RECLAMATION OF ALUMINUM FINISHING SLUDGES
                        by
               F.  Michael Saunders
           School  of Civil Engineering
        Georgia Tech Research Corporation
         Georgia Institute of Technology
              Atlanta,  Georgia 30332
       Cooperative Agreement CR810290-01-0
                 Project Officer
                 Thomas J.  Powers
Industrial Wastes and Toxics Technology Division
      Water Engineering Research Laboratory
              Cincinnati, Ohio M5268
               in cooperation with

          The Aluminum Association,  Inc.
                 Washington,  D.C.

          The Aluminum Extruders Council
             Rolling Meadows, Illinois

-------
                                DISCLAIMER
     The information in this document has been funded wholly or in part by
the United States Environmental Protection Agency under Cooperative
Agreement CR 810290-01-0 to the Georgia Tech Research Corporation.  It has
been subject to the Agency's peer and administrative review and it has been
approved for publication as an EPA document.  Mention of trade names or
commercial products does not constitute endorsement or recommendation for
use.
                                     11

-------
                                 FOREWORD
     The U.S. Environmental Protection Agency is charged by Congress with
protecting the Nation's land, air, and water systems.  Under a mandate of
national environmental laws, the Agency strives to formulate and implement
actions leading to a compatible balance between human activities and the
ability of natural systems to support and nurture life.  The Clean Water
Act, the Safe Drinking Water Act, and the Toxics Substances Contrl Act are
three of the major congressional laws that provide the framework for
restoring and maintaining the Integrity of our Nation's water, for
preserving and enhancing the water we drink, and for protecting the
environment from toxic substances.  These laws direct the EPA to perform
research to define our environmental problems, measure the impacts, and
search for solutions.

     The Water Engineering Research Laboratory is that component of EPA's
Research and Development program concerned with preventing, treating, and
managing municipal and Industrial wastewater discharges; establishing
practices to control and remove contaminants from drinking water and to
prevent its deterioration during storage and distribution; and assessing the
nature and controllability of releases of toxic substances to the air,
water, and land from manufacturing processes and subsequent product uses.
This publication Is one of the products of that research and provides a
vital communication link between the researcher and the user community.

     This report includes research results on the reclamation of aluminum-
finishing sludges as commercial products.  Results are presented for
enhanced dewatering of these industrial sludges and the acidic extraction of
aluminum from them.  Project results Indicate that sludge reclamation can be
achieved, producing a commercially marketable product and eliminating a
solid waste disposal problem for the industry.
                                       Francis T. Mayo, Director
                                       Water Engineering Research Laboratory
                                     111

-------
                                 ABSTRACT
     Aluminum anodizing plants may typically produce up to 500 metric tons/
month of finished aluminum extrusions.  In finishing aluminum extrusions,
approximately 3 to 5 percent of the mass of the extrusions is discharged as
soluble aluminum metal in plant wastewaters.  This waste aluminum is
typically neutralized and results in the production of a highly-gelatinous,
aluminum-hydroxide sludge suspension which, when thickened and dewatered,
can equal or exceed the mass of extruded aluminum products produced at a
plant.  This solid waste residue must then be disposed of in a landfill or
by other acceptable methods.  Solid waste reduction has an extremely high
priority in this industry and can be addressed through alterations in
aluminum finishing or waste treatment procedures or reclamation of the solid
waste.  This project investigated reclamation of dewatered sludges as
commercial-strength, aluminum sulfate solutions, i.e., liquid alum.

     Sludge dewatering with filter press systems was the focus of the
initial phase of the project.  Successful reclamation of dewatered sludges
is dependent on obtaining a minimum dry solids content of approximately 21
percent, if a commercial-strength liquid alum is to be produced.  Fixed-
volume, recessed-chamber filter presses were shown to be capable of
dewatering conventional-neutralization, aluminum-anodizing sludges to 25
percent and 27 percent at low (6-7 bar) and high (U-15 bar) pressures,
respectively.  Segregated neutralization sludges could be dewatered to
solids contents of U9 percent and 51 percent at low and high pressures,
respectively.  Blending small volumes (5 to 30 percent) of segregated-
neutralization suspensions with thickened segregated neutralization
suspensions markedly improved sludge dewatering properties.  A variable-
volume recessed-chamber filter press produced dewatered conventional-
neutralization sludges with solids contents of 25-31 percent.  It was,
therefore, demonstrated that fixed-volume and variable-volume filter presses
were capable of producing dewatered sludge cakes with solids contents of
greater than or equal to 21 percent and could be effectively used in
conjunction with the reclamation of aluminum-anodizing sludge as liquid
alum.

     Extraction of aluminum from dewatered aluminum-anodizing sludge cakes
is possible with addition of stoichioraetric quantities of sulfuric acid.
The exothermic, acidic extraction is extremely rapid and can be taken to
completion at temperatures of 50-100°C in 30 to 60 rain.  Commercial-strength
products of liquid alum (i.e., >_ 8 percent as A^Oj) can be produced from
conventional-neutralization, segregated-neutralizalton, and etch-recovery
solids.  Trace metal concentrations are of limited concern, with the
possible exception of nickel and tin concentrations in conventional-
neutralization sludges.  Reclamation of aluminum-anodizing sludges as  liquid

                                     iv

-------
alum was demonstrated to be an effective method for elimination of a-major
industrial solid-waste-disposal problem.

     This report was submitted in fulfillment of cooperative agreement
CR810290-01-0 by the Georgia Tech Research Corporation under the sponsorship
of the U.S. Environmental Protection Agency.   This report covers a period
from September 1982 to April 1987 and work was completed as of April 1987.

-------
                                 CONTENTS

                                                                        Page

Foreword	   iii
Abstract	    i v
Figures	  viii
Tables	   xii
Acknowledgment	  xv i i

   1 .   Introduction	     1
   2.   Conclusions	     3
   3.   Recommendations	     6
   H.   Materials and Methods	     7
            Aluminum-anodizing-plant sludges and wastewaters	     7
            Experimental equipment	     8
                 Pressure filtration systems	     8
                      Netzsch press	     8
                      JWI press	    10
                 Aluminum extract ion reactor	    11
                 Segregated neutralization reactor	    11
            Operational procedures	    11
                 Pressure filtration	    11
                      Netzsch press	    11
                      JWI press	    13
                 Aluminum extraction reactor	    13
                      Sulfuric acid dosage	    13
                      Aluminum extraction	    1 4
            Analyt i cal methods	    15
                 Suspended solids concentration	    15
                 Cake solids content	    15
                 Specific resistance	    15
                 Capillary Suction Time (CST)	    16
                 Specific gravity of aluminum extracts	    16
                 Aluminum concentration	    16
                 Trace metal analyses	    17
   5.   Sludge Dewatering by Pressure Filtration	    18
            Sludge suspensions	    18
                 Conventional neutralization suspensions	    18
                      Plant A suspensions	    18
                      Plant X suspensions	    22
                 Segregated neutralization suspensions	    22
            Fixed-volume pressure filtration	    2*4
                 Conventional neutralization suspensions	    2M
                      Plant A clarifier underflow (AC3)	    2U
                                     vi

-------
                        High-pressure filtration	    24
                        Low-pressure filtration	    31
                   Plant A neutralization basin effluent (AGO	    33
                        High-pressure filtration	    33
                        Low-pressure filtration	    41
                   Plant X clarifier underflow (ACS)	    ^8
                        High-pressure filtration	    48
                        Low-pressure filtration	    49
              Segregated neutralization suspensions	    59
                   Blends with Plant A suspensions (ACS)	    60
                        High-pressure filtration	    60
                        Low-pressure filtration	    62
                   Blends with Plant X suspensions (XC3)	    64
                        High-pressure filtration	    64
                        Low-pressure filtration	    66
         Variable-volume pressure filtration	    70
         Summary analysis of pressure filtration	    79
              Fixed-volume pressure filtration	    80
              Variable-volume pressure filtration	    87
6.  Liquid Alum Production from Dewatered Sludge Solids	    88
         Acidic extraction of aluminum from dewatered
         sludge cakes	    89
              Conventional neutralization sludge cakes (CN)	    89
              Segregated neutralization sludge cake (SN-1)	    98
              Etch recovery sludge cake (ER-1)	   101
              Blended sludge cakes	   106
                   Conventional neutralization/segregated
                   neutralization sludge cake (CN-2/SN-1)	   106
                   Conventional neutralization/etch recovery
                   sludge cake (CN-2/ER-1)	   109
         Metal content of sludges and liquid alum extracts	   111
              Selected aluminum-finishing suspensions and
              resi dues	   ill
              Liquid alum products	   11 6
7.  References	   119
                                  vii

-------
                                 FIGURES

Number                                                                  Page
  1       Schematic diagram of wastewater treatment plant
           at Plant A	      9

  2      Cumulative filtrate volume for AC3 suspensions
           in runs 6-9 during high-pressure filtration	     26

  3      Cumulative filtrate volume for ACS suspensions
           in runs 1U, 16 and 17 during high-pressure
           filtration	     27

  J4      Cumulative filtrate volume for ACS suspensions
           in runs 21-23 during high-pressure filtration	     28

  5      Cumulative filtrate volume for undiluted (runs 24-26)
           and diluted (runs 27-29) ACS suspensions during
           high-pressure filtration	     29

  6      Cumulative filtrate volume for ACS suspensions
           in runs 71, 101,  142 and 172 at high-pressure
           filtration	     30

  7      Cumulative filtrate volume for ACS suspensions
           in runs 10-13 during low-pressure filtration	     35

  8      Cumulative filtrate volume for undiluted (runs 31~33)
           and diluted (runs 34-37) ACS suspensions during
           low-pressure filtration	     36

  9      Cumulative filtrate volume for ACS suspensions
           in runs 72 and 102 at low-pressure filtration	     37

 10      Cumulative filtrate volume for ACS suspensions
           in runs J-21, J-61,  J-81 and J-91 during
           low pressure filtration	     38

 11       Cumulative filtrate volume for AC1 suspensions
           in runs 18-20 during high-pressure filtration	     42

 12      Cumulative filtrate volume for an AC1  suspension
           in run 41  during high-pressure filtration	     U3
                                   Vlll

-------
13      Cumulative filtrate volume for ACT suspensions
          during high-pressure (run 62) and low-pressure
          (run 61 ) filtration
        Cumulative filtrate volume for ACT suspensions
          in runs J-31 and J-32 during low-pressure
          filtration
15      Cumulative filtrate volume for AC1  suspensions
          in run J-71  during low-pressure filtration .................    U7

16      Cumulative filtrate volume for XC3  suspension
          in run 51  during high-pressure filtration ..................    50

17      Cumulative filtrate volume for XC3  suspensions
          in runs 81  and 82 during high-pressure
          filtration .................................................    51

18      Cumulative filtrate volume for XC3  suspension
          in run 121  during high-pressure filtration .................    52

19      Cumulative filtrate volume for XC3  suspension
          in run 52 during low-pressure filtration ...................    5^
20      Cumulative filtrate volume for XC3 suspension
          in runs 73 and 1^ during low-pressure
          filtration on Netzsch press ................................    55

21      Cumulative filtrate volume for XC3 suspension
          in runs J-1 1  and J-12 during low-pressure
          filtration on JWI press ....................................    56

22      Cumulative filtrate volume for XC3 suspension
          in run J-41 during low-pressure filtration
          on JWI press ...............................................    57

23      Cumulative filtrate volume for XC3 suspension
          in run J-1 21  during low-pressure filtration
          on JWI press ...............................................    58

24      Cumulative filtrate volume for AC3 and AS7 suspensions
          and Mended suspensions in runs 171, 173, 17*», 177
          and 178 during high-pressure filtration ....................    63

25      Cumulative filtrate volume for AC3 and AS7 suspensions
          and blended suspensions in runs J-81 through J-85
          during low-pressure filtration .............................    65

26      Cumulative filtrate volume for an AS7 suspension
          (run 93) and blends of XC3 and AS7 suspensions
          in runs 91-93 during high-pressure filtration ..............    67

-------
27      Cumulative filtrate volume for an AS7 suspension
          and blends of XC3 and AST suspensions in runs
          J-51 through J-53 during low-pressure filtration ...........    69

28      Cumulative filtrate volume for ACS suspension
          during run 141  using Netzsch diaphragm press ...............    72

29      Cumulative filtrate volume for AC3 suspension
          during runs 151-153 using Netzsch diaphragm
          press [[[    73

30      Cumulative filtrate volume for AC3 and blends
          of AC3 and AS7 suspensions during runs 172,
          175, 176 and 179 using Netzsch diaphragm
          press [[[    74

31      Cumulative filtrate volume for AC3 suspension
          during run 181  using Netzsch diaphragm press ...............    75

32      Cumulative filtrate volume for AC3 suspension
          during run 182 (filter portion only) using
              ich diaphragm press ....................................    76

                   filtrate volume for AC1 suspension
                 run 161  using Netzsch diaphragm press ...............    77

        Filtration rate data for series-20 runs with AC3
          suspensions for projection of ultimate filtrate
          volume [[[    81

35      Filtration rate data for series-30 runs with AC3
          suspensions for projection of ultimate filtrate
          volume [[[  °  82

36      Filtration rate data for selected runs with XC3
          suspension for projection of ultimate filtrate
          volume [[[    83

37      Filtration rate data for selected AC3 and AS7
          suspensions and blends of each for projection
          of ultimate filtrate volume ................................    84
38      Temperature of reactor contents during acidic

-------
        Filtrate aluminum concentration for sulfuric-acid
          extractions of sludge cake CN-1 during runs 1-1,
          1-2 and 1-3
        Filtrate aluminum concentration for sulfuric-acid
          extractions of sludge cake CN-2 during runs 2-1
          and 2-2
42      Filtrate aluminum concentration for sulfuric-acid
          extractions of sludge cake SN-1  during runs 3~1 •
          3-2 and 3-3 ................................................   100

43      Filtrate aluminum concentration for sulfuric-acid
          extractions of sludge cake ER-1  during runs 4-1
          through 4-4 ................................................   100

44      Filtrate aluminum concentrations for sulfuric-acid
          extraction of sludge cake CN-2/SN-1  during run 5-1 .........   1 08

45      Filtrate aluminum concentration for sulfuric-acid
          extractions of sludge cake CN/ER-1 during run 6-1 ..........   108
                                   XI

-------
                                 TABLES

                                                                       Page

        Characteristics of neutralization basin effluent,
          clarifier underflow and vacuum-filter cake
          during filtration studies ..................................     19

        Characteristics of neutralization basin effluent at
          Plant A prior to polymer conditioning ......................     20

        Characteristics of clarifier underflow suspension (XC3)
          from Plant X as used during filtration studies .............     23

        Characteristics of segregated-neutralization suspension
          produced from spent caustic etch and acid at Plant A
          for filtration studies .....................................     23

        Characteristics of clarifier underflow suspensions
          for high-pressure, fixed-volume pressure
          filtration on Netzsch press ................................     25

        Results for high-pressure, fixed-volume filtration
          of clarifier underflow suspensions (AC3)
          on Netzsch press ...........................................     32

        Characteristics of clarifier underflow suspensions (AC3)
          for low-pressure, fixed-volume, pressure filtration
          on Netzsch and JWI presses .................................     3U

        Results for low-pressure, fixed-volume filtration
          of clarifier underflow suspensions (ACS)
          on Netzsch and JWI presses .................................     39

        Characteristics of neutralization basin effluent
          suspensions (AC1) for high-pressure, fixed-volume
          filtration on Netzsch press ................................     UO
10      Results for high-pressure, fixed-volume filtration
          of neutralization basin effluent suspensions (AC1 )
          on Netzsch press ...........................................    U i

11      Characteristics of neutralization basin effluent
          suspensions (AC! ) for low-pressure, fixed-volume
          filtration on Netzsch and JWI presses ......................    U5
                                   xii

-------
12      Results for low-pressure, fixed-volume filtration
          of neutralization basin effluent suspensions (AC1 )
          on Netzsch and JWI presses .................................    48

13      Characteristics of clarifier underflow suspensions (XC3)
          for high-pressure, fixed-volume filtration
          on Netzsch press ...........................................    49

11      Results for high-pressure, fixed-volume filtration
          of clarifier underflow suspensions (XC3)
          on Netzsch press ...........................................    49

15      Characteristics of clarifier underflow suspensions (XC3)
          for low-pressure,  fixed-volume pressure filtration
          on Netzsch and JWI presses .................. •. ..............    53

16      Results for low-pressure, fixed-volume filtration
          of clarifier underflow suspension (XC3)
          on Netzsch and JWI presses .................................    59

17      Characteristics of clarifier underflow (AC3) and
          segregated neutralization (AS?) suspensions and
          blends of these suspensions used in high-pressure
          filtration on Netzsch press ................................    60

18      Results for high-pressure, fixed-volume filtration
          of clarifier underflow (AC3) and segregated-
          neutrallzation (AS7) suspensions and blends
          of these suspensions .......................................    62

19      Characteristics of clarifier underflow (AC3) and
          segregated neutralization (AST) suspensions and
          blends of these suspensions used in low-pressure
          filtration on JWI  press ....................................    64
20      Results for low-pressure, fixed-volume filtration
          of clarifier underflow (AC3) and segregated-
          neutralization (AST) suspensions and blends
          of these suspensions .......................................    64

21      Characteristics of a segregated neutralization (AST)
          suspension and blends with a clarifier underflow
          suspension used in high-pressure filtration
          on Netzsch press ...........................................    66

22      Results for high-pressure, fixed-volume filtration
          of a segregated neutralization (AST) suspension
          and blends with a clarifier underflow (XC3)
          suspension .................................................    68
                                   Xlll

-------
23      Characteristics of a segregated neutralization (AST)
          suspension and blends with a clarifier underflow
          (XC3) suspension used in low-pressure filtration
          on JWI press	    68

24      Results for high-pressure, fixed-volume filtration
          of a segregated neutralization (AS7) suspension
          and blends with a clarifier underflow (XC3)
          suspension	    68

25      Comparison of low- and high-pressure filtration
          for blends of clarifier underflow (AC3; XC3)
          and segregated neutralization (AST) suspensions	    70

26      Characteristics of clarifier underflow (AC3) and
          neutralization basin (AC1) suspensions for
          dewatering by variable-volume pressure filtration	    71

27      Results for variable-volume, pressure filtration
          of AC3 and AC1 suspensions on Netzsch press	    78

28      Characteristics of clarifier underflow (AC3) and
          blends with a segregated neutralization (AS7)
          suspension for variable-volume,  pressure
          filtration	    79

29      Results for variable-volume, pressure filtration
          of AC3 and blends of AC3 and AS7 suspensions	    79

30      Predicted solids contents of dewatered cakes
          for suspensions from Plant A during high-
          and low-pressure filtration	    85

31      Predicted solids contents of dewatered cakes for XC3   »
          suspensions during high- and low-pressure
          filtration	    86

32      Predicted solids contents of dewatered cakes for an
          AC3 suspension and blends of AC3 and AS7
          suspensions during high-pressure filtration	    87

33      Characteristics of dewatered conventional neutrali-
          zation- sludge cakes CN-1 and CN-2	    89

31*      Initial experimental conditions for extractions
          of conventional neutralization sludge cake CN-1	    90

35      Initial experimental conditions for extractions
          of conventional neutralization sludge cake CN-2	    91
                                    XIV

-------
36      Actual and control temperatures for extractions
          of conventional neutralization sludge cakes
          CN-1 and CN-2	    93

37      Material balances on total mass and aluminum
          for extractions of conventional neutralization
          sludge cakes CN-1  and CN-2	    95

38      Characteristics of liquid alum produced by acidic
          extraction of conventional neutralization sludge
          cakes CN-1 and CN-2	    97

39      Characteristics of dewatered segregated neutrali-
          zation sludge cake SN-1	    98

40      Initial experimental conditions for extractions
          of segregated neutralization sludge cake SN-1	    98

U1      Material balances on total mass and aluminum
          for extractions of segregated neutralization
          sludge cake SN-1	    99

42      Characteristics of liquid alum produced by acidic
          extraction of segregated neutralization sludge
          cake SN-1	   101

43      Characteristics of dewatered etch recovery sludge
        cake ER-1	   1 02

44      Initial experimental conditions for extractions
          of etch recovery sludge cake ER-1	   103

45      Material balances on total mass and aluminum
        for extractions of etch recovery sludge cake ER-1	   1 04

46      Characteristics of liquid alum produced by acidic
          extraction of etch recovery sludge cake ER-1	   105

47      Characteristics of combined dewatered conventional
          neutralization/segregated neutralization sludge
          cakes CN-2 and SN-1	   106

48      Initial, experimental conditions for extraction
          of combined conventional neutralization/
          segregated neutralization sludge cakes
          CN-2 and SN-1	   1 06

49      Material balances on total mass and aluminum
          for extraction of combined conventional
          neutralization/segregated neutralization
          sludge cakes CN-2 and SN-1	    107


                                    XV

-------
50      Final product quality data for liquid alum produced
          by acidic extraction of combined conventional
          neutralization and segregated neutralization
          sludge cakes CN-2 and SN-1	   109

51      Characteristics of combined conventional neutralization
          and etch recovery sludge cakes CN-2 and ER-1	   109

52      Initial experimental conditions for extraction
          of combined conventional neutralization and
          etch recovery sludge cakes CN-2 and ER-1	   110

53      Material balances on total mass and aluminum
          for extraction of combined conventional
          neutralization and etch recovery sludge
          cakes CN-2 and ER-1	   111

51*      Characteristics of liquid alum produced by acidic
          extraction of combined conventional neutralization
          and etch recovery sludge cakes CN-2 and- ER-1	   112

55      Metal content of dewatered conventional neutralization
          sludge cakes from selected aluminum-anodizing plants	   113

56      Projected increases in metal concentrations in
          coagulated drinking water treated with alum,
          produced from dewatered conventional neutralization
          sludge cakes, at an aluminum concentration of 10 mg/L	   115

57      Metal concentrations for a segregated neutralization
        suspension produced at Plant H	   11 6

58      Metal content of liquid-alum samples produced
          from CN-2, SN-1 and ER-1 sludge cakes and
          three commercial alum products	   117
                                   XVI

-------
                              ACKNOWLEDGMENT
     The research in this report was conducted with funding provided by the
U.S. Environmental Protection Agency and the School of Civil Engineering at
the Georgia Institute of Technology-  This is the third report in a series
focused on characterization, treatment, reclamation and disposal of
aluminum-anodizing wastewaters and sludges.  The continuing support of the
Aluminum Extruders Council of Rolling Meadows, IL and The Aluminum
Association of Washington, D.C. as cosponsors of this report and sponsors of
two previous project reports is acknowledged.  This study would not have
been possible without their excellent assistance in contacting numerous
aluminum-anodizing plants regarding participation and providing information
about the industry.

     The assistance provided by personnel at the plant where pilot-scale
dewatering studies were conducted was critical to the successful completion
of the study.  Their continued assistance and support in providing space,
supplies, personnel support and changes in operational practices were
appreciated.

     Filter press systems used on the project were provided by DMP
Corporation of Fort Mill, SC and Netzsch, Inc. of Exton, PA.  Special
appreciation goes to Mr. David Masters of DMP Corporation for the loan of
the JWI press and for his excellent assistance with liquid-alum production
systems.  Mr. Paul L. Scholtyssek of Netzsch, Inc. provided considerable
technical assistance in conjunction with the low-pressure, high-pressure and
diaphragm filter press systems.

     The data presented in this report were collected and analyzed with the
assistance of numerous research assistants.  Messrs. Steven Cox, Mark
Bradley, Randall Shaw, Michael Roeder and Gray Saunders assisted with
monitoring and operating the filter presses during the pilot-scale
dewatering studies.  Messrs. Randall Shaw and Jamshid Havash coordinated and
conducted the acidic aluminum extraction studies, while Mr. Michael Roeder
assisted in graphical presentation and analysis of project data.  Finally,
Ms. Henrietta Bowman and Mrs. Elaine Sharpe typed and organized the report
manuscript.  The support and assistance provided by these individuals was
greatly appreciated.

     Finally, appreciation is expressed to Ms. Lynn Apel and Mr. Thomas
Powers, both of whom served at times as Project Officer.  Special appre-
ciation is noted for the understanding, assistance and diligent support
provided during the development and review of this final project report by
Mr. Thomas Powers.
                                    xvii

-------
                                 SECTION 1

                               INTRODUCTION
     The use of extruded aluminum products for architectural and structural
purposes in domestic and industrial markets in the United States is well
established and continues to expand.  Approximately 1  million metric tons of
extruded products are annually anodized, painted or finished prior to use.
Of the finishing processes employed, anodizing systems present some unique
environmental problems.  To anodize an extruded aluminum alloy, surface
oxides, imperfections and deposits, resulting from high-temperature
extrusion and manual handling, must be removed.  This is commonly
accomplished by chemical etching and results in the removal of a thin layer
of the alloy metal from the surface of an extrusion.  These alloy-metal and
other finishing wastes are continuously discharged for treatment and
disposal.

     Conventional neutralization to neutral pH values and subsequent treat-
ment of anodizing wastewaters results in formation of aluminum-hydroxide and
other metal precipitates.  Due to the highly hydrophilic characteristics of
aluminum hydroxide, thickened sludge suspensions are usually very dilute
(e.g., 1-2 percent dry solids content).  In addition,  these suspensions are
difficult to dewater and typically contain high levels of residual moisture.
The mass of dewatered sludge solids produced for disposal is extremely high
relative to production rates for finished aluminum.  For example, the mass
of wet dewatered sludges produced at two anodizing plants, with production
capacities of approximately 500 metric tons/month, was estimated to be 55
and 85 percent of the mass of finished aluminum products, assuming highly
effective sludge dewatering could be accomplished (Saunders e_t al., 1982;
Saunders e_t al_., 198*0.  Without highly efficient dewatering, the final
sludge mass would approach, or exceed, the mass rate of production of
finished product.  Therefore, regardless of issues related to environmental
effects of chemicals contained in the dewatered sludges, the high relative
mass, and the resulting high relative volume, of dewatered sludge solids
produced constitute one of the major environmental problems with which the
industry is faced.  The research conducted and presented herein was focused
directly on this issue, with major emphasis placed on reduction in the mass
of sludge solids for disposal through use of extraction of aluminum from the
sludges to produce a reclaimed, marketable product.

     This report is the third in a series reporting on research focused on
characterization, treatment, reclamation and disposal of aluminum anodizing
wastewaters and sludges.  The first report by Saunders e_t al.  (1982) was
focused on wastewater characterization and evaluation of conventional
treatment technology.  The focus of the second report by  (Saunders e_t  al. ,

                                      1

-------
(1984) was on evaluation of innovative treatment technologies with potential
for application in the industry-  Of the technologies investigated, sludge
reclamation using acidic extraction to produce a commercial-strength
solution of aluminum sulfate was shown to have great potential for
elimination of, or dramatic reduction in, sludge disposal requirements.
This topic was therefore selected for further analysis and research and is
the subject of this report.

     The research project was conducted in two phases.  The initial phase
was focused on the high moisture content of dewatered sludges.  Previous
research (Saunders et al.,  1984) indicated that a sludge moisture content of
less than approximately 79  percent (i.e., sludge solids content greater than
21 percent) was required to be able to produce a commercial-strength
solution of aluminum sulfate.  Earlier research indicated that this level of
dewatering was not routinely achieved at full-scale facilities and therefore
presented a major impediment to implementation of reclamation of dewatered
sludge as aluminum sulfate.

     Pressure filtration using recessed-plate filter presses was reported to
be the most effective approach to producing dewatered sludges with elevated
solids contents (EPA, 1982; EPA, 1986).  With this objective being central
to the overall success of sludge reclamation, the initial phase of the
project included a pilot-scale evaluation of the use of filter press systems
to dewater thickened aluminum-finishing sludge suspensions.  Two pilot-scale
presses were used to investigate the performance of fixed-volume,
recessed-plate filter presses at low and high pressures and variable-volume,
recessed-plate filter presses, i.e., diaphragm filter press.  This phase of
the study was conducted at  "an aluminum-anodizing plant (Plant A) using
thickened sludge suspensions routinely produced at the plant as feed
suspensions for the filter  presses.  Thickened sludge suspensions were also
obtained from a second aluminum-anodizing plant (Plant X) and transported to
Plant A for dewatering studies with the filter presses.

     The second phase of the project was focused on extraction of aluminum
from dewatered sludges using sulfuric acid.  This phase was conducted using
laboratory-scale equipment  with sludges produced by pressure filtration at
the local aluminum-anodizing plant.

     The overall objective  of the project was to firmly establish the
feasibility of the production of commercial-strength solutions of aluminum
sulfate from aluminum-finishing sludges.  Issues related to the use of
pressure filtration for dewatering; incorporation of other innovative
treatment technologies; chemical and physical requirements for aluminum
extraction; and quality of  aluminum-sulfate products were also investigated.

-------
                                 SECTION 2

                                CONCLUSIONS
     The research was focused on filter-press dewatering and reclamation of
three types of aluminum-anodizing sludges as commercial-strength liquid alum.
Numerous conclusions were drawn from the study.

     1.   Conventional neutralization sludge suspensions obtained from Plant
         A had suspended solids concentrations of 0.76 to 5.12 g/L and
         averaged 2.5 g/L, and following gravity thickening had suspended
         solids concentrations which ranged from 18 to 62 g/L and averaged
         36 g/L.   Gravity thickened suspensions from the clarifier underflow
         had an average specific resistance of 3.7 Tin/kg and a specific-
         resistance cake solids content of 13-2 percent.  Dewatered sludge
         cakes obtained from the vacuum filter at Plant A had solids
         contents which ranged from 9.9 to 15.9 percent and averaged 12.4
         percent.

     2.   A conventional neutralization suspension following gravity thicken-
         ing at Plant X had an average suspended solids concentration of
         20.7 g/L, a specific resistance of 2.8 Tm/kg and a specific-
         resistance cake solids of 7.1 percent.

     3.   Segregated neutralization suspensions produced experimentally by
         direct neutralization of spent caustic etch with conventional
         anodizing acid at Plant A had suspended solids concentrations
         ranging from 80.1 to 180.1 g/L; an average specific resistance of
         1.4 Tm/kg; and an average specific-resistance cake solids of 45
         percent.

     4.   A devatered etch-recovery sludge cake had a solids content of 91.6
         percent.

     5.   Pressure filtration studies were conducted with two pilot-scale,
         fixed-volume, recessed-chamber filter presses at low (6-7 bar) and
         high (14-15 bar) pressures.  Replicate runs of variable filtration
         time were virtually identical with respect to cumulative filtrate
         volume collected.

     6.   With data for filtration rate and mass and solids contents of
         dewatered cakes it was projected that the density of the aluminum-
         anodizing solids was 3200 kg/m3.

     7.   The solids content of dewatered sludge cakes at the ultimate

-------
     completion of a filter press run were projected from filtration
     rate data.  Conventional neutralization suspensions from Plant A
     had ultimate, dewatered-cake, solids contents of 23-4 percent and
     27 percent, respectively, at low (6-7 bar) and high (14-15 bar)
     pressure filtration for suspensions with influent suspended solids
     concentrations of approximately 16-57 g/L.  Plant X suspensions
     with an average suspended solids concentration of 20.7 g/L had
     ultimate dewatered-cake solids contents of 17.6 percent and 24.5
     percent at low and high pressure filtration,  respectively.

 8.  Segregated neutralization suspensions could be effectively
     dewatered separately and resulted in major improvement in the
     dewatering of conventional neutralization suspensions when blended
     with them.  At low and high pressures,  ultimate cake solids
     contents of 48.6 and 51.3 percent,  respectively, were achieved with
     segregated neutralization suspensions.   A batch, segregated-
     neutralization suspension produced  at Plant A following a full-
     scale dump of a caustic etch tank was dewatered to 42 and 45
     percent with low- and high-pressure filtration, respectively.
     Blends of segregated neutralization suspensions at 5 to 30 percent
     volumetric ratios with conventional neutralization suspensions
     resulted in ultimate solids contents of 33 to 39 percent with
     high-pressure filtration and 31  to  37 percent with low-pressure
     filtration.

 9.  A diaphragm press was used effectively to dewater all aluminum
     anodizing suspensions.  Conventional neutralization suspensions had
     final cake solids contents of 25.4  to 31.2 percent, while 5 to 30
     percent volumetric blends of segregated neutralization suspensions
     with conventional neutralization suspensions had solids contents of
     31  to 43 percent.

10.  Commercial-strength solutions of aluminum sulfate, i.e., liquid
     alum, can be effectively and rapidly produced with the addition of
     sulfuric acid.  Addition of stoichiometric quantities of acid,
     based on sludge aluminum content, resulted in virtually complete
     extraction within 30-60 min.

11.  Following the addition of acid and  the initial elevation of
     temperature to > 95°C, maintenance  of temperatures between 50-90°C
     had minimal impact on the extent of aluminum extraction.

12.  Conventional neutralization sludge  cakes with solids contents of
     17.4 to 18.1 percent were extracted to produce liquid alum with
     concentrations of 7.4 to 8.8 percent as A^C^.  A total of 95 to 99
     percent of the initial suspended solids were destroyed in the
     extraction and 93 to 97 percent of  the aluminum was extracted.

13.  Segregated neutralization sludge cakes with solids contents of 36.8
     percent were extracted to produce liquid alum with concentrations
     of "8.1  to 9.0 percent as A^Og with the addition of water equal to
     80 to 100 percent of the mass of wet sludge extracted.

-------
"U.  An etch recovery sludge with a solids content of 91.6 percent
     produced a liquid alum with a concentration of 8.3 to 9.2 percent
     as A^Oj, following addition of water equal to 200 to 370 percent
     of the wet mass of etch recovery sludge added.  Acidic extraction
     resulted in destruction of 5^ to 85 percent of the suspended solids
     and extraction of 70 to 85 percent of the aluminum.

15.  Addition of segregated-neutralization or etch-recovery solids to
     conventional neutralization solids increased the aluminum content
     of the blended sludge and could be effectively extracted to easily
     produce commercial-strength liquid alum.

16.  A survey of numerous aluminum-anodizing sludges indicated that
     aluminum, sodium and iron were the major metals, accounting for the
     following respective percentages of total dry .solids:  22 to 48
     percent; 0.6 to 8.9 percent;  and 0.17 to 1.13 percent.  These three
     metals accounted for 27 to 57 percent of sludge dry solids.

17.  Major heavy metals were nickel and tin with concentrations ranging
     from trace levels to 13,300 mg/kg (or 1.33 percent of dry solids).
     Other metals found occasionally at concentrations of 300 to 1000
     mg/kg were chromium, copper,  lead and zinc.  Arsenic, cadmium,
     mercury, silver and selenium were only found at trace «2 mg/kg)
     levels.

18-  The cadmium, chromium and iron concentrations of liquid alum
     produced from conventional neutralization  sludges were less than
     those of commercial products, while lead,  silver and zinc concen-
     trations were slightly above those for commercial product samples.
     The concentrations of nickel and tin were  6- to  17-fold higher  than
     those  in commercial product samples.  The  high nickel and tin
     concentrations were attributed to dragout  from the two-step
     anodizing process and to the use of nickel  in seal tanks.
     Segregation of these wastes from plant wastewaters may be needed  to
     eliminate them from the sludges produced  for extraction and
     reclamation.

-------
                                 SECTION 3

                              RECOMMENDATIONS
     The experimental results presented herein confirm that virtually
complete reclamation of aluminum-anodizing sludges as a commercial strength
product is possible.  Effective dewatering of aluminum-anodizing sludges is
a necessity,  as is control of heavy metal wastewaters discharged to the
plant treatment systems.  It is recommended that a national inventory of
aluminum-finishing sludges be made to establish the quantity of aluminum
available for reclamation as aluminum sulfate.  The intensive survey should
be focused on characterization of the sludges with respect to solids
content, extractable aluminum and other extractable metals, including toxic
heavy metals.

     It is recommended that a treatment technology series be developed to
summarize advancements in treatment and reclamation of aluminum-finishing
sludges and wastewaters.  The issues to be addressed should include
wastewater characterization, sludge thickening and dewatering, lime recovery
of etching wastes, advances in filter press systems and acidic extraction
for reclamation of sludge aluminum as liquid alum.

-------
                                 SECTION 4

                           MATERIALS AND METHODS


ALUMINUM-ANODIZING-PLANT SLUDGES AND WASTEWATERS

     The initial phase of the project was focused on evaluation of pressure
filtration of thickened aluminum-anodizing suspensions using pilot-scale and
laboratory-scale pressure filtration equipment.  These studies were
conducted at the site of an aluminum-anodizing plant in the metropolitan
Atlanta area.  This plant, referred to herein as "Plant A", is an integrated
aluminum extrusion and anodizing facility with extensive fabrication
capabilities for use in the production of prefabricated architectural
components.

     The aluminum finishing line at Plant A employed a system designed for
production of anodized extrusions using conventional sulfuric-acid and
electrolytic-coloring anodizing procedures to produce a range of clear,
bronze and black finishes.  The anodizing line was a single-pass system
which included the following sequential finishing steps:  alkaline cleaner;
caustic etch; desmut; sulfuric-acid anodize; electrolytic coloring; and seal.
Each finishing step, except the final seal, was followed by a reclaimed-
water rinse.  Finishing solutions were indicated as being standard
proprietary formulations typically employed in the industry.  Further
specific elaboration was not provided.  The caustic etch finishing solution
was a proprietary "no-dump" formulation which resulted in minimal direct
wastage of spent etch.  However, the tank contents were annually discharged
to waste because of the accumulation of impurities.  In addition, when
required for wastewater treatment, 1- to 3~n^ (260 to 800 gallon) quantities
of caustic etch were discharged to a small holding tank for use in
wastewater neutralization.  Spent concentrated acids from desmut, anodizing
and electrolytic coloring were accumulated in a large, 19-m3 holding tank
for use with spent caustic etch in wastewater neutralization.

     Wastes discharged from the anodizing line were contained in one of
three separate flows, i.e., spent finishing acids; spent caustic etch; and
rinsewaters.  The rinsewater flow, in addition to rinse tank overflows,
contained all spills and dragout from finishing tanks which were discharged
to the collection pit immediately below the finishing-line tanks.  Spent
finishing-acid and caustic-etch flows were collected and stored separately
for use in wastewater neutralization.

     The wastewater treatment system included wastewater neutralization;
pump transfer; polymer flocculation; and gravity sedimentation.  Clarified
wastewater was discharged to a storage tank for recycle and use as a

-------
reclaimed rinsewater on the anodizing line.  Thickened underflow suspensions
from the sedimentation basin were discharged to a rotary-drum vacuum filter.
Filtrate water was discharged to the neutralization system and dewatered
sludge solids were accumulated in dumpster tanks for ultimate disposal in a
contract landfill.  A schematic diagram of the wastewater treatment system
is presented in Figure 1.

     The suspensions routinely used at Plant A were those collected from the
underflow of the gravity sedimentation basin.  No preferential operational
control over the wastewater treatment system or the sedimentation basin was
provided by plant personnel for the study because the treatment system was
operated near the maximum available capacity throughout the experimental
study.  Therefore, wastewater suspensions were those available in the
influent line to the vacuum filter.

     On selected occasions it was necessary to examine sludge suspensions
which were not conditioned with polymer.  In these few instances,
neutralized suspensions from the third stage of the neutralization basin
were pumped into numerous 0.2-m3 containers and allowed to gravity thicken.
Clarified supernatant liquid was withdrawn by siphon and the thickened
sludge samples were combined for testing purposes.

     On one occasion, a thickened sludge suspension from another aluminum
anodizing plant, Plant X, was examined in the pressure filtration portion of
the study.  This local plant was similar to Plant A with respect to both the
finishing line and the wastewater treatment system.  Plant X, however, did
not employ electrolytic-color anodizing but used integral-color anodizing to
produce bronze finishes.  A 1.5-m3 volume of a gravity thickened suspension
was transported in 0.2-m3 containers to Plant A for examination.

     Finally, wastewater samples were provided by numerous plants for metal
analysis and characterization with respect to alum production.  In addition
to Plants A and X, other plants (i.e., Plants B, C, D, E, F and I) provided
conventional-neutralization, segregated-neutralization and etch-recovery
suspensions and dewatered sludge cakes for analysis.
EXPERIMENTAL EQUIPMENT

Pressure Filtration Systems

     Two filter presses were used to examine the dewatering properties of
thickened suspensions from aluminum anodizing lines.

Netzsch Press—
     A Model 470 filter press manufactured by Netzsch, Inc. of Exton, PA was
used extensively throughout the initial phase of the project.  The filter
press system was composed of a progressive-cavity pump; an In-line
accumulator tank with a separate air-compressor system; a set of U?0 mm x
U70 mm ungasketed polypropylene plates with filter cloths; and a manual,
hydraulic-pump closure system.  The fixed-volume recessed plates formed a
chamber with a dewatered-cake thickness of 30 mm; had a total filter area  of

-------
                  Influent
                Wastewater-
           Combined Rinsewaters
                 and Spills
NEUTRALIZATION
     BASIN
      Polymer
    CLARIF1ER
                              Spent Acid
                              Spent Etch and
                              Virgin Caustic
  Filtrate
    to
Wastewater
  Influent
            Dewatered
             Sludge
                                                            Landfill
                                 ROTARY
                                 VACUUM
                                  FILTER
                 Clarified
                 Effluent
                  Plant
                 Recycle
                        STORAGE
                          TANK
                                   Sewer
                                 Discharge
                                 Overflow
   Figure  1.    Schematic  diagram of wastewater  treatment
                system at  Plant  A.

-------
0.346 m2/chamber; and had a chamber volume of 4.6 x 10~3 m3/chamber.  The
filter cloth used throughout the study was a multifilaraent polypropylene
cloth (style number:  42614-3-HS; weight:  475 g/m2 (14 oz/yd2); weave:
oxford;  count:  75 x 21;  permeability: 0.6-0.9 m3/m2-min (2-3 cfm/ft2);
manufacturer:  C. Goodman and Company, Inc., Paterson, NJ).  Initial testing
by the manufacturer of the filter press indicated this to be the most
appropriate media for the sludge suspensions being examined.  The
performance of the media was excellent throughout the study and no further
evaluation of media was necessary.

     The pressure regulation system for the filter press allowed for
operation over a broad range of pressures and within narrow limits.
Automatic adjustment of the internal pressure of the influent sludge
suspension using the accumulator tank allowed for operation within pressure
limits of ± 0.5 bar (± 50 kPa).  When operated to simulate a filter press at
low pressure, the range of pressure was manually set at 6-7 bar (600-700
kPa), while high-pressure operation was at 14-15 bar (T400-1500 kPa).  These
ranges are typical of those used in practice for similar systems.   For
example, maximum operation pressures for low- and high-pressure systems were
indicated by EPA (1986) to be 6.9 bar (690 kPa) and 15.5 bar (1550 kPa).  It
is important to note that unlike many earlier plate-and-frame and filter
press systems, the operating pressure for these systems was constant
throughout a run and did not increase with time through a run.  This was
accomplished by use of a pressure sensor on the accumulator tank to control
the feed pump.  When the pressure in the accumulator tank feeding the press
dropped to the lower set-point pressure (i.e., 6 bar or 14 bar), the feed
pump was activated and the accumulator tank and filter chambers .were
gradually pressurized to the higher set-point pressure (i.e., 7 bar or 15
bar).  This operating mode minimized energy requirements for the feed pump
and provided for constant-pressure filtration.

     In addition to operation of the Netzsch press with fixed-volume,
recessed-chamber plates at low and high pressures, variable-volume,
recessed-chamber filtration was examined.  The variable-volume plates, or
diaphragm plates, were similar in construction to the fixed-volume plates
with the exception that each chamber contained an inflatable bladder used to
compress or "squeeze" a sludge cake prior to discharge.  The internal
bladder  was pressurized using compressed nitrogen to a maximum pressure of
15 bar (1500 kPa).  The filter cloth employed was identical to that used on
fixed-volume plates.

JWI Press—
     A laboratory-scale filter press, manufactured by JWI, Inc. of Holland,
MI and provided for use on the project by DMP Corporation of Fort Mill, SC,
was used in parallel with the larger Netzsch filter press.  The JWI press
was composed of a two-stage, air-driven diaphragm pump; a set of 250 mm x
250 mm,  ungasketed polypropylene plates with filter cloths; and a manual
hydraulic-pump closure system.  The fixed-volume recessed plates formed a
chamber  with a dewatered-cake thickness of 25 mm; had a total filter area of
7.63 x 10~2 m2/chamber; and had a chamber volume of 9.5 x 10"^ m^/chamber-
The filter cloth used was identical in composition to the raultifilament
polypropylene cloth used on the Netzsch press.  The JWI press did not have

                                     10

-------
an accumulator tank or any other pressure-regulation system.  Sludge
suspensions were fed into the filter press under the direct pressure
provided by the diaphragm pump, which was driven using an in-plant
compressor system.  Filtration pressures ranged from 6 bar (600 kPa) to
7 bar (700 kPa).

Aluminum Extraction Reactor

     Acidic extractions of dewatered sludge cakes were conducted in a
laboratory-scale batch reactor.  An insulated 2-L pyrex beaker with a
watch-glass cover and placed on a laboratory hot plate was the basic reactor.
A variable-speed laboratory mixer was used to continuously mix the contents
of the reactor during an extraction.  The hot plate was controlled by a
thermostat-probe placed in the reactor and was used to maintain reaction
temperatures of 50 to 90°C.

Segregated Neutralization Reactor

     On several occasions, segregated neutralization of small volumes of
caustic etch and spent acid was conducted at Plant A using a plastic-lined,
0.2-m3 container.  An approximate volume of 0.1 m3 of spent caustic obtained
directly from the caustic etch tank was placed in the container.  Spent
acid, obtained directly from a sulfuric-acid anodizing tank, was pumped with
peristaltic tubing pumps into the container.  A high-speed mixer was used to
mix the container.  Pumping was discontinued when the pH of the neutralized
suspension reached a value of approximately 9.5.  The neutralized suspension
was then used directly in pressure filtration studies.
OPERATIONAL PROCEDURES

Pressure Filtration

     The Netzsch press was used as the primary filtration system throughout
the study.   The operational procedures were virtually identical for low- and
high-pressure operation with fixed-volume, recessed-chamber plates and
operation as a diaphragm press was initially similar to the low-pressure
system.  The JWI press was used as a secondary system to examine
low-pressure, fixed-volume, recessed-chamber filtration during the initial
phase of the study.  The JWI press was the only unit used for preparation of
dewatered sludge cakes for the subsequent aluminum extraction studies.

Netzsch Press—
     To initiate a filtration run, a sample of a thickened sludge suspension
from a clarifier" underflow was placed in a 0.2-m3 plastic-lined container
and mixed with two high-speed mixers placed at different depths.  A sample
of the mixed suspension was taken for analysis of suspended solids concentra-
tion, specific resistance and Capillary Suction Time (CST) to characterize
the feed suspension.  Typically, a large volume (e.g., 0.5 to 1 m3) of a
sludge suspension was collected in three to five 0.2-m3 containers at one
time to allow for a series of runs on a similar sludge suspension to be
conducted over a 1 - to 3-day period.  The similarity of the various samples


                                      11

-------
included in such a series was confirmed by collection of an aliquot of a
suspension in the feed reservoir immediately before each run and by analysis
of the sample as indicated above.

     To initiate filtration runs, minimum and maximum set points for
filtration pressure were set at 6 and 7 bar and 14 and 15 bar for low- and
high-pressure runs, respectively.  The chamber plates were aligned and the
press was closed using the manual hydraulic jack.  The main control switch
was activated and the pump was automatically started.  The sludge suspension
was pumped simultaneously into the accumulator tank and the filter press.
In less than 30 seconds after starting the pump, filtrate began to be
discharged from the press.  At this time a stopwatch was activated to
monitor elapsed filtration time.  Within 2 to 3 min of starting the pump,
the maximum set-point pressure was reached and the pump was automatically
stopped.  A check valve on the pump discharge prevented backflow through the
pump and the pressure in the filter press was maintained with the
pressurized suspension in the accumulator tank.  When the pressure in the
filter press dropped below the minimum set-point pressure, the pump was
automatically activated and additional sludge was pumped into the press
until the maximum set-point pressure was reached.  This pumping cycle was
continued throughout the length of a filtration run.

     The volume of filtrate produced was recorded at frequent intervals
initially (e.g., at 30-s to 2-rain intervals) and at longer intervals (e.g.,
at 10- to 20-min intervals) as the filtration rate decreased through a run.
The turbidity of the filtrate was monitored during the initial fifteen runs.
The filtrate was virtually free of suspended solids as indicated by
turbidity readings of less than 6 NTU for all runs.  No further monitoring,
other than visual confirmation of the lack of turbidity (i.e., <5 NTU) in
the filtrate produced, was necessary or conducted for subsequent runs.  The
solids capture throughout all studies conducted was greater than 99.9
percent.

     At the completion of a filtration run, internal pressure in the press
was reduced to ambient pressure and the hydraulic closure device was
released.  Dewatered sludge cakes (usually two) were removed intact from the
press and weighed on an industrial balance.  The dewatered cakes had masses
ranging from approximately 5 to 6 kg and were weighed to the nearest 30 g
(I.e., ± 0.5 to 0.6 percent).  After weighing a cake, portions of it were
collected from each quadrant, being sure to avoid collecting samples from
the tapered edges, the internal dimples and the central feed-port area.
These samples were placed in a polyethylene sample bag and sealed for
analysis for solids content.

     For operation of the Netzsch press as a diaphragm press, the
fixed-volume plates were replaced with membrane (i.e., variable-volume)
plates and an auxiliary air-driven, duplex, diaphragm pump was installed to
feed sludge suspensions into the diaphragm chambers.  The pump discharge was
located such that the accumulator tank was not used with the diaphragm
system.  Initial operation of the press proceeded as indicated above for low
pressure filtration.  After filling the press with solids by low-pressure
filtration,  the feed pump was stopped and the inlet valve to the chambers
                                      12

-------
was closed.  The bladders internal to the membrane plates were then
gradually pressurized using compressed nitrogen gas over a 5-min interval to
a final bladder pressure of 15.5 bar.  This latter phase using the internal
bladder to compress the cake solids was referred to as the "squeeze" portion
of the cycle.  The time interval between termination of low-pressure
filtration and initiation of the squeeze cycle was typically 2 to 3 min.
Filtrate was collected and its volume was recorded throughout the run.  The
squeeze cycle was terminated when the filtrate production rate was zero.
The dewatered solids were then collected, weighed, sampled and analyzed.

JWI Press —
     The procedures used to initiate a filtration run with this bench-scale
system were similar to those for the Netzsch press.  A sludge suspension was
placed in a 20-L plastic bucket used as a feed reservoir and vigorously
mixed.  A sample was collected for analysis for suspended solids, specific
resistance and CST.  Since this system was only used for low-pressure
filtration and the pump was driven with compressed air from the plant
system, no pressure adjustment was necessary.  The chambers were closed with
a manual hydraulic Jack and the pump was activated.  Filtrate volume was
monitored with time throughout the filtration cycle.  At the end of a
filtration run, the press pump was stopped and the dewatered cakes were
removed, weighed and sampled for analysis.

Aluminum Extraction Reactor

     Aluminum extractions were conducted by addition of sulfuric acid to
dewatered sludge cakes.  In order to extract all available aluminum, as well
as produce a product which was not excessively acidic, a procedure was
developed to establish the quantity of sulfuric acid required to complete an
extraction.

Sulfuric Acid Dosage —
     Earlier research by Saunders et al. (1982, 1984) indicated that
aluminum-anodizing sludges could be assumed to be composed of three major
components:  (i) aluminum, as aluminum oxides and hydroxides; (li) water;
and (ill) trace quantities of metals, dissolved salts and other contaminants.
It was also indicated that when aluminum extraction was being considered the
sludge could be viewed as being composed of aluminum hydroxide (AKOH)^) and
water.  With this simplification, the acidic extraction of aluminum from
dewatered aluminum-anodizing sludges was shown to be indicated by the
following equation (Saunders and Harmon, 1984):
         2A1(OH)3 * x H20 * 3H2SOi|  ---- +  Al^SO^ + (6 * x)H20         (1)

In this equation- the dry solids of the dewatered sludge cake are represented
by AKOH)^ and the quantity of moisture in the dewatered sludge cake  is
represented by x H20.

     From equation (1), the acid requirement for the extraction of aluminum
from a sludge cake is 1.89 g H2SOi4/g AKOH)^.  From previous research  it  was
shown that the aluminum content of the fixed solids in conventional
aluminum-anodizing sludge cakes varied from 32.4 to 39.7 percent, and  was in


                                      13

-------
general agreement with a theoretical value of 34.6 percent for the aluminum
content of A1(OH)3 (Saunders et al., 1982, 1984).  Therefore, using the
fixed solids content of the dewatered sludge cakes and an aluminum content
of 0.346 g Al/g fixed solids, the acid requirement for extractions would be
equal to 1.89 g HjSOjj/g fixed solids.  Based on measured concentrations of
dry and fixed solids, as well as the added dosage of sulfuric acid, the
final product alum concentration could be estimated.  Commercial-grade
liquid alum strength is typically presented in terms of A^Og.  As an
example, a 1000 g (wet) aliquot of a sludge cake with a dry solids content
of 21.3 percent and a fixed solids content of 16.8 percent would contain 832
g of moisture and would require 318  g of f^SOij (1000 g * 0.168 g/g * 1.89
g/g).  Conserving all mass components, the final product would be 28 percent
as Al2(SOtj)3 or 8.3 percent as A^Og.  In these calculations, it was assumed
that the additional loss in mass following exposure to 550°C in the analysis
for fixed solids was indicative of the loss in bound moisture associated
with highly gelatinous aluminum-hydroxide precipitates.

     An alternative to using fixed-solids as a measure of AKOH)^ content
consisted of analytically determining aluminum concentrations in dewatered
sludge cakes.  From equation (1), the acid requirement would be 5.44 g/g Al.
Because it was based on aluminum content alone and no assumption was made
regarding the chemical form of the aluminum precipitates, this acid dose is
herein referred to as the "stoichlometrlc" acid requirement.

Aluminum Extraction—
     In order to investigate the acidic extraction process with regard to
reaction kinetics, stoichioraetry, final product quality, and overall process
feasibility, the following experimental strategy was developed.  Sulfuric-
acid extractions were conducted in the experimental reactor at various
temperatures and levels of sulfuric  acid addition.  The kinetic progress of
a reaction was monitored by withdrawal of samples at various intervals
throughout an experimental run.  A material balance on total mass, as well
as aluminum, was conducted on each extraction to account for the fate of
various constituents (e.g., sludge solids, residual aluminum, soluble
aluminum, and acid) throughout an extraction.  For example, the mass of
aluminum is conserved throughout an  extraction and at any point aluminum may
exist in the solid phase (i.e., contained in unextracted sludge-cake solids)
and the liquid phase (e.g., as soluble aluminum sulfate).  Therefore, a mass
balance on aluminum contributed from both phases would provide an accurate
description of the conversion of solid-phase aluminum to soluble aluminum
during an extraction.  A material balance across the system would also
provide a check on analytical procedures and techniques.

     To initiate an extraction, a wet sludge cake was removed from
refrigerated storage (2-5°C), weighed to the nearest 0.5 g, and placed into
the reactor.  Based upon a fixed solids or an elemental aluminum analysis,
the quantity of required sulfuric acid was calculated according to equation
(1).  Due to the vigorous nature of  the acid-sludge reaction, acid was
slowly added to the reactor over several minutes.  Because an excessive rate
of acid addition resulted in the production of foam which would overflow the
reactor, avoidance of excessive foam production was used to regulate the
rate of acid addition.  Following acid addition, the reactor was covered a-.d


                                      14

-------
mixing was initiated.  It is noted that initial temperatures reached
approximately 90°C due to the heat of hydration from sulfuric acid addition.
The contents of the reactor was allowed to, cool to and regulated at, the
desired temperature (e.g., 50° to 90°C).  The reactor cover was wrapped with
parafilm wax to minimize evaporation and heat losses during an extraction.

     Samples were withdrawn at various intervals throughout the reaction for
analysis.  The sampling procedure consisted of terminating mixing and dis-
connecting the stirring paddle shaft; removing the parafilm wax; and
manually mixing the reactor contents while pouring a sample aliquot of
approximately 100 mL into pre-tared sample containers.  The reactor was re-
assembled and mixing and heat addition were continued within approximately
2-3 min. Sample preparation consisted of weighing the tared sample container
with contents, followed by filtration of approximately 15 mL through a 0.^5-
ym membrane filter.
ANALYTICAL METHODS

     In general three types of samples were analyzed during the study:
(i) suspensions containing precipitated aluminum solids from wastewater
treatment systems at full-scale aluminum-finishing plants; (ii) dewatered
cakes produced by pressure filtration of wastewater suspensions; and
(iii) highly-acidic, aluminum-rich suspensions and solutions produced by
sulfuric-acid extraction of dewatered sludge cakes.  Each type of sample
dictated some different handling and analytical procedures, as indicated
below.

Suspended Solids Concentration

     Wastewater suspensions were examined for suspended solids concentrations
using Gooch crucibles and glass-fiber filters as specified in Method  209 of
Standard Methods (APHA, 1985).  Fixed solids were determined by ignition of
dried solids at 550°C in a muffled furnace for 30 min as  indicated in Method
209D (APHA, 1985).

Cake Solids Content

     Dewatered sludge cakes were examined for dry solids  content using
aluminum-foil weighing dishes.  The dishes were used once and thrown  away
and were much easier to manage and handle than evaporating dishes for the
large number of analyses required.  All analyses were conducted in duplicate
at 103-105°C in accordance with Method 209D (APHA, 1985).  Drying'periods  of
at least 18 h were shown to result in a constant mass for dewatered cakes;
periods of 18-21 hours were used throughout the study.  In addition,
dessication times of greater than 3 h at ambient balance  temperatures were
used prior to determination of dish mass on an analytical balance.  Fixed
solids were determined by ignition of dried solids at 550°C.

Specific Resistance

     The specific resistance of suspensions used  in pressure  filtration

                                     15

-------
studies was determined using standard procedures (O'Connor, 1984).  The 9-cm
Buchner funnel utilized had an effective filtration area of 6.4 x 10~3 m2.
A differential pressure of 0.5' bar (50 kPa) was maintained with a
laboratory-scale vacuum pump and monitored with a 1-m mercury manometer.  A
Whatman No. 1  filter pad placed on a fine-mesh wire screen was used as a
filtration medium.  Filtrate was measured with time of filtration and
dewatered solids were examined for total wet mass and cake solids content.
A linear regression analysis was used with the linear form of the filtration
equation to establish a value for specific resistance, which was expressed
In units of terameters/kilogram (Tm/kg or 1012 m/kg).

Capillary Suction Time (CST)

     Capillary Suction Time was measured with a standard CST apparatus (Type
92/1, Triton Electronics Limited; Dunraon, Essex, England) using an 18 mm
reservoir.  Whatman No. 40 chromatography paper was used as a standard
medium throughout the studies.  All measurements were taken in duplicate at
the suspended solids concentration of the suspension.

Specific Gravity of Aluminum Extracts

     Determination of the specific gravity of aluminum extracts was made
immediately after collection of extract samples at elevated temperatures.
Volume measurements for other analyses were made at these same temperatures,
making subsequent temperature corrections for material balances unnecessary.
The procedure followed was that in Method 213E of Standard Methods (APHA,
1985.).

Aluminum Concentration

     Aluminum determinations were conducted using several pretreatment and
analytical procedures.  Suspensions and dewatered cakes of aluminum-
finishing solids were pretreated using a standard nitric-acid digestion
(Method 302D in Standard Methods, APHA (1985)).  Aluminum determinations
were then made using atomic absorption spectrometry.  A Perkin Elmer Model
303 spectrometer was used with a calibration curve, developed with
primary-standard solutions, to establish aluminum concentrations.

     All samples of suspensions extracted with sulfuric acid to produce
high-strength aluminum-sulfate solutions were filtered through 0.45-ym
membrane filters prior to analysis for aluminum.  Two procedures were
employed for determination of aluminum on these filtered samples.  Atomic
absorption spectrometry as indicated above was used.  Because of the acidic
properties of the filtrates, the high aluminum concentrations (e.g., Al =
50-60 g/L), and the necessity to use extensive (e.g., 1000-fold) dilutions,
experimental investigation proved that nitric-acid digestion was not
necessary and was not employed prior to atomic absorption analyses.

     A second analysis procedure was also used with acid extraction samples.
This procedure was that used by alum manufacturers for use in determination
of the aluminum content of liquid and crystalline aluminum sulfate (Allied
Chemical,  1976).  The procedure was based on complexation of aluminum with a


                                      16

-------
slight excess of EDTA and back tltration of excess EDTA using a standard
zinc sulfate solution.  Comparison of the two procedures was highly
favorable.

Trace Metal Analyses

     Samples of suspensions,  dewatered cakes and aluminum extracts were
examined for trace heavy metals Including silver (Ag), cadmium (Cd),
chromium (Cr),  nickel (Ni),  lead (Pb), tin (Sn) and zinc (Zn).  Analyses for
these metals were preceeded  by nitric acid digestion (Method 302D in
Standard Methods, APHA (1985)).  Atomic absorption spectrometry was used for
metal analysis  in accordance  with Methods 303A and B (APHA,  1985).
                                     17

-------
                                 SECTION 5

                 SLUDGE DEWATERING BY PRESSURE FILTRATION
     Sludge suspensions from Plants A and X were examined.  Two types of
suspensions from Plant A were examined, as well as one produced experi-
mentally using finishing wastes from Plant A, while only one suspension from
Plant X was examined.  The suspensions were examined using two filter
presses (i.e., Netzsch and JWI presses).  The Netzsch press was operated in
three modes, i.e.,  as a low- and a high-pressure, fixed-volume system and a
variable-volume system, while the JWI press was operated as a low-pressure,
fixed-volume system.
SLUDGE SUSPENSIONS

Conventional Neutralization Suspensions

Plant A Suspensions—
     At Plant A,  influent wastewaters are conventionally neutralized in a
3-stage system using additions of spent acid and virgin or spent caustic-
etch solutions in the two initial stages.  Neutralized suspensions are
collected in the  variable-volume third stage, from which they are pumped to
a clarifier with  concurrent polymer addition.  Thickened underflow
suspensions from  the clarifier are dewatered on a rotary vacuum filter and
disposed of in a  landfill.

     Prior to and during pilot-scale pressure filtration studies at Plant A,
the characteristics of selected wastes were monitored and are presented in
Table 1. With one exception (i.e., Day 303), the suspended solids (SS)
concentration of  the neutralization basin effluent (designated AC1; i.e.,
Plant A; a Conventional-neutralization sludge; point ^ in the plant) ranged
between 0.76 and  6.3 g/L and averaged 2.M g/L.  Two grab samples of
neutralization basin effluent (i.e., ACT) were collected to study the
dewatering of a sludge suspension without the aid of polymer conditioners.
The characteristics of these two samples are presented in Table 2.  The
samples were concentrated to the levels indicated in Table 2 by gravity
sedimentation. The relatively low suspended solids concentrations achieved,
however, were consistent with previous thickening studies for unconditioned
sludges (Saunders, et al., 1982) indicating poor thickening and dewatering
properties for similar unconditioned suspensions.

     On Day 303 at noon, a large volume (e.g., 10-60 m3) of spent caustic
etch containing elevated levels of aluminum (e.g., 100-150* g/L) was
discharged to the treatment plant for neutralization and disposal.  All

                                     18

-------
TABLE 1.  CHARACTERISTICS OF NEUTRALIZATION BASIN EFFLUENT, CLARIFIER
          UNDERFLOW AND VACUUM-FILTER CAKE AT PLANT A DURING
          FILTRATION STUDIES
Neutralization
Date Basin (AC1 )


257O)a
258(2)
259(3)
262(5)
265(8)
269(12)
270(13)
271(11)
272(15)
273(16)
276(19)
279(22)
280(23)
284(27)
285(28)
290(33)
291(3*0
292(35)
293(36)
294(37)
297(40)
298(41)
299(42)
300(43)
301(44)
302(45)
303(46)C
304(47)
305(48)
308(51)
309(52)
312(55)
315(58)
319(62)
321(64)
325(68)
326(69)
342(85)
SS
g/L
4.0
6.3
-
1.23
2.49
1.4
-
1.53
1.7
-
1.4
-
0.76
-
4.2
-
5.12
1.2
-
-
1.41
0.98
-
-
-
-
46.0
-
-
-
-
2.1
_
-
-
-
-
-
PH

8.6
8.0
-
-
8.6
-
-
8.6
8.4
-
8.3
-
8.6
7.7
-
-
-
8.4
-
-
-
8.4
-
-
8.4
8.2
—
-
-
8.2
8.1
-
8.4
8.1
-
-
-
8.2
Clarifier Underflow (AC3)

Specific Resistance
SS r Ck
g/L Tm/kg %
27 -
30.7
_
_
39.5
18.0
-
19.9
32.0
- -
38.8
_
26.4
37.4
53.2
_
_
39.1
_
_
-
62.2
_
-
56.7 6.0 13.4
51.3 4.9 12.7
_ . *
91.5 1.6 19.8
_
72.8 3.7 15.4
63.9 4.1 13.7
-
32.8
20.7 4.5 12.2
46.4
17.1 2.5 11.1
24.8 3-2 10.7
39.1 3.2 12.6
CST
s
.
20b
-
-
-
-
-
21 b
-
-
-
-
-
33b
-
-
-
-
-
-
-
-
-
-
237
191
_
182
-
196
257
-
103
63
-
50
-
110
Vacuum
Filter
ck
%

-
12.6
12.9
10.7
10.2
12.2
-
13.4
12.3
12.4
7.4
-
10.4
9.9
11.1
-
-
15.2
14.4
-
-
14.0
15.9
-
-
»
18.8
14.8
-
-
-
_
13.6
-
-
-
-
                                                             (Continued)
                                 19

-------
TABLE 1  (Continued)
Neutralization
Date Basin (AC1 )

343(86)
350(93)
SS
g/L
-
Clarifier Underflow (ACS)
Specific Resistance
PH
8.2
8.8
SS
g/L
38.9
39.3
p Ci
Tm/kg %
3.3 12.2
3.6 11.7
CST
s
139
149
Vacuum
Filter
ck
-
AVERAGE
8.46   40.8
3.7
13.2
12.7
a( ) - Elapsed time, days.
bCST values for 18-mm cylinder; others are for 10-mm cylinder.
GSample taken following batch dump of contaminated caustic-etch tank; for a
 concentrated sample of the suspension:  SS » 104 g/L,  r « 2 Tm/kg,
 Ck - 33.3%,  and CST - 59 s.
dAverage value, exclusive of value for Day 303.
eMedian value.
     is a function of SS and was therefore not averaged.
        TABLE 2.   CHARACTERISTICS OF NEUTRALIZATION BASIN EFFLUENT
                  AT PLANT A PRIOR TO POLYMER CONDITIONING

Date

312
318

PH

8.4
8.0

SS
g/L
12.1
16.6
Specific Resistance
p Ci
Tm/kg %
1.2 11.9
4.7 11.3

CST
s
38
81
rinsewater flows were reduced to minimal values and the third stage of the
neutralization basin was pumped to its minimum liquid volume prior to the
batch dump of caustic etch.  The concentrated waste was treated and neutral-
ized on a semi-batch basis in the neutralization basin over a 12-h period.
This neutralization of a concentrated waste increased the solids loading on
the  system and, as shown by segregated neutralization studies of spent
caustic etch by Saunders et al. (1984), improved the overall thickening and
dewatering properties of the resulting mixed suspension in the clarifier.

     The majority of the pressure filtration studies were conducted with
suspensions obtained from the clarifier underflow without alteration or
adjustment of waste characteristics.  These suspensions were designated as
"AC3".   On a limited number of occasions these suspensions were diluted to
examine concentration effects; dilution water was obtained directly from the
                                     20

-------
recycle water system which was supplied by effluent from the clarifier.  As
indicated in Table 1, the pH of the suspensions throughout the study was
between 7.7 and 8.8, with a median of 8.4.  The suspended solids
concentration of the underflow varied considerably throughout the study,
ranging between 17.1 and 91.5 g/L.  The batch etch dump on Day 303 and the
resulting discharge of a large quantity of segregated-neutralization-like
solids into the clarifier significantly increased the suspended solids
concentration at the underflow.  On the day following the dump (i.e., on Day
304), the suspended solids concentration was at its maximum value for the
study, i.e., 91.5 g/L.  These elevated values continued five and six days
later (i.e., 72.8 g/L on Day 308 and 63.9 g/L on Day 309).  After a
twelve-day period the underflow concentrations returned to typical values,
ranging from 17.1 to 46.4 g/L.  Therefore, for evaluation of dewatering
results using clarifier underflows, three evaluation periods were used:  (i)
prior to the etch dump on Day 303; (ii) immediately following the etch dump
on Day 303 to Day 314; and (iii) the period following the etch dump on Days
315 to 350.  Over the initial period, the suspended solids concentration of
the -underflow ranged from 18 to 62.2 g/L and averaged 38 g/L (14 values).
Over the final period the suspended solids concentration ranged from 17.1 to
46.4 g/L and averaged 32.4 g/L (8 values).  Within these limits, the initial
and final periods appeared to be similar in terms of the performance of the
clarifier.  It is, however, impossible with the data available, given the
highly variable operational procedures employed on the finishing line and
wastewater treatment system at the plant, to establish this point
conclusively.  It is appropriate, however, to state that the initial and
final periods were generally typical of the routine operation of the
wastewater treatment system.

     Pressure filtration studies were conducted over a 50-day period (i.e.,
Days 301-350).  During this interval, grab samples of clarifier underflow
were collected for filtration studies and were characterized using specific
resistance and CST measurements.  Specific resistance, which is theoretical-
ly independent of suspended solids concentration, varied from 1.6 to 6.0
Tm/kg (i.e., 1.6 x 1012 to 6.0 x 1012 m/kg) and averaged 3-7 Tm/kg (11
values), as indicated in Table 1.  Specific resistance values did not
appreciably change during the period immediately following the etch dump
(e.g., Days 303-309).

     As a part of the specific resistance analytical procedure, the solids
content of the resulting dewatered cake, formed at a differential pressure
of 0.5 bar (50 kPa) was determined.  These values are also presented in
Table 1.  Using data prior to and following the etch dump, the specific-
resistance cake solids ranged from 10.7 to 13.4 percent and averaged 12.1
percent (8 values).  On the day following the etch dump (Day 304), the
specific-resistance cake solids was at a high value of 19.8 percent,
indicating a significant impact of the etch dump on sludge dewaterability-
The values decreased to 15.4 and 13.7 percent over the following 5-day
period, indicating the impact of continual removal of etch-dump solids from
the clarifier and continual addition of conventional-neutralization solids
to the clarifier.  By the sixteenth day after the etch dump, the specific-
resistance cake solids was back to a typical value of 12.2 percent.  In
addition to specific resistance, CST was monitored for each sludge


                                     21

-------
suspension examined.  As indicated in Table 1, these values ranged from 50
to 257 seconds during the period of the filtration studies.  However, since
CST varies with suspended solids concentration and solids characteristics,
it is difficult to use these data to identify changes in sludge
characteristics.

     Finally,  a rotary vacuum filter was used at Plant A to dewater the
clarifier underflow.  Grab samples of dewatered sludge solids were collected
from the unit  and monitored for solids content.  As presented in Table 1,
the vacuum-filter cake solids ranged from 7.^ to 18.8 percent and averaged
12.7 percent (19 values).  The average solids content of 12.7 percent is in
general a low  value for a dewatered sludge but was not atypical of values
for aluminum-anodizing systems (Saunders £t al.,  1982).   The vacuum-filter
cake solids of 18.8 percent on Day 304 was the highest obtained throughout
the study period and compared favorably with the specific-resistance cake
solids of 19.8 percent on the same day.

Plant X Suspensions—
     A 1.5~m3  volume of clarifier underflow was collected on Day 313 and
transported to Plant A for use in the filtration studies.  The characteris-
tics of samples of the sludge suspension used in the various pressure
filtration runs are presented in Table 3.  Examination of the data indicated
that the parameters monitored in the various samples did not significantly
deviate from average values over the 29-day period (Day 31U-3U2).  The
average suspended solids concentration of 20.6 g/L was well below that for
Plant A and the pH was slightly lower.  The average specific resistance
value was slightly lower than that for Plant A (i.e., 2.8 Tm/kg vs. 3-7
Tm/kg), while  the specific-resistance cake solids value of 7.1 percent was
drastically lower than the 13.2 percent average value for Plant A.
Therefore, the clarifier suspension from Plant X was relatively dilute,
compared to similar suspensions from Plant A, and had dewaterability
characteristics similar to and slightly poorer than those from Plant A.

Segregated Neutralization Suspensions

     To examine the impact of segregated-neutralization suspensions on
pressure filtration of conventional-neutralization suspensions, segregated
neutralization of caustic etch was conducted on a batch basis.  'Typically a
0.1-m3 volume  of caustic etch was obtained from the finishing line at Plant
A and was placed in a 0.2-m3 mixed reactor.  Spent acid from the on-site
storage container was then pumped into the reactor at the rate of
approximately  1 L/rain for a period of 1-2 h.  The temperature of the
reaction mixture ranged between 40-70°C.  The neutralization reaction was
typically terminated when the pH of the suspension was consistently below 10,
The suspensions were then continuously mixed until they were to be used in
pressure filtration runs that same day.

     The characteristics of the four segregated-neutralization suspensions
examined are presented in Table U.  The suspended solids concentration of
each suspension was dictated by the aluminum content of each batch of spent
etch and acid  available, which was dictated by the operation of the finish-
ing line.  For undiluted suspensions, the suspended solids concentration

                                     22

-------
    TABLE 3-   CHARACTERISTICS OF CLARIFIER UNDERFLOW SUSPENSION (XC30
              FROM PLANT X AS USED DURING FILTRATION STUDIES
specific Resistance
Datea
314
314
314
320
320
321
321
341
PH
7.5
7.9
7.9
7.6
7.6
7.7
7.7
7.2
SS
g/L
19.5
20.5
20.6
21 .0
20.2
19.7
21.5
21 .1
r
Tm/kg
2.8
2.9
3.3
2.5
2.3
2.4
3-1
2.7
*k
7.3
6.6
7.1
-
7.0
7.2
7.5
7.1
CST
s
60
75
70
54
64
68
74
60
 342
21.7
 AVERAGE
20.6
2.8
7.1
66
aSample collected on Day 313 and held in 0.2-m3 containers at ambient
 temperature for use over a 30-day period.
^Median value.
    TABLE 4.  CHARACTERISTICS OF SEGREGATED-NEUTRALIZATION SUSPENSIONS
             PRODUCED FROM SPENT CAUSTIC ETCH AND ANODIZING ACID AT
             PLANT A FOR FILTRATION STUDIES
Specific Resistance
Evaluation
Date
294
297
322a





343a


PH
9.0
10.1
8.8
8.7
8.8
8.8
8.8
8.8
10.6
10.6
10.6
SS
g/L
80.1
137.3
173.2
157.8
155.1
64.7
58.5
34.0
180.1
90.7
41 .2
r
Tm/kg
-
-
1.5
1.4
1.9
1.5
1 .4
1.6
1.3
2.3
1 .1
Ck
-
-
49.5
45.0
47.0
47.7
41 .9
45.2
43.8
42.5
41.5
CST
s
109b
84b
581
-
345
181
-
141
485
-

aReplicates  are analyses of diluted samples; clarified supernatant liquid
 was used  as dilution water.
bCST using 18-rran cylinder;  others are with 10-rnm cylinder.
                                     23

-------
ranged from 80.1 to 180.1 g/L, indicating the concentrated nature of these
suspensions.  The suspensions produced on Days 322 and 343 were used in
filtration runs.  As indicated by comparison with data in Tables 1-3, these
suspensions had specific resistance values (1.1-2.3 Tm/kg) slightly lower
than those for conventional neutralization sludges and had specific-
resistance cake solids (41.5-49.5 percent) which were dramatically higher
than those for conventional-neutralization sludges.
FIXED-VOLUME PRESSURE FILTRATION

Conventional Neutralization Suspensions

Plant A Clarifier Underflow (AC3)~
     High-Pressure Filtration—Numerous filtration runs were made with AC3
suspensions at Plant A.   The characteristics of the suspensions are
summarized by filtration run in Table 5.

     Runs 6-17 were made prior to the etch dump on Day 303.  Suspended
solids concentrations for underflow samples varied from 41.6 to 57.4 g/L
(note: run 16, at a suspended solids concentration of 30.9 g/L, was
conducted following dilution of the sample from run 15) and were near the
highest experienced in the filtration studies, with the exception of those
immediately following the etch-dump of Day 303-  Runs 21-29 were conducted
during this etch-dump period and were typically higher in suspended solids
concentration than any of the conventional sludge suspensions employed.  For
the period following the etch dump, Runs 71 and 101 had low suspended solids
concentrations (20.6 g/L and 17-3 g/L). while runs 142 and 171 had
near-average concentrations of 38.4 and 37.9 g/L, respectively.

     The fixed suspended solids concentrations for all suspensions averaged
75 percent of the total  suspended solids concentrations.  Previous data
(Saunders, et al.,  1982; 1984) indicate this value is typical of these
gelatinous suspensions throughout the industry.  Furthermore, while the
wastewater contained organic matter at measured TOC concentrations of
170-230 mg/L, the aluminum-hydroxide precipitates formed by conventional
neutralization contain high levels of bound and mechanically-occluded water
that is not removed by drying at 103-105°C (APHA, 1985).  Therefore, the 25
percent mass loss of total suspended solids upon exposure to 550°C was
attributable to bound and mechanically-occluded water, as well as volatile
organic matter.  Finally, the fixed suspended solids, based on previous data
presented by Saunders ejt al. (1982, 1984), were indicative of dry aluminum
hydroxides.  For purposes of estimation, it can be assumed that 34.6 percent
of the fixed suspended solids are equal to aluminum using the ratio
A1/[A1(OH)3] - 0.3^6 (i.e., 27/C27 + (17)(3)]}.  Therefore, estimated
aluminum concentrations  for the suspensions in Table 5 ranged from 5.3 to
24.2 g/L.

     During pressure filtration runs, the volume of filtrate produced witn
time was monitored.  These data are presented graphically in Figures 2-6.
Filtrate was produced rapidly during the initial chamber-filling process.
                                     24

-------
    TABLE 5.  CHARACTERISTICS OF CLARIFIER UNDERFLOW SUSPENSIONS FOR
              HIGH-PRESSURE,  FIXED-VOLUME PRESSURE FILTRATION ON
              NETZSCH PRESS
Run


6
7
8
9
14
15a
16
17
21 b
22
23
24
25
26
27C
28
29
71
101
142
171
Type


AC3
AC3
AC3
AC3
AC3
AC3
AC3
AC3
AC3
AC3
AC3
AC3
AC3
AC3
AC3
AC3
AC3
AC3
AC3
AC3
AC3
Date


301
301
301
301
302
302
302
302
304
304
304
308
308
308
308
308
308
319
325
342
343
Temp.

°C
23
22
25
25
18
21
22
22
25
24
24
25
24
23
23
21
21
16
17
16
13
PH


8.4
8.3
8.4
8.3
8.3
8.3
8.3
8.4
—
-
-
8.2
8.2
8.2
-
-
-
8.1
7.8
8.2
8.2
Suspended Solids
Total
g/L
56.5
56.7
56.8
57.4
50.7
44.5
30.9
41 .6
93-0
93.0
88.5
71.6
73.7
73.2
39.1
36.1
36.8
20.6
17.3
38.4
37.9
Fixed
g/L
42.
42.
42.
43.
36.
34.
22.
31.
69.
69.
66.
53.
54.
54.
29.
28.
28.
15.
—
_
28.
0
2
3
0
7
5
8
2
8
8
9
3
8
5
1
3
7
3


2
Specific
Resistance
Evaluation
r
Tm/kg
5.7
7.0
5.8
5.5
5.5
4.9
4.0
3.6
1.8
1.8
1.3
4.1
3.5
3.3
3.1
2.6
2.2
6.0
2.3
3-2
3.6
C

13
14
13
12
13
11
12
12
20
20
19
13
16
16
14
14
14
11
11
12
11
k

.0
.3
.5
.6
.1
.9
.3
.5
.1
.1
.5
.0
.4
.7
.8
.7
.6
.8
.5
.6
.7
CST

s
245
240
225
244
195
138
84
141
178
178
189
214
176
199
95
90
100
59
53
1 10
139
aNo filtrate-volume data were collected for run 15.
^Collected from clarifler underflow on day following etch dump.
cRuns 27-29 were-conducted with a diluted sample of suspension
 used with runs 24-26.
The rate of production of filtrate decreased with increasing time as the
fixed-volume chamber(s) filled with suspended solids and approached an
asymptotic value with extended time of filtration.  Within common sets,
e.g.,  runs 6-9,  21-23, 24-26 and 27-29, the procedure was to make multiple
                                      25

-------
                             40            60
                             Time,  min.
80
100
lri^urc' 2.  Cumulative filtrate volume for AC3 suspensions In runs 6-9 during high-pressure
         f11tratlon.

-------
to
--J
            .,,.  I I  I I  I I I I I  I I  I I  I I  I I  I I  I I  I I  I I  1 I I  1 I  I I  I I  I I  I I  ,.  I ....
                                        40            60

                                         Time,  min.
80
100
             Figure  3. Cumulative filtrate volume  for AC3 suspensions in runs 14,  16 and 17 during

                     high-pressure filtration.

-------
I  I I  I I  I  I I  I I  I  I I  I I  I I  I I  I  I I  I I  I I  I I  I  I I  I I  I I  I I  I I  I  I I  I I  I I  I  I I
                             40             60
                              Time,   min.
                                                   80
100
PiI'.ure
Cumulative filtrate volume for AC3 suspensions  in runs 21-23 during high-pressure
f 111ration .

-------
80         120         160
      Time,  min.
                                                             200
                                                                 240
Figure 3.
Cumulative filtrate volume for undiluted (runs 24-26) and diluted (runs 27-29)
AC3 suspensions during high-pressure filtration.

-------
80
                                    120         160        200        240
                               Time,  min.
Figure  6.  Cumulative filtrate volume  for AC3 suspensions  In runs 71, 101, 142 and 172 at
         high-pressure filtration.

-------
filtration runs over various time intervals to be able to establish the
effects of filtration time on filter performance.  In these instances., it is
shown in Figures 2, 4 and 5 that replicate runs closely coincided with
previous runs.  Data in Figures 3 and 7 were for numerous suspensions at
various concentrations and did not overlap, as expected.

     The data for each of the runs in Figures 2-6 are summarized in Table 6,
including data for the dewatered cakes produced in each run.  In this and
all subsequent tables, the cakes produced were numbered 1 and 2, with 1
being the cake at the filter influent (i.e., face chamber) and cake 2 being
the next one (i.e., the tail cake).  In all runs made throughout the study,
the maximum number of chambers was two, with only one chamber being used in
some instances.

     Inspection of the data in Table 6 indicated a slight increase in total
mass of wet cakes and cake solids content with increasing time of
filtration, although the correlation was very low, due, in part, to
variations in initial sludge concentrations and solids characteristics.  For
runs prior to and well after the etch dump (i.e., runs 6-17 and 71-171),
cake solids ranged from 21.6 to 28.8 percent, with fixed solids being equal
to an average of 74 percent of total solids content.

     For runs 21-29, clarifier underflow samples were affected by the
addition of etch dump solids to the clarifier.  Cake solids content ranged
from 24.7 to 32.4 percent for undiluted suspensions (i.e., runs 21-26) and
23.6 to 28.9 percent for diluted suspensions (i.e., runs 27-29).  It is
obvious that the etch dump resulted in improved filter performance, with the
improvement being attributable to increased suspended solids concentrations
of the influent suspensions "and improved dewaterability of the suspensions.

     Low-Pressure Filtration—Low pressure filtration of AC3 suspensions was
examined using the Netzsch and JWI filter presses.  The characteristics of
the suspensions studied are included in Table 7.  The suspensions collected
prior to the etch dump on Day 303 had relatively high suspended solids
concentrations, while those for suspensions collected immediately following
the etch dump (runs 31-3*0 were similarly high, although not as high as the
88.5-93.0 g/L values for runs 21-23 (see Table 5).  Suspensions for runs 72,
102, J-21, J-61, J-81 and J-91 were collected after the etch dump of Day 303
and had suspended solids concentrations ranging from 16.9-39.4 g/L.  These
suspensions were generally typical of the clarifier underflow suspensions
produced at Plant A, as indicated by comparison with data in Table 1.

     The volume of filtrate collected during low-pressure filtration with
the Netzsch and JWI presses are presented in Figures 7 to 10.  Replicate
sets in runs 10-13, 31-33 and 34-37 were virtually identical for similar
sludge suspensions.  Filtration data presented in these figures are
summarized in Table 8.  Prior to and following the etch-dump periods, the
cake solids for dewatered cakes ranged from 13.6 to 24.7 percent.  However,
with the exception of runs conducted for short filtration times (i.e., _< 60
min), cake solids ranged from 20.1 to 24.7 percent and averaged 22.6 percent
(14 values).  Filtration of underflow samples collected during the period
immediately following the etch dump resulted in production of cake solids of

                                      31

-------
TABLE 6.  RESULTS FOR HIGH-PRESSURE, FIXED-VOLUME FILTRATION
          OF CLARIFIER UNDERFLOW SUSPENSIONS (AC3)
          ON NETZSCH PRESS
Filtration
Run
Pressure
Time of
Filtration
Dewatered
Mass
Cake(s)
Solids Content
Total
bar
6

7

8

9

14

15

16

17

21a

22

23

24

25

26

27b

28

29

14-1

13-1

13-1

13-1

13-1

13-1

13-1

13-1

13-1

13-1

13-1

14-1

14-1

14-1

5

4

4

4

4

4

4

4

„

4

4

5

5

5



.6

.6

.6

.6

.6

.6

.6

.6

.6

.6







14-15

14-1

14-1


5

5






min
83

60

40

20

60

150

91

79

60

80

40

75

50

35

156

90

68

Fixed
kg % %
1)
2)
1)
2)
1)
2)
1)
2)
1)
2)
1)
2)
1)
2)
1)
2)
1)
2)
1)
2)
1)
2)
1)
2)
1)
2)
1)
2)
1)
2)
1)
2)
1)
2)
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5

5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
.56
.48
.39
.45
.28
.86
.05
.14
.25
.28
.48
.53
.31
.45
.51
-
.85
.85
.9
.9
.85
.85
.71
.71
.68
.68
.68
.56
.73
.68
.53
.62
.45
.53
23
23
22
20
20
20
15
16
22
21
23
23
21
-
22
23
31
31
-
-
32
31
28
27
25
27
24
25
28
28
24
25
23
23

.3
.6
.9
.2
.6
.5
.1
.3
.6
.4
.7
.7

.2
.1
.7
.6


.4
.7
.6
.6
.8
.0
.7
.8
.8
.9
.7
.4
.6
.6
16
17
16
15
1 4
1 4
11
11
16
17
17
17
16
-
16
17
24
24


24
24
21
20
18
21
18
20
20
22
18
19
17
17
.8
.0
.5
.3
.9
.8
.5
.8
.3
.0
.6
.5
.0

.2
.0
.3
.1
-
-
.9
.3
.9
.6
.9
.6
.3
.2
.9
.5
.3
.0
.5
.5
                                                         (Continued)
                              32

-------
Table 6 (Continued)

Run
Filtration
Pressure
Time of
Filtration
Dewatered
Mass
Cake(s)

Solids Content
Total

71

101

142

bar
14-15

14-15

14-15

min
203

160

135

kg
1)
2)
1)
2)
1)
2)
5
5
5
5
5
5
.57
.62
.61
.66
.80
.70
I
24
26
24
25
28
28

.9
.3
.9
.5
.7
.8
Fixed
%
^
-
.
-
_
-
171'
14-15
103
5.62
26.3
18.8
Collected from clarifier underflow on day following etch dump.
bfiuns 27-29 were conducted with a diluted sample of suspension used
 for runs 24-26.
°Run conducted with only 1 chamber.
15.9 to 21.8 percent over very short operational periods, i.e., 18 to 75 min
as opposed to 80 to 201  min for the runs with cake solids of 20.1 to 24.7
percent.

Plant A Neutralization Basin Effluent Suspensions (AC1)—
     To examine the dewaterability of the neutralized suspension at the
influent  to the clarifier without the aid of polymer conditioning, samples
were collected directly from the third stage of the neutralization basin
for filtration studies.

     High Pressure Filtration—As indicated in Table 9,  three suspensions
were examined by high-pressure filtration over a total of five runs.  The
suspension collected on Day 303 was that produced during the period of
neutralization of the caustic etch dump.  The suspension was furthermore
concentrated by a factor of 2 by gravity sedimentation and was the thickest
suspension examined during the study, with a suspended solids concentration
of 101.1  - 109.4 g/L.  The specific resistance values (i.e., 0.17 - 0.25
Tm/kg) were approximately an order of magnitude below those for all other
suspensions and the specific-resistance cake solids were exceptionally high
(32.9 - 33.8 percent), indicating excellent dewaterability, even without the
aid of polymer conditioning.
                                      33

-------
     TABLE 7.  CHARACTERISTICS OF CLARIFIER UNDERFLOW SUSPENSIONS (ACS)
               FOR LOW-PRESSURE, FIXED-VOLUME. PRESSURE FILTRATION ON
               NET2SCH AND JWI PRESSES
Run
NETZSCH PRESS
10
11
12
13
31a
32
33
3*
35*
36
37
72
102
JWI PRESS
J-21
J-61 (141-2)°
J-8K171-2)
J-91 (181-2)
Type
AC3
AC3
AC3
AC3
AC3
AC3
AC3
AC3
AC3
AC3
AC3
AC3
AC3
AC3
AC3
AC3
ACE
Date
301
301
302
302
309
309
309
309
309
309
309
319
325
315
342
343
350
Temp.
°C
-
18
18
24
24
22
22
20
20
21
16
17

15
13
16
PH
8.3
8.3
8.2
8.2
8.1
8.1
8.1
8.1
8.2
8.2
8.3
8.1
7.8
8.4
8.2
8.2
8.8
Suspended Solids
Total
g/L
57-4
57.4
51.9
50.7
64.5
64.5
63.2
63.2
31 .8
33.3
32.3
20.8
16.9
32.8
39.4
37.9
39.3
Fixed
g/L
43.0
43.0
38.6
36.7
50.1
50.1
48.8
48.8
24.5
25.7
25.5
15.5
_
25.1
29.3
28.2
29.2
Specific
Resistance
Evaluation
r
Tm/kg
5.5
5.5
3.8
5.5
3.7
3-7
4.5
4.5
3.3
3.4
2.8
3.0
2.7

3.2
3.6
4.2
ck
%
12.6
12.6
12.7
13.1
12.6
12.6
14.7
14.7
14.7
12.7
12.9
12.5
10.7

12.6
11 .7
12.8
CST
s
24U
24U
187
195
251
251
263
263
104
103
92
66
46
102
103
139
149
Suspension collected from clarifier underflow six days following
 etch dump.
bRuns 35-37 were conducted with a diluted sample of suspension used
 with runs 31-34.
c( )  - runs with Netzsch press which were conducted with identical
 suspension.
                                     34

-------
   80
 0)

-60

 CD
-4-»
 O


Si 40

M—
 O

 0)

 E20
JD

 o
     0
0
                                                     •  10
                                                     *  11
                                                     x  12
                                                     O  13

                     20
40            60

 Time,  min.
                                                          80

100
Figure 7.
                 Cumulative  filtrate volume for AC3 suspension in runs 10-13 during low-pressure

                 f i 1 tration.

-------
0
              80
      120
Time,  min.
160
200
240
 Figure  8.
Cumulative filtrate volume for  undiluted (runs 31-33) and diluted (runs 34-37)
AC3 suspensions during low-pressure filtration.

-------
   200 r
.t± 120
         ' I I I I I  I I I I I  I I I I I  I I I I I I  I I I I I  I I I I I I  I I I I I I  I I I I I I  I I I I I I I I I I I  I I I I
                    40
             80        120        160        200        240
                  Time,  min.
         Figure
Cumulative filtrate volume for AC3 suspensions in runs 72 and 102 at low-pressure
f11tratIon.

-------
       14
       12
OJ
Co
 0)
-t-'
—  8
  *»
 (D
 E  6
_D
 O
>  4
       -1 I I » I I I I  I I I I  I I I  I I I I  I I I I  I I I  I I I I  I I I I  I I I I  I I I  I I I I  I I I I  I » I I  I I I I  I I -
        0
                                                                  •  J21
                                                                  •  J61
                                                                  ^  J81
                                                                  X  J91
            I I I  I I I  I I I  I I I I  I I I I  I I I I  I I I  I I I I  I I I  I I I I  I I I I  I
                                                               I I I  I I I  I I
           0
                   40
80        120        160
      Time,   min.
200         240
            Figure  10. Cumulative filtrate volume for AC3 suspensions In runs J-21, J-61, J-81 and J-91
                     during low-pressure filtration.

-------
TABLE 8.  RESULTS FOR LOW-PRESSURE, FIXED-VOLUME FILTRATION
          OF CLARIFIER UNDERFLOW SUSPENSIONS (AC3)
          ON NETZSCH AND JWI PRESSES
Filtration
Run

Pressure
bar
Time of
Filtration
min
Dewatered Cake(s]
Mass
kg
I
Solids Content
Total
5
Fixed
%
NETZSCH PRESS
10

1 1

12

13

3la

32

33

3^

35"

36

37

72

102

JWI PRESS
J-2H71-2)

J-6K141-2)

6-7

6-7

6-7

6-7

6.5-7.5

6-7

6-7

6-7

6-7

6-7

6-7

6-7

6-7


6-7

6-7

80

40

60

20

75

58

30

18

120

66

30

201

160


124

97

1) 5.31
2) 5.34
1) -
2) 5.28
1) 5.25
2) 5.34
1) 4.88
2) 4.82
1) 5.34
2) 5.34
1) 5.34
2) 5.39
1) 5.28
2) 5.39
1) -
2) 5.22
1) 5.34
2) 5.42
1) 5.31
2) 5.22
1) -
2) 5.02
1) 5.42
2) 5.42
1) 5.4U
2) 5.43

1) 0.915
2) 0.945
1) 0.955
2) 0.960
20.1
20.1
17.3
18.7
19.3
19.5
-
13.6
21.5
21.8
21 .1
21 .1
19.8
19.7
16.1
15.9
22.4
21.3
18.0
19.2
15.3
14.8
22.4
22.8
24.7
22.3

21.3
21 .7
23.4
23.1
14.6
1U.7
13-0
13.0
14.1
14.1
-
10.6
15.4
15.6
15.9
14.9
15.1
14.6
12.4
12.2
16.0
16.4
12.9
14.5
11 .0
10.7
_
-
_
-

_
-
_
-
                                                          (Continued)


                              39

-------
TABLE 8 (Continued)
Filtration
Run Pressure
Time of
Dewatered
Filtration Mass
Cake(s)
Solids Content
Total
bar
J-81 (171-2) 6-7

J-91 (181-2) 6-7

min
100 1)
2)
97 1)
2)
kg
0
0
0
0
.965
.970
.985
.985
*
23
23
24
23

.8
.6
.1
.1
Fixed
}
16.7
16.7
_
™
Collected from clarifier underflow six days following etch dump.
bfiuns 35-37 were conducted with a diluted sample of suspension used
 for runs 31-34.
  TABLE 9.  CHARACTERISTICS OF NEUTRALIZATION BASIN EFFLUENT SUSPENSIONS
            (AC1) FOR HIGH-PRESSURE FIXED-VOLUME FILTRATION ON NETZSCH
            PRESS
Run

Type

Date

Temp.
°C
Specific
Resistance
pH Suspended Solids Evaluation
Total
g/L
18*
19*
20*
41
62
AC1
AC1
AC1
AC1
AC1
303
303
303
312
318
21
21
20
19
14
101
101
109
8.4 12
8.0 16
.4
.2
.4
.1
.7
Fixed r
CST
Ck
g/L Tm/kg J
0.
0.
0.
8.9 1 .
2.
19
25
17
2
1
33
32
33
11
11
.2
.9
.8
.9
.1
s
64
63
51
38
56
'Suspension for runs 18-20 was collected immediately after etch dump and
 concentrated by a factor of approximately 2; no polymer conditioner was
 added.
     The two remaining suspensions examined were conventional sludges and,
even with extensive gravity thickening on a batch basis, were low in
suspended solids concentration.  The specific resistance values were not
significantly different from those for AC3 suspensions, although
specific-resistance cake solids were relatively low at 11.1 - 11.9 percent,

     The volumes of filtrate collected during high-pressure filtration are


                                      40

-------
presented in Figures 11-13.  Replicate runs 18, 19 and 20 were virtually
identical.  In addition, the rate of production of filtrate after the'
initial 10-min period decreased drastically.  This was significantly
different from the responses demonstrated during other runs (e.g., see
Figures 2-6) of even longer duration.

     As indicated by data in Figure 12, run 41 was terminated very early in
its filtration cycle due to the lack of sufficient volume of influent
suspension.  Data for run 62 are presented in Figure 13 in conjunction with
that for run 61, a low-pressure run.

     Data in Table 10 for runs 18-20 confirmed the excellent dewaterability
      TABLE 10.  RESULTS FOR HIGH-PRESSURE, FIXED-VOLUME FILTRATION
                 OF NEUTRALIZATION BASIN EFFLUENT SUSPENSIONS (AC1)
                 ON NETZSCH PRESS

Run


18

19

20

41

Filtration
Pressure

bar
13-14.6

13-1^.6

13-14.6

14-15

Time of
Filtration

min
40

15

27

54

Dewatered Cake(s)
Mass


1 )
2)
1)
2)
1)
2)
1)
2)

kg
6.58
6.30
6.44
6.38
6.58
6.23
5.13
5.32
i
Solids Content
Total
%
42.8
41 .6
41 .6
44.7
42.5
41 .8
18.1
17.5
Fixed
%
34.0
32.3
32.4
32.1
32.9
32.1
13.6
13.2
 62*
14-15
180
5.34
19.8
14.5
*0nly 1 chamber was used for run 62.
of the suspensions produced during the batch etch dump.  In time periods of
15-40 rain, dewatered cake solids of 41.6 - 44.7 percent were achieved.  The
dewatered cake solids for runs 41 and 62 were low (17.5 to 19.8 percent),
indicating the dramatic difference between conventional- and segregated-
neutralization suspensions.

     Low-Pressure Filtration — Three AC1 suspensions were examined during
four filtration runs.  The characteristics of the suspensions are summarized
in Table 11 and were similar to those for the suspensions collected at times
other than during the etch dump period.  The suspension for run J-32, which
was the same one used for run 61 on the Netzsch press, was polymer con-
ditioned on a batch basis prior to the filtration run to examine the  impact
of conditioning on dewatering.  While the suspended solids concentration
                                      41

-------
   80
 CD
'^60

 Q)
-+-•
 O
LI 40
 o
 (D
 E20
_D
 O
QLL_L_1.
  0
                         •  18
                         •  19
                         •  20
                                .  ... I  i ... I  . .. .  I ...  . t  i i  i •  I  i i  • i  I.
                      10
20            30
 Time,   rnin.
40
50
         Kigure 11.  Cumulative  filtrate volume for ACl suspensions  In runs 18-20 during high-
                   pressure filtration.

-------
120 M "ML"-I
      I • • ' • I
I • ' •  • I
                                                         41
                 I  I I I I I  I I
                      I .... I .... I .... I .... I .... I .... I
                                                            I I I I I  I I I
                           80        120       160
                                 Time,  min.
                                              200        240
      Figure 12.
Cumulative filtrate volume for an AC1 suspension in run 41 during high-pressure
filtratIon.

-------
  80
.' ' ' ' I  ' ' ' ' I  ' ' ' ' I '  ' ' ' I '  ' ' ' I '  ' ' ' I ' '  ' ' I ' ' ' '  I ' ' ' ' i  ' ' ' '  I ' ' ' '  I
  60
0>
oJ40
E
o
   20
    0
                                                   •  61
                                                   •  62
                  » . ... I .... I .... I
                                                     I .... I .... I
      0
           40
80        120       160        200        240
     Time,  min.
       Figure 13.
          Cumulative filtrate volume for AC1 suspensions during high-pressure (run 62) and
          low-pressure  (run 61) filtration.

-------
 TABLE 11.   CHARACTERISTICS OF NEUTRALIZATION BASIN EFFLUENT SUSPENSIONS
            (AC1)  FOR LOW-PRESSURE FIXED-VOLUME FILTRATION ON NETZSCH AND
            JWI  PRESSES


Run




Type




Date




Temp. pH

°C
Specific
Resistance
Suspended Solids Evaluation
Total Fixed r C^
g/L g/L Tm/kg J


CST

s
NETZSCH PRESS
61
JWI PRESS
J-3K61)
J-32(6Da
J-71 (61)

AC1

AC1
AC1
AC1

318

318
328
303/
(342)b
13 8.0

11 8.2
11 8.1
18 9.1

16.5 - 4.8

17.1 12.7 4.7 11.9
40.1 - 4.6 10.9
82.3 61.7

70

66
108
_

Suspension  was  polymer  conditioned with 2 percent (by volume) of polymer
 solution used at  plant  during a 30-s rapid mix,  followed by 150-s slow mix,
 gravity settling  and decantation of clarified water.
Dotation (303/(342)) indicates sample was collected on Day 303 and was
 dewatered on Day  342.
was dramatically  increased as a result of polymer conditioning and gravity
settling,  the  specific  resistance of the suspension was not altered
significantly  but the specific-resistance cake solids decreased from 11.9
percent to 10.9 percent.   This in general was an expected and predictable
response for polymer conditioning.

     The AC1 suspension for run J-71  was similar to that for run 161 and was
collected during  the etch dump on Day 303, but was not examined by
low-pressure filtration until Day 342.  The suspended solids concentration
was exceptionally high  at 82.3 g/L and well above the average value of 2.4
g/L for AC1 suspensions,  as presented in Table 1.

     Filtrate  volume data for low pressure runs are presented in Figures
13-15.   The rate  of filtrate production for run 61 was lower than that for
high-pressure  run 62, as indicated in Figure 13-  The combined effects of
(i) polymer conditioning and (ii) concentration by gravity settling are
apparent in Figure 14.   Although run J-32 was terminated prematurely, it is
obvious that the  rate of filtration for the thicker, polymer-conditioned
suspension was much higher, indicating a dramatic improvement in sludge
dewaterability-   The filtration response for run J-71 (Figure 15) was
excellent initially, as indicated by a rapid filtration rate, followed very
quickly by an  abrupt decrease in rate as the press was filled with  solids.

                                     45

-------
i i l  l I i l I  I I I I I I I I I  I I I I I  I I I I I  I I I I I  I I I I I  I I I I I  I I I I I  I I I I  I I I I I I I I
                      80        120        160
                            Time,  min.
                                               200        240
 Figure 14
Cumulative filtrate volume for ACl suspensions in runs J-31 and J-32 during
low-pressure filtration.

-------
   14
   12
 0)
-»->
-  8
 (U
 E  6
J3
 O
>  4
    0
      0
       T '  ' • ' I ' ' ' ' I  ' ' ' ' I  ' ' ' ' I ' ' ' •  I ' ' ' '  I ' ' '  ' I
                                                             J-71
             I I I I  I I I I I I I I I  I I I I  I I I I I  I I I I  I I I I I  I I I I I  I I I I I I I I I  I I I I  I I I I I  I '
40
80        120        160
      Time,  min.
200        240
        Figure 15.
Cumulative filtrate volume for ACl suspension in run J-71 during low-pressure
f111rat Ion.

-------
     Data in Table 12 for runs 61  and J-31  indicate that the two presses
produced similar dewatered cake solids contents when operated at similar
intervals and pressures.   Comparison of data for runs J-31  and J-32 confirm,
in part, previous conclusions from Figure 14.   The polymer conditioned
sludge was dewatered to 14.5  percent solids in 29 min while the uncondi-
tioned suspension required 171  min to achieve  18.1 percent solids.   Run J-71
was conducted with an etch-dump suspension  which was dewatered to an average
cake-solids content of 46.2 percent in a 30 min period,  indicating  the
dramatic improvements of   an  etch-dump on suspension dewatering
characteristics.
       TABLE 12.   RESULTS  FOR LOW-PRESSURE,  FIXED VOLUME FILTRATION
                  OF NEUTRALIZATION BASIN EFFLUENT SUSPENSIONS (AC1)
                  ON NETZSCH AND JWI PRESSES
Run
Filtration
 Pressure

   bar
 Time of
Filtration

   min
                                                  Dewatered Cake(s)
Mass
                                             kg
Solids Content
            Total
              %
                                                                       Fixed
NETZSCH PRESS
61
JWI PRESS
J-31 (61)
J-32C61)
J-71 (161)

6-7

6-7
6-7
6-7

165

171
29*
30

5.15

0.95
0.895
1) 1.16
2) 1.15
17.3

18.1
14.5
47.1
45.3
12.9

13.2
10.2
_

 Run terminated prematurely due to lack of sufficient volume of the
 influent sludge suspension.
Plant X Clarifier Underflow (XC3)~
     A single sample of a clarifier underflow suspension from Plant X was
examined over a 29-day period (Day 314-342;  Table 3).  The characteristics
of suspension aliquots and results of pressure filtration studies are
presented below.

     High-Pressure Filtration — The characteristics of the suspensions
examined by high-pressure filtration are presented in Table 13.  No major
differences in suspension characteristics are apparent, with the small
exception of a decrease in suspension pH with increased storage time.  This
variation was, however, not indicative of a well-defined trend, as indicated
by data in Table 3.
                                     48

-------
   TABLE 13.   CHARACTERISTICS OF CLARIFIER UNDERFLOW SUSPENSIONS (XC3)
              FOR HIGH-PRESSURE, FIXED-VOLUME FILTRATION ON NETZSCH
              PRESS
Run


51
81
82
Type Date Temp.


XC3
XC3
XC3


311
311/321
311/321
•C

16.5
11
12
PH


7.9
7.7
7.7
Suspended Solids
Total
g/L
20.5
19.7
21.5
Fixed
g/L
16
15
17
.1
.6
.1
Specific
Resistance
Evaluation
r
Tm/kg
2.9
2.1
3.1
ck
%
6.
7.
7.


6
16
5
CST

s
75
68
71
121    XC3 311/311  11
7.2
21 .1
16.8
2.7
7-1
60
     The volumes of filtrate collected during high-pressure filtration are
presented in Figures 16-18.   Data for replicate runs 81  and 82 were
virtually identical and all  four runs were carried well  into the declining
filtration phase.   Performance data for the four runs are presented in Table
11.   Cake solids contents of 17-7 to 20.3 percent were achieved for runs
with elapsed filtration times of 170 to 225 min.
      TABLE 11.   RESULTS FOR HIGH-PRESSURE,  FIXED-VOLUME FILTRATION
                 OF CLARIFIER UNDERFLOW SUSPENSIONS (XC3) ON NETZSCH
                 PRESS
Filtration
Run Pressure
bar
51 11-15

81 11-15

82 11-15

121 11-15

Time of
Dewatered Cake(s)
Filtration Mass
min
225 1)
2)
170 1)
2)
90 1)
2)
170 1)
2)
kg
5.36
5.11
5.28
5.33
5.21
5.25
5.20
5.30
Solids Content
Total
20.3
19.6
17.7
-
16.2
15.8
18.8
18.5
Fixed
15.1
11.9
_
-
-
-
_

     Low-Pressure Filtration — Characteristics of the XC3 suspensions used
during low-pressure runs are presented in Table 15 and are consistent with
those presented in Tables 3 and 13.   Cumulative filtrate volumes for the

-------
          100
Ul
o
                                                                     51
11 I  I«111 111 1111111111
            0
                i i i  i I i i  i i I i i  i i I i i  i i I i  i i i I i  i i i I i  i i i I i  i i i I » i i i I i  i i i I i  i i i I  i i i	i"
               0          40         80        120        160        200        240


                                            Time,  min.

                Figure  16.  Cumulative filtrate volume for XC3 suspension in run 51 during high-pressure
                         f iltrat ion.

-------
         100
         80
     .±  60


      CD
      _,  40
(—•    	
      o
          20
           0
                   i.... i
             0
 40
             111111.11111.11111111111111111111111
80        120        160

     Time,  min.
200        240
               Figure 17.
Cumulative filtrate volume for XC3 suspensions In runs 81 and 82 during high-
pressure f 11 tr.it Ion .

-------
   100
O)
        0
             80        120         160
                  Time,  min.
200        240
         Figure 18.
Cumulative filtrate volume for XC3  suspension In run 121 dulrng high-pressure
fIItrat ion.

-------
low-pressure runs are presented in Figures 19-23.   Cumulative filtrate data
were,  however,  not taken for run J-42,  an extremely short (14 rain) run.
Data for replicate runs 73 and 74, and  J-11  and J-12 were virtually
identical,  as experienced in previous replicate runs.  It should be
recognized  that runs 52, J-11  and J-12;  and  runs 73, 74,  J-41 and J-42 were,
respectively, performed on the same days with virtually identical aliquots
of sludge suspensions.   It is, however,  difficult  here to confirm the
similarities and differences between the various runs; this will be examined
later.
   TABLE 15.   CHARACTERISTICS OF CLARIFIER UNDERFLOW SUSPENSIONS (XC3)
              FOR LOW-PRESSURE,  FIXED-VOLUME PRESSURE FILTRATION ON
              NETZSCH AND JWI PRESSES
Run
Type Date Temp.
pH
°C
Suspended
Total
g/L
Solids
Fixed
g/L
Specific
Resistance
Evaluation
r Ck
Tm/kg %
CST
s
NETZSCH PRESS
52
73

74

XC3
XC3

XC3

314 1
314/ 1
(320)*
314/ 1
(320)*
6
4

3

7.9
7.6

7.6

20
21

20

.6
.0

.2

16
16

16

.7
.7

.0

3
2

2

.3 7.1
.5

.3 7.03

70
54

64

JWI PRESS
J-11

J-12

J-41
J-42
J-121

XC3C51/2)

XC3(51/2)

XC3(73/4)
XC3(73/4)
XC3O21/2)

314/ 1
(320)*
314/ 1
(320)*
314 1
314 1
314/ 1
(341)*
7

7

5
5
8

7.8

8.0

7.5
7.5
7.2

20

20

19
19
21

.5

.3

.5
.5
.1

16

16

16
16
16

.5

.2

.4
.4
.8

2

3

2
2
2

.9 7.5

.2 7.5

.8 7.3
.8 7.3
.7 7.1

67

70

60
70
60

'Notation indicates suspension was collected on first day indicated
 and filtration run was conducted on second day indicated.
     The results for the cakes produced during the low-pressure runs are
presented in Table 16.   Cake solids data for runs 52, J-11 and J-12
indicated minor improvements in performance with increased filtration time,
Run times of 100,  176 and 224 rain resulted in production of average cake
                                     53

-------
     111 i  11 11111111 111 11111
80        120        160
      Time,   min.
                                                           200        240
Figure 19.  Cumulative filtrate volume for XC3 suspension In run 52 during low-pressure
          filtration on Netzscli press.

-------
100 pr^T
         lllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllI 111II11111111
     0    20    40    60    80  100   120  140  160   180  200  220  240


                                      Time,   min

       Figure  <-T).  Cumulative filtrate volume for XC3 suspensions In runs  73 and  74 during low-
                  pressure filtration on Netzsch press.

-------
   14
   12
 0>
-+-•
-  8
  •s
 0)
 E  6
_D
~o
>  4
    0
                                                               I '  ' ' ' I
            . I
 I  I I I I  I I I I  I I I I  I I I I  I I I I  I I I I  I I I I  I I I I  I I I  I I I I  I I I I  I I I I  I	I-
       0
 40
80        120         160

      Time,  min.
200        240
        figure 21,
Cumulative filtrate volume for XC3 suspensions In runs  J-ll and J-12 during  low-

pressure filtration on .IWI press.

-------
          40
             80        120        160
                   Time,  min.
200         240
FlRure  22.
Cumulative filtrate volume for XC3 suspension in run J-Al during low-pressure
filtration on JWI press.

-------
Ln
Co
                ................ I ............ I ................ I  ....-
                          40
80        120        160

      Time,   min.
200         240
                Figure 2'J.  Cumulative filtrate volume for Xc3 suspension In run .1-121 during low-pressure

                          filtration on  tlie .IWI press.

-------
solids of 15.4, 16.3 and 16.2 percent, respectively.  Cake solids data for
runs 73 and 74 indicate improved performance with an increase in filtration
time from 101  min to 172 min as did runs J-41  and J-42 for an increase from
17 tnin to 135  min.  Since these four runs were conducted on the same day, it
       TABLE 16.   RESULTS FOR LOW-PRESSURE,  FIXED VOLUME FILTRATION
                  OF CLARIFIER UNDERFLOW SUSPENSION (XC3) ON NETZSCH
                  AND JWI PRESSES
Run
Filtration
 Pressure

   bar
 Time of
Filtration

   min
                                                  Dewatered Cake(s)
Mass
Solids Content
                                                        Total
                                                          %
                           Fixed
                             %
NETZSCH PRESS
52
73

74

6-7
6-7

6-7

224
172

101

1)
2)
1)
2)
1)
2)
5
5
5
5
5
5
.18
.23
.2
.0
.14
.14
16
15
15
15
14
14
.5
.9
.6
.1
.4
.0
12.6
12.1
_
-
-
-
JWI PRESS
J-11

J-12

J-41

J-42

J-121

6-7

6-7

6-7

6-7

6-7

176

100

135

17

130

1)
2)
1)
2)
1)
2)
1)
2)
1)
2)
0
0
0
0
0
0
0
0
0
0
.905
.905
.905
.91
.905
.905
.805
.855
.875
.880
16
16
15
15
16
16
9
7
15
15
.2
.4
.5
.3
.4-
.1
.4
.8
.4
.6
12.1
12.4
11 .8
11.8
.
-
-
-
_

is furthermore apparent that the data, as expected, indicate that increased
filtration time,- over a broad range, produced drier cakes, although run J-41
appeared to indicate higher performance at 135 rain than did run 73 at 172
min.   Subtle differences between the two presses, as well as analytical
variations, could have produced this difference.

Segregated Neutralization Suspensions

     Filtration studies with segregated neutralization suspensions were
generally conducted with suspensions formed by batch neutralization of spent
                                     59

-------
caustic-etch and anodizing-acid suspensions collected from the finishing
line at Plant A.  These segregated-neutrallzation suspensions, designated
AS7, were then generally dewatered on a filter press in a series of
experiments in which the suspensions were blended with conventional-
neutralization suspensions from Plants A and X (i.e., suspensions ACS and
XC3, respectively).   Each series typically included one run in which a
segregated-neutrallzation suspension was dewatered without blending with
another suspension.   The results of these filtration studies are presented
below with respect to suspensions with which the segregated-neutralization
suspensions were blended.

Blends with Plant A  Suspensions (ACS)—
     High-Pressure Filtration—The characteristics of a segregated-neutrali-
zation suspension and its blends with ACS suspensions examined with high
pressure filtration  are presented in Table 17.  Data for runs 171 and 173
indicate, respectively, the properties of the two suspensions.  Suspended
solids concentrations and pH values indicate two major differences for the
suspensions.  The segregated neutralization suspension (see run 173 in Table
17) was not thickened prior to analysis, while the conventional neutraliza-
tion suspension (see run 171  in Table 17) was collected from the underflow
   TABLE 17-   CHARACTERISTICS OF CLARIFIER UNDERFLOW (ACS)  AND SEGREGATED
              NEUTRALIZATION (AS7)  SUSPENSIONS AND BLENDS OF THESE
              SUSPENSIONS USED IN HIGH-PRESSURE FILTRATION  ON
              NETZSCH PRESS
Run
Blend
Date
Temp.
, pH
% by Volume

171
173
171
177
178
AS7
0
100
5
15
30
ACS
100
0
95
85
70

3*3
3*6
3*6
3*6
3*6
°C
13
13
12
12
9

8.2
10.6
8.7
9.0
"
Suspended Solids
Total
g/L
37.9
180.1
*7.0
59.9
78.0
Fixed
g/L
28
13*
35
U4
58
.2
.0
.0
.6
.0
Specific
Resistance
Evaluation
r
Tm/kg
3.6
1.3
3.9
2.5
2.5
Ck
%
11 .7
*3.8
10.8
21 .9
28.9
CST

s
139
U85
172
186
25*
of the gravity thickener at Plant A.  The suspended solids concentrations of
37.9 g/L (run 171)  and 180.1  g/L (run 173) indicate a wide disparity in
potential solids loadings for a filter press.  The specific resistance data
indicated that the  resistances to filtration were similar (i.e., 3.6 and 1.3
Tm/kg) but that the solids in the segregated-neutralization suspensions were
less hydrophilic or gelatinous, as indicated by a specific-resistance solids
content of *3.8 percent (run 173) vs.a value of 11.7 percent (run 171) at a
differential pressure of 0.5 bar (50 kPa).

-------
     The disparity in suspended solids concentrations for the two
suspensions created a problem in evaluating blends of the suspensions.
Blended suspensions with higher portions of segregated-neutralization solids
would be more concentrated and would require less filtration time to achieve
a specified degree of dewatered solids.  Notwithstanding this problem,
however, it was decided that the most appropriate method of analysis would
be through directly blending the two suspensions.  From a plant viewpoint, a
logical operational procedure would be to blend segregated-neutralization
suspensions with influent wastewaters discharged to the neutralization basin
and form one combined suspension that would be collected for dewatering from
the clarifier underflow.  Due to the limited volume of the experimentally-
produced segregated-neutralization suspension, this option was not available
at Plant A or Plant X and direct formation of blended suspensions on an
experimental level was required.  With regard to the origin of suspended
solids discharged to an aluminum-finishing plant, Saunders et al. (1984)
indicated that 65.6 percent of the aluminum mass flow Into the wastewater
treatment system at one plant was contained in the concentrated spent etch
and acid suspensions.  These two spent solutions constituted only 6 percent
of the total wastewater flow at the plant.  On this basis a blend of
approximately 6 percent (i.e., 6 volumes of segregated-neutralization
suspension to 94 volumes of conventional-neutralization suspension) would
appear to be appropriate at this plant.  However, the operational procedures
of the caustic etch system will significantly affect the mass flow of
aluminum discharged in the spent etch.  For example, 29 percent of the
overall aluminum mass flow at the above plant (Saunders e_t al., 198*0 was
contained in caustic-etch rinsewaters, indicating a high dragout rate from
the etch tanks and an apparent high-level of aluminum in the etching
solution.  Conversion of this etch tank to a system employing limited
chemical additives, or extenders, could be more economical, but would result
in increased mass flow of aluminum in spent etch.  Use of segregated
neutralization could, however, result in the production of less wet mass of
dewatered sludge solids, which would be attributable to a higher solids
content of the final dewatered sludge.  It was therefore deemed appropriate
to investigate blends of from 5 to 15 percent by volume to represent typical
values for those systems using segregated neutralization on a routine,
steady-state basis.  In some instances where batch treatment of spent etch
is practiced, blends of higher volumetric ratios could be utilized and for
this reason blends of 30 percent by volume were examined in some experiments.
Therefore, a series of experimental runs typically included runs with each
of the two suspensions and three blends of the two suspensions, as indicated
for example by runs 171, 173, 171. 177 and 178 in Table 17.

     The data indicating percent by volume for each blend were indicative of
the actual volumes employed, while the suspended solids data were based on
actual measurements of each blended or neat suspension, not calculations
using concentrations of initial suspensions.  The suspended solids concentra-
tions for blended and neat suspensions ranged from 37.9 to 180.1 g/L,
indicating considerable variation in the solids loading for each run.  The
specific resistance values did not vary considerably but followed a trend
consistent with blend variations.  Specific-resistance cake solids increased
dramatically as the percent of AS7 solids increased, ranging from 10.8
percent (run 174) to 28.9 percent (run 178).  CST values generally increased


                                      61

-------
as suspended solids concentration increased,  as expected.

     Cumulative filtrate volume data for the  high-pressure runs are
presented in Figure 24.   As the percentage of AST suspension increased,  the
volume of filtrate collected decreased considerably.   As shown in Table  18,
the solids content of the dewatered cake Increased (i.e., from 26.6 percent
(run 174) to 34.8 percent (run 178)) with increasing  percentage of AS7
suspension in a blend.   The addition of 5 percent AS7 suspension to the  ACS
suspension resulted in improved performance in that similar dewate^ed cake
solids concentration was achieved in 70 min,  as opposed to 103 rain for the
conventional-neutralization suspension alone.
        TABLE 18.   RESULTS FOR HIGH-PRESSURE,  FIXED-VOLUME FILTRATION
                   OF CLARIFIER UNDERFLOW (ACS)  AND SEGREGATED-
                   NEUTRALIZATION (AS7)  SUSPENSIONS AND BLENDS
                   OF THESE SUSPENSIONS

Run


171
173-
174
177
178


Blend
I by
AS7
0
100
5
15
30
Volume
AC3
100
0
95
85
70
Filtration
Pressure

bar
14-15
14-15
14-15
14-15
14-15
Time of
Filtration

min
103
35
70
87
70

Mass

kg
5.62
6.97
5.65
5.94
6.10

Solids
Total
%
26.3
53
26.6
33.1
34.8

Content
Fixed
%
18.8
_
20.1
-
"
     Low-Pressure Filtration—Characteristics of the various blends of AS7
and AC3 suspensions are presented in Table 19.   The five runs (J-81 through
J-85)  were conducted in parallel with the high-pressure runs in Table 17
with aliquots of the same suspensions and suspension blends.  Cumulative
filtrate volumes for the low-pressure runs are presented in Figure 25.  It
is apparent that filtration rates decreased with increasing levels of AS7 in
blended suspensions, as did the total volume of filtrate produced.  These
responses are reflective of the increased solids loadings and of the
improved dewatering properties of the AS7 suspensions.

     The results of the filtration studies are included in Table 20.  The
results indicated the positive impact of Increased addition of
segregated-neutralization solids to the blended suspension.  At similar
filtration times, the average dewatered-cake solids content increased from
23.7 percent for the conventional-neutralization suspension to 26.4, 30.8,
and 36.5 percent for 5, 15 and 30 percent blends, respectively.  These
values were similar to those achieved with high-pressure filtration at
shorter filtration times (see Table 18).
                                     62

-------
   80
   60
0)
v_
0)
 E
JD
 O
   20
     0
                         i  ....  i .... i.
       0
40
                                         •  171
                                         •  173
                                         *  174
                                         X  177
                                         O  178
80        120        160
      Time,  min.
200
240
        Figure 24. Cumulative filtrate volume for AC3 and AS7 suspensions and blended suspensions
                 in runs 171,  173,  174,  177 and 178 during li ^It-pressure filtration.

-------
   TABLE  19.   CHARACTERISTICS OF CLARIFIER UNDERFLOW (AC3)  AND SEGREGATED
              NEUTRALIZATION  (AS7)  SUSPENSIONS  AND BLENDS OF THESE -
              SUSPENSIONS  USED IN LOW-PRESSURE  FILTRATION ON JWI PRESS
Run
Blend
Date
Temp
. pH
% by Volume

J-8K17D*
J-82O73)
J-83M7*)
J-8*(177)
J-85(178)
AS7
0
100
5
15
30
AC3
100
0
95
85
70

3*6
3*6
3*6
3*6
3*6
°C
13
22
12
22
10
Suspended Solids
Total
g/L
8.2
10.6
8.7
9.0
—
37
180
*7
59
78
.9
.1
.0
.9
.0
Fixed
g/L
28.2
13*
35.0
**..6
58.0
Specific
Resistance
Evaluation
r
Tm/kg
3.6
1.3
3.9
2.5
2.5
ck
*
11.7
*3.8
10.8
21.9
28.9
CST

s
139
*85
172
186
25*
"Companion runs  performed  at  high  pressure  on  Netzsch  press.
        TABLE  20.   RESULTS  FOR  LOW-PRESSURE,  FIXED-VOLUME  FILTRATION
                   OF  CLARIFIER UNDERFLOW (AC3)  AND SEGREGATED-
                   NEUTRALIZATION  (AS7)  SUSPENSIONS AND BLENDS
                   OF  THESE SUSPENSIONS
Run
Blend
Filtration
Pressure
Time of
Filtration Mass
% by Volume

J-8K

J-82(

J-83(

J-8*(

J-85(


171)*

173)

17*)

177)

178)

AS7
0

100

5

15

30

AC3
100

0

95

85

70

bar
6-7

6-7

6-7

6-7

6-7

min
100

*5

110

100

107

Kg
1)
2)
1)
2)
1)
2)
1)
2)
1)
2)
0
0
1
1
0
0
1
1
1
1
.965
.970
.195
.195
.995
.985
.03
.03
.10
.085
Solids Content
Total
t
23
23
38
39
26
26
30
30
36
36

.8
.6
.7
.7
.7
.1
.7
.9
.5
.5
Fixed
%
16.7
16.7
_
-
19.5
19.2
23.6
-
-

 Companion  runs  performed  at  high  pressure on Netzsch press.
Blends  with  Plant  X  Suspensions  (XC3)~
     High-Pressure Filtration—Data for  characteristics of the blends
examined  by  high-pressure  filtration on  the Netzsch press are presented in

-------
Ln
                                                                    •  J-81
                                                                    •  J-82
                                                                      J-83
                                                                    X  J-84
                                                                   O J-85
           0 taj-1J
              0
40
80        120         160
      Time,  min.
200
240
                Figure 25.  Cumulative  filtrate volume for AC3 and AS7 suspensions and blended suspensions
                          in runs J-81 through .1-85 during low-pressure  filtration.

-------
Table 21.   A sample of the conventional-neutralization suspension alone was
not included in the 90-series  since this  suspension had been examined
extensively in previous studies  (see Tables 13 and 14 and Figures 16-18).
 93
       TABLE 21.   CHARACTERISTICS OF  A SEGREGATED NEUTRALIZATION (AS?)
                  SUSPENSION  AND  BLENDS WITH A CLARIFIER UNDERFLOW
                  SUSPENSION  USED IN  HIGH-PRESSURE FILTRATION ON
                  NETZSCH  PRESS
Specific
Resistance
Run


91
92
Blend
% by
AS7
15
35
Volume
XC3
85
65
Date


322
322
Temp.

°C
18
-
pH


8.3
8.6
Suspended
Total
g/L
42.2
71..7
Solids
Fixed
g/L
33.5
56.9
Evaluation
r
Tm/kg
1.5
1.4


1
1
ck
%
6.9
9.3
CST

3
90
134
100
322   36    8.8  173.2
137.5
1.5
49.5   581
Data in Tables  13  and  14  for  runs  81  and 82 were collected on the day prior
to the 90-series and will be  used  for comparative purposes.  As with AC3
suspensions,  the suspended solids  concentration increased dramatically for
the blended suspensions.   The specific resistance of the blends decreased to
that of the segregated-neutralization suspension but specific-resistance
cake solids increased  with increasing percentage of segregated-
neutralization  solids.

     The filtrate-volume  data for  runs 91-93 are presented in Figure 26 and
other results are  presented in Table 22.  For a 15 percent blend, the
dewatered cake  solids  increased from an average value of 16 percent (run 82
in Table 14)  to approximately 28 percent, while a 35-percent blend produced
a cake with an  approximate solids  content of 37 percent.

     Low-Pressure  Filtration—Characteristics of the blended suspensions
examined by low-pressure  filtration are presented in Table 23 and data in
Table 15 are  representative of the XC3 suspension used to make the blended
suspensions.

     Data in  Figure 27  and Table 24 indicate the filtration results for the
three low-pressure runs.   The dewatered solids for the 15-percent blend was
26.4 percent, while that  for  the XC3 suspension at a comparable time of
filtration (I.e.,  run  75  in Table 16) was 14.2 percent.  The 35-percent
blend produced  a dewatered-cake solids of 35.7 percent, while that for the
segregated-neutralization suspension was 50.8 percent.

     A comparison  of low- and high-pressure filtration for the AC3 and XC3
blends is presented in  Table  25.  In the majority of the instances,
high-pressure filtration  resulted in higher cake solids contents or
comparable solids  contents in less filtration time.  However, the
                                      66

-------
                       80        120        160
                             Time,   min.
200        240
Klgure 2t>.  Cumulative filtrate volume for an AS7 suspension (run 93) and Mends of XC3
          and AS7 suspensions In runs 91 and 92 during high-pressure filtration.

-------
        TABLE  22.   RESULTS  FOR  HIGH-PRESSURE,  FIXED-VOLUME  FILTRATION
                   OF  A  SEGREGATED  NEUTRALIZATION  (AST)  SUSPENSION
                   AND BLENDS WITH  A  CLARIFIER UNDERFLOW (XC3)
                   SUSPENSION

Run


91

92

93



Blend
% by
AS7
15

35

100

Volume
AC3
85

65

0

Filtration
Pressure

bar
14-15

14-15

^ i( — 1 5

Time of
Filtration

min
90

97

50




Mass


1 )
2)
1 )
2)
1 )
2)

kg
5.
5.
6.
6.
7.
7.


69
75
15
31
36
12

Solids
Total
%
26.9
28.9
37.4
36.8
53.8
54.2

Content
Fixed
5
_
-
-
-
-

       TABLE  23.   CHARACTERISTICS  OF  A  SEGREGATED  NEUTRALIZATION  (AS7)
                  SUSPENSION  AND BLENDS WITH  A  CLARIFIER  UNDERFLOW
                  (XC3)  SUSPENSION USED IN  LOW-PRESSURE FILTRATION
                  ON  JWI PRESS
Specific
Resistance
Run


J-51
J-52
Blend
% by
AS7
15
35
Volume
XC3
85
65
Date


322
322
Temp.

°C
20
20
pH Suspended Solids
Total
g/L
8.3 46.4
65.0
Fixed
g/L
36.8
51 .6
Evaluation
r
Tm/kg
1.3
0.9
ck

18.3
26.5
CST

s
88
100
J-53    100
322   20
8.7  155.1
123.1
1.9
47.0   345
        TABLE  24.   RESULTS  FOR HIGH-PRESSURE,  FIXED-VOLUME  FILTRATION
                   OF A SEGREGATED  NEUTRALIZATION  (AS7)  SUSPENSION
                   AND BLENDS WITH  A  CLARIFIER UNDERFLOW (XC3)
                   SUSPENSION
Run
Blend
Filtration
Pressure
Time of
Filtration Mass
J by Volume

J-51

J-52

J-53

AS7
15

35

100

XC3
85

65

0

bar
6-7

6-7

6-7

min
112

105

47 '

kg
1) 0.985
2) 0.945
1) 1.06
2) 1.05
1) 1 .22
2) 1.22
Solids Content
Total
*
26.0
26.8
35.8
35.6
51.6
50.0
Fixed
%
—
-
-
-
-
-
                                     68

-------
(T-
VO
                                                         I I  I I I  I I I I I I  I I I  I I I  I
I I  I I I  I I I  I I I  I I I  I I I  I I
                                        80         120         160

                                               Time,  min
                                                               200
240
                Fi'Mire 27.   Cumulative  filtrate volume for an AS7 suspension anil l>lends of XC3 and AS7
                           suspensions in runs .1-51  through J-53 during low-pressure filtration.

-------
    TABLE 25.   COMPARISON OF LOW- AND HIGH-PRESSURE FILTRATION FOR BLENDS
               OF CLARIFIER UNDERFLOW (ACS;  XC3)  AND SEGREGATED
               NEUTRALIZATION (AST)  SUSPENSIONS
Blend
% by Volume
ACS or
AST XC3
ACS Suspensions
0 100
0 100
5 95
5 95
15 85
15 85
30 70
30 70
XC3 Suspensions
0 100
0 100
15 85
15 85
35 65
35 65
100 0
100 0
Filtration
Pressure
bar

6-7
14-15
6-7
14-15
6-7
14-15
6-7
14-15

6-7
14-15
6-7
14-15
6-7
14-15
6-7
T4-15
Time of
Filtration
min

103
100
110
70
100
87
107
70

100
90
112
90
105
97
47
50
Average Dewatered
Cake Solids
%

23.7
26.3
26.4
26.6
30.8
33.1
36.5
34.8

15.4
26.0
26.4
27.9
35.7
37.1
50.8
54.0
differences  between  low-  and  high-pressure filtration only became
significant  with  the higher blends  with segregated-neutralization
suspensions.

VARIABLE-VOLUME PRESSURE  FILTRATION

     Selected  suspensions were  examined using variable-volume filtration
plates  (i.e.,  diaphragm plates)  on  the Netzsch press.  The chambers were
filled  with  sludge solids at  low pressure (i.e.,  6-7 bar)  and then
pressurized  to 15.5  bar.   Filtrate  volume was collected and monitored
throughout a run, which was terminated when filtrate production ceased.  All
of the  data  collected were for  suspensions from Plant A.
                                      70

-------
     Characteristics of clarifier underflow (AC3) suspensions examined using
a variable-volume filter press (i.e., diaphragm press) are presented in
Table 26.   No discernible variations in characteristics from previous
suspensions are apparent.  The data for run 161 were for a suspension (AC1)
collected  from the neutralization basin following a massive dump of
caustic-etch suspension at Plant A.  The sample had an exceptionally high
suspended  solids concentration and an elevated pH, both as expected for the
etch dump.

     Cumulative filtrate volumes are presented in Figures 28-33.  No
filtrate-volume data were collected for initial runs 111-113.  Data in
Figure 30  for runs 175, 176 and 179 are for blends of AC3 and AS7
suspensions and will be discussed later.  The data for run 172 were,
however, for an unaltered AC3 suspension.  Filtrate data for runs conducted
with the diaphragm press were divided into "filter" and "squeeze" cycles.
     TABLE 26.  CHARACTERISTICS OF CLARIFIER UNDERFLOW (ACS) AND NEUTRA-
                LIZATION BASIN (AC1) SUSPENSIONS FOR DEWATERING BY
                VARIABLE-VOLUME PRESSURE FILTRATION
Run
Type
Date Temp.
pH Suspended
Total

11
11
11

1
2
3

AC3
AC3
AC3

326
326
326
8C
17
17
17
g/L
24
24
24
.8
.8
.8
Specific
Resistance
Solids Evaluation
Fixed r
CST
ck
g/L Tm/kg
3
3
3
.2
.2
.2
10
10
10
f
.7
.7
.7
3
_
-
-
141   AC3
16
8.2
                                39.7
3.2
12.6
110
151
152
153
AC3
AC3
AC3
342
342
342
13
13
13
8.0
8.0
8.0
37.1
36.9
37.2
26
27
27
.7
.5
.1
3-3
3.3
3.3
12
12
12
.2
.2
.2
120
120
120
172   AC3   343    13
      8.2
       39.9
                                          28.8
3.6
11 .7
139
181    AC3    350     16    8.8     39.3       29.2
182    AC3    350     16    8.8     39.3       29.2
                                                     4.2
                                                     4.2
                                         12.8
                                         12.8
                                               149
                                               149
161   AC1* 303/3^1  11     9.1     82.3
 AC1 suspension collected on Day 303 immediately following caustic-etch dump
 to neutralization basin (see discussion of Table 1)


The squeeze cycle was initiated with the application of pressure  to  the
internal diaphragms in the recessed chambers.  As indicated by data  for the
squeeze cycle, each filter run was conducted until the filtration rate was
                                      71

-------
    100
    80
    60
 (U
    40
 o
>
     20
      0
        0
40
                                 141
                              •  Filter.
                              o  Squeeze.
80
                                       Time,  min.
120
           Figure 28.  Cumulative filtrate volume for AC3 suspension during run 141 using Netzscli
                    diaphragm press.

-------
    100
    80
 0)
.t±   60
 CD
     40
 o
     20
      0
         0
                                 t   i  I  i  i  r
                         •  151 Filter.
                         o  Squeeze.
                         •  152 Filter.
                         O  Squeeze.
                         *  153 Filter.
                         A  Squeeze.
                                                                        I  .   .  .   .  •
40                      80
      Time,  min.
120
          Figure 29.  Cumulative filtrate volume for AC3 suspension during runs  151-153 using Netzsch
                   diaphragm press.

-------
ou
c/i60
(U
+-»
cu"40
E
ZJ
o
•>
20
0
(
I'lVI'll-llllllllllllllllll
: • 172 Filter
1 o Squeeze
I -175 Filter
I D Squeeze
7 * 176 Filter
I ^ Squeeze
1 X 179 Filter
I X Squeeze

-" ^ - ^-l-^x'^^*-*'*
i i i i 1 i i i i 1 i i i i 1 i i i i 1 i i i i | i
3 40 80
i i 4
—
—
—
-
ill
12
                     Time,   min.
Figure  30
Cumulative filtrate volume for  AC3 and blends of AC3 and AS? suspensions during
runs  172, 175, 176 and 179 using Netzsch diaphragm press.

-------
   80
   60
 0)
 v_
 0)
 E
J3
 O
   20
     0
       0
                                 161
                              •  Filter.
                              o  Squeeze.
40                      80
      Time,   min.
                 I  I  J	1	I  I   I  I   I  I  I  I  I   I  I   I  I   I  I  I   I  I  I   I  I   I
120
         Figure 31.  Cumulative filtrate volume for AC3 suspension during run 181 using Netzscli
                  diaphragm press.

-------
   80
   60
 (D
 E
JD
~o
   20
     0
 i • • •  • i
       i • •  • • i • •  • • i • • • • i
                                                            •  182
        I  I I I I  I I I I i I I I I  I I I I  I I I I  I I I I  I I I I  I I I I I  I I I I I  I I I I  I I I I  I I I I  I I I I  I I I
       0
40
80        120        160

      Time,  min.
200        240
        Figure 32.  Cumulative  filtrate volume for AC3 suspension during run 182 (filter portion

                  only)  using Netzsch diaphragm press.

-------
ou
.60
en
^
(D
+-•
(D-40
>
20

161 :
• Filter. J
o Squeeze. I
^ —


0 20 40 60 80 100 12




0
                               Time,   min.
Figure 33.  Cumulative filtrate volume for ACl  suspension during run  161 using Netzsch
          diaphragm press.

-------
zero,  i.e.,  no additional water could be squeezed from the sludge ca.ke at a
pressure of  15.5 bar.

     Results of each run on the diaphragm press are presented in Table 27.
For AC3 suspensions, the solids content of dewatered cakes ranged from 25.4
to 31.2 percent and averaged 28.7 percent, for a broad range of operational
conditions.   While no attempt was made to systematically evaluate the
variables involved in diaphragm filtration, the results indicated the
relative insensitivity of the final cake solids content to these variables.
        TABLE 27.   RESULTS FOR VARIABLE-VOLUME, PRESSURE FILTRATION
                   OF AC3 and ACT  SUSPENSIONS ON NETZSCH PRESS
Run
Time of Filtration, min
Mass

111

112

113

141

151

152

153

161
172
181
182
Filter
11

32

50

49

81

58

33

30
20
70
90
Squeeze
8

-

17

17

20

20

20

30
50
20
5
kg
1)
2)
1)
2)
1)
2)
1)
2)
1)
2)
1)
2)
1)
2)




1
1
2
2
3
2
3
3
3
3
3
3
2
2
4
2
3
5
.93
.81
.30
.21
.13
.94
.2
.3
.5
.46
.5
.33
.16
.72
.98
.10
.46
.58
Dewatered Cake(s)
Thickness
mm
8
8
-
-
-
-
17
17
20
20
-
-
15
15
_
»
—
29
Solids Content
Total
27
27
25
25
25
25
30
31
30
30
29
30
30
30
49
31
30
26
.4
.1
.4
.4
.4
.4
.9
.2
.1
.3
.8
.3
.0
.6
.6
.1
.5
.1
Fixed
_
-
-
-
-
-
—
-
_
-
-
-
-
-
_
_
_

The results for run 161  for an AC1 suspension produced following a massive
batch dump of spent caustic etch indicated the excellent dewatering
properties of this segregated-neutralization-like suspension.

     Data in Tables 28 and 29 are for a series of runs conducted on blends
of AC3 and AS7 suspensions to determine the impact of segregated neutraliza-
tion solids on dewatering of conventional suspensions.  Despite broad
variations in solids loadings and time of filtration, it was apparent  that
                                     78

-------
      TABLE 28.   CHARACTERISTICS OF CLARIFIER UNDERFLOW (AC3)  AND BLENDS
                 WITH A SEGREGATED NEUTRALIZATION (AST) SUSPENSION FOR
                 VARIABLE-VOLUME,  PRESSURE FILTRATION
Run
Blend
Date
Temp.
PH
% by Volume

172
175
176
179
AS7
0
5
15
30
AC3
100
95
85
70

3*3
3*3/3*6
3*3/3*6
3*3/3*6
8C
13
12
13
10
Specific
Resistance
Suspended Solids Evaluation
Total
g/L
8.2
8.7
9.0
™
39
*7
59
78
.9
.0
.9
.0
CST
Fixed r C|<
g/L Tm/kg
28.8 3.
3.
**.6 2.
58.0 2.
6
9
5
5
%
11 .7
10.8
21.9
28.9
3
139
172
186
25*
        TABLE 29.   RESULTS FOR VARIABLE-VOLUME,  PRESSURE FILTRATION
                   OF AC3 AND BLENDS OF AC3 AND AS7 SUSPENSIONS
Run
Blend
Time of Filtration, min
J by Volume

172
175
176
179
AS7
0
5
15
30
AC3
100
95
85
70
Filter
20
58
70
60
Squeeze
50
18
20
20
Dewatered Cake(s)
Mass Thickness
kg
2.10
3.56
3.96
3.20
mm
16
_
20
18
Solids
Total
31.1
32.1
37.6
*2.9
Content
Fixed
_
.
-

segregated neutralization enhanced the extent to which a suspension could be
dewatered.  Comparison with data in Table 18 for high-pressure fixed-volume
filtration of other  aliquots of the blended suspensions indicated that
diaphragm filtration produced cakes with significantly higher solids
contents  (e.g.,  for  blends of 5, 15 and 30 percent,  high-pressure filtration
gave solids contents of 26.6, 33.1 and 3*.8 percent, while diaphragm
filtration gave  values of 32.1, 37.6 and *2.9 percent, respectively).
Diaphragm filtration therefore produced superior dewatering performance when
compared  to fixed-volume pressure filtration.

SUMMARY ANALYSIS OF  PRESSURE FILTRATION

     The  primary objective for investigation of pressure filtration systems
was to establish the extent to which aluminum-finishing sludges could be
dewatered in preparation for acidic aluminum extraction.  To produce a
commercial-strength  product of aluminum sulfate [A^tSOij^] In the liquid
                                     79

-------
form, i.e., "liquid alum", previous studies indicated that a dewatered-
cake solids content of >_21 percent was required (Saunders _et al. ,  1934).

Fixed-Volume Pressure Filtration

     The pressure-filtration data collected during fixed-volume filtration
generally indicated that it was possible, in many instances, to achieve a
cake solids content of >_21 percent.  However,  the filtration studies were
infrequently run to completion and, in many series, replicate runs were
terminated prematurely to determine the extent of dewatering during a run.
It was therefore necessary to examine the data and predict, if possible,
performance of the filtration runs as they would be conducted in practice.

     Numerous investigators have indicated that pressure filtration for
dewatering proceeds through three phases.  In  a companion study, Johns
(1987) indicated that during the final consolidation phase of a pressure
filter run, the rate of filtration became linear with the time of filtration.
Johns (1987) furthermore indicated that the ultimate volume of filtrate to
be produced on a given run could be predicted  in accordance with such a plot.
To illustrate this point, filtration rate data for several runs on the
Netzsch press are presented in Figures 34-37.   As indicated in Figures 34
and 35, replicate runs (e.g., runs 24-26 and 27-29 in Figure 34) were highly
reproducible.  Using the terminal linear portion of each curve, or sets of
replicate curves, the ultimate volume of filtrate was estimated.  For
example, the ultimate filtrate volume for runs 24-26 was equal to 0.052
m^/m2 of filtration area or a total filtrate volume of 36 L for the
2-chamber run.  Using procedures developed by  Johns (1987) in a companion
study, the density of the dry aluminum-finishing solids was estimated to be
3200 kg/m3.  Using these data and the projected volumes of filtrate, the
maximum values for the solids content, C^, of  the dewatered cakes were
estimated using the following equation (Johns, 1987).

                                   V C
                       k   V * (CoVps * p)/Ps)


where V0 - Volume of influent suspension processed
      C0 - Suspended solids concentration of influent suspension
      Vc - Total volume of press chambers
      p  - Density of filtrate
      ps - Density of dry sludge solids.

In addition,  since filter presses are not typically run to produce the
projected ultimate volume of filtrate, the solids content at the point of
production of 90 percent of the ultimate filtrate volume was estimated.

     Data for Plant A suspensions are presented in Table 30.  It was
projected that high-pressure filtration of AC3 suspensions could produce
dewatered cakes with solids contents of from 26.5 to 34.9 percent.  The data
indicated that solids content generally increased with influent suspended
solids concentration.  However, three of the data sets (i.e., 36.3 g/L, 72.3


                                     80

-------
00
        1.00
       0.80
     £
     o
     (D
   0.60
 o
o:
 c
 o
       0.40
     o
    u.
        0.20
                   I I  T I I I  I
                                       T  I I  T
                                                               II  I  I I  I  1 I
                 t
        0.00
         0.000
• 24
• 25
» 26
x 27
O 28
+ 29
                                                                I  .
                      0.050           0.100           0.150
                               Volume,   m3/m2.
               0.200
              higure 34.  Filtration rate data for serles-20 runs with AC3 suspensions for projection
                       of ultimate filtrate volume.

-------
oo
       1.00 fT"1 ''  '| i  i  'r
       0.80
    E
    o
      **
    CD
    -*-»
    O
0.60
     C
     o
    -*-'
     o
       0.40
       0.20
       0.00
         0.000
                                fy  .
• 31
• 32
» 33
x 34
O 35
+ 36
X 37
                    0.050           0.100           0.150
                             Volume,   m3/m2.
                0.200
               Figure 35.
                  Filtration rate data for series-30 runs with AC3 suspensions for projection
                  of ultimate filtrate volume.

-------
oo
U)
        1.00
       o.8o
     E
     o

      •*
     0)
    -»-•
     o
   0.60
 c
 o
-+-'
 o
       0.40
        0.20
              *f • I  r~l '  '  I  ' r
        0.00 r '  ' '  '
         0.000
                                                           •  51

                                                           •  81

                                                           •  82

                                                           x  121
                                                                   I I  I  I
                      0.050           0.100           0.150

                                Volume,  m3/m2.
0.200
              Figure 36.  Kt 1 lr.it i on rate data for selected nins wltli XC3 siispons ions for projection

                       of  ultimate filtrate volume.

-------
CD
        1.00

      •
     c

     £ 0.80
     £
     o


     (D
    -4-*
     O
   0.60
c
o
        0.40
     o
        0.20
        0.00
         0.000
                                      • 171
                                      • 173
                                      » 174
                                      x 177
                                       178
                                      + 182
                                                                I  I  I  I I
0.050           0.100

          Volume,
                                                           0.150
0.200
               Figure 37.   Filtration rate data for selected AC3- and AST-suspensions and blends of each
                         for projection of ultimate  filtrate volume.

-------
   TABLE 30.   PREDICTED SOLIDS CONTENTS OF DEWATERED CAKES FOR SUSPENSIONS
              FROM PLANT A DURING HIGH- AND LOW-PRESSURE FILTRATION

 Average          Predicted Cake Solids Content      Number of Runs Included
Suspended         	            in Analysis
 Solids           Ultimate    at 90J Completion           (run numbers)
   g/L	%	%	

High-Pressure Filtration
20.6
30.9
36. 3a
56.9
72. 8a
93. Oa
27
27.2
28.4
26.5
29.0
34.9
2^.9
25.2
25.6
24.7
26.1
32.7
1
1
3
4
3
3
(71)
(16)
(27-29)
(6-9)
(24-26)
(21-23)
  104.pt>	45	42.4	3  (18-20)

Low-Pressure Filtration
16.9
20.8
32. 5a
54.0
63. 9a
24.8
24.6
23.8
20.7
21 .8
22.7
22.6
21 .4
19.3
19.6
1
1
3
4
4
(102)
(72)
(35-37)
(10-13)
(31-34)
aAC3 suspensions collected from clarifier underflow during period
 (Days 303-314)  in which etch dump of Day 303 affected sludge
 properties (see Table 1)
bSuspension collected directly from neutralization basin during
 etch dump on Day 303 (see Table 1)
g/L and 93 g/L)  were for AC3 suspensions which were influenced by an etch
dump (see Table  1).   Therefore,  ignoring these data, high-pressure
filtration of typical AC3 suspensions should produce an ultimate dewatered
cake solids content  of approximately 27 percent and, at 90 percent
completion of a  filtration run (I.e., collection of 90 percent of projected
ultimate filtrate volume), a solids content of approximately 25 percent.
Therefore, high-pressure filtration should be adequate for producing a
dewatered cake suitable for production of liquid alum.

     The influence of the etch dump of Day 303 on sludge dewatering
properties is apparent from the  data for a suspended solids concentration of
104 g/L.  This AC1 suspension was collected directly from the neutralization
basin during the etch dump and was not polymer conditioned.  The ultimate
solids content of 45 percent was 1.67-fold higher than that typical of ACS
suspensions and  would result in  a 40 percent reduction in volume of wet
sludge for disposal.

                                     85

-------
     The ultimate solids content projected for low-pressure filtration of
ACS suspensions ranged from 20.7 to 24.8 percent.  Again, ignoring those
data influenced by the etch dump, low-pressure filtration would produce a
sludge with an ultimate solids content of approximately 23.4 percent and, at
90 percent completion of a filter-press run, a value of 21.5 percent.
Therefore, low-pressure filtration would theoretically be suitable for use
in conjunction with alum production but has no margin for variation in
solids content if a minimum solids content of 21  percent is required.

     Data for the single, XC3 suspension from Plant X are presented in Table
31.  The data for high-pressure filtration were slightly lower than those
for AC3 suspensions, i.e.,  ultimate values of 27 percent (AC3) vs 24.5
percent (XC3).  Low-pressure filtration was similarly lower than that for
AC3, i.e., 23.5 percent (AC3) vs 17.6 percent (XC3).  Therefore, the XC3
suspension could not be as effectively dewatered as could the AC3 suspension
and only high-pressure filtration could be used with an alum production
system.
     TABLE 31.   PREDICTED SOLIDS CONTENTS OF DEWATERED CAKES FOR XC3
                SUSPENSIONS DURING HIGH- AND LOW-PRESSURE FILTRATION

 Average          Predicted Cake Solids Content      Number of Runs Included
Suspended         	            in Analysis
 Solids           Ultimate    at 90* Completion           (run numbers)
   g/L	I	%	

High-Pressure  Filtration

   20.7             24.5            22.5                4  (51 ,  81 , 82, 121 )

Low-Pressure Filtration

   20.7             17.6            16.2                3  (52,  73, 74)
     To indicate the impact of implementation of segregated neutralization
on overall  sludge dewatering performance, data for the 170-series (i.e.,
runs 171,  173,  174,  177 and 178)  are presented in Table 32.  First, the
conventional-neutralization suspension examined in run 171 had excellent
dewatering  properties,  as compared to those for other AC3 suspensions, with
a projected ultimate solids content of 28.7 percent.  As the volumetric
percentage  of segregated-neutralization suspension increased, the ultimate
solids content increased from 33.0 to 39.2 percent, with a projected
ultimate value of 51.3  percent for the neat segregated-neutralization
suspension.  Therefore, segregated-neutralization solids have a dramatic
impact on the ultimate  volume of  sludge produced and can enhance the
potential of dewatering systems with respect to alum production.
                                     86

-------
    TABLE 32.  PREDICTED SOLIDS CONTENTS OF DEWATERED CAKES FOR AN ACS
               SUSPENSION AND BLENDS OF ACS AND AS? SUSPENSIONS DURING
               HIGH-PRESSURE FILTRATION

Suspended
Solids
8/L
37.9
47.0
59.9
78.0
180.1

Portion as AS7 Suspension
by Volume
%
0
5
15
30
100
Predicted

Ultimate
I
28.7
33.0
36.6
39.2
51.3
Cake Solids Content

at 90% Completion
%
26.6
30.7
34.2
36.7
48.6
Variable-Volume Pressure Filtration

     The limited studies conducted with the diaphragm press indicated
excellent performance in dewatering all aluminum-finishing sludges.  For AC3
suspensions,  the solids content of dewatered cakes, following removal of all
filtrate at a squeeze pressure of 15.5 bar, ranged from 25.4 percent to 31.2
percent and averaged 29.4 (see Table.27).  This was slightly higher than the
average value of 27 percent achieved with high-pressure (14-15 bar)
filtration of AC3 suspensions, indicating the limited utility of the more
complex diaphragm press over a high-pressure, fixed-volume press.

     With blends of conventional (AC3) and segregated (AS7) neutralization
suspensions,  the diaphragm press produced sludges with solids contents
ranging from 31.1 percent (5 percent by volume of AS7) to 42.9 percent (30
percent by volume of AS7), while the ultimate solids contents projected for
high-pressure filtration were 33 and 39.2 percent, respectively (see Tables
29 and 30).  The diaphragm press, therefore, may have potential for
application when overall system design is examined.  However, based on
potential for production of a high-solids-content sludge it appears to have
only minimal  benefit.
                                      87

-------
                                SECTION 6

           LIQUID ALUM PRODUCTION FROM DEWATERED SLUDGE SOLIDS
     Aluminum-finishing plants produce numerous wastewater residues in
conjunction with the etching,  anodizing and associated surface-treatment
processes used to apply finishes to extruded aluminum products.  These
residues are produced in several distinctively different ways; three of the
most common residues were examined herein.

     Gelatinous aluminum hydroxide suspensions are conventionally produced
by neutralization of dilute plant rinsewaters at ambient temperatures in a
treatment facility.   The pH of rinsewaters are typically adjusted with spent
caustic etch and finishing-acid suspensions.  These conventional neutraliza-
tion (CN) sludges are produced at virtually all aluminum finishing plants.

     An innovative treatment process developed by Saunders _e_t al. (1984) to
reduce sludge volumes is referred to as segregated neutralization.  Spent
caustic etch is neutralized with spent finishing acid to an alkaline pH
value (e.g., pH - 8-10) at elevated temperatures producing crystalline,
aluminum-hydroxide precipitates such as pseudoboehmite.  These crystalline,
segregated-neutralization (SN) solids can be dewatered separately or
combined with conventional neutralization suspensions and collected as a
mixture.

     Finally, the introduction of etch recovery systems by Alcoa, Inc. and
Fugi Sash,  Ltd. has  resulted in a third type of sludge residue.  Etch
recovery solids are  produced at elevated temperatures (e.g., 60°C) following
seeding of  caustic-etch solutions with aluminum trihydrate crystals.  The
etch recovery (ER),  or aluminum-trihydrate, suspension produced is periodi-
cally removed from the crystallizer and dewatered.  The sludge suspension
can be mechanically  dewatered to solids contents in excess of 90 percent.

     Conventional-neutralization, segregated-neutralization and etch-
recovery sludges were examined separately and in mixtures as indicated below.
The conventional neutralization sludges were obtained from clarifier under-
flow (i.e.,  AC3) at  Plant A and were dewatered at low-pressure (6-7 bar) on
the JWI press, due to the unavailability of the high-pressure Netzsch press
during this  portion  of the study.  Segregated neutralization suspensions
were produced by batch neutralization of caustic etch solutions from Plant A
with conventional anodizing acid and dewatered on the JWI press.  Etch
recovery solids were provided by personnel from Plant A through a companion
aluminum-finishing facility.  The identification codes used for these
dewatered sludges in subsequent tables and figures are: CN - conventional
neutralization sludge; SN » segregated neutralization sludge; and ER * etch

                                     88

-------
recovery sludge.  In addition to each individual sludge, two blends were
examined, i.e., CN/SN and CN/ER.  A blend of segregated neutralization  (SN)
and etch recovery (ER) sludges was not examined since these sludges are
produced in processes which are focused on treatment of spent caustic etch
and are not mutually compatible.

ACIDIC EXTRACTION OF ALUMINUM FROM DEWATERED SLUDGE CAKES

     Acidic extraction of dewatered sludge samples was accomplished in a
covered batch reactor with a temperature-control system.  Sulfuric acid,
water and sludge additions were carefully controlled and the extraction
mixtures were closely monitored throughout the extractions.  The results
from all extraction runs completed are presented below.

Conventional Neutralization Sludge Cakes (CN)

     Two separate samples of dewatered conventional-neutralization sludge
cakes were examined.  The characteristics of these cakes are presented  in
Table 33.  The solids contents of the two cakes were well below 21 percent
but were typical of those produced by low-pressure filtration of conven-
tional neutralization sludges (see Table 8).  The fixed (550°C) solids
contents were 73 and 80 percent of the dry (103°C) solids and the aluminum
contents were 35.6 and 39.2 percent (on a fixed-solids basis).  By
^Tiparison,  aluminum hydroxide precipitates represented by the formula
   "H)^ theoretically contain 34.6 percent aluminum.  The dewatered cakes
   e, therefore, rich in aluminum and the fixed solids were similar in
aluminum content to AKOH)^.  The 20 to 27 percent of dry solids lost
following ashing at 550°C was attributed to moisture retained by the
gelatinous aluminum precipitates; was not indicative of volatile organic
matter; and was typical of aluminum-finishing sludges examined previously by
Saunders etal. (1982, 1984).
     TABLE 33-  CHARACTERISTICS OF DEWATERED CONVENTIONAL NEUTRALIZATION
                SLUDGE CAKES CN-1 AND CN-2

Parameter                                                       Sludge Cake
	CN-1     CN-2

Dry (103°C) Solids Content, g dry solids/100 g wet cake	  18.1     17.4

Fixed (550°C) Solids Content, g fixed solids/100 g wet cake... 13-3     13.9

Aluminum Content, g Al/100 g fixed solids	  35.6     39.2
     A total of three extractions, or runs, were made with the CN-1 sludge
cake and two runs were made with the CN-2 cake.  The initial experimental con-
ditions employed with cake CN-1 are presented in Table 34.  Between 0.54 and
0.74 kg of wet sludge was employed in the runs.  The dose of acid added to each
run was based on the fixed solids at the rate of 1.89 g f^SO^/g fixed solids


                                     89

-------
          TABLE 3U.  INITIAL EXPERIMENTAL CONDITIONS FOR EXTRACTIONS
                     OF CONVENTIONAL NEUTRALIZATION SLUDGE CAKE CN-.1

Parameter                                                       Run
                                                        1-1     1-2     1-3
SLUDGE ADDITION

  Total Mass, g	  709.5   5^1.0
      Fixed Solids, g	   9H.1    56.8    98.7
      Moisture, g	  615.1   U8M.2   645.8

SULFURIC ACID

  Sulfuric Acid Addition, g	  163.5   107.1   186.1
  Percent of Stoichiometric Acid Dosage, Percent	   89.2    97.2    97.2
CONTROL TEMPERATURE, °C 	
ESTIMATED PRODUCT CONCENTRATION,
Percent as Al20o 	

	 90
	 7.3

70
5.9

50
7.1

(see equation (1) in Section 4).  After conducting these initial runs,
aluminum-content values were obtained and the acid dose was then evaluated
relative to the Stoichiometric requirements for acid (i.e., an acid dose of
5.1*1* g HgSOij/g Al).  Based on these post-experiment measurements, the
percent of the Stoichiometric acid dosage was calculated.  As indicated in
Table 3^t the sludges were dosed with 89 to 97 percent of the Stoichiometric
quantity of acid, indicating a slight acid deficiency.  The nominal control
temperatures for the reactors ranged from 50°C to 90°C.  Furthermore, the
estimated product concentrations, based on the fixed solids and acid
quantities employed, ranged from 5.9 to 7.3 percent as AljO^.  These values
were well below the desired concentration of 8 percent as A1203 and were
reflective of the low solids content of the sludge cakes employed.  Although
this could not be improved upon with the available low-pressure press, the
purpose was to monitor the kinetics of the extraction and develop a material
balance for it.  In this light, the low solids content (i.e., high moisture
content) was not viewed as a deterrent to the study.

     As indicated in Table 35, two runs were conducted with cake CN-2 at
nominal control temperatures of 90°C (run 2-1) and 50°C (run 2-2) to confirm
results obtained with cake CN-1 and to conduct runs at 100-percent,
stoichiotnetrlc-acid doses.  As indicated in Table 33, the aluminum content
of cake CN-2 was higher than that for cake CN-1, allowing for runs with
estimated product concentrations of 7.9 to 8.0 percent as A1203.  Further-
more, the acid doses used for runs 2-1 and 2-2 were based on aluminum
content and not on fixed solids, allowing for the 100-percent, stoichio-
metric-acid dose.
                                     90

-------
          TABLE 35.  INITIAL EXPERIMENTAL CONDITIONS FOR EXTRACTIONS
                     OF CONVENTIONAL NEUTRALIZATION SLUDGE CAKE CN-2

                                                                   Run
Parameter                                                     2-1       2-2
SLUDGE ADDITION

  Total Mass,  g	  611.5     641 .5
      Fixed Solids,  g	   84.9      89.4
      Moisture, g	  526.6     552.1

SULFURIC ACID

  Sulfuric Acid Addition, g	  181.4     191.0
  Percent of Stoichiometric Acid Dosage, Percent	  100.0     100.0

CONTROL TEMPERATURE,  °C	   90        50

ESTIMATED PRODUCT CONCENTRATION
  Percent as AljO	    7.9       8.0
     As indicated in Tables 3^ and 35,  the control temperatures (i.e., the
desired reaction temperatures) for the  runs with sludge cakes CN-1  and CN-2
were either 508C, 70°C or 90°C.  These  temperatures were to be maintained
throughout each run with the exception  of the initial reaction period in
which sulfuric acid was added to dewatered sludge cakes.  During this
initial period, temperature could not in all cases be controlled due to the
highly exothermic reaction resulting from the addition of sulfuric acid to
the moist sludge cakes.  Data presented in Figures 38 and 39 indicate the
temporal variation in temperature for runs with sludge cakes CN-1  and CN-2,
respectively.   Although not presented in the figures, the sludge cakes were
at ambient temperature (e.g., 25°C ± 5°C) prior to acid addition.   As shown
in Figure 38,  initial reaction temperatures for runs 1-1 and 1-3 were
between 83 and 858C, while run 1-2 was  at 66°C.  The temperature of runs 1-1
and 1-2 remained near the initial values, while run 1-3 cooled to the
desired control temperature of 50°C over a 1.1-h period before the
temperature control system was activated.

     As indicated in Table 36 for the time interval over which temperature
could be controlled at the desired level, average temperatures were 89°C,
71 °C and 51°C  for runs 1-1, 1-2 and 1-3, respectively.  Within the limits of
one standard deviation of the average value, the desired control
temperatures were achieved.  Data in Figure 39 for runs 2-1 and 2-2 indicate
that both runs were initially at temperatures of 89 to 90°C.  Run 2-1 was
effectively controlled near 90°C throughout the run, while run 2-2 cooled to
and was controlled at near 50°C after a 1-h period.  The average
temperatures for the time period of control were 88°C and 52°C,
respectively,  for runs 2-1 and 2-2, as  shown in Table 36.  In addition,  the
desired control temperature was within  the limits of one standard deviation


                                    91

-------
  100

   90

<->  80

I  70
"o
£60
E
£  50

   40

   30
                                               RUN
                                              • 1-1
                                              • 1-2
                                              T 1-3
                             Time, hours
Figure 38.  Temperature of reactor contents during acidic extraction
            of sludge cake CN-1 in runs 1-1 through 1-3.

0
l_
"o
V
a
E
•-

IUU
90
80
70
60
50
40
VI
i x«
[ *^S "~"~~-."~~~— •^___ — — %
-\ RUN
i \ 'i~i
I \
i- \
>w - - - •
. ... i .... i .... i .... i .... i .... i ,,., i ..
-
•i
•i
-i
-•
-i
                                                          6
                            Time,  hours
Figure 39.  Temperature of reactor contents during acidic extraction
            of sludge cake CN-2 in runs 2-1 and 2-2.
                                 92

-------
         TABLE 36.  ACTUAL AND CONTROL TEMPERATURES FOR EXTRACTIONS
                    OF CONVENTIONAL NEUTRALIZATION SLUDGE CAKES
                    CN-1  AND CN-2

Parameter

TIME INTERVAL, hours 	
CONTROL TEMPERATURE, a °C...
ACTUAL TEMPERATURE5
Average, °C 	
Standard Deviation, °C...


1-1
0.13-6
90

89
2.7
CN-1
Run
1-2
0.2-6
70

71
3.0
CN-2
Run
1-3
1.1-6
50

51
1.6
2-1
0.1-6
90

88
1.9
2-2
1-6
50

52
3.7
aDesired temperature selected for temperature-control system.
^Temperature data for indicated time interval.
of the actual average temperature for runs 2-1  and 2-2.   For this primary
reason, desired control temperatures will be used in all subsequent runs to
indicate reactor temperatures and will also be  used for  runs with sludges
CN-1  and CN-2.

     The aluminum concentration of extract filtrates (filter pore diameter =
0.45  ym) was used to monitor the progress of each run,  as presented for CN-1
and CN-2 cakes  in Figures 40 and 41.  In each ef the extraction runs, the
initial sample  was collected after approximately 0.5 h.   This interval was
selected since  it was virtually impossible to collect a representative
sample earlier  because of the initial time required to produce a uniformly
mixed slurry from the dewatered-cake and acid mixture.   However, the
extraction of aluminum from cakes CN-1 and CN-2 was, in all instances, near
completion after 0.5 h.  Over the remaining period of 5.5 h, filtrate
aluminum concentration increased gradually, further indicating the rapid
rate  and extent of the initial reaction.

     The effects of temperature on the extractions were relatively minor.
The runs at control temperatures of 90"C, however, resulted in the highest
rates of extraction although those for control  temperatures of 50°C and 70°C
were  only slightly lower.  This apparent insensltivlty to the effects of
temperature was attributable, in part, to the initial robust reaction which
occurred upon the addition of acid.  As shown in Figures 38 and 39. initial
reaction temperatures actually measured ranged  from 82°C to 90°C, with the
exception of 66°C for run 1-2.  Localized temperatures at the point of acid
addition presumedly exceeded these values.  Furthermore, reaction tempera-
tures after 0.5 h ranged from a low of 69°C (runs 1-2,  1-3 and 2-2) to 92°C
(run  1-1), indicating that during the period of initial reaction,
temperatures remained relatively high.  Therefore, since filtrate aluminum
concentrations, as shown in Figures 40 and 41,  indicated the extractions
were  at or near completion within the initial 0.5 h, temperatures at  or


                                     93

-------
                                     CONVENTIONAL NEUTRALIZATION
                                       DEWATERED SLUDGE CAKE

                                                 STOICHIOMETRIC
                                        RUN  TEMP  ACID DOSAGE
                                        1-1  90 "C  89 PERCENT
                                        1-2  70 "C  97 PERCENT
                                        1-3  50 *C  97 PERCENT
                                                                     8
                                    TIME, HOURS
Figure 40.   Filtrate  aluminum  concentration  for sulfuric-acid extractions of
             sludge cake CN-1 during runs  1-1,  1-2 and  1-3.
                                     CONVENTIONAL NEUTRALIZATION
                                       DEWATERED SLUDGE CAKE
                                                STOICHIOMETRIC
                                      RUN  TEMP   AGIO DOSAGE
                                      2-1  90 X  100 PERCENT
                                      2-2  50 "C  100 PERCENT
                                   TIME,  HOURS

Figure 41.   Filtrate  aluminum  concentration  for sulfuric-acid  extractions  of
             sludge cake  CN-2 during runs  2-1 and 2-2.
                                       94

-------
above 708C were apparently sufficient to assure completion of the extraction.
As shown in Figures 38 through 41, this was valid for runs 1-3 and 2-2 which
were set at control temperatures of 50°C, although this temperature was not
achieved until approximately 1.0 to 1.2 h into the extraction runs.
Therefore, the effect of reaction temperature within the limits of the
experimental conditions used were minimal due, in part, to the lack of
temperature control in the initial reaction period.  It is therefore
possible that temperature could play a significant role in the extraction
if, for example,  the environmental conditions associated with the reaction
vessel suppressed reaction temperature throughout an extraction.  While this
may be possible under some conditions, it is not expected to occur within
the context of the environmental conditions associated with typical indus-
trial process and wastewater-treatment facilities.  Furthermore, in a larger
batch reactor system employed in an industrial plant with a lower surface-
area-to-voluae ratio, less heat of reaction would be lost (compared to that
for the laboratory beaker) and elevated reaction temperatures would be
expected for a longer time period than that indicated in Figures 38 and 39.

     Material balances for total mass of the reactor contents and aluminum
mass for all runs with cakes CN-1 and CN-2 are included in Table 37.  During

   TABLE 37.  MATERIAL BALANCES ON TOTAL MASS AND ALUMINUM FOR EXTRACTIONS
              OF CONVENTIONAL NEUTRALIZATION SLUDGE CAKES CN-1 AND CN-2
Parameter
1-1
Run
1-2
Run
1-3
2-1
2-2
TOTAL MASS

  Input:
      Fixed Solids,  g...     94.4       56.8       98.7       84.9       89.4
      Moisture,  g	    615.1      484.2      645.8      526.6      552.1
      Acid, g	    163.5      107.1      186.1      181.4      191 .0
      Total,  g	    873-0      648.1      930.6      792.9      832.5

  Output,  g	    837.0      522.5      913-3      667.2      814.4

  Percent  Recovered	     96         81         98         84         98

ALUMINUM

  Input, g	     33.6       20.2       35.1       33-3       35.1

  Output:
      Soluble, g	
      Residual, g	
      Total,  g	

  Percent Recovered	
                                     95

-------
each run, mixed aliquots of reactor contents were withdrawn at 0.5- to 1 -h
intervals.  The mass and aluminum content of these aliquots were determined
and accumulated with similar values for the volume remaining at the end of
the run.   The data for runs 1-1, 1-3, and 2-2 indicated that the material
balance on total mass closed (i.e., output mass was equal to 96-98 percent
of input  mass).  Material balances for runs 1-2 and 2-1 were not as
complete, as indicated by recoveries of 81 to 8M percent.  A major portion
of the mass loss could be attributed to evaporation.  Although the reactor
was covered, it was not permanently sealed, allowing for the loss of water
at the elevated temperatures employed.  With respect to an aluminum balance,
within the typical limits of detection for aluminum analyses, the aluminum
recoveries were excellent, ranging from 99 to 10M percent.

     The  extent of the extractions are graphically indicated in Figures UO
and Ml.  To more precisely indicate the quality of the extracted product,
the characteristics of the terminal sample taken for each run are presented
in Table  38.  Commercial-strength, liquid alum has a nominal aluminum
concentration of 8 percent as A^O^.  This minimal criterion was met or
exceeded  in runs 1-1, 2-1 and 2-2 and was approached in runs 1-2 and 1-3.
Given that the moisture contents of the initial sludge cakes were noted
previously as being higher than desired (i.e., desired moisture content of
<79 percent) and that the estimated product concentrations ranged between
5.9 and 8.0 percent as AljO^ (see Tables 3^ and 35), these results were
considered to be excellent and indicative of successful extractions.  The
data for  free A^Oj, free acid and Congo Red measurements indicated the
products  were near neutral or slightly alkaline, relative to A^O^
dissolution, and additional acid was required to completely extract the
remaining aluminum.  Given that the acid doses for runs 1-1, 1-2 and 1-3
were less than the stoichiometric dose, the results indicating excess free
aluminum  (i.e., free A^O^ > 0; free acid - 0; and a light-brown color with
the Congo Red test) were expected.  Furthermore, the results for runs 2-1
and 2-2 were indicative of stoichioraetric addition of acid as confirmed by
the near  neutral conditions, relative to Al2C>  dissolution, for each run.
     Total iron in the extracts ranged from 0.03 to 0.09 percent as
While no standards exist for iron in commercial alum, a value of 0.2 percent
as F&2^2 *s typical.   These alum products were then of excellent quality
relative to iron.

     The suspended solids concentrations of 1.9 to 9.1 g/L in Table 38 were
for terminal extraction samples which had not been gravity clarified and no
attempt was made to remove suspended matter.  Informal studies indicated
that these samples could be clarified by gravity settling but this was not
quantified.  Comparison of the final suspended solids concentrations with
projected initial  values indicated 95 to 99 percent of the initial suspended
matter was destroyed  by acidic extraction, leaving a suspended residue of
only 1 to 5 percent of the initial mass.

     Acidic extraction of conventional-neutralization sludge cakes was,
therefore, rapid and virtually complete within a 1- to 2-h period.  Product
strengths were near commercial levels and indications are that commercial
                                     96

-------
         TABLE 38.  CHARACTERISTICS OF LIQUID ALUM PRODUCED BY ACIDIC EXTRACTION OF CONVENTIONAL
                    NEUTRALIZATION SLUDGE CAKES CN-1 AND CN-2
Parameter
  CN-1
  Run
                         CN-2
                         Run
                                         1-1
 1-2
 1-3
    2-1
  2-2
Total A^O^, percent	   8.0

Free A^C^, percent	   1.2

Free Acid, percent	

Total Iron, percent as Fe20^	   0.09

Suspended Solids:
    Concentration, g/L	   5.6
    Percent Reduction	  97

Specific Gravity	   1.3

pH (2% dilution)	   3.8

Congo Red Teat, color*	  light
                                        brown
 7.5

 1.0



 0.03


 7.1
95

 1.3

 M.O

light
brown
 7.4

 0.5



 0.07


 5.9
97

 1.3

 3.7

light
brown
    8.8

    0.1



    0.07
    1.9
   99

    1.3
light brown/
   purple
  8.1



  0.3

  0.08


  9.1
 95

  1.3

  3.0

light
purple
 'Qualitative acidity teat:   light brown - free alum; purple = neutral; blue - free acid.

-------
strengths could be routinely achieved with an adequate dewatering system
producing cakes with solids contents of approximately 21 percent and, higher.

Segregated-Neutralization Sludge Cake (SN-1)

     As indicated in Table 39, the segregated neutralization cake had a
solids content of 36.8 percent following low-pressure (6-7 bar) dewatering
and a fixed solids content of 29.9 percent, both well above the minimal
acceptable solids content of 21 percent.  The aluminum content of 31 percent
on a fixed solids basis further indicated the high potential for alum
production.

      TABLE 39.  CHARACTERISTICS OF DEWATERED SEGREGATED-NEUTRALIZATION
                 SLUDGE CAKE SN-1

Parameter                                                       Sludge Cake
	SN-1

Dry (103°C) Solids Content, g dry solids/100 g wet cake	      36.8

Fixed (550°C) Solids Content, g fixed solids/100 g wet cake...     29.9

Aluminum Content, g Al/100 g fixed solids	      31.0
     Three extractions of SN-1 cakes were conducted, as indicated in Table
40, and a total of 0.3 to 0.33 kg of wet sludge solids were extracted.

          TABLE 40.  INITIAL EXPERIMENTAL CONDITIONS FOR EXTRACTIONS
                     OF SEGREGATED NEUTRALIZATION SLUDGE CAKE SN-1

Parameter                                                       Run
                                                        3-1     3-2     3-3
SLUDGE ADDITION

  Total Mass, g	  333.5   300.0   307.0
      Fixed Solids, g	   99.7    89.7   107.9
      Moisture, g	  233.8   210.3   199.1

WATER ADDITION, g	  264.5   238.0   340.0

SULFURIC ACID

  Sulfuric Acid Addition, g	  188.0   169.1   182.2
  Percent of Stoichiometric Acid Dosage, Percent	  111.6   111.6   100.0

CONTROL TEMPERATURE,  °C	   90      70      90

ESTIMATED PRODUCT CONCENTRATION,
  Percent as A1203	    7.4     7.4      7.6
                                     98

-------
Because the solids content was well above 21 percent, It was necessary for
water to be added to the extraction mixture.  Since liquid alum has a
maximum solubility near 9 percent as A1203 and will crystalize at higher
concentrations at ambient temperatures, water addition was necessary to
prevent formation of a crystalline alum.  The quantities of water added
resulted in estimated product concentrations of 7.H to 7.6 percent as A^O^.
Sulfuric acid addition was at, or slightly in excess of, the stoichiometric
dose, as indicated as being necessary from previous experience.

     Filtrate aluminum data in Figure ^2 indicated the extraction of
aluminum in runs 3-1, 3-2 and 3-3 was rapid but that extraction at 70°C  (run
3-2) was slightly slower than those at 90°C, although similar levels of
filtrate aluminum were ultimately achieved.  An abrupt increase in aluminum
between U and 6 hours for runs 3-2 and 3-3 was unexplained but was
attributed, in part, to evaporational losses.  The extractions of SN-1
solids were therefore very similar to those for CN-1 and CN-2 solids (see
Figures UO and M1). indicating that, although SN-1 solids were highly
crystalline and CN solids were highly amorphous and gelatinous, acidic
extraction of both types of sludge cakes proceeded rapidly to completion,
producing a commercial-strength product.

     Data in Table U1 indicate that 95 to 96 percent of total mass was

   TABLE U1.  MATERIAL BALANCES ON TOTAL MASS AND ALUMINUM FOR EXTRACTIONS
              OF SEGREGATED NEUTRALIZATION SLUDGE CAKE SN-1
Parameter
Run
                              3-1
3-2
3-3
TOTAL MASS

  Input:
      Fixed Solids, g	    99.7
      Moisture, g	   198.3
      Acid, g	   188.0

      Total, g	   786.0

  Output, g	   7^3.8

  Percent Recovered	    95

ALUMINUM
  Input,  g	    30.9

  Output:
      Soluble, g	    31 .0
      Residual, g	     0.0

      Total, g	    31.0

  Percent Recovered	   100
27.8
33.5
                                     99

-------
            10
          8 8
         LU
         O

         LU
         Q.
6  -
         I 4
         2
         3
                                       SEGREGATED NEUTRALIZATION
                                       DEWATERED SLUDGE CAKES
                                       RUN  TEMP
                                    •  3-1  90 "C
                                    •  3-2  70 "C
                                    T  3-3  90 «C
                                    STOCHlOMETRiC
                                     ACID DOSAGE
                                     112 PERCENT
                                     112 PERCENT
                                     100 PERCENT
               02468

                                     TIME. HOURS
Figure 42.   Filtrate aluminum concentration  for sulfuric-acid  extractions
             of sludge cake  SN-1 during runs  3-1,  3-2 and 3-3.
                                           ETCH RECOVERY
                                        OEWATERED SLUDGE CAKE
                                                STOICHlOMETRlC
                                                  ACID DOSAGE
                                                   88 PERCENT
                                                   79 PERCENT
                                                  100 PERCENT
                                                  100 PERCENT
                                     TIME. HOURS

Figure 43.   Filtrate aluminum concentrations  for sulfuric-acid extractions
             of sludge cake  ER-1 during runs  4-1 through 4-4.
                                      100

-------
recovered experimentally in runs 3-1, 3-2 and 3~3 and that residual unex-
tracted aluminum was negligible (I.e., 0 to 0.1  g of Al).  The acid doses
which were slightly in excess of stoichloraetric requirements (i.e., runs 3-1
and 3~2) resulted in complete extraction of aluminum (i.e., no residual non-
soluble aluminum in the output), while that conducted at a stoichiometric
dose of 100 percent (run 3-3) resulted in virtually complete extraction of
available aluminum.  In addition,  aluminum recoveries,  relative to input and
output aluminum, were excellent.

     Product quality data in Table 42 indicated aluminum concentrations
ranged from 8.1  to 9.0 percent as AljO}, and were well  within acceptable
limits for commercial products.  The liquid alum products were virtually
free of suspended solids and were near neutral or slightly acidic, with
respect to dissolution of A^O^.  Total iron was 0.03 to 0.11 percent as
F6203 and was well within acceptable limits.  In summary, the alum products
produced were of excellent quality, Indicating segregated neutralization
solids were excellent sources of extractable aluminum.
   TABLE 42.  CHARACTERISTICS OF LIQUID ALUM PRODUCED BY ACIDIC EXTRACTION
              OF SEGREGATED NEUTRALIZATION SLUDGE CAKE SN-1
Parameter
         SN-1
          Run
                                                        3-1
          3-2
          3-3
Total A^Oj,  percent

Free A1203 ,  percent

Free Acid, percent

Total Iron,  percent as
Suspended Solids:
    Concentration, g/L
    Percent Reduction, %
  8.1
  1.4
  8.7



  1.3

  0.03
Specific Gravity.

pH (2% dilution).
  0.0
100

  1.4

  2.5
  1.4

  2.5
9.0

0.3



0.11
  0.1      0.0
>99.9   100
1.4

3.7
Congo Red Test,  color*	  purple  purple  brown
 Qualitative acidity test:  light brown - free alum; purple - neutral;
 blue - free acid.
Etch Recovery Sludge Cake (ER-1)

     The etch-recovery solids examined during the study were  provided  as
                                    101

-------
dewatered solids by a companion plant of Plant A.  The ER solids were
produced from a caustic etching process similar in function to that used at
Plant A.  The characteristics of the etch-recovery solids are inducted in
Table 43 and indicate the solids were exceptionally low in free moisture, as
Indicated by a dry solids content of 91.6 percent and a fixed solids content
of 66.5 percent.  Approximately 27 percent of the dry solids were volati-
lized at 550°C which was typical of the CN and SN solids as well.  Because
the etch-recovery process must be operated without the addition of organic
chelatlng and sequestering agents, It was assumed that the loss of solids at
550°C was attributable to the loss of bound water.  The aluminum content of
the fixed solids was 43.6 percent and was the highest of any of the sludge
solids included In the study.  This value was exceptionally higher than the
theoretical value of 34.6 percent for A1(OH)3 but below the theoretical
value of 52.9 percent for A^Oj.  In examination of this point, aluminum
hydroxides are frequently characterized as hydrated forms of aluminum oxide.
For example, A1(OH)3 is equivalent to A1203'3H20 in terms of aluminum
content.  In examination of the 43.6 percent value in Table 43, this
aluminum content could be representative of a compound with the formula
A1203«1.21H20.  Regardless, the sludge cake was rich in aluminum and an
excellent candidate for production of alum.  However, because of its
apparent crystalline nature and more aluminum-oxide-like form, there was
concern that It would be more difficult to extract under acidic conditions.
   TABLE 43.  CHARACTERISTICS OF DEWATERED ETCH-RECOVERY SLUDGE CAKE ER-1
Parameter

Dry (103°C) Solids Content,
Fixed (5508C) Solids Content
Aluminum Content, g Al/100 g




, g fixed solids/100 g wet cake...

Sludge Cake
ER-1
91 .6
66. U
43.6

     Four runs were conducted with the ER-1 cake, as presented in Table 44,
in which 0.14 to 0.33 kg of ER-1 cake were extracted.  Because of the
exceptionally high aluminum content of the sludge cakes It was necessary to
add between 0.45 to 0.6 kg of water to the extractions.  In all but run 4-3,
the water was added prior to the addition of acid.  In run 4-3, 0.1 kg and
0.025 kg of water were added to the extraction mixture immediately following
the collection of samples at the 1-h and 6-h time Intervals.  These
additions were made to prevent the formation of a crystalline product and
decreased filtrate aluminum concentrations slightly-  Runs 4-1 and 4-2 were
conducted with less than stoichiometrlc addition of acid while runs 4-3 and
4-4 were conducted at stoichiometric levels.  Furthermore, all runs were
conducted at 90°C for periods of 8 to 24 h because of the refractory nature
of the aluminum precipitates (Harmon and Saunders, 1985).  The estimated
product concentrations ranged from 6.9 to 10.5 percent as Al^.  The low
value of 6.9 percent as A1203 for run 4-4 was the result of an error made  in

                                     102

-------
          TABLE  44.   INITIAL EXPERIMENTAL CONDITIONS FOR EXTRACTIONS
                     OF  ETCH RECOVERY SLUDGE CAKE ER-1
Parameter
ER-1
Run
4-1 4-2 4-3

4-4
 SLUDGE ADDITION

  Total Mass,  g	   140.5
      Fixed Solids,  g	    93.3
      Moisture,  g	    47.2

 WATER ADDITION,  g	   478.6

 SULFURIC ACID

  Sulfuric Acid  Addition,  g	   194.6
  Percent of Stoichiometric Acid  Dosage,
     Percent	    87.9
 CONTROL TEMPERATURE,  °C.
 ESTIMATED PRODUCT  CONCENTRATION,
   Percent as
90


 9.4
       153.8
       101 .0
        52.8
 10.5
        139.0
         87.4
         51.6
190.3   207.6

 79.4   100.0

 90      90
8.4
      119.1
       74.9
       44.2
       451.7   509.9   599.8
177.9

100.0

 90


  6.9
 the mass of water  added  to  the  extraction  mixture,  creating  a  dilute  product.
 The estimated  product  concentrations  for runs  4-1 ,  4-2  and 4-3 were therefore
 all above  the  desired  value of  8  percent as
      Filtrate  aluminum  data  in  Figure  43  indicate  that  the  acidic  extraction
 of  cake ER-1 proceeded  at  rates slower than  those  demonstrated  for CN  and  SN
 cakes, with  the  exception  of run 4.3.   Filtrate  aluminum  concentration for
 runs  4-1  and 4-2 reached a plateau  value  of  approximately 6  percent in one
 hour  and  proceeded  to values in excess of 8  percent  after 4  hours,  while run
 4-3 was at approximately 9 percent  in  one hour.  The accelerated rate  of
.extraction for run  4-3, as depicted in Figure  43,  was attributed to the
 elevated  strength of acid  in the extraction  mixture  during  the  initial por-
 tions of  the run.   Because 0.125 kg of the total of  0.45  kg  of  water to be
 added was withheld  initially, but was  added  at the 1-h  and  6-h  intervals,
 the effective  strength  of  the sulfuric acid  added  to the  mixture was initi-
 ally  28.4 percent- as H2SOn (cone.)  as  opposed  to 23.9 percent for  runs 4-1
 and 4-2 and  19.8 percent for run 4.4.   This  elevated acid strength enhanced
 the extraction of aluminum and  would appear  to be  the preferred means  for
 extraction where possible.  Filtrate aluminum  data were not  collected  for
 run 4-4 during the  initial 4 h  of the  extraction.   At 4 h,  filtrate aluminum
 was 6.6 percent  and was virtually at completion  since the projected
 concentration  for the run  was 6.9 percent.  After  the initial 4-h  sample,
 filtrate  aluminum concentration varied slightly  and trended  towards higher
 values due to  continued extraction  of  aluminum and evaporation  of  water.
                                     103

-------
     The material balances conducted on reaction-mixture and aluminum mass
are presented in Table 45.  The recovery of total mass varied from 82 to 91
percent.  Given that three of the four runs were conducted over a 24-h
period and evaporational losses would be expected to be high, these
recoveries were acceptable.  Aluminum recoveries ranged between 89 to 109
percent and were excellent.  Residual aluminum mass, however, ranged from 5
to 12.4 g and accounted for approximately 32 percent of output aluminum for
runs 4-1 and 4-2 and 15 to 19 percent for runs 4-3 and 4-4.  The lack of
complete extraction of aluminum for runs 4-1 and 4-2 was expected since acid
doses were 88 and 79 percent of stoichiometric values, respectively.
Stoichiometric acid doses were used for runs 4-3 and 4-4, accounting for the
lower portion of unextracted aluminum.  However, there was apparently
additional aluminum that could have been extracted had higher acid doses
been employed or had initial acid strength been increased by delayed
addition of water.
   TABLE 45.  MATERIAL BALANCES ON TOTAL MASS AND ALUMINUM FOR EXTRACTIONS
              OF ETCH RECOVERY SLUDGE CAKE ER-1
Parameter
4-1
ER-1
Run
4-2 4-3

4-4
TOTAL MASS

  Input:
      Fixed Solids, g..    93-3
      Moisture, g	   525.8
      Acid, g	
      Total, g	

  Output, g	

  Percent Recovered...

ALUMINUM
  Input, g.
  Output:
      Soluble, g,.
      Residual, g.
      Total, g	

  Percent Recovered	
40.7
44.0
38.1
 74.9
64U.O
177.9

896.3

737.2

 82



 32.7
                                    104

-------
     The characteristics of selected terminal or near-terminal samples from
extraction runs are presented in Table 46.  Data presented for runs 4-1, 4-2
and 4-3 were those for samples collected after 6 or 8 h of extraction and
were equivalent to those presented in earlier tables.  Data for run 4-4 were
collected after 24 h.  Aluminum concentrations ranged from 8.3 to 9.2
percent as A^O^ and were well within acceptable limits.  The samples were
all near neutral conditions with respect to free-acid and free-A^O^
concentrations and contained low levels of total iron.  Residual suspended
solids concentrations were high, ranging from 25 to 94 g/L, and indicated
incomplete extraction of aluminum.  The free aluminum values of 0.1 to 0.4
percent as A^O^ were not in agreement with this conclusion but the residual
aluminum data in Table 45 confirmed the incomplete extraction of aluminum.
It therefore appeared that a portion of the aluminum was resistant to
extraction and was not measured in the "free A^O^" procedure.  No runs
were,  however, conducted with excess acid doses to establish the extent to
which  the residual aluminum could be extracted.
       TABLE 46.  CHARACTERISTICS OF LIQUID ALUM PRODUCED BY ACIDIC
                  EXTRACTION OF ETCH RECOVERY SLUDGE CAKE ER-1
Parameter
ER-1
 Run
Total A120-3 , percent 	
Free A^Oo , percent 	

Total Iron, oercent as FeoO? 	
	 8.3
	 0.1


9.2
0.2
8.5
0.2
0.1
8.6
0.4
0.1
Suspended Solids:


pH (2% dilution) 	
Congo Red Test color' ' 	 	 	

	 93.7
	 54
	 1.4
	 2.9
. . . . purple

66.9
71
1 .4
3.2
brown

38.9
81
1 .3
2.9
purple

25.0
85
1 -3
3.6
brown

        from sample collected after a 6-h reaction period.
        from sample collected after an 8-h reaction period.
        from sample collected after a 24-h reaction period.
(^Qualitative acidity test:  light brown - free alum; purple
   blue « free acid.
     neutral;
                                    105

-------
Blended Sludge Cakes

     Two runs were made in which sludge cakes were prepared by blending a
conventional neutralization sludge cake (CN-2) with either a segregated-
neutralization sludge cake (SN-1) or etch-recovery solids (ER-1).

Conventional Neutralization/Segregated Neutralization Sludge Cake
(CN-2/SN-1)--
     The segregated neutralization cake employed in studies reported
previously (see Tables 39-42) was blended with a conventional neutralization
sludge cake (CN-2) at a fixed solids ratio of 43 g of CN-2 to 54 g of SN-1.
As indicated in Table 47, the net dry (103°C) solids content of the blend
was 25.2 percent, fixed solids content was 21.4 percent and aluminum content
was 34.6 percent.  All of these values were indicative of excellent
potential for production of commercial-strength liquid alum.  Run 5-1 was
made using a 100 percent stoichiometric acid dose at 90°C (Table 48).

TABLE 47.  CHARACTERISTICS OF COMBINED DEWATERED CONVENTIONAL NEUTRALIZATION/
           SEGREGATED NEUTRALIZATION SLUDGE CAKES CN-2 AND SN-1
Parameter                                                       Sludge Cake
	CN-2/SN-1

Dry (103°C) Solids Content, g dry solids/100 g wet cake	     25.2

Fixed (550°C) Solids Content, g fixed solids/100 g wet cake...     21.4

Aluminum Content, g Al/100 g fixed solids	     34.6


    TABLE 48.  INITIAL EXPERIMENTAL CONDITIONS FOR EXTRACTION OF COMBINED
               CONVENTIONAL NEUTRALIZATION/SEGREGATED NEUTRALIZATION
               SLUDGE CAKES CN-2 AND SN-1
Parameter                                                            Run
                                                                     5-1
SLUDGE ADDITION

  Total Mass, g	    451.5
      Fixed Solids, g	     96.8
      Moisture, g	    354.7

WATER ADDITION, g	    178.6

SULFURIC ACID

  Sulfuric Acid Addition, g	    182.6
  Percent of Stoichiometric Acid Dosage, Percent	    100.0

CONTROL TEMPERATURE, °C	     90

ESTIMATED PRODUCT CONCENTRATION, Percent as A1203	      7.8
                                     106

-------
With the addition of a slight excess of water, the projected alum concen-
tration was 7.8 percent as A1203.  Of the total of 178.6 g of water atided, a
total of 50 g was added after collection of the fourth sample at U h.  This
resulted in dilution of the remaining extract as shown by a decrease in
aluminum concentration in Figure M4.

     Filtrate aluminum data in Figure H4 indicated the extraction proceeded
to completion rapidly within the initial 0.5 h.  This was in general
agreement with results obtained previously for CN-1,  CN-2 and SN-1 sludges
(see Figures 40-42).  The recovery of total mass was 90 percent while
aluminum was recovered at the 99 percent level, as indicated in Table U9.
In addition, the residual aluminum was only 0.6 percent of the total,
indicating virtually complete extraction of aluminum from the sludge cake.
Data in Table 50 indicated that the concentration of the liquid alum was 8.2
percent as Al203 and that the product was near neutral with respect to
extraction of AljO^.  Iron and suspended solids concentrations were low
with the suspended solids being reduced by 98.4 percent, relative to the
solids placed in the reactor.  It was therefore concluded that blended CN
   TABLE 49.  MATERIAL BALANCES ON TOTAL MASS AND ALUMINUM FOR EXTRACTION
              OF COMBINED CONVENTIONAL NEUTRALIZATION/SEGREGATED
              NEUTRALIZATION OF SLUDGE CAKES CN-2 AND SN-1

Parameter                                                            Run
                                                                     5-1
TOTAL MASS

  Input:
      Fixed Solids, g	     96.8
      Moisture, g	    533-3
      Acid, g	    182.6

      Total,  g	    812.7

  Output,  g...:	    730.1

  Percent  Recovered	     90

ALUMINUM

  Input, g	     33.5

  Output:
      Soluble, g	     33.0
      Residual, g	       0.2

      Total,  g	     33.2

  Percent Recovered	     99
                                     107

-------
                                                MIXTURE OF
                                       CONVENTIONAL  NEUTRALIZATION/
                                        SEGREGATED NEUTRALIZATION
                                         DEWATERED SLUDGE  CAKES
                                                   STOICHIOMETRIC
                                         RUN  TEMP  ACID DOSAGE
                                         5-1  90X  100 PERCENT
                                     TIME. HOURS

Figure 44.   Filtrate aluminum concentrations for sulfuric-acid extraction
             of sludge  cake  CN-2/SN-1  during run 5-1.
                                                 MIXTURE OF
                                       CONVENTIONAL NEUTRALIZATION/
                                             ETCH  RECOVERY
                                         DEWATERED SLUDGE  CAKES
                                                    STOICHIOMETRIC
                                          RUN  TEMP  AGO DOSAGE
                                          6-1  90"C  100 PERCENT
                                      TIME, HOURS
Figure 45.   Filtrate aluminum concentrations for  sulfuric-acid  extraction
             of sludge  cake  CN-2/ER-1  during run 6-1.

                                       108

-------
  TABLE 50.  CHARACTERISTICS OF LIQUID ALUM PRODUCED BY ACIDIC EXTRACTION
             OF COMBINED CONVENTIONAL NEUTRALIZATION AND SEGREGATED
             NEUTRALIZATION SLUDGE CAKES CN-2 AND SN-1

Parameter                                                            Run
                                                                     5-1
Total A1203 , percent ............................................     8.2

Free A1203 , percent .............................................     0.2

Free Acid , percent ..............................................

Total Iron , percent as Fe20 ....................................     0 . O
Suspended Solids:
    Concentration, g/L ................................ ; .........     3.0
    Percent Reduction, percent ..................................    98

Specific Gravity ................................................     1.3

pH (2t dilution) ................................................     3.5

Congo Red Test , color* ..........................................    brown/
_   _  _ purple

'Qualitative acidity test:  light brown » free alum; purple • neutral;
 blue - free acid.


and SN sludges could be effectively extracted with acid to produce
commercial-strength liquid alum.

Conventional Neutralization/Etch Recovery Sludge Cake (CN-2/ER-1 )--
     A dewatered conventional neutralization cake was mixed with etch-
recovery cake at a fixed solids ratio of 1 . 1 to  1 (i.e., 50 g CN-2 fixed
solids to 45 g of ER-1 fixed solids).  The net characteristics of the
blended sludges are presented in Table 51 .  The  extraction was conducted at

        TABLE 51 .  CHARACTERISTICS OF COMBINED DEWATERED CONVENTIONAL
                   NEUTRALIZATION AND ETCH RECOVERY SLUDGE CAKES CN-2
                   AND ER-1

Parameter                                                       Sludge  Cake
_ _      CN-2/ER-1

Dry (103°C) Solids Content, g dry solids/100 g wet cake ......      30.2

Fixed (550°C) Solids Content, g fixed solids/100 g wet cake...     22.5

Aluminum Content, g Al/100 g" fixed solids ....................      *»1 .3
                                     109

-------
90°C using a stoichiometric acid dose with a projected alum concentration of
8.9 percent as AJ^C^ (Table 52).  A total of 50 g of water was withheld from
the initial extraction mixture and was added after collection of a sample
at the 2-h interval, resulting in a slight dilution of the extraction
product.
    TABLE 52.  INITIAL EXPERIMENTAL CONDITIONS FOR EXTRACTION OF COMBINED
               CONVENTIONAL NEUTRALIZATION AND ETCH RECOVERY SLUDGE CAKES
               CN-2 AND ER-1

Parameter                                                            Run
                                                                     6-1
SLUDGE ADDITION

  Total Mass, g	    423.5
      Fixed Solids, g	     95.1
      Moisture, g	    328.4

WATER ADDITION, g	    196.8

SULFURIC ACID

  Sulfuric Acid Addition, g	    213.9
  Percent of Stoichiometric Acid Dosage, Percent	    100.0

CONTROL TEMPERATURE, °C	     90

ESTIMATED PRODUCT CONCENTRATION, Percent as A1203	      8.9
     Filtrate aluminum data in Figure 45 indicated that the extraction
proceeded relatively rapidly but was not at completion until about 2 h, due
to the presence of the etch recovery solids.  Approximately 94 percent of
the total mass was recovered during the extraction and aluminum recovery was
102 percent (Table 53).  The residual aluminum was, however, equal to 18.8
percent of the initial aluminum added, indicating incomplete extraction of
the aluminum.  This was consistent with results obtained previously with
etch recovery solids (see Table 45).

     Finally, the product produced had an aluminum concentration of 8.8
percent as A^O^-and contained a slight excess of free acid, as indicated in
Table 54.  Iron concentration was exceptionally low but suspended solids
concentration was high (i.e., 32.7 g/L).  A total of 16 percent of the
initial suspended solids remained indicating the incomplete extraction of
sludge aluminum.
                                    110

-------
   TABLE 53.  MATERIAL BALANCES ON TOTAL MASS AND ALUMINUM FOR EXTRACTION
              OF COMBINED CONVENTIONAL NEUTRALIZATION AND ETCH RECOVERY
              SLUDGE CAKES CN-2 AND ER-1

Parameter                                                            Run
                                                                     6-1
TOTAL MASS

  Input:
      Fixed Solids, g	     95.1
      Moisture, g	    525.2
      Acid, g	    213.9

      Total,  g	    83^.2

  Output, g	    780.2

  Percent Recovered	     914

ALUMINUM

  Input, g	     39.3

  Output:
      Soluble,  g	     32.8
      Residual, g	      7.4

      Total,  g	     40.2

  Percent Recovered	    102
METAL CONTENT OF SLUDGES AND LIQUID ALUM EXTRACTS

     In the initial phase of the study when sludge dewatering properties
were being examined, the metal content of numerous sludges and aluminum
residues were examined in conjunction with studies to determine the extent
of sludge dewatering achieved at various full-scale plants.  These data are
presented herein in conjunction with experimental data collected for liquid
alum extracts produced as a part of this study.

Selected Aluminum-Finishing Suspensions and Residues

     The metal contents of dewatered sludge cakes provided by seven
aluminum-anodizing plants are presented in Table 55.  The solids contents
for the dewatered cakes ranged from 9.8 to 25.9 percent.  The high value of
25.9 percent was produced with a high-pressure filter press, while the
remainder were produced with low-pressure presses or rotary vacuum filters.
                                    Ill

-------
         TABLE 5H.  CHARACTERISTICS OF LIQUID ALUM PRODUCED BY ACIDIC
                    EXTRACTION OF COMBINED CONVENTIONAL NEUTRALIZATION
                    AND ETCH RECOVERY SLUDGE CAKES CN-2 AND ER-1

Parameter                                                            Run
                                                                     6-1
Total A1203, percent ............................................     8.8

Free A1203 ,  percent .............................................

Free Acid, percent ..............................................     0.8

Total Iron,  percent as Fe20 ....................................     O.OM
Suspended Solids:
    Concentration, g/L ................................ . .........    32.7
    Percent Reduction , percent ..................................    8A

Specific Gravity ................................................     1 . H

pH (2% dilution) ................................................     2.6

Congo Red Test ,  color* ..........................................    purple


"Qualitative acidity test:  light brown - free alum; purple - neutral;
 blue - free acid.

     The aluminum content for the cakes varied from 18.4 percent to M8.2
percent of the total dry mass (i.e., 1.84 x 10^ to U.82 x 10^ mg/kg of dry
solids).  Sodium was the next most prevalent metal and ranged from 0.6 to
8.9 percent, followed by iron, which ranged from 0.17 to 1.13 percent.  The
sludge cakes contained measurable quantities of many of the potentially-
toxic metals examined.  On an overall basis, the predominant trace metals
detected were chromium (Cr). copper (Cu), lead (Pb), nickel (Ni) and tin
(Sn), although there was considerable variation in the concentrations of
these.

     In examination of the metal content of the sludges relative to
production of liquid alum, it was assumed that the alum product was to be
used for the purposes of treating drinking waters.  This, however, is a very
strict assumption, since less than 5 percent of the alum produced is used in
treatment of water and wastes.  Furthermore, it was assumed that all metals
in a sludge cake would be solubillzed during acid extraction and would
appear as a contaminant with the extracted aluminum.  In addition, no
consideration was given to the extent to which the contaminant metals would
be removed with the aluminum-hydroxide floe during the coagulation process.
Just as aluminum is removed in an effective coagulation process, trace
metals are commonly removed through adsorption to or complexation with the
aluminum hydroxide formed.  This was ignored to provide a conservative
estimate of metal impact.

                                     112

-------
             TABLE 55.  METAL CONTENT OF DEWATERED CONVENTIONAL NEUTRALIZATION SLUDGE CAKES
                        FROM SELECTED ALUMINUM-ANODIZING PLANTS
Metal


Aluminum, AJ.
Calcium, Ca
Iron, Fe
Magnesium, Mg
Potassium, K
Sodium, Na
Arsenic, As
Cadmium, Cd
Chromium, Cr
Copper, Cu
Lead, Pb
Mercury, Hg
Nickel, Ni
Selenium, Se
Silver, Ag
Tin, Sn
Zinc, Zn
Concentration, mg/kg (dry mass)
Plant
ACl-3161
1.1x105
77
2993
1667
51
23073
1.03
0.37
13.9
381
26
0.05
6198
1.38
0.28
13218
0.01
A
AC1-3201
2.2x105
915
1713
2110
13
82518
1.65
0.29
36.5
108
18
0.11
8532
0.96
0.05
16513
32
Plant
BC1-3181
2.55x105
3208
6918
1853
201
5983
1.85
0.59
108
328
107
0.1
12352
1.62
0.023
7722
100
B
BC1-3201
3.26x105
571
11301
21?1
30
21731
1.31
0.77
853
261
9.78
0.05
8261
2.71
0.035
9566
221
Plant C
CC1-3201
1.81x105
190
20100
5100
11
30585
2.35
0.53
357
210
21
0.81
118
0.21
0.011

-------
     To evaluate the impact of liquid alum to be produced from the sludge
cakes, a hypothetical aluminum dose was to be selected.  Concentrations of
aluminum reported as being used in coagulation of drinking waters include:
0.27 to 2.7 mg/L (Stumm and O'Melia, 1968; Mangravite et al., 1975); 0.8 to
4.0 mg/L (Parkham, 1962) and 0.05 to 9 mg/L (Edwards and Amirtharajah, 1985).
To be conservative in the evaluation, an aluminum concentration of 10 mg/L
was selected for analysis of the sludge metal data.  This value is analagous
to an "alum" concentration of 111 mg/L as AljCSOn^-1 4.3^0.  Using these
assumptions and the data in Table 55, metal concentrations in drinking water
coagulated with aluminum at a concentration of 10 mg/L were calculated and
are presented in Table 56.

     The metals controlled through federal drinking water regulations are
identified with their maximum contaminant levels (MCL) in Table 56.
Projected arsenic, cadmium, lead, mercury, selenium and silver concentra-
tions were all below their respective MCL values.  With the exception of two
values for lead (i.e., 2.7 ug/L and 4.2 ug/L), all concentrations of these
metals were less than 3 percent of the MCL value of 50 ug/L, indicating
minimal impact on the quality of "the drinking water, even without giving
allowance for reduction in contaminant concentrations through sedimentation
and filtration processes following coagulant addition.  Projected chromium
concentrations were all below the MCL of 50 ug/L, although two projected
values (i.e., 19.4 ug/L and 26.2 ug/L) were near the limit concentration.
The remaining samples, however, had projected concentrations which ranged
from 0.3 to 4.2 ug/L and were well below the MCL value.

     Copper and zinc are not regulated but values of 1000 ug/L and 5000 ug/L,
respectively, have been suggested for use in evaluation of drinking-water
quality.  All projected concentrations of these metals were well below these
suggested values.

     The remaining heavy metals, nickel and tin, are not currently (1987)
regulated by federal drinking-water limits, although nickel is listed as a
contaminant for which a MCL must be established by 1989.  Concentrations for
these contaminants were highly variable and were the highest for any of the
heavy metals examined.  Nickel ranged from 2.2 ug/L to 485 ug/L, and tin
ranged from <0.0001 ug/L to 751 ug/L.  While these are not regulated, the
elevated concentrations for nickel (e.g., 158 ug/L to 485 ug/L) and tin
(e.g., 293 ug/L to 751 ug/L) in sludge cakes from Plants A and B indicate
caution should be exercised in the ultimate use of alum produced from these
sludge cakes.

     It must, however, be emphasized that the foregoing analysis does not
indicate the products to be produced from these sludge cakes would not meet
current standards.  For example, the Water Chemicals Codex developed by the
National Research Council (1982) has proposed recommended maximum impurity
content (RMIC) values for water treatment additives.  The RMIC values for
alum are, however, based on contaminant concentrations remaining in water
samples which have been dosed with an appropriate aluminum concentration
(e.g., 10 mg/L of aluminum); adjusted to a pH of 6; flocculated for 1 h;  and
then filtered with an 0.45um filter.  This RMIC value thereby takes into
consideration the removal of alum contaminants during coagulation,


                                    114

-------
      TABLE 56.   PROJECTED INCREASES IN METAL CONCENTRATIONS IN COAGULATED DRINKING HATER  TREATED
                 WITH ALUM,  PRODUCED FROM DEWATERED CONVENTIONAL NEUTRALIZATION  SLUDGE  CAKES,
                 AT AN ALUMINUM CONCENTRATION OF 10 mg/L
Metal

Calcium, Ca
Iron, Fe
Magnesium, Mg
Potassium, K
Sodium, Na
Arsenic, As
Cadmium, Cd
Chromium, Cr
Copper , Cu
Lead, Pb
Mercury, Hg
Nickel, Nl
Selenium, Se
Silver, Ag
Tin, Sn
Zinc, Zn
MCL
Mg/L

-
-
-
-
-
50
10
50
(1000)*
50
2
-
10
50
-
(5000)*
Concentration, ug/L
Plant A
ACl-3161
2
73
1
1.3
562.8
0.03
0.01
0.3
9.1
0.6
0.001
158
0.11
0.006
323
0.0002
AC1-3201
13
79
96
2
3752
0.08
0.01
1.7
1.9
0.8
0.02
388
0.01
0.002
751
1.5
Plant B
BC1-3181
126
273
73
8
235
0.07
0.02
1.2
12.9
1.2
0.001
185
0.06
0.0009
303
3.9
BC1-3201
18
317
67
1
667
0.01
0.02
26.2
8.0
0.3
0.002
253
0.08
0.001
293
6.9
Plant C
CC1-3201
27
1111
278
2
1666
0.13
0.03
19.1
13.1
1.3
0.01
22.8
0.01
0.002
<0.0003
5.6
Plant D
DC1-3059
6
99
10
5
1209
0.021
0.009
1.7
11.2
2.7
0.001
16.7
0.003
0.001
61.2
12.3
Plant E
EC1-3201
7
182
61
2
1556
0.01
0.07
2.2
3.3
0.8
0.01
2.2
0.01
0.001
<0.0002
7.8
Plant F
FC1-3201
86
72
278
2
1389
0.03
0.02
1.9
2.5
0.9
<0.0003
3.3
0.03
0.002
<0.0002
0.9
Plant I
ICl-3201
1
18
32
1
1852
0.03
0.0001
0.8
1.5
0.6
0.0002
2.2
0.001
0.008
<0.0001
1.1
"Secondary MCL values;  all others are primary MCL values.

-------
sedimentation and filtration in a treatment plant.  The above analysis did
not, however, allow for removal of contaminants by subsequent treatment.
The foregoing analysis, however, does highlight concern with respect to
selected metals and indicates a need to control or reduce them if alum
production is to be pursued.  For example, nickel and tin originate from
two-step anodizing baths which could be isolated and treated separately to
minimize the concentrations of these metals in the sludges produced.

     Data in Table 57 are for a segregated neutralization suspension
produced at Plant H.  The concentrated suspension was produced by addition
of spent acids to a spent caustic suspension and, as expected, alumuminum
and iron were the major metallic constituents, followed by tin at 1200 mg/L,
       TABLE 57.  METAL CONCENTRATIONS FOR A SEGREGATED NEUTRALIZATION
                  SUSPENSION PRODUCED AT PLANT H

                    Metal	Concentration, mg/L
Aluminum, Al
Calcium, Ca
Iron, Fe
Magnesium, Mg
Potassium, K
Sodium, Na
17,500
39
50
120
49
13,100
Arsenic, As
Cadmium, Cd
Chromium, Cr
Copper , Cu
Lead, Pb
Mercury, Hg
Nickel, Ni
Selenium, Se
Silver, Ag
Tin, Sn
Zinc, Zn
0.06
0.01
1.58
50
0.27
0.01
30
<0. 1
0.025
1200
1.8
                    Suspended Solids            83.1 g/L
Liquid Alum Products

     Based on initial characterizations of sludge solids to be extracted,
selected trace metals were examined in liquid alums produced by acidic
extraction of conventional-neutralization (CN-2), segregated-neutralization
(SN-1) and etch-recovery (ER-1) sludge cakes.  Because copper, mercury,
arsenic and selenium were not major-constituents, they were not examined.
Metal data for three alum samples produced from sludge cakes CN-2, SN-1  and
ER-1 are presented in Table 58.  In addition, data for two samples of  a

                                    116

-------
standard grade of commercial liquid alum,  as obtained directly from two
metropolitan Atlanta drinking-water treatment plants, and one sample of an
"iron-free" grade of commercial liquid alum are included in Table 58.


  TABLE 58.  METAL CONTENT OF LIQUID-ALUM SAMPLES PRODUCED FROM CN-2,  SN-1
             AND ER-1  SLUDGE CAKES AND THREE COMMERCIAL ALUM PRODUCTS
Metal
Concentration, mg/L
Sludge-Cake Alum


Cadmium, Cd
Chromium, Cr
Iron, Fe
Lead, Pb
Nickel, Ni
Silver, Ag
Tin, Sn
Zinc, Zn
CN-2

0.4
30.0
60.0
20
752
0.7
914
12
SN-1

0.3
4.4
18.3
19
8
0.2
28
6
ER-1

0.3
3.7
21 .0
26
7
0.08
17
7
Commercial
Standard
1
0.03
78
1845
6.6
44
0.25
155
8.5
2
0.3
0.9
2080
4.1
44
0.2
-
8.5
Alum
Iron-Free

0.3
0.9
10.1
5.6
-
0.15
2.5
1 .0
     For the three samples of liquid alum produced from sludges,  the
concentrations of cadmium (Cd), chromium (Cr), iron (Fe), silver  (Ag) and
zinc (Zn) were at, or well below, concentrations contained in standard
commercial liquid alum.  The lead (Pb) concentrations ranged from 19 to 26
mg/L in alums from sludges and ranged from 4.1 to 6.6 mg/L in standard
commercial products.

     The conventional neutralization sludge (CN-2) produced a liquid alum
with excessive concentrations of nickel (Ni - 752 mg/L) and tin (Sn -914
mg/L).   It was speculated that the two-step (Ni/Sn-based) anodizing process
employed at Plant A resulted in elevated concentrations of nickel and tin in
the alum products.

     At an aluminum dose of 10 mg/L for treatment of a drinking water, as
discussed earlier, these levels would result in approximate concentrations
of Ni - 137 yg/L and Sn - 166 ug/L in a coagulated water.  These
concentrations could, however, be reduced significantly by gravity
sedimentation and filtration.  Despite the fact that no standards are
currently applicable to concentrations of nickel and tin, it is apparent
that the indicated levels warrant further investigation on a plant-by-plant
basis.   The levels of nickel and tin in the alum products produced with SN-1
and ER-1 sludges were well below concentrations in standard commercial
liquid alum.

     In summary, the trace metal analyses for the liquid alum produced from
conventional neutralization sludge (CN-2) indicated that only lead,  nickel
and tin were of concern when compared to standard commercial products.
Concentrations of cadmium, chromium, iron, silver and  zinc were of no


                                     117

-------
concern in this regard.  Finally, only lead concentrations in liquid alum
were of marginal concern when examining segregated neutralization (SN-1 ) and
etch recovery (ER) sludges.  It is recommended that waste segregation be
considered when addressing the control of nickel and tin.
                                    118

-------
                                 SECTION 7

                                REFERENCES
Allied Chemical Corporation, 1976, "Alum", Allied Chemical Corporation,
     Syracuse, NY.

APHA, 1985, Standard Methods for the Examination of Water and Wastewater,
     15th Edition, American Public Works Association,  Washington,  D.C.

Edwards, G. A., and Amirtharajah, A., 1985, "Removing  Color Caused by Humic
     Acids," Jour. American Water Works Assoc.,  77. 50.

EPA, 1982, Process Design Manual for Dewatering Municipal Wastewater
     Sludges, EPA-625/1-82-014, U.S. Environmental Protection Agency, MERL,
     Cincinnati, OH.

EPA, 1986, Design Information Report - Recessed Plate  Filter Presses, EPA
     600-M-86-017, WERL, U.S. Environmental Protection Agency. Cincinnati,
     OH.

Harmon, C. B. and Saunders, F- M., 1985, "Production of Liquid Alum from
     Aluminum-Anodizing Sludges," Water Science and Technology, 17, 5^1-550.

Johns, P. J., 1987, "Design and Operation of Fixed-Volume Filter Presses,"
     A Special Problem Report to School of Civil Engineering, Georgia
     Institute of Technology, Atlanta, GA.

Mangravite, F. J., e_t al.,  1975, "Removal of Humic Substances by Coagulation
     and Microflotation",  Jour. American Water Works Assoc., 67, 88.

National Research Council,  1982, Water Chemicals Codex,  National Academy
     Press, Washington, D.C.

O'Connor, J. T., 1984, Environmental Engineering Unit  Operations and Unit
     Processes Laboratory Manual, Third Edition, Association of Environ-
     mental Engineering Professors, The University of  Texas, Austin, TX.

Parkham, R. F., 1962, "The Theory of the Coagulation Process - A Survey of
     the Literature II.  Coagulation as a Water Treatment Process", Proc.
     Soc. Water Treat. Exam., 11, 106.

Saunders, F. M., e_t al., 1982, Characterization, Reclamation and Final
     Disposal of Aluminum Bearing Sludges, SCEGIT-82-102, Report to The
     Aluminum Association,  Inc., Washington, D.C.

                                     119

-------
Saunders, p. M., et al.,  1984, "Evaluation of Process Systems for Effective
     Management of Aluminum Finishing Wastewaters and Sludges," U.S.
     Environmental Protection Agency, IERL, Cincinnati, OH; PB 84-170 661,
     Springfield, VA.

Stumm, w.f and O'Melia, C.  R., 1968, "Stoichiometry of Coagulation", Jour.
     Amer. Water Works Assoc., 60,  514.
                                    120

-------
TECHNICAL REPORT DATA
fi'kasc read /nslnicl:/.'its nn l/ic went: be/ore cum/ildi/if/
1. REPORT NO. 2.
4. TITLE AND SUBTITLE
RECLAMATION OF ALUMINUM FINISHING SLUDGES
7. AUTHORISI
F. Michael Saunders
9. PERFORMING ORGANIZATION NAME AND ADDRESS
School of Civil Engineering
Georgia Tech Research Corporation
Georgia Institute of Technology
Atlanta, GA 30332
12. SPONSORING AGENCY NAME AND ADDRESS
Water Engineering Research Laboratory- Cine
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45258
3 RECIPIENT'S ACCESSION NO.
5. REPORT DATE
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
CR1102PO-01-0
13. TYPE OF REPORT AND PERIOD COVERED
innati , OH Prnjpct. Rppnrt
14. SPONS'&RING AGENCY CODE
EPA/600/14
15. SUPPLEMENTARY NOTES
Project Officer, Thomas J. Powers, 513-569-7550/FTS 684-7550
16. ABSTRACT
The research study of the reclamation of aluminum-anodizing sludges was
conducted in two sequential phases focused on enhanced dewatering of
aluminum-anodizing sludges to produce commercial-strength solutions of
aluminum sulfate, i.e., liquid alum. The use of high-pressure (14 to 15 bar)
and a diaphragm filter press were shown to be effective in dewatering aluminum
anodizing sludges to cake solids contents of 27% to 29% and 25% to 31%,
respectively. These values were well above the ?l% value required to justify
pursuit of direct acidic extraction of aluminum. Commercial-strength
solutions of aluminum sulfate with concentrations of 8% as A^fh were
produced using conventional-neutralization, segregated-neutral ization, and
etch-recovery sludge cakes. The trace metal contents of the alum products
were, in general, typical of commercial products.
17 KEY WORDS AND DOCUMENT ANALYSIS
.1. DESCRIPTORS

18 DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
b. IDENTIFIERS OPEN ENDED TERMS L. COS ATI 1 idd 'Croup

19. SECURI TY CLASS i this Hi'purti 21 NO OF PAGES
UNCLASSIFIED
20 SECURITY CLASS - rins pjwi 22 PRICE
UNCLASSIFIED
EPA Form 2220-1 (R«v. 4-77)    PREVIOUS  EDITION is OBSOLETE
                                                       121

-------