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

-------