anionic or
non-ionic.
Filter Break Test
The filter break test is an extension and refinement of the Buchner
funnel test. A flocculated sample of sludge is filtered until it reaches
dryness. Air then enters the system through drying cracks or breaks and the
vacuum decreases. The'time required until the vacuum drops is recorded.
This test, however, did not prove to be very precise. The time of break could
not in general be determined with high accuracy because the vacuum did not drop
noticeably; or, in some cases, did not drop at all. The failure of these tests
can probably be attributed to a breaking down of the floes during filtration
and a subsequent "blinding" of the filter media by the smaller (1.0 micron or
less) ash particles. An approximate break time was generally measurable and
it appeared that the break time for the flocculants examined was between one
and three minutes.
i
The filter break tests were discontinued as the results obtained were
not considered precise enough to be meaningful.
Filter Leaf Test
The filter leaf test is designed to reproduce, on a lab-bench scale,
the operating cycle of a commercial vacuum filter. Using a series of commercial
-------
TABLE 2
VARIATION OF ZETA POTENTIAL VALUES AS A FUNCTION OF INORGANIC
COAGULANT DOSAGE ON WET COLLECTED LIMESTONE MODIFIED FLYASH
FROM KANSAS POWER AND LIGHT
Collected October, 1970
Additive
Natural
A12(S04)3
BaCl2
KC1
Concentration
0
1 ppm
5 ppm
10 ppm
25 ppm
100 ppm
1 ppm
5 ppm
25 ppm
100 ppm
500 ppm
Zeta
Potential
+3
+3
+3
+3
+2
+2
+3
+3
+3
+•2
+5
Specific
Cond.
2900
2600
2500
2500
2500
2500
2700
2600
2600
2400
3600
EH
11.3
11.2
10.8
10.7
10.7
10.7
11.0
10.9
10.9
10.9
10.7
-------
Flocculant
Magnifloc 571C
Magnifloc 836A
Calgon 2425
Magnifloc 835A
Magnifloc 560C
Magnifloc 905N
Magnifloc 837A
Separan AP30
Magnifloc 570C
Separan AP 273
Magnifloc 521C
Hereofloc 827
Nalcolyte 675
Hereofloc 831
Percol 292
Percol 156
Calgon 240
Nalcolyte 675
Nalcolyte 670
Reagent S-3595
Drewfloc 230
Hercofloc 827
Hercofloc 818
Hercofloc 815
Hercofloc 831
Hercofloc 821
Reagent S-3595
Superfloe 127
Superfloc 16
Superfloe 20
Superfloc' 84
Aerofloc 550
Superfloc 202
Hercofloc 818
Zeta Floe S
Average % Recovery = 30.4%
Average Optimum Dosage = 3.19 ppm
Flocculant Type
Anionic
Cationic
Nonionic
(1) Optimum Concentration
Average
Opt. Cone.
PPM
4.0
1.9
2.4
TABLE 3
VACUUM FILTRATION TEST RESULTS
Type
Cationic
Anionic
Anionic
Anionic
Cationic
Nonionic
Anionic
Anionic
Cationic
Anionic
Cationic
Nonionic
Anionic
Anionic
Cationic
Anionic
Anionic
Anionic
Nonionic
Anionic
Anionic
Nonionic
Anionic
Cationic
Anionic
Anionic
Anionic
Nonionic
Nonionic
Nonionic
Nonionic
Anionic
Anionic
Anionic
Cationic
% Recovery
(2)
0.6
3.8
5.8
5.8
2.9
0.19
2.9
2.9
0.1
4.8
1.0
0.19
3.8
1.9
2.9
5.8
2.9
4.8
1.9
4.8
4.8
2.9
5.8
3.8
1.9
1.9
1.9
5
4
3
2
4.8
5
2
2
10%
55%
55%
72%
27%
3%
26%
48%
18%
32%
22%
40%
39%
15%
23%
30%
30%
43%
23%
62%
21%
60%
23%
18%
33%
19%
39%
21%
16%
4%
42%
47%
18%
23%
7.1%
Average 3
Recovery
36.5
17.9
26.1
(2) The Percentage Increase of Water Filtered from the Flocculated
Slurry in 0.5 Minutes
9
-------
TABLE 4
DESCRIPTION OF FILTER MEDIA EXAMINED
O
Eitnco
Media
Number
NY-529F
NY-527F
SFYC-175-211B
NY-420
NY-319F
DY-453
NY-306F
NY-432
NY-384
OR-593F
NY-372
POPR-858F
NY- 301
CO- 12
PO-801-RF
Weave
2/2 Twill
2/2 Twill
Non-Woven
1/1 Plain
3/1 Twill
2/2 Twill
1/1 Plain
1/1 Plain
1/1 Plain
3/1 Twill
7/1 Satin
2/2 Twill
2/2 Twill
3/1 Twill
1/1 Plain
Thread
Count
144 x 54
157 x 62
23 mil
35 x 35
240 x 120
70 x 38
52 x 39
60 x 34
79 x 40
48 x 46
60 x 40
70 x 32
54 x 37
62 x 36
105 x 40
Finish
Heat-Set
Col lander ed
Stabilized
Heat-Set
Grey
Heat-Set
Heat-Set
Heat-Set
Heat-Set
Spun
Heat-Set
Greige
Grey
Napped 1/5
Grey
Weight
oz/yd2
8.00
5.75
4.14
5.20
4.50
14.75
5.5
10.00
9.00
—
12.00
12.80
13.30
10.00
9.00
Air Flow
cfm/ft2
39.65
176.00
44
4-5
6.97
35/40
2.00
6.00
73.00
1400
53.37
100/150
Yarn
Type
Multifilament
Mono filament
Cotton
Multifilament
Multifilament
Spun Staple
Multifilament
Multifilament
Multifilament
Spun Staple
Filament Warp
Spun Staple
Multifilament
Spun Staple
Mono filament
-------
TABLE 5
FILTER LEAF TEST RESULTS USING CONCENTRATED (10.5 g/100 ml) SLURRY AND
15" Hg VACUUM FILTRATION TIME 11/2 MINUTES
Eimco
Media
Number
NY-529F
NY-527F
SFYC-175-211B
NY-420
NY-319F
DY-453
NY-306F
NY-432
NY- 384
OR-593F
NY-372
POPR-858F
NY- 301
CO-12
PO-801-RF
Approximate
Cake
Thickness
in
Inches
0.200
0.160
0.200
0.120
0.200
0.200
0.240
0.240
0.28
0.200
0.240
0.240
0.200
0.08
Percent
Solids
39
36
40
—
37
39
40
40
41
39
38
35
36
37
34
Comments
350 ml cloudy filtrate
430 ml cloudy filtrate
360 ml slightly cloudy filtrate
entire sample passed through media
200 ml clear filtrate
230 ml slightly cloudy filtrate
350 ml milky filtrate
350 ml with large amount of solids
310 ml clear filtrate
400 ml clear filtrate
320 ml slightly cloudy filtrate
460 ml clear filtrate
430 ml milky filtrate
390 ml very clear filtrate
200 ml clear filtrate
-------
2
filter media and a filter leaf apparatus of 0.1 ft area obtained from
Eimco Corporation of Salt Lake City, Utah; data was obtained concerning the
percent solids in the filter cake, cake-thickness and the volume of filtrate
recovered from a slurry containing 10.5 percent solids as recommended by the
apparatus manufacturer.
The slurry formed medium sized floes which disintegrated into fine
particles. The cake formed tended to be compact and did not exhibit the
degree of cracking which might be expected. Rather, one or two major cracks
developed and exhibited slow growth with little branching. A brief description
of the various filter media examined is included as Table 4 and the resulting
data from these tests are presented in Table 5. This data indicated that the
percent moisture in the filter cake is fairly constant regardless of the
filter cloth used. The major effects which the different cloths had lay in the
areas of filtrate clarity and ease of removal of the filter cake. Also, as
Indicated in Table 6, a sample of raw sludge subjected to the filter leaf test
gave a percentage of water nearly the same as that of sludge treated with
flocculant. Another sample of flocculated sludge air-dried for an additional
10 minutes underwent no further appreciable loss of moisture indicating that
further dewatering would be difficult.
TABLE 6
COMPARISON OF % MOISTURE IN RAW,
835A-FLOCCULATED AND AIR-DRIED
SLUDGES
Sample Vacuum Cake Thickness % Solids
In. Hg (inches)
Average for all cloths 15 0.200 38
tested on flocc. slurry
Raw sludge 15 0.200 37
Air-dried (10 min) 15 0.240 38
-------
In order to further define the variables involved in vacuum filtration of
modified flyash, a series of tests designed to simulate a commercial vacuum
filter were performed. The portions of a normal vacuum filter operating cycle
investigated included the dip or filter submergence time, the drying time and
the cake removal time. The tests consisted of submerging the filter leaf
apparatus (see Appendix A) into a 1.3 percent solids modified flyash slurry
and determining the volume of filtrate recovered as a function of submergence
time. The thickness and percent solids of the filter cake were also monitored.
These tests were performed using spun staple and multifilament filter media
CO-3 and NY-301 and a vacuum of 28 to 29 inches of mercury. It was found that
the filter cake had to accumulate to a thickness of 0.125 inches during the
submerged period to permit effective cutting and removal of the dewatered
cake. This minimum cake thickness corresponded to a minimum submergence period
of four minutes. Included as Table 7 is data which indicates that increasing the
submergence period or dip time did not have a significant effect on either the
cake thickness or the percent solids of the cake. However, the filtrate
volume or the amount of water recovered did increase with time.
TABLE 7
Cake
Thick-
Cloth
CO- 3
CO-3
CO-3
CO-3
CO-3
NY-301
NY-301
NY-301
NY-301
NY-301
Dip
Time
4 min
4 min
6 min
8 min
10 min
4 min
6 min
8 min
4 min
4 min
Filtrate
Vol.
2310
2150
2840
3140
4170
2440
2740
3240
1920
1990
ness
inches
0.18
0.12
0.12
0.12
0.19
0.12
0.19
0.12
0.12
0.19
Drying
Time
10 sec
60 sec
60 sec
60 sec
60 sec
15 sec
15 sec
15 sec
15 sec
120 sec
%
Solids
30.7
30.3
31.2
30.4
30.5
30.4
29.4
30.2
30.9
-------
In an effort to further examine the effects of reagents upon dewatering,
a flocculated and unflocculated sample were examined together. The resulting
data are shown in Table 8. It can be seen that the filtrate volume which could
be recovered from the flocculated slurry was approximately 50 percent greater
than from the raw slurry while the percent solids in the filter cake dropped
slightly. Both factors are due to the bridging nature of the flocculants. In
the slurry the floes are light and fluffy allowing the passage of water; while
upon drying, the floes tend to trap and hold water.
TABLE 8
A COMPARISON OF FILTRATE RECOVERY AND CAKE
% SOLIDS FOR FILTRATION OF AN UNFLOCCULATED AND
FLOCCULATED SLURRY
Filter Slurry Filtrate %
Media Type Vol. Thickness Solids
NY-301 Unflocculated slurry 2235 ml 0.12 inches 31.3
CO-12 Unflocculated slurry 1850 0.19 inches 31.1
NY-301 Flocculated with 835 A 3235 0.16 inches 30.7
CO-12 Flocculated with 835 A 2770 0.22 inches 30.5
These tests indicated that vacuum filtration and possibly flocculation
methods are usable with modified flyash slurries. However, due to the large
volumes of water passing through the scrubbing system, a preliminary thickening
or concentration of the slurry should be effected.
E. Sand Filtration or porous bed filtration was examined as a means of
dewatering large volumes of slurry which might exceed the capabilities of
conventional equipment. In the laboratory an 11 inch diameter, packed and
graded* filter bed of sand and gravel was prepared and approximately 30 liters
*Graded is a geological term meaning vertically sorted particles increasing
in size from top to bottom.
-------
of a one percent flyash slurry was introduced on top of the bed. Over 95 percent
of the slurry water passed through the filter bed in a short period of time and
the cake was allowed to air dry in place. The solids content of the cake with
respect to drying duration is shown in Figure 2. The samples taken for percent
solids determinations were removed from different locations and care was taken
to disturb the drying cake as little as possible. Drying cracks appeared on
the 8th day; and by the 20th day, the cake contained 60 percent solids and
appeared to have sufficient strength to allow removal on a large scale by
common earth moving equipment. In actual practice, this time period would be
affected by such positive factors as sunshine and wind and such negative factors
as rain and snow. Analyses of the waters were not performed as the quality of
the filtrate water would be dependent upon specific slurry and filter bed
compositions.
F. Impoundment. Lagoons or settling ponds are likely to be the most
commonly used dewatering method for modified flyash. Draining of the lagoon
should allow the production of an ash-water slurry containing 50 percent by
weight solids. Excess calcium ions in the slurry water react with carbon
dioxide in the air and the slurry pH gradually drops from 11.0 to approximately
8.0. The resultant calcium carbonate precipitates out of the water and is
collected with the flyash. The settling rate, shown in Figure 3, for KPL slurry
was monitored and it was observed that a major portion of the settling occurs
within the first 15 minutes of impoundment.
If insufficient land is available for a series of settling lagoons, use
of a clarifier-thickner is recommended. Conversations held with engineers of
Combustion Engineering, Inc. indicated that slurries with solids contents up
to 30 percent by weight could be obtained from the clarifier discharge.
-------
70
60
Figure 2
SAND FILTRATION TESTS ON KPL SLURRY
50
r
40
u
20
10
8 10
DRYING TIME-DAYS
12
14
16
18
20
-------
Figure 3
SETTLING TEST OF KPL SLURRY
25 Grams Solid/100 Mis. of Slurry
1000
800
v>
a
£j 600
H
UJ
(/)
Ik
O
iu 400
200
10
20
30 40
TIME-MINUTES
-------
G. Conclusions. A series of graded filter beds prepared from gravel and
sand appears to be the most efficient and practical dewatering method for use
with wet-collected, limestone-modified flyash. If water-soil interaction
is no problem, then a normal settling lagoon such as is currently being used
at Kansas Power and Light will provide partial dewatering. Care must be
taken, however, to remove all of the water present in the lagoon before
cleaning operations begin as remixing with the solids will occur otherwise.
Strongly anionic or non-ionic flocculants were effective and could
prove useful if a final water clarification step is desired.
STRUCTURAL PRODUCTS
Potential utilization methods involving wet-collected modified flyash
must include some means of either binding the sulfur components in a non-
reactive form or recapturing them if an evolution step is involved. Investigations
into the pozzolanic activity of wet-collected, limestone-modified flyash indicated
that some cementitious activity did occur during such normal sedimentation
processes as would exist in a slurry settling pond. A close study of the settled
materials indicated that the binding takes place within relatively thin layers
of the partially compacted sediments. The binding layers consisted of flyash,
growing calcium sulfate crystals and sub-crystalline calcium carbonate.
Calcium silicates may have been present as a coating on some flyash particles,
but absolute identification by X-ray methods was not possible. Although the
binding strength given by the formation of these minerals is quite low, it has
proven sufficient in several cases to plug process lines when the power plant
slurry flow was stopped for a short time period. Tests performed using 100
percent modified flyash as a concrete mix showed that compressive strengths in
the range of 30 to 40 pounds per square inch (psi) occurred. The technical
literature was examined to determine whether processes existed which could
-------
enhance the slight natural binding properties of the modified flyash. Because
the ash is similar, at least superficially, to the constituents used in the
production of autoclaved "sand-lime" (eg, calcium-silicate) structural materials
in Europe, it was decided to investigate this potential utilization area. The
two major advantages to be found in autoclave processing of modified flyash
are:
1. Sulfur gases do not have to be captured and marketed.
All of the processing occurs below sulfur dioxide
regeneration temperatures; and
2. No pollution problems due to the presence of soluble sulfur
compounds have been noted. Die sulfur components in auto-
claved modified flyash appear to be either chemically
combined or bound within the matrix structure of the
final structural products.
Prior research, reported in detail in the final report for Contract
CPA 70-66 ("Pilot Scale Up of Processes to Demonstrate Utilization of Pulverized
Coal Flyash Modified by the Addition of Limestone-Dolomite Sulfur Dioxide
Removal Additives") , in the area of utilizing wet-collected modified flyash
showed that a very satisfactory calcium-silicate type brick or block could
be produced. The raw materials used in the process, shown schematically in
Figure 4, consisted, o" • dry basis, of 50 percent wet-collected modified
flyash (at 34 percent moisture), 11 percent calcium oxide and 39 percent
sand. The resultant brick exceeded 4500 psi compressive strength and had an
average cold water absorption value of 23 percent with no measurable shrinkage
or expansion occurring during autoclaving. 'These bricks meet ASTM specifications
for Grade SW calcium silicate face brick as per specification C73-67.
Under the subject contract, structural shapes were produced using, on
a dry basis, 100 percent wet-collected, limestone-modified ash. A flowsheet
of the process, Figure 5, which yielded the maximum compressive strength
is included. Samples produced using this method ranged in compressive
strength between 875 psi and 960 psi. The blocks had a green strength
before curing of approximately 400 psi and could withstand normal handling
-------
TABLE 9
Al
Slurry Extraction
Na Ion Leach Time Liquor Concentration Efficiency
Autoclave g/liter hours Composition g/1 %
20.2
22.4
26.9
30.3
31.4
26.9
30.3
36.0
(1)
12.7
43.8
13.4
37.0
3.8
44.9
10.8
17.5
33.4
18.5
47.1
12.5
32.3
11.0
36.1
37.0
43.7
37.fi
37.9
37.0
32.5
30.3
26.4
27.5
25.2
29.5
9.0
3.9
8.87
17.1
0.02
1.95
4.97
4.20
1.14
0.54
3.82
8.42
3.91
3.19
3.1
1.78
(1) Sample contaminated - test not rerun
£3
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
No
Yes
No
Yes
No
Yes
No
Yes
No
No
No
No
Yes
Yes
Yes
Yes
Yes
No
Yes
No
Yes
No
Yes
No
Yes
Yes
Yes
Yes
Yes
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes <•
Yes
Yes
Yes
Yes
16
16
41
41
16
16
41
41
35.7
35.7
35.7
35.7
35.7
35.7
35.7
35.7
35.7
35.7
35.7
35.7
39.7
39.7
39.7
39.7
35.7
38.4
38.4
41
41
41
85.9
41
41
41
41
41
41
41
41
H20
41
41
85.9
85.9
75.4
112
41
41
85.9
80.6
58.4
1/2
1/2
1/2
1/2
2
2
2
2
1/2
1/2
1
1
1/2
1/2
1
1
1/2
1/2
1
1
1/2
1/2
1
1
2
2
2
4
2
2
2
2
2
1
2
1
1/2
1/2
24
48
48
12
24
48
24
24
24
48
24
24
24
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH, Na2C03
NaOH, N32C03
NaOH, Na2C03
NaOH, Ns2C03
NaOH, Na2C03
NaOH, N32C03
NsOH, Na2C03
NaOH, Na2C03
NsOH, N32C03
NaOH, Na2C03
NaOH, Na2C03
NaOH, Na2C03
9NaOH, !Na2C03
9NaOH, !Na2Cn3
9NaOH, 1N32C03
9NaOH, !Na2C03
NaOH, N32C03
3NaOH, !Na2C03
3NaOH, 1N32C03
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH, Na2C03
4i
-------
Figure•4
Silica Sand
(30x100 Mesh)
Wet-Collected Limestone
Modified Flyash (1% Slurry)
Dewater Ash
to 34% H20
50% Flyash + 11% CaO +
39% Silica Sand
Muller-type Mixer (10 Min.)
@ 20.4% H0
Slaking Reactor
(1 Hour)
Pressing Load
3750 psi, 17.7%
H20
47 (lour Humidity Storage
95% Rel. Humidity at
Room Temperature
Autoclave, 8 Hours
190 psi, 185°C
72 Hour Air
Drying
Lime (CaO)
96% Pure
Flow Sheet: Optimum Conditions for the Production of Calcium-Silicate
Brick from Wet-Collected Modified Ash.
-------
Figure 5
RAW K.P.L SLUDGE
44% H20 - 5.8% CaO
Air
Drying
17.7% H20- 2.l%CaO^*
PADDLE MIXER
Dry
3000
Press
PSI
GREEN BRICK STRENGTH
380 PSI
Humidity
24 HRS-
Cure
95% Pel/Hum
AUTOCLAVE
195°C -190 PSI - 16 MRS
Air
5
Drying
Days
CURED BRICK STRENGTH
.1
925 PSI- 67.4 LBS/FT3- O.I4%CaO
$ "Fr0e"or Reactive Lime Reported as CaO
FINAL FLOW CHART FOR THE PRODUCTION OF STRUCTURAL
PRODUCTS FROM 100% WET-COLLECTED LIMESTONE MODIFIED FLYASH
-------
procedures. The low final strength, approximately 3600 psl lower than those
required by ASTM specifications, prompted a further study to determine whether
all of the calcium constituents were being utilized during autoclaving. The
study showed that the modified ash examined had 5.8 percent "free-lime" or
reactive lime present before processing and that only 0.14 percent remained
after curing. This data indicates that almost all of the "free-lime" originally
available was chemically combined during processing and that very little
increase in corepressive strength could be obtained.
Conclusions. Although very satisfactory calcium-silicate products can
be produced using modified flyash, sand and lime, the use of 100 percent
wet-collected modified ash does not appear feasible. The maximum compressive
strength obtained was less than 1000 psi although "free-lime" tests indicated
that almost all of the reactive calcium was chemically combined during process-
ing.
ALUMINUM EXTRACTION
A bench-scale extraction program designed to investigate the potential
recovery of aluminum values present in KFL wet-collected, limestone-modified
flyash was performed. The basic approach taken considered the modified ash
as being similar in nature to the materials produced using conventional lime-
sintering processes. The basis for such an approach lies in the direct
similarities between the mode of formation and the initial compounds present
in lime-sintering processes and modified ash. Lime-sintering involves the
mixing of alumino-silicates, eg, clay or flyash, with lime and then roasting
to form calcium aluminates and insoluble dicalcium silicates. It was hoped
that the presence of excess lime during the coal combustion process would
allow the formation of sinter products. Extraction performed using
sodium hydroxide, sodium carbonate, or a mixture of these as solvents for the
-------
calcium alumlnates showed that less than one half of the aluminum present in
the modified ash could be extracted. It was assumed that the sintering
process did not go to completion due to the relatively short contact time
between the lime and the coal ash materials in the boiler. Potential
sintering time in the boiler flame envelope would generally be measurable in
seconds or fractions of seconds compared to a normal sintering time of one
half hour or more. In addition, a portion of the calcium present is used up in
sulfur oxide capture. A further sintering step might provide the desirable
calcium aluminates; however, such a process would re-release the sulfur
components in gaseous form.
Autoclaving was examined as a means of further solubilizing the
aluminum containing constituents of the modified ash. The variables which
were monitored included: solvent type, solvent concentration, contact time,
solid-to-liquid ratio, and such autoclaving factors as temperature, pressure,
and duration.
A. Literature Studies. Currently, less than ten percent of the aluminum
produced in the U. S. is derived from domestic ores and the increasing use and
strategic importance of this metal has led to a re-examination of locally
available materials such as kaolin, other aluminum bearing clays, ferrugineous
bauxite and anorthosite as potential ores for the near future. In an effort
to determine the types of reactions involved in .the processing of such ores, a
partial technical discussion and literature survey was prepared (see Appendix C).
In addition, discussions were held with personnel of the College Park, Maryland,
Station of the U. S. Bureau of Mines who are currently investigating the
recovery of aluminum from unmodified coal flyash.
-------
B. Experimental Methods and Results. Extraction tests were performed using
sodium hydroxide (NaOH), sodium carbonate (Na2C03>, and combinations of the two in
varying leach liquor concentrations, slurry concentrations and contact durations.
The criterion chosen as a basis for comparison between tests was the extraction
efficiency (amount of aluminum in leach liquor/total amount of aluminum
available in the sample). In order to obtain the maximum information
possible from a limited number of experiments, all test series were system-
atically planned following established factorial design methods. All
samples were leached in glass beakers at 95 to 100°C and agitation was
supplied by Teflon coated magnetic stirring bars.
1. Autoclaving - The conditions within the autoclave for the
majority of the tests were 190 + 5 psig of steam pressure at
188 + 3°C for 24 hours. This temperature and resultant
pressure were selected as they are easily reached and
monitored by most commercial autoclaves. Figure 6 illus-
trates graphically the effect of autoclaving upon aluminum
extraction efficiency using paired tests in which auto-
claving is the only variable within each pair. A listing
of the conditions for each numbered test may be found in
Table 9. As can be seen from the figure, autoclaving
did not significantly increase the extraction efficiency
in any test pair and in four of the test pairs significantly
depressed the efficiency. These four test pairs were all
leached with a combination of NaOH and Na2C03 . The auto-
claving time of 24 hours was selected for the tests because
structural material testing had indicated that almost all
of the autoclaving reactions went to completion within 12 to
18 hours. It is possible that higher steam temperature and
pressure combinations might improve the extraction
efficiency, but the costs in equipment and energy require-
ments of such a program would probably preclude any industrial
acceptance.
2. Leach Contact Time - The effect of increasing leach solution-
modified ash contact time upon the extraction of aluminum
is shown graphically in Figure 7. As the contact time
Increased from 1/2 to four hours, the extraction
efficiency generally increased. As can be noted from
the figure, the amount of increase for a set of specific
test conditions would not be predictable for a given
time increase; and specific process tests would have to
be performed before a relative increase could be calculated.
No tests were performed to accurately determine an
optimum contact time for each set of test conditions;
however, it was noted that the percentage of available
-------
50
TEST
Figurt 6
THE EFFECTOR AUTOCLAVING UPON ALUMINUM
EXTRACTION EFFECIENCY
PAIRED TESTS DIFFER ONLY IN AUTOCLAVING AND THE
DIFFERENCE BETWEEN SETS OF TESTS IS SHOWN IN TABLE 9
D-
HOT AUTOCLAVED
I
- AUTOCLAVED
-------
50
o
>
o
X
0>
> 30
UJ
o
UJ
I 20
O
tr
»-
uj 10
TEST NO-
Figure 7
THE EFFECT OF LEACHING TIME UPON ALUMINUM
EXTRACTION EFFECIENCY
PAIRED TESTS DIFFER ONLY IN LEACHING TIME AND THE
DIFFERENCE BETWEEN SETS OF TESTS IS SHOWN IN TABLE 9.
;_
y
\t
£•
^
\
/
1
•9
(/t
—
\
/
^^H
4 HOURS
i
0
CVJ
58 56 47
QC
O
00
I
48
1
D
1/2 HOUR
I
I HOUR
2 HOURS
-------
aluminum extracted lowered as the contact time increased
past 24 hours. Such an effect could have been caused by the
formation and precipitation of hydrated calcium aluminates
from the leach liquor.
3. Sodium Ion Concentration - The aluminum extraction efficiency
was also related to the sodium concentration (Na g/liter)
available to the leach liquor. As shown in Figure 8, an
increase in Na concentration from 16 to 41 g/liter led
to a 20 to 30 percent increase in extraction efficiency.
However, a further increase in Na concentration to 85.9
g/liter led to decreased efficiency. Further optimization
tests were not performed.
4. Slurry Concentration - Slurry concentration, expressed as
grams of flyash per liter of leach liquor, had the most
definite effect on extraction of the variables monitored.
As shown in Figure 9, a decrease in the solids content of
the leach slurry led to an increase in the percentage
of aluminum extracted. Such a relationship is to be
expected as extraction is partially dependent upon the
number of solvent ions which contact the aluminum present
and also upon the availability of water for solution.
No tests were performed to determine an optimum slurry
concentration due to the large number of other related
variables.
5. Physical Factors - The most important physical characteristic
of limestone modified flyash, as related to extraction of aluminum,
is its particle size. By weight, greater than 95 percent
will pass through a 74 micron sieve and approximately 85
percent will pass through a 38 micron sieve. An advantage
of the extremely fine particle size lies in the resultant
extremely large surface area per unit weight of the flyash
which allows a large contact area between leach liquor and
particles. However, this large surface area also makes release
and recovery of the leach liquor after extraction very difficult
as the leach liquor tends to adhere to the particles and is very
difficult to filter or wash off the very fine material.
The glassy nature of the flyash is also an important factor in that
the aluminum present is probably derived from coal-associated shales and
clays. Fusing of these silicates during combustion probably further
incorporates a large amount of the aluminum present within the silicate
structure. Therefore, complete recovery of the metal would require a complete
dissolution of the glassy siliceous matrix.
-------
Figure 8
THE EFFECT OF SODIUM ION CONCENTRATION UPON ALUMINUM
EXTRACTION EFFECIENCY
PAIRED TESTS DIFFER ONLY IN No* ION CONCENTRATION AND
THE DIFFERENCE BETWEEN SETS OF TESTS IS SHOWN IN TABLE 9.
TEST NCX
D-
g/i
-41 g/l
I
• 85.9 g/l
-------
e ••
Figure 9
THE EFFECT OF SLURRY CONCENTRATION UPON ALUMINUM
EXTRACTION EFFECIENCY
PAIRED TESTS DIFFER ONLY IN SLURRY CONCENTRATION AND
THE DIFFERENCE BETWEEN SETS OF TESTS IS SHOWN IN TABLE 9 .
D
100 g/l
•20 9/\
I
• K> g/l
-------
C. Conclusions. Recovery of aluminum from wet-collected, limestone-
modified flyash by methods similar to those employed in lime-sintering processes
is a highly complex process made more difficult by the extremely fine particle
size and low aluminum content of the ash. Inter-relationships were found between
all of the process variables monitored and the aluminum extraction efficiency.
In effect, a very close monitoring of all aspects would have to be performed
to achieve maximum extraction. In addition, sufficient contact time is not
available during combustion to completely convert the aluminum present into
the desired calcium aluminates; but further sintering is probably inadvisable
due to the re-release of the trapped sulfur gases. Autoclaving of the ash,
although it did not re-release the sulfur, was considered unsuccessful as a
method of forming the desired calcium aluminates.
-------
-------
APPENDIX A
-------
INTRODUCTION
Dewatering is defined as the separation of a mixture of solids and
water into two parts, one of which is relatively solid free, and the other
relatively liquid free.
Such a process requires that the forces binding the liquid to the
solid be overcome.
These forces include:
1. Physical and chemical forces which directly bind the liquid to the
solid.
2. External forces such as gravity, electrical fields, and thermal
energy which bind the liquid and the solid.
3. Properties of resistance in flowing fluids and media.
The problem of solid-liquid separation is one which occurs in
practically every industry. The sludges from industry may create a
health problem or they may contain materials which are of value
economically. Thus, it is essential that efficient, inexpensive methods
of dewatering sludge be available.
o
The practice of utilizing materials from waste water is not new. As
early as 1740, the French were extracting chemicals from waste waters. In
1884, the English first used filter presses to dewater waste sludges. Between
1880-1900, the use of chemicals and filtration in treating wastes became
established in the United States.
DEWATERING METHODS
The major methods of dewatering are gravity thickening, vacuum filtration,
and centrifugation. Other methods which find limited utilization include:
-------
electrophoresis, vibration sieving, filter pressing, pressure filters, precoat
filters and hot air drying.
Gravity Thickening;
Concentration can be affected by allowing the sludge to settle with or
without stirring or chemical conditioning. The cheapest and simplest means
of dewatering is gravity thickening. In this technique the solids are allowed
to settle out of suspension under the influence of gravity. The success of
this method is dependent upon the nature of the solids, their ability to form
floes (combination of small solid particles into larger ones), the degree of
compaction (final settled sludge volume) and the ease of elimination of water
from the settling material. The sludge concentrate should be gently stirred
so as to break down structures which have formed during sedimentation.
There are several advantages to thickening before further treatment.
These include:3
(a) reduction of sludge volume
(b) reduction of conditioning cost (ie, addition of chemical conditioners)
(c) reduction of water content and
(d) reduction of over-all treatment cost (ie, secondary treatment processes)
The thickened sludge is compacted and a great deal of the water is
removed but interstitial water is still present and is not easily removed. Thus,
further treatment is often required.
Vacuum Filtration:
Vacuum filtration of process industry slurries has been practiced for
many years. Since 1960, the use of vacuum filtration has risen for the
following reasons:
-------
(a) development of self-cleaning media such as polyfilament fibers,
(b) increased cost of alternate methods and
(c) applicability to all sewage wastes and to many industrial wastes.
In vacuum filtration a rotary drum passes through a slurry tank and
solids are retained on the drum surface under vacuum. As the drum rotates
through the slurry, a cake is built up and the water is removed by vacuum.
The emerged solids are dewatered under vacuum and then are removed from
the drum, generally by scraping at the end of each cycle. There are
several advantages to vacuum filtration which include:
(a) variety of sludges dewatered,
(b) low floor space requirement,
(c) relatively dry filter cake produced, and
(d) good solids capture.
These advantages are somewhat offset by possible disadvantages which
include:
(a) frequent blinding (clogging of pore spaces by fine particles)
of filter media,
(b) operators required, and
(c) lack of precise scientific control.
In choosing the filter media to be used, several factors must be
considered. If the filter medium is a cloth fabric, the individual fibers
may embed in the filter cake causing the cake to adhere to the drum. The
filter medium must be strong enough to withstand the strain created by the
vacuum without rupturing. Also, the pore size should be such that the fine
solids are retained but blinding is kept at a minimum. The materials which
are commonly used include: cotton, synthetic fibers, metallic cloths, and
punched rubber sheeting. Cotton cloth is usually used in milling since it has
the advantages of great tensile strength, flexibility, durability, and is
easily cleaned.
IB
-------
Chemical conditioning of the slurry to enhance the efficiency of
devatering is often a necessary step before sludge vacuum filtration.
Pressure Filtration:
Pressure filtration has been found to be effective in dewatering
and is simple, flexible and can be used either with or without chemical
reagents.
The pressure from the slurry pump is adequate for dewatering the slurry
in most cases.
There are several variables which need to be considered in relationship
to each specific sludge.
Process Variables:
(a) feed rate or slurry concentration
(b) feed consistency
(c) particle size
Machine Variables:
(a) capacity of filter cake area
(b) amount of time required to remove filter cake and replace
filter on stream
Miscellaneous Methods:
There are many other methods which may be used to affect sludge dewater-
ing. Briefly, they include:
(a) Electrophoretic dewatering in which the particles migrate
in an electric field to a point of concentration or collection;
(b) Vibrational sieving, used with coarse suspensions:
(c) Filter presses which may be used for sludges which are rich in
minerals and may be desiccated without previous treatment;
(d) Centrifugation which can normally be used with nonabrasive
slurries:
(e) Precoat filters which involve an auxiliary layer of diatomaceous
earth, sawdust, or an ash covering a membrane filter. The need
C. i
-------
for prior conditioning is generally eliminated by precoating
the filter; and
(f) Drying by hot air, drying drum, rotary dryer, pressure drying
multistage drying, or thermal spray drying.
DEWATERING AIDS
In many instances sludges can be dewatered by various combinations of
the above methods to yield a product of proper moisture content for utilization
or disposal. However, in numerous other slurries, including wet collected
modified flyash, the particles are of such minute dimensions that they settle
out of suspension very slowly and can pass through ordinary filters. These
materials may require the use of inorganic chemical conditioners to coagulate
or organic chemical conditioners to flocculate the fine particles prior
to dewatering by the above methods.
Coagulation:
The stability of fine particles that resist coagulation is a result of
electrostatic forces and hydration which involve the charging of surface
components and the subsequent adsorption of ions from solution.
The influence of charge on the stability of dispersed systems of particles
is a function of the zeta potential, which is the magnitude of the potential
difference of the boundry between the volume of liquid held by the particle
Q
and the solution.
The value of the zeta potential in millivolts (mv) may be determined by
use of a zeta meter.
The stability of a colloidal system decreases as the zeta potential
9
of the component particles approach a value of zero mv.
The coagulation of dispersed particles (particles that have zeta
potentials greater than + 15 millivolts-) by charge reduction is often induced
by the addition of chemical electrolytes which reduce the electrostatic forces.
-------
The slurry is generally agitated to insure collision between particles.
With inorganic electrolytes the coagulation power of bi-valent ions is
about 20 to 80 times greater than that of uni-valent ions and the coagulation
power of tri-valent ions is many times greater than that of bi-valent ions.^
As a rule, the higher the charge of an ion the smaller its solvated
volume rendering it more useful in charge reduction. Thus, hydrolyzed tri-
valent metal salts such as A^CSO^, A1(N03>3 and Fe2(S04)3 have been
investigated as coagulants for lyophobic sols. 12 The results of these
experiments indicate that aging of the salt solution has a strong effect
on the coagulation of the sol. In all cases studied using Al^CSO.^, aging was
found to reduce the critical coagulation concentration (C.C.C.). When a
common ion (I^SO^) was added, the C.C.C. was observed to decrease slightly.
Flocculation :
Observations of flocculation phenomena have led to the formulation of
three possible mechanisms of flocculation.
1. Charge Effect Neutralization: The net effect of the
electrostatic repulsive forces is neutralized between
particles having a double layer. The polyelectrolytes
compress the double layer allowing the attractive
forces to act.
2. Polymer Bridging: The process Involves the use of a
flocculant to interfere with the free movement of sol
particles . This may occur as a result of the adsorption
of a number of particles on the polymer network or the
formation of a bridge between two particles which would
not readily interact.
3. Mutual Dehydration: Anon-ionic flocculant reacts
with a sol to form an insoluble precipitate.
Starch was the first long chained molecule used as a flocculant.
Other natural products which have been used include: locust bean gum,
cactus extract, linseed, glue, gelatin, etc.
Gelatin has been thought to be unique in that it is applicable to any
suspension under the appropriate conditions.15 Flocculation with gelatin
-------
is also pH dependent and the best results are obtained under conditions where
the gelatin carries a small positive change.
Since the early 1950's, a wide variety of synthetic polymers have
become available for use as flocculants.
In order to be effective, a flocculant must have certain characteristics.
1. The molecules must contain polar groups such as -COOH or -OH
which can be adsorbed on the particle surface.
2. The molecule must be of sufficient length to form
a bridge between particles producing uniform
floes of a proper size and shape to give the
filter cake the proper porosity.
3. The flocculant must be water soluble and the
effective concentration must be very small.
4. The molecules should have a good scavenging
and binding ability in order to gather up the
sol particles and resist deformation of the
filter cake under vacuum filtration.
1 ft
Healy and LeMer have developed a quantitative theory of flocculatlon
in which the optimum flocculation occurs with 50 percent of the surface being
covered with polymers. When the amount of surface covered reaches 100 percent,
the floes become metastable and may re-disperse due to the insulative effects
of excess flocculant.
Molecular weight of the polymer is important in the flocculation
process. Flocculation is improved by using polymers of high molecular weight,
but too high a molecular weight will cause fast settling of the floe leaving
fines in suspension.
Non-Ionic polymers have been studied as flocculating agents for dilute
20
clay suspensions. It was found that polyethylene oxide will be absorbed
on clay sols and form a floe by a bridging mechanism. The process is pH
dependent and also is dependent upon the counter ions present.
-------
Studies of the flocculation of silica suspensions by cationic polymers
21
indicate that flocculation can occur with polymers of low molecular weight.
Much interest has developed in these cationic polymers. Polyacrylamide
22
has been investigated as a flocculate for the mining industry. It has been
found to be remarkably versatile. Suspended solids from sewage, coal mines,
and chemical precipitates have been studied.
Thus, it would appear that the use of flocculants or coagulants is the
best approach for removing fines from suspension. The concentrations of the
conditioners involved are small enough that, for most processes, the additional
electrolytes would not present a water purity problem. With polymers, heating
to an elevated temperature should remove the organic materials as oxidized
products. Polymeric flocculation will increase the water content of the
floe over that of an untreated floe.
-------
FOOTNOTES
1. Taggart, A. F., Handbook of Mineral Dressing, John Wiley and Sons,
New York, 1945.
2. Stanley, W. E., et al Sludge Dewatering. Water Pollution Control Federation,
Washington, D. C., 1969.
3. Burd, R. S., A Study of Sludge Handling and Disposal, Federal Water
Pollution Control Administration, Pub. WP-2Q-4, Washington, D. C.,
May, 1968.
4. Ibid.
5. Eckenfelder, W. W., Industrial Water Pollution Control, McGraw-Hill,
New York, 1966.
6. Taggart, A. F., Handbook of Mineral Dressing, John Wiley and Sons,
New York, 1945.
7. Gaudin, A. M., Principles of Mineral Dressing, McGraw-Hill, New York,
1939.
8. Riddick, T. M., Control of Colloidal Stability Through Zeta Potential,
Livingston, Wynnewood, Pa., 1968.
9. Ibid.
10. Jirgerson, B., Strumanis, W. E., A_jShort Textbook of Colloid Chemistry,
MacMillian, New York, 1962.
11. McCarty, M. F., and Olson, R. S., Mining Engineering, January, 1959,
61-65 (June, 1959).
12. Matijevic, E., Stayker, L. V., Colloid and Interfact Sci: 22, 68-77,
(1966).
13. Welles, W. E., J. Colloid and Interface Science: 27, No. 4, 797-803
(Aug., 1968).
14. Purchas, D. B., Industrial Filtration of Liquids, Chemical Rubber Co.,
Cleveland, 1967.
15. Ibid.
16. Kuagli, A. M. and Longston, W. B., J. Colloid Science- 17, 101-123.
(1962).
17. LaMer, V. K., Smallie, R. H., Lee, P. K., J^_Colloid_Sci• 12, 230-239,
(1957). ~ ~
-------
18. Healy, T. W., LeMer, V. K., J. Colloid Sci: 19, 323-332 (1964).
19. Linke, W. F. and Booth, R. D., "Physical Chemical Aspects of
Flocculation by Polymers," All-IE Convention, San Francisco, Calif.,
February, 1959.
20. Birkner, F. B., Edzweld, J. K., Annual Conf. AHWA, 1969.
21. Dixon, J. K., LaMer, V. K., Messinger, C. S., Linsford, H. B.,
J. Colloid and Interface Sci: 23, 456-473.
22. McCorly, M. F., Olson, R. S., Mining Engineering, 61-65, (Jan., 1959).
-------
-------
APPENDIX B
-------
FLOCCULANT FILTERABILITY TESTING
These techniques include the Buchner funnel test, the vacuum break
test, and the filter leaf test, and constitute a laboratory testing program
to provide information on optimum additive concentration and specific vacuum
filter cycle segments. Using this information, a slurry dewatering program can
be designed.
A. Buchner Funnel Test;
The Buchner funnel test is a common method by which optimum flocculant
concentration may be obtained.
Procedure; Flocculant solutions are prepared by dissolving 0.1 gram
of flocculant in 99.9 ml of water to form a stock solution. 12.5 ml of the
stock solution is then diluted to 500 ml for use in the flocculation
studies.
Samples are prepared for filtration by mixing 100 ml of concentrated
slurry with flocculant and distilled water in the proportions presented in
Table 1.
TABLE 1
mis of
Flocculant
0.0
0.5
1.0
2.0
3.0
5.0
10.0
15.0
20.0
25.0
30.0
mis of Sample
Water Total Vol.
30.0 130 ml
29.5 "
29.0
28.0
27.0
25.0
20.0 "
15.0
10.0
5.0
0.0
(ppm)
Floe. Cone.
0.000
0.096
0.193
0.386
0.579
0.965
1.930
2.895
3.860
4.825
5.790
4'S
-------
The samples are then vacuum filtered under 5 inches of mercury vacuum and
the filtrate volume is recorded at the end of 0.5 minutes. The optimum
flocculant concentration is selected on.the basis of maximum filtrate volume
and clarity.
B. Filter Break Test;
The filter break test is an extension or refinement of the Buchner
funnel test. A flocculated sample of sludge is filtered until it reaches
dryness. Air then enters the system through drying cracks or breaks and the
vacuum decreases. The time required until the vacuum drops is recorded.
Procedure: The optimum amount of flocculant determined in the Buchner
funnel test is completely mixed with a 100 ml sample of sludge. The
flocculated sludge is then filtered under a vacuum of 10 inches Hg and the
time required for the vacuum break to occur is determined. At this point the
vacuum would no longer be effective as air would be passing through the cake.
C. Filter Leaf Test
• ~™^~""~~~~~^~~~~~~~
The filter leaf test is designed to reproduce the operating cycle of
a commercial vacuum filter.
Procedure: A series of filter media to be examined along with a filter
leaf apparatus having a leaf area of 0.1 ft? are used.
The optimum flocculant concentration determined from Buchner funnel
tests is added to 900 ml of slurry. The filter leaf with attached cloth is
submerged in the flocculated slurry and a vacuum of 15 inches Hg is applied
to the system for 1.5 minutes. The leaf is then removed from the slurry and
held upright for three minutes. The percent moisture, thickness of the
filter cake, and the volume of filtrate are recorded. The resultant information
can then be used in the design of a field size vacuum filter dewatering operation
by using scale-up factors determined by specific equipment manufacturers.
-------
43
-------
APPENDIX C
-------
ALUMINA LEACHING
When World War II began, the United States Government realized the
seriousness of the situation that could arise if the existing supply of
bauxite were cut off. Since then, extensive studies have been made of
methods to extract alumina from clays and other alumina-silicates. The
methods to be discussed in this summary include sinter processes , acid
processes, and alum processes.
Sintering Processes:
Two general approaches have been taken in sintering processes .
These methods are generally referred to as the lime sinter and the lime-
soda sinter processes.
The lime sintering process is based upon reactions which convert
aluminum silicates to calcium orthosilicates and calcium aluminates.
In the lime-sinter process, alumina bearing materials are mixed with
limestone and are sintered to form calcium aluminate (2CaO • Si02) from which
the alumina is extracted as sodium aluminate (NaA102) by leaching with NaOH,
Na2C03 or mixtures of these, d)
1370°C
3(A1203 • 2Si02 • 2H20) + 17CaC03 5CaO • 3A1203 + 6(2CaO • SiD2)
+ 17C02 + 6H20. (1)
The calcium orthosilicate formed by the lime sinter process ties up nearly
all of the silica present in a form which is insoluble in the leach liquor. As
the sinter cools below 675°C, it undergoes a crystallographic change with a
subsequent increase in volume resulting in powdering or "dusting" of the
sinter. This provides a much greater surface area and the alumina becomes
more accessible to the leach liquor.
Lundquist and Leitchv ' have investigated the solubility of monocalcium
aluminate in water, sodium hydroxide, sodium carbonate, and mixtures of these.
They reported that calcium aluminate exibited a very limited solubility
(llg/liter) in water even though it was completely hydrolyzed.
Soluble alumina did appear in solution, probably due to the presence of
(OH)~ from the Ca(OH)2 generated during the hydrolysis reaction. The end
result of the hydrolysis was represented by:
3[CaO . A1203] + 12H20 = Ca3 [A1(OH)6]2 + 4A1(OH)3
-------
The maximum soluble alumina concentration occurred within the first hour.
The temperature of the leaching tests had a negligible influence on the initial
extraction of soluble alumina, moderate effect on the precipitation of
A1(OH)3 and a strong influence on the crystal structure of hydrated calcium
alumina tes formed.
The solubility of CaO • A1203 in NaOH was observed to be similar to its
solubility in water. The reaction with NaOH may be represented by:
3[CaO • A1203], + ANaOH + 12H20 Ca3 [Al(OH)s]2 + *Na+ + 4A1(OH)4~. (3)
The presence of NaOH provides additional (OH)~ ions to inhibit the
precipitation of A1(OH)3.
The amount of A1(OH)3 precipitated or dissolved is dependent upon
the following reaction:
A1(OH)3 + (OH)~ A1(OH>4~. (4)
The amount of (OH) in solution available during the reaction depends upon
the composition of the leach liquor, the extent of hydrolysis, and the amount
of calcium alumina te initially in the material to be leached. The formation of
soluble sodium alumina te is offset by the precipitation of hydrated calcium
alumina tes .
The reaction which occurs when sodium carbonate is used is :
CaO . A1203 + Na2C03 + 4H20 CaC03 + 2Na+ + 2A1(OH)4~. (5)
In this reaction, CaC03 becomes the principle precipitate instead of the less
desirable insoluble Ca3r/^l(OH)g] . Thus, a combining of carbonate, to prevent
precipitation of the aluminates, and sufficient OH~ to hold alumina in solution
should provide maximum alumina extraction.
The lime-soda process is not significantly different from the lime sinter
process. Ground siliceous ore is mixed with limestone and sodium carbonate. .
to give a mixture corresponding to dicalcium silicate and sodium aluminate. '
Peters and co workers have evaluated a lime-soda process for
producing alumina from clay. The sintering reaction is represented by:
(A1203 • 2Si02 • 2H20) + 4CaC03 + NaoCO, Na,0 • A1203 + 2H20 + 2(2CaO .
Si02) + 5C02 (6)
The temperature requirements, however, 'are not quite so high as they
are in the lime sinter process. The dicalcium silicate doesn't undergo a
phase change and "dusting." However, the sinter is usually less dense and is
readily friable, porous and easily ground if it is not over burned.
Sodium aluminate is the principle aluminum compound formed in the
lime-soda sinter process.
Lundquist and Leitch™) have reported the solubility characteristics
of sodium aluminate. They found that nearly all of the sodium aluminate
added to H20, NaOH, or Na2C03 dissolved. However, the resultant solution
was unstable; the amount of aluminum in solution is dependent upon the
amount of free OH.~ The early precipitation, however, of gibbsite is a
51
-------
detrimental factor in the lime-soda process:
A1(OH)A~ = A1(OH)3 (Gibbsite) + OH.~ (7)
Thus, it is necessary that an excess of OH~ be present in order to
prevent excess precipitation of gibbsite. Also, rapid removal of the
tailing solids from the leach liquors aids in the inhibitation of gibbsite
formation.
NATURE OF SODIUM ALUMINATE AND STRUCTURAL PHASES FORMED IN LIME-SINTER
AND LIME-SODA SINTER PROCESSES.
Sodium Aluminate:
The dissolution of aluminum compounds in NaOH solutions takes place
with the formation of sodium aluminate which is considered to be fully
dissociated in solution.(5)
According to Pearson, the precipitation of aluminum hydroxide may be
written:
A1(OH)4 . (H20)2~ A1(OH)3 + 2H20 + OH~ (8)
/o\
Lundquistv has suggested that the reaction is one of polymerization
in the precipitation of an alumina gel or crystalline aluminum hydroxide.
The Initial polymerization reaction might be written:
2[A1(OH)4 . (H20)2r = A12(OH)8 . (H20)2= + 2H20 (9)
The reaction builds longer ion units by addition of other monovalent
ions. The growth of polyvalent ions continues througli colloidal stages into
the crystalline aluminum hydroxide.
Hydrated Calcium Aluminates Encountered in
Lime-Sinter and Lime-Soda Sinter Processes:
The hydrated calcium aluminate Ca3[Al(OH)5]2 is formed in the leaching step
of both the lime-sinter and the lime-soda sinter processes. .-Also, in the lime-
soda process the hydrate 3CaO . A1203 . (8-12)H20 is formed.™'
The hydrates which form may have very different crystallographic
structures. The compound 3CaO . Al20-j . (8-12)H20 belongs to a group of
compounds with the general formula of !ICa(OH)2 . nAL(OH)-) . plUO and forms
hexagonal plates in a layered structure. The layers are thought to be built
up of Ca(OH)2 and Al(OH)-} with variable water content between the layers.
The compound Ca3[Al(OH)g]2 consists of compact cubic crystals with
octahedral ions, Al(OH)g~ held together by Ca""" in positions of eipht fold
coordination. It is ionic in nature and corresponds to a liydrogarnet mineral.
Nitric and Hydrochloric Acid Processes:
A. Nitric Acid Processes
Johnson and coworkers* ' have reported a nitric acid process for
producing alumina from siliceous ores.
-------
In this process, the clay is first calcined at 700°C and dehydrated to
remove moisture and combined water. The aluminum is converted from insoluble
aluminum silicates to an acid soluble compound upon dehydration.
(A1203 . 2 Si02 . 2H20) (A1203 + 2Si02) + 2H20 . (10)
A less than stoichiometric amount of 30% nitric acid is used to digest
the clay at 165°C and 85 psig, thus forming a slightly basic aluminum nitrate
solution in which iron is nearly insoluble,
(A1203 + 2Si02) + 6HN03 2A1(N03)3 + 2Si02 + 3H20. (11)
The iron and silica are removed by filtration. The aluminum nitrate
solution is then concentrated by vacuum evaporation and aluminum nitrate
nonahydrate is crystallized.
This hydrate is then fed to a fired rotary kiln operating at 500°C where
the crystals decompose,
2[A1(N03)3 . 9H20 A1203 + 3N205 + 18H20. (12)
The advantages of this method over other less expensive acid processes
include: the lower solubility or iron in basic aluminum nitrate solutions,
ease of decomposition of nitrates, simple recovery of acids. The disadvantages
include: large cooling requirements, pressure digestion, comparatively
expensive make up acid.
B. Hydrochloric Acid Processes
Peters and coworkers reported a series of hydrochloric acid processes
for extraction of alumina. dl)
The major differences are found in the method of extraction of the
alumina after leaching.
1. Isopropyl Ether Extraction
Calcined clay is leached with 20% HC1 and the resulting slurry filtered
to remove the silica residue,
(A1203 + 2Si02) + 6HC1 2A1C13 + 2Si02 + 3H20. (13)
Isopropyl ether is contacted with the leach liquor countercurrently
and the iron removed through the formation of a complex:
[FeCl3 . HC1 . 10.3 H20] x [(CH3)4 C2H20]. (14)
.The purified aluminum chloride solution is evaportated to form crystalline
[A1C13 . 6H20].
The aluminum chloride is then decomposed at temperatures between 9 25- 1100 "C
to yield aluminum,
2[A1C13 . 6H20] A1203 + fiHCL + 3H20 . (15)
-------
2- Gas^ ^Precij) itatipn^ Process
In this process calcined clay is digested with HC1 and the slurry is filtered
to remove silica residue. The filtrate is then treated with hydrogen chloride
gas to precipitate AlCl-j . 6H20 leaving the iron in solution. The A1CJ3 . 6ll20
solution is then evaporated to form the hexahydrate which is then calcined
to give alumina.
3• Gas^ Precipitation -__ Isppropy1_ Process
This is a modification of the previously described gas precipitation
purification process. It has been studied because it offers a method by
which iron precipitated with the aluminum chloride may be removed by
extraction with isopropyl ether, Up to 95% of the iron contamination
may be removed from the crude aluminum chloride by this method.
4. Caustic Purification Process
Calcined clay is leached with 20% HC1 to form AlCl-j. The iron present
also dissolves under the reaction conditions (95°C, 1 atm). The filtrate
is concentrated to a 30% solids (by weight) slurry. The hydrated aluminum
and iron chlorides are then removed by vacuum filtration. The aluminum
chloride is decomposed at 540°C to give crude alumina. The crude alumina
is then digested with NaOH at 120°C resulting in nearly complete dissolution
of the alumina and leaving the iron as solids.
Pearson'"' has formulated the aluminate ion to consist of a central ion
surrounded by four hydroxyl ions and two water molecules in octahedral
coordination with a simple negative charge.
Here, the two water molecules may readily be replaced by two OH~ ions to
form a triply negative aluminate ion Al(l)H)(j~3 which would be consistent with
the precipitation of Ca3[Al(UH)^]2.
However, Roman studies of sodium aluminate solutions'^' and studies of density-
volume changes during the formation of sodium aluminate(7) support a
tetrahedral coordination with a single negative charge.
This Al (Oll)^" sPecies exists in low to moderate concentrations only. At
concentrations of 15% Na20 or more, the tetrahyclroxyaluminate ion may undergo
dehydration to the meta-aluminate ion,
A1(OH)4- A10 (OH)2~ + H20 A102~ + 21I20 (16)
Between 25°C and 350°C, an equilibrium exists between the univalent ions
of different degrees of hydra t ion shown in equation (10) .
-------
The solids are removed by filtration and the filtrate is cooled and
seeded with A1C13 . 3H20 crystals. Alumina trihydrate is then formed
according to:
2NaA102 + 4H20 A1203 . 3H20 + 2NaOH. (17)
The hydrate is then decomposed at 1100°C,
A1203 . 3H20 A1203 + 3H20. (18)
5. Sinter Purification Process
Calcined clay is leached with 20% HC1 at 95°C producing crude A1C13
and FeCl3. The solids are removed by vacuum filtration and the filtrate
is concentrated to a 30% slurry of aluminum and iron chlorides (AlClo .
6H20, FeCl3. 6H20). The crystals, are then calcined at 540°C to yield
crude alumina.
The crude alumina is then sintered with soda ash and limestone at
1040°C. The alumina is converted to sodium aluminate while the iron
remains as solid iron oxide (Fe203).
A1203 + Na2C03 2NaA102 + C02. (19)
The aluminate is then leached from the sinter and alumina is precipitated
from the filtrate with carbon dioxide,
2NaA102 + C02 + 3H20 A1203 . 3H20 + Na2C03. (20)
The hydrated alumina is then calcined to produce alumina.
There are several advantages and disadvantages to the hydrochloric
acid processes. Among the advantages are rapid filtering rates, removal
of iron, and recovery of acid.
The disadvantages include need for hydrochloric acid proof equipment,
large amounts of water required, and large material requirements.
Sulfuric Acid Processes:* '
In these processes, the silica in the raw material is not dissolved and
no insoluble compounds containing silica and alumina as sulfates are formed
which might cause alumina loss. However, the presence of iron will cause
problems and it must be removed.
a. Electrolytic Iron-Removal Process
The starting material is calcined at 700°C to dehydrate the aluminum
silicates,
(A1203 . 2Si02 . 2H20) (A1203 + 2Si02) + 2H20. (21)
The dehydrated material is leached with a 40% solution of sulfuric acid,
-t
*j»O
-------
The iron present in tin- filtrate it. removed In merr.ury cdthodc cells
as metallic iron. The silica is removed from tlit Al ?(S()/() 3 solution L>y healing
it to 8!»°(.' and mixing witli a clay residue. The solution is then concentrated
by evaporation to a concentration of 40% Al 2^(^)3. The solution is then
filtered and the filtrate concentrated by t-vnporat ion to form a liydrated aluminum
sulfate containing 18 moles IM) per mole Al2(S()/,)3.
This hydr;ited material is dehydrated at MM)°C and decomposed
-------
2) Chemical Iron Removal:
a) large amounts of ozone required
b) cost of ozone •
c) large utility requirements
3) Ethanol Purification:
a) large amounts of make-up ethanol
b) utility requirements
c) cost of ethanol
Sulfurous Acid-Caustic Process :
A sulfurous acid-caustic process for recovering alumina has been reported
by Peters and coworkers.*^)
Raw clay is calcined at approximately 700°C. It is then leached with
sulfurous acid to dissolve alumina as a sulfite,
3H2S03 + (A1203 + 2Si02) A12(S03)3 + 3H20 + 2Si02. (25)
Iron is also converted to sulfite whereas the silica remains insoluble
and is discarded.
The solution is then autoclaved at atmospheric pressure and 100°C. 502
is given off and a basic aluminum sulfite precipitates. The iron and other
impurities may be kept in solution by having 2 moles of free SOo per mole of
alumina.
The precipitate will be a well defined compound with the general
formula:
A1203 . 2S02 . 5H20. (26)
The monobasic sulfite may be decomposed by either calcination at 485-595°C
or by a hydrothermal method.
The hydrothermal method consists of removing S02 by heating the slurry
to between 110-160°C under 4-6 atm.
The precipitate formed is dissolved in sodium hydroxide.
2NaOH 4- A1203 . (x)H20 2NaA102 + (l+x)H20. (27)
From this step, the trihydrate is formed then to give alumina,
2NaA102 + 4H20 A1203 . 3H20 + 2NaOH, (28)
A1203 . 3H20 A1203 + 3H20. (29)
-------
The advantages of the sulfurous-caustic process are:
a. acid recovery
b. the A12(S03)3 may be decomposed hydrothermally at 160°C
c. sulfurous acid is relatively inexpensive
Disadvantages:
a. requires pressure, digestion and decomposition
b. long reaction time for reasonable yield
c. requires closed equipment
d. acid proof equipment required
Ammonium Alum Process;
(14)
Peters and coworkers have reported the evaluation of ammonium alum
processes for the extraction of alumina.
The raw material is calcined at 700°C. The dehydrated clay is treated
with an 85% solution of NH4HS03 and 15% (NH)4S04 at 93°C and 1 atm.
The aluminum and iron dissolve according to:
A1203 + 6NH4HS04 A12(S04)3 + 3(NH4)2S04 + 3H20. (30)
Fe203 + 6HN4HS04 Fe2(S04)3 + 3(NH4)2S04 + 3H20. (31)
The iron is removed by reduction in a 50% mole solution of NH.HSO,
and 50%'NH4S03.
S02 + NH4OH NH4HS03. (32)
S02 + 2NH4OH (NH^jjSO-j + H20. (33)
The iron reduction may be represented by the following:
Fe,(S04)3 + NH4HS03 + H20 2FeS04 + NH,HS04 + H2S04, (34)
Fe2(S04)3 + (NH4)2S03 -f H20 2FeS04 + 2NH4HS04. (35)
The reduced iron remains in solution allowing the impure alum to
crystallize out of solution. Ammonium alum is purified and dissolved
in 40% NH4S04 solution. It is treated with 20% NH4OH in an autoclave
at 100°C to precipitate the trihydrate which then is filtered, dehydrated
and decomposed.
(A12(S04)3 . (NH,)_S04 . 24H20) + 6N1LOH (A1203 . 3H20) + 4(NH4)2
S04 + 24H20.
-------
Potassium Alum [K9SOA . Al0(SOA)i . 24H90] Process;
_ — * 'I • •! ^ • --
A potassium alum process for aluminum extraction has been evaluated by
Peters and coworkers . '")
Raw clay is dehydrated and leached with a sulfuric acid-potassium sulfate
solution counter currently at approximately 90°C.
(A120, + 2Si02) + 3H2S04 + K2S04 + 21H20 K2S04 . A1203 . 3S03 .
+ 2Si02. (37)
The solution is then filtered to remove the silica residues. Iron removal
is affected by reduction of the iron with sulfur dioxide. The ferrous iron
formed then remains in solution during alum crystallization. Iron present in
the recycle solution is oxidized and precipitates during leaching when excess
clay is added to the leach system.
The potassium alum is crystallized in vacuum crystallizers at 20°C, leaving
Impurities such as Ti, Ca, P, Ca, Fe, Na and Mg in solution. The alum crystals
are then separated from the mother liquor in centrifugal filters and washed.
The alum is decomposed in autoclaves at 200 °C for 30 minutes with 4 1/2
pounds of 250 psig steam for each pound of alum present,
3(K2SOA . A1203 . 3S03 . 24H20) K2S04 . 3A1203 . 4S03 . 9H20
+ 5H2S04 + 2K2S04 + 58H20. (38)
The basic potassium alum is fed to furnaces where the cyrstals are first
dehydrated at temperatures less than 600 °C,
(K2S04 . 3A12)3 . 4S03 . 9H20) K2S04 . 3A1203 . 4S03 + 9H20. (39)
The decomposition is continued at temperatures between 600°C-800°C,
(K2S04 . 3A1203 . 4S03) 3A1203 + K2S04 + 4S03. (40)
The remaining iron is removed by reduction with S02. The ferrous ion
remains in solution during alum crystallization.
-------
RiIFERF.NCES
1. Henn, J. J., Johnson, P. W., Amoy, E. B., Peters, F. A.. Me thqds for
Producing Alumina from Clay: An Evaluation of TVq_Lime_ Sinter Processes,
USBH Rept. of Inv. 7299, 1969.
2. Lundquist, R. V., and Leitch, H., Solubility Characteristics of
Monocalciuni Aluminate. USBM Rept. of Inv. 629~4, 1963.
3. Peters, F. A., Johnson, P. W. , Henn, J. J., Kirby, R. C., Methods for
Producing Alumina from Clay; An Evaluation of A Lime-Soda Process,
USBM Rept. of Inv. 6927, 1967.
4. Lundquist, R. V., Leitch H., Solubility Characteristics of Sodium Aluminate
USBM Rept. of Inv. 6504, 1964.
5. Glastonbury, V. R., Chemistry and Industry, February, 1969, pp 121-125.
6. Pearson, T. G., The Chemical Background of the Aluminum Industry,
Royal Institute of Chemistry (London). Lectures, Ifonongraphs and Reports,
No. 3, 1955, pp 22-28.
7. Sakamoto, K., "Mechanism of Decomposition of Sodium Aluminate
Solutions," in The Extractive Metallurgy of Aluminum, Gerard, G. Stroup,
R. T., eds., Vol. 1, Alumina, AIME, N. Y.7 1963.~
8. Lundquist, R. V., Specific Conductance, pll, Density, and Viscosity of
Sodium Aluminate Solutions and Some Properties of the Aluminate Ion,
USBM Rept. of Inv. 6582, 1965.'
9. Lundquist, R. V., Leitch, H., Ttoo Hydrated Calcium AluminatesEncountered
in the Lime-Soda Sinter Process, USBM Rept. of Inv7 6335,~1963.
10. Johnson, P. W., Peters, F. A., Kirbys R. C., Methods for Producing Alumina
from Clay; An Evaluation of a Nitric Acid Process, USBM Rept. of Inv.
6431, 1964.
11. Peters, F. A., Johnson, P. W., Kirby, R. C., Methods for Producing Alumina
from Clay; An Evaluation of Five Hydrochloric Acid Processes, USBII
Rept. of Inv. 6133, 1962.
12. Peters, F. A., Johnson, P. W., Kirby, R. C., Methods for Producing Alumina
from Clay; An Evaluation of Three Sulfuric Acid Processes, USBM
Rept. of Inv. 6133, 1962.
13. Peters, F. A., Johnson, P. W., Kirby, R. C., Methods for Producing Alumina
from Clay: An Evaluation of the Sulfurous-Acid-Caustic Purific:ation Process ,
USBM Rept. of Inv. 5~997, 1962.
14. Peters, F. A., Johnson, P. W., Kirby, R. C., Methods for Producing
Alumina from Clay: An Evaluation of Two Ammonium Alum Processes,
USBM Rept. of Inv. 6593, 1965.
15. Peters, F. A., Johnson, P. W., Kirby, R. C., Methods for Producing Alumina
from Clay: An Evaluation of a Potassium Alum Process, USBI! Rept. of Inv.
6290, 1963.
-------
BIBLIOGRAPHIC DATA
SHEET
1. Report No.
EPA-R2-73-262
3. Recipient's Accession No.
4. Title and Subtitle.
Evaluation of Dewatering of Limestone Wet Scrubbing
Process Sludges
5. Report-Dace
May 1973
6.
7. Author(s)
N.A.
8. Performing Organization Rept.
No.
9. Performing Organization Name and Address
Coal Research Bureau
West Virginia University
Morgantown, West Virginia 26505
10. Project/Task/Work Unit No.
11. Contract/Grant No.
EHSD 71-11
12. Sponsoring Organization Name and Address
EPA, Office of Research and Monitoring
NERC/RTP, Control Systems Laboratory
Research Triangle Park, North Carolina 27711
13. Type of Report & Period
Covered .
Final
14.
15. Supplementary Notes
16. Abstracts The repOrt presents the results of studies during which several methods of
dewatering solid materials were applied on bench-scale to wet-collected, limestone-
modified fly ash from a coal-fired electric power plant using limestone-injection and
wet-scrubbing methods to control the emission of gaseous sulfur oxides. Porous-bed
sand filtration, lagooning, and possibly pressure filtration appear to hold the most
promise. Aluminum extraction tests using sodium hydroxide, sodium carbonate, and
combinations of the two yielded less than 50 percent of the aluminum available in the
leach liquor. Structural materials testing indicated that there is insufficient free lime
available in the modified ash to act as a suitable binding agent.
17. Key Words and Document Analysis. 17a. Descriptors
Air Pollution
Economic Analysis
Chemical Analysis
Dewatering
Aluminum Oxide
Limestone
Dolomite (rock)
Fly Ash
Coagulation
17b. Idcntifiers/Opcn-l£nded Terms
Air Pollution Control
Stationary Sources
Wet Scrubbing Processes
-Injection Processes
Sand-Lime Brick
Coal
Combustion
Electric Power Plants
Flocculating
Injection
Washing
Sulfur Oxides
Sand Filtration
Lagoons (ponds)
Pressure Filtration
Vacuum Filtration
Aluminum
Sodium Hydroxide
Sodium Carbonates
Leaching
Liquids
Calcium Oxides
Sludge
T7c. COSATI F.-el.i/Groun'
13B, 7A
Unlimited
19..Security Cla.ss (This
Report)
UNCLASSIFIED
20. Security Class (This
Page
UNCLASSIFIED
21. .'v-o. ot Kigcs
66
22. Price
FORM NTIS-35 (REV. 3-721
61
USCOMM-DC 149S2-P72
------- |