<%BANNER%>

Design of Electrolytic Dewatering Systems for Phosphatic Clay Suspensions

Permanent Link: http://ufdc.ufl.edu/UFE0041569/00001

Material Information

Title: Design of Electrolytic Dewatering Systems for Phosphatic Clay Suspensions
Physical Description: 1 online resource (103 p.)
Language: english
Creator: Mckinney, James
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: clay, consolidation, dewatering, electrokinetics, phosphate, waste
Chemical Engineering -- Dissertations, Academic -- UF
Genre: Chemical Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Phosphatic clays arise as a waste product of the Florida phosphate mining industry. The clays exist initially as a 2-6 solids weight percent slurry which is pumped to large impoundment areas for natural settling. The clay settling process takes as long as 25 years to reach a value of 40wt% solids. The clay settling areas currently cover an area of over 100 square miles in Central Florida, representing 40 percent of the land that has been mined. The Florida Institute of Phosphate Research (FIPR) has supported bench- and pilot-scale studies to determine the technical and economic feasibility of a variety of processes to speed the dewatering of clays associated with phosphate mining. The approaches considered included using a freeze-thaw cycle to remove water, adding sand layers to enhance drainage, and adding flocculants to enhance settling. While flocculants are used commercially, the other techniques have not been permanently implemented on a large scale, and settling to an acceptable solids content still requires as much as 25 years. Application of an electric field provides an alternative approach for accelerating the removal of water from clay. In this concept, direct electrical current is applied to induce movement of either clay particles or water. In dilute suspensions, the electric field induces the movement of clay particles suspended in water which is known as electrophoresis. Upon formation of a solid matrix, the electric field further induces the movement of water in a process known as electro-osmosis. The objective of this work was to use small-scale electrokinetic experiments to develop parameters that can be used for large-scale design. A bench-top Plexiglas cell was built to perform the experiments. Clay slurry samples were obtained from a phosphate mine located in Central Florida. A set of experimental results were used to calculate scaling parameters to aid in predicting the solids content as a function of operating time and the electric field applied. This was done by scaling the change in solids content by the applied electric field. A linear relationship was found at short times while at longer times a maximum solids weight percent was reached as a function of the electric field. A constitutive relationship was established which relates the increase in solids content to time and the applied electric field. A mathematical model previously developed at the University of Florida was used to model and evaluate varying electrode configurations in a one-square-mile clay settling area. The electrical current generated from the applied voltage was calculated to project electrical power and energy requirements for such a process. For a given electrode configuration, the associated electric field can also be calculated. The experimental work suggests a relationship between the solids content of the clays with the electric field. The model results can then be used to identify regions where the electric field is nonuniform, which indicates regions where the clays would have a nonuniform solids weight percent. Therefore, the model allows selection of an optimal electrode configuration based on electrical power requirements as well as the most uniform drying of the clays.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by James Mckinney.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Orazem, Mark E.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041569:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041569/00001

Material Information

Title: Design of Electrolytic Dewatering Systems for Phosphatic Clay Suspensions
Physical Description: 1 online resource (103 p.)
Language: english
Creator: Mckinney, James
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: clay, consolidation, dewatering, electrokinetics, phosphate, waste
Chemical Engineering -- Dissertations, Academic -- UF
Genre: Chemical Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Phosphatic clays arise as a waste product of the Florida phosphate mining industry. The clays exist initially as a 2-6 solids weight percent slurry which is pumped to large impoundment areas for natural settling. The clay settling process takes as long as 25 years to reach a value of 40wt% solids. The clay settling areas currently cover an area of over 100 square miles in Central Florida, representing 40 percent of the land that has been mined. The Florida Institute of Phosphate Research (FIPR) has supported bench- and pilot-scale studies to determine the technical and economic feasibility of a variety of processes to speed the dewatering of clays associated with phosphate mining. The approaches considered included using a freeze-thaw cycle to remove water, adding sand layers to enhance drainage, and adding flocculants to enhance settling. While flocculants are used commercially, the other techniques have not been permanently implemented on a large scale, and settling to an acceptable solids content still requires as much as 25 years. Application of an electric field provides an alternative approach for accelerating the removal of water from clay. In this concept, direct electrical current is applied to induce movement of either clay particles or water. In dilute suspensions, the electric field induces the movement of clay particles suspended in water which is known as electrophoresis. Upon formation of a solid matrix, the electric field further induces the movement of water in a process known as electro-osmosis. The objective of this work was to use small-scale electrokinetic experiments to develop parameters that can be used for large-scale design. A bench-top Plexiglas cell was built to perform the experiments. Clay slurry samples were obtained from a phosphate mine located in Central Florida. A set of experimental results were used to calculate scaling parameters to aid in predicting the solids content as a function of operating time and the electric field applied. This was done by scaling the change in solids content by the applied electric field. A linear relationship was found at short times while at longer times a maximum solids weight percent was reached as a function of the electric field. A constitutive relationship was established which relates the increase in solids content to time and the applied electric field. A mathematical model previously developed at the University of Florida was used to model and evaluate varying electrode configurations in a one-square-mile clay settling area. The electrical current generated from the applied voltage was calculated to project electrical power and energy requirements for such a process. For a given electrode configuration, the associated electric field can also be calculated. The experimental work suggests a relationship between the solids content of the clays with the electric field. The model results can then be used to identify regions where the electric field is nonuniform, which indicates regions where the clays would have a nonuniform solids weight percent. Therefore, the model allows selection of an optimal electrode configuration based on electrical power requirements as well as the most uniform drying of the clays.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by James Mckinney.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Orazem, Mark E.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041569:00001


This item has the following downloads:


Full Text

PAGE 1

DESIGNOFELECTROLYTICDEWATERINGSYSTEMSFORPHOSPHATICCLAY SUSPENSIONS By JAMESPATRICKMCKINNEY ADISSERTATIONPRESENTEDTOTHEGRADUATESCHOOL OFTHEUNIVERSITYOFFLORIDAINPARTIALFULFILLMENT OFTHEREQUIREMENTSFORTHEDEGREEOF DOCTOROFPHILOSOPHY UNIVERSITYOFFLORIDA 2010

PAGE 2

c r 2010JamesPatrickMcKinney 2

PAGE 3

Tomyparents,sister,andgrandparents 3

PAGE 4

ACKNOWLEDGMENTS Ithankmyadvisor,ProfessorMarkOrazem,forhissupportandguidance.Hehasshown menotonlyhowtoimprovemyabilitiesinresearch,buthowtoimprovemyabilitiesasaperson. IwouldliketothankCharlotteBrittainofMosaicFertilizerLLCforherinvolvementinsponsoring thisproject.IappreciateMrs.Brittain'swillingnesstobeofhelpanytimethatIhadquestionsor neededassistance.Shenotonlybenettedtheproject,butalsobenettedmyunderstandingin howthisprojecttintothescopeofthephosphateminingbusiness.Iwouldalsoliketothank CharlesGuan,RickGrove,andBrandonRussellofMosaicFertilizerLLCfortheirsupport.I thankProfessorGuerryMcClellan,ProfessorDavidBloomquist,andProfessorHassanEl-Shall fortheiradvisementonissuesrelatedtoclaysandclaydewatering.Iwouldliketothankallof thestudentswhohaveworkedinProfessorOrazem'sresearchgroupduringmytimeatthe UniversityofFlorida.ThisincludesNelliannPerez-Garcia,Mei-WenHuang,MichaelMatlock, SunilRoy,ShaolingWu,BryanHirschorn,ErinPatrick,DanielRood,RuiKong,andRodneyDel Rio.IwouldalsoliketothankthestaffmembersintheDepartmentofChemicalEngineering. ThisincludesShirleyKelly,DeborahSandoval,DennisVince,JimHinnant,andSeanPoole. Finally,Iwouldliketothankmyparents,mysister,andmygrandparentsfortheirloveand supportthroughoutmylife.Iappreciatetheminstillingthevalueofeducationinmeatanearly age. 4

PAGE 5

TABLEOFCONTENTS page ACKNOWLEDGMENTS .......................................4 LISTOFTABLES ...........................................7 LISTOFFIGURES ..........................................8 LISTOFSYMBOLS .........................................12 ABSTRACT ..............................................14 CHAPTER 1INTRODUCTION ........................................16 2THEPHOSPHATEINDUSTRY ................................19 2.1PhosphateMineralogy ..................................20 2.2PhosphateBeneciation .................................23 2.3ClaySettlingAreas(CSAs) ...............................25 3STUDYOFCLAYS .......................................29 3.1ClayPropertiesandStructures .............................29 3.2ChargeonClays .....................................30 3.3TypesofClays ......................................31 3.3.1SmectiteClay ...................................31 3.3.2KaoliniteClay ...................................32 3.3.3PalygorskiteClay .................................32 4ELECTROCHEMICALENGINEERINGANDELECTROKINETICTHEORY .......33 4.1ElectrochemicalEngineeringTechniques .......................33 4.2MethodstoOptimizeElectrochemicalCells ......................34 4.3PrinciplesofElectrokinetics ...............................35 4.3.1ElectricalDoubleLayer ..............................35 4.3.2ApplicationofElectricField ...........................38 5LITERATUREREVIEW ....................................40 5.1ElectrokineticDewatering ................................40 5.1.1Small-scaleBench-topCells ...........................40 5.1.2Large-scaleFieldApplications ..........................41 5.2OtherDewateringMethods ...............................42 5.2.1Flocculation ....................................42 5.2.2Large-scaleFieldApplication ..........................43 5.3AdditionalElectrokineticApplication ..........................44 5.4ElectrokineticParameters ................................44 5.4.1SelectionandDesignofElectrodes .......................45 5.4.2pHandZetaPotential ..............................45 5.4.3UseofIntermittentCurrent ............................46 5

PAGE 6

5.4.4ParticleSize ....................................47 5.5AssessmentoftheLiterature ..............................47 6EXPERIMENTALAPPROACH ................................49 6.1Ex-situAnalysisofClaySuspensions .........................49 6.1.1X-RayDiffraction .................................49 6.1.2ParticleSizeDistribution .............................50 6.1.3ZetaPotentialMeasurements ..........................51 6.2ElectrokineticStudies ..................................52 6.2.1ExperimentalInstrumentation ..........................52 6.2.2Bench-topCell ..................................52 6.2.3ExperimentalOperation .............................54 7EXPERIMENTALRESULTSANDDISCUSSION ......................57 7.1ProofofConceptwithBench-topCell ..........................57 7.2ConstitutiveRelationshipatShortTimes ........................57 7.3UniformityofWaterRemoval ..............................59 7.4ConstitutiveRelationshipatLongTimes ........................64 7.5EnergyUsageModel ...................................69 7.6ElectrochemicalCharacterization ............................75 8SIMULATIONSFORLARGE-SCALEDEWATERINGSYSTEMS .............80 8.1IntroductiontoCP 3 DandApplicationforCathodicProtection ............80 8.1.1MathematicalDevelopment ...........................81 8.1.2BareSteel .....................................84 8.1.3CoatedSteel ...................................84 8.1.4SacricialandImpressedCurrentAnodes ...................86 8.2ApplicationforClayDewatering .............................86 8.2.1Large-scaleSimulations .............................86 8.2.2EconomicImplications ..............................90 9CONCLUSIONSANDFUTUREWORK ...........................93 9.1Conclusions ........................................93 9.2FutureWork ........................................94 REFERENCES ............................................97 BIOGRAPHICALSKETCH .....................................103 6

PAGE 7

LISTOFTABLES Table page 7-1ResultsofcalculationsusedtomodeltheenergyrequirementspresentedinFigure717.................................................74 7-2Valuesofparametersandvariablesusedinenergyusagemodel..............74 7-3Conductivitiesofclaysuspensionsfromthisworkarecomparedwiththosereported intheliterature.EISandDCwerethemethodsusedtoexperimentallydeterminethe conductivityofclaysuspensionsusedwithinthiswork....................79 8-1Resultsofpowerandenergycalculationsfordewateringofsimulatedone-square-mile claysettlingarea.Solidsweightpercentwasincreasedfrom10to25wt%........90 8-2Resultsofpowerandenergycalculationsfordewateringofsimulatedone-square-mile claysettlingarea.Solidsweightpercentwasincreasedfrom10to20wt%........90 7

PAGE 8

LISTOFFIGURES Figure page 2-1Regions(takenfromBloomquist 1 )ofFloridaphosphatedepositsareidentied.The insetpresentsageneralviewofthemakeupoftheCentralFloridadepositswithvaryingdepth.............................................21 2-2Landstructuralregions(takenfromKauwenbergh etal. 2 afterUSGSbulletin1914)in Floridaareindicated.Theinsetpresentsthelocationsofseveralphosphateminesin theCentralFloridaregion.Eachminelocationismarkedbytheicon b X ..........22 2-3Photographofelectricallyoperateddraglineshownminingthephosphatematrix(photographbyMarkOrazem,usedwithpermission)......................24 2-4Photographofbucketattachedtothedraglineusedtodisplacethephosphatematrix (photographbyMarkOrazem,usedwithpermission)....................24 2-5Illustrationofclaysettlingareashowingwatermigrationtowardsonecornerforrecyclingtothebeneciationplant(imageaccessedfromMosaicwebsite). ..........26 2-6Photographsofclaysettlingareaslookingdownfromraisedembankment;A)oneend ofthepondwhereclearwaterhassurfacedtothetop;andB)theotherendofthepond whereclayhasnotsettledaswell(photographsbyMarkOrazem,usedwithpermission). ..............................................27 2-7Photographsofclaysettlingareasshowingtheinteriorslopedembankment;A)one cornerofthepondwhereclearwaterhassurfacedtothetop;andB)anothercorner ofthepondwhereclayhasnotsettledaswell(photographsbyMarkOrazem,used withpermission). .......................................28 3-1Molecularrepresentations(reproducedfromCraig 3 )ofthebasicstructuresofclayminerals.Themolecularstructurescombinetoformthedepictedsheetstructures. .....30 4-1SchematicofelectricaldoublelayerindicatingtheinnerHelmholtzplane(IHP),the outerHelmholtzplane(IHP),andthethicknessofthediffusepartofthedoublelayer .Thepositivecirclesindicatecationsandthenegativecirclesindicateanions. .....37 4-2Schematicillustratingthedirectionoftheelectriceld,thediffusedoublelayer,and thevelocityprolethatarises. .................................38 6-1X-raydiffractionpatterngeneratedfromanorientedaggregatemountofaphosphatic claysample.Peaksareidentiedwiththeirassociatedclayorsandmineral. .......51 6-2Schematicofbench-topcellwithlabeledlocationsoftheelectrodesandthetemperatureandvoltagemeasurements.Darkershadedareawithinthecellrepresentswhere clayslurryisloadedforexperiments. .............................54 6-3Schematicillustratingconnectionsanduseofresistorsbetweenpotentiostatandbenchtopcell.Resistorsareincludedtoamplifytheappliedpotentialtothecell. ........55 7-1Photographsofthebench-topPlexiglasshown:A)Beforeexperimentwasstartedthe cellwasloadedinitiallywitha9wt%solidssuspension,andB)aftertheexperiment endedthecellwasoccupiedbyathickenedlumpofclayconsistingof33.5wt%solids. .58 8

PAGE 9

7-2Resultsofbench-topdewateringexperimentsillustratingthechangeinsolidsweight percentasafunctionofoperatingtimewithappliedpotentialasaparameter.Thedata marked1.5Vwerecontrolledgalvanostaticallytoyieldaconstantelectriceld......59 7-3SolidsweightpercentresultsfromFigure7-2arescaledbytheappliedelectriceld E .Atrendlinettothedatasuggestsalinearrelationshipwhichisgivenasequation (7)...............................................60 7-4Cellpotentialof10Vwasappliedforthedurationoftheexperiment.TheinsetrepresentsaschematicoftheverticallyorientedcellwithzoneDatthetopofthecellrepresentingthelocationofthesupernatantwaterseparatedfromthebulkclay.......61 7-5Cellpotentialof20Vwasappliedforthedurationoftheexperiment.TheinsetrepresentsaschematicoftheverticallyorientedcellwithzoneDatthetopofthecellrepresentingthelocationofthesupernatantwaterseparatedfromthebulkclay.......61 7-6Cellpotentialof40Vwasappliedforthedurationoftheexperiment.TheinsetrepresentsaschematicoftheverticallyorientedcellwithzoneDatthetopofthecellrepresentingthelocationofthesupernatantwaterseparatedfromthebulkclay.......62 7-7Cellpotentialof80Vwasappliedforthedurationoftheexperiment.TheinsetrepresentsaschematicoftheverticallyorientedcellwithzoneDatthetopofthecellrepresentingthelocationofthesupernatantwaterseparatedfromthebulkclay.......62 7-8Cellpotentialof20Vwasappliedforthedurationoftheexperiment.ThedatainFigure7-5arereproducedhereinadditiontoanexperimentoperatingfor48hours.....63 7-9Photographofcellafterthecompletionofa4day,20Vexperimentanditscomparisontothesettlingofacontrol(onright)sampleintheabsenceofanelectriceld. Thecellpotentialwasappliedfor12hourseachday.Thearrowindicatesthedistance betweenelectrodes.Aclearlayerofwaterisindicatedabovetheclay. ..........64 7-10Additionaldataatlongertimesandlargersolidsweightpercentareincludedwiththe datapresentedinFigure7-3.Thehorizontaldashedlinesrepresentthesuggested plateaureachedatthreedifferentelectriceldsizes. ....................65 7-11Maximumchangeinsolidscontentplottedasafunctionoftheelectriceld.Therelationshipdevelopedfromthelineartrendlineisgivenbyequation(7). .........66 7-12DatafromFigure7-10withtheconstitutiverelationshipforlong-times(eq.(74))tto thedataforthreedifferentelectriceldsizes. ........................67 7-13DataincludedinbothFigures7-10and7-12arepresentedwithoutscalingbytheelectriceld.InB),theconstitutiverelationshipforlongtimes(eq.(7))isttothedata atthreeelectriceldsizes. ..................................67 7-14Constitutiverelationshipforlongtimes(eq.(7))plottedforseveralelectriceldsizes. 68 7-15Constitutiverelationshipforlongtimes(eq.(7))plottedforseveralelectriceldsizes withthechangeinsolidsweightpercentscaledbytheelectriceld. ...........68 7-16Energyrequiredpermassofwaterremovedisgivenasafunctionoftheelectriceld. .70 9

PAGE 10

7-17DatapresentedinFigure7-16comparedwithamodeledenergyrequirementcurve. Energyrequiredpermassofwaterremovedisgivenasafunctionoftheelectriceld..71 7-18Energyrequirementisgivenasafunctionofoperatingtime.InB),thetransitionfrom lineartononlinearbehaviorisindicated............................75 7-19Energyrequirementasafunctionoftheelectriceldforthemaximumseparationachievableattheappliedelectriceld E .ValuesweredeterminedusingtheterminaloperatingtimeindicatedbythemethodpresentedinFigure7-18A................76 7-20DatapresentedinFigure7-17areincludedforcomparisontothemodelbasedupon theterminaloperatingtime...................................76 7-21Twosetsofoperatingtimesusedforthemodelaregivenasafunctionoftheelectric eld................................................77 7-22Polarizationcurvegeneratedfrombench-topcellloadedwithclaysuspension.Each datapointwasindividuallymeasuredfromseparateexperimentsforeachappliedpotential...............................................78 7-23Impedancescangeneratedatopencircuitpotentialonbench-topcellloadedwithclay suspension............................................78 7-24Re-scaledperspectiveofimpedancescanpresentedinFigure7-23withtheaxisfor therealimpedancesubtractedbytheOhmicresistance R e .................79 8-1CP 3 Dimageshowingthephysicalorientationofthesoilsurfacewithrespecttothe pipeline.Thedarkerareaonthepipelinerepresentsthecoatingaworholiday. .....81 8-2Representationofalargeclaysettlingareawithonemilelongcylindricalelectrodes spacedequallyalongtopandbottomsurfaces.Thezoomedinportionrepresentsa cross-sectionofthecylindricalelectrodes.Althoughonlyrowsofthreecylindersare shown,thesimulationwasscaledforrowsthatextendtoadistanceofonemile. ....87 8-3IllustrationofhorizontallyorientedelectrodecongurationpresentedinFigure8-2with thecalculatedpotentialdistributionfromCP 3 Dpresentedasafalse-colorimage. ....87 8-4Verticallyorientedelectrodesarepresented:A)CP 3 Dimageofverticallyorientedelectrodeconguration;andB)photographofageosyntheticelectrodecoveredwithalterclothrequiredasaseparatorforremovalofwaterusingaverticallyorientedcathode(takenfromFourie etal., 4 Copyright c r 2008NRCCanadaoritslicensorsandreproducedwithpermission). ..................................88 8-5IllustrationoftheelectriceldcalculatedfromthesimulationpresentedinFigure8-4A: A)threedimensionalrepresentation,andB)twodimensionalrepresentation. ......89 8-6Energyrequirementforwaterremovalisgivenasafunctionoftheelectriceld.ExperimentaldatapresentedinFigure7-16areincludedwithbothsimulationandadditionalexperimentaldata. ....................................91 8-7Atopviewoftwodifferentone-square-mileclaysettlingareas.Themodieddesignis picturedontherightdemonstratingthetreatmentofanisolatedsectionoftheCSA. ..92 10

PAGE 11

8-8Schematicrepresentationofthickeneradaptedforelectrokineticallyenhancedseparation...............................................92 9-1Schematicrepresentationofbench-scalesettlingbasinadaptedforelectrokinetically enhancedseparation......................................95 11

PAGE 12

LISTOFSYMBOLS Roman A c cross-sectionalareaofbench-topexperimentalcell,cm 2 b aa kineticparameterdenedinequation(7),V )Tj /T1_4 7.97 Tf (1 b ac kineticparameterdenedinequation(7),V )Tj /T1_4 7.97 Tf (1 b ca kineticparameterdenedinequation(7),V )Tj /T1_4 7.97 Tf (1 b cc kineticparameterdenedinequation(7),V )Tj /T1_4 7.97 Tf (1 c i concentrationofspecies i,mol/cm 3 D i diffusioncoefcientofspecies i ,cm 2 /s d distancebetweenelectrodes,cm E electriceld,V/cm E Fe equilibriumpotentialforoxidationofiron,V E H 2 equilibriumpotentialforhydrogenevolution,V E O 2 equilibriumpotentialforoxygenreduction,V E req energyrequirementforremovalofwater,Wh/kgofwaterremoved F Faraday'sconstant, 96; 487 C/equiv I current,A i currentdensity,mA/cm 2 i lim;O 2 masstransferlimitingcurrentdensityforoxygenreduction,mA/cm 2 N i uxofspecies i,mol/cm 2 s q charge,C/cm 2 R resistance, ncm 2 or n (1n=1Vs=C) R universalgasconstant, 8:314 J/molK R e electrolyteorOhmicresistance, n or ncm 2 R i rateofhomogeneousproductionofspeciesi,mol/cm 3 s T temperature,K t experimentaloperatingtime,h u i mobilityofspecies i,relatedtodiffusioncoefcientbyequation(8) V potential,V v velocity,cm/s 12

PAGE 13

w c solidscontentoftheslurry,wt% w max maximumsolidscontentoftheslurry,wt% z i chargeassociatedwithspecies i Greek symmetryfactorusedinelectrodekinetics,dimensionless f Tafelslope,V/decadeofcurrent thickness,cm permittivity,F/cm 0 permittivityofvacuum, 8:8542 10 )Tj /T1_3 7.97 Tf (14 F/cm zetapotential,mV a anodicoverpotential,V c cathodicoverpotential,V conductivity,S/cm Debyelength,nm uidviscosity,g/cms electricalresistivity, ncm 2 e electrolyteresistivity, ncm b potential,V w c changeinsolidscontentoftheslurry,wt% Subscripts i pertainingtochemicalspecies i Z j pertainingtotheimaginarypartoftheimpedance Z r pertainingtotherealpartoftheimpedance 13

PAGE 14

AbstractofDissertationPresentedtotheGraduateSchool oftheUniversityofFloridainPartialFulllmentofthe RequirementsfortheDegreeofDoctorofPhilosophy DESIGNOFELECTROLYTICDEWATERINGSYSTEMSFORPHOSPHATICCLAY SUSPENSIONS By JamesPatrickMcKinney May2010 Chair:MarkE.Orazem Major:ChemicalEngineering PhosphaticclaysariseasawasteproductoftheFloridaphosphateminingindustry.The claysexistinitiallyasa2-6solidsweightpercentslurrywhichispumpedtolargeimpoundment areasfornaturalsettling.Theclaysettlingprocesstakesaslongas25yearstoreachavalueof 40%solids.Theclaysettlingareascurrentlycoveranareaofover100squaremilesinCentral Florida,representing40percentofthelandthathasbeenmined. TheFloridaInstituteofPhosphateResearch(FIPR)hassupportedbench-andpilot-scale studiestodeterminethetechnicalandeconomicfeasibilityofavarietyofprocessestospeed thedewateringofclaysassociatedwithphosphatemining.Theapproachesconsideredincluded usingafreeze-thawcycletoremovewater,addingsandlayerstoenhancedrainage,andadding occulantstoenhancesettling.Whileocculantsareusedcommercially,theothertechniques havenotbeenpermanentlyimplementedonalargescale,andsettlingtoanacceptablesolids contentstillrequiresasmuchas25years. Applicationofanelectriceldprovidesanalternativeapproachforacceleratingtheremoval ofwaterfromclay.Inthisconcept,directelectricalcurrentisappliedtoinducemovementof eitherclayparticlesorwater.Indilutesuspensions,theelectriceldinducesthemovement ofclayparticlessuspendedinwaterwhichisknownaselectrophoresis.Uponformationofa solidmatrix,theelectriceldfurtherinducesthemovementofwaterinaprocessknownas electro-osmosis. Theobjectiveofthisworkwastousesmall-scaleelectrokineticexperimentstodevelop parametersthatcanbeusedforlarge-scaledesign.Abench-topPlexiglascellwasbuiltto performtheexperiments.Clayslurrysampleswereobtainedfromaphosphateminelocated 14

PAGE 15

inCentralFlorida.Asetofexperimentalresultswereusedtocalculatescalingparametersto aidinpredictingthesolidscontentasafunctionofoperatingtimeandtheelectriceldapplied. Thiswasdonebyscalingthechangeinsolidscontentbytheappliedelectriceld.Alinear relationshipwasfoundatshorttimeswhileatlongertimesamaximumsolidsweightpercent wasreachedasafunctionoftheelectriceld.Aconstitutiverelationshipwasestablishedwhich relatestheincreaseinsolidscontenttotimeandtheappliedelectriceld. AmathematicalmodelpreviouslydevelopedattheUniversityofFloridawasusedtomodel andevaluatevaryingelectrodecongurationsinaone-square-mileclaysettlingarea.The electricalcurrentgeneratedfromtheappliedvoltagewascalculatedtoprojectelectricalpower andenergyrequirementsforsuchaprocess.Foragivenelectrodeconguration,theassociated electriceldcanalsobecalculated.Theexperimentalworksuggestsarelationshipbetweenthe solidscontentoftheclayswiththeelectriceld.Themodelresultscanthenbeusedtoidentify regionswheretheelectriceldisnonuniform,whichindicatesregionswheretheclayswould haveanonuniformsolidsweightpercent.Therefore,themodelallowsselectionofanoptimal electrodecongurationbasedonelectricalpowerrequirementsaswellasthemostuniform dryingoftheclays. 15

PAGE 16

CHAPTER1 INTRODUCTION Electrokineticdewateringtakesplacethroughinteractionbetweenuidow,electrical current,andchargeadsorbedontheclayparticles. 5 Inordertoapplyanelectriceldtheremust beatleasttwoelectrodesimmersedintotheclayslurryorelectrolyte.Oneelectrodefunctions asthepositivelychargedanodewhereoxidationreactionsoccurandtheotherfunctionsasthe negativelychargedcathodewherecathodic(orreduction)reactionsoccur.Attheanode,the electrochemicalreactionsincludeoxygenevolution 2H 2 O O 2 +4H + +4e )Tj /T1_0 10.909 Tf 154.339 -4.505 Td ((1) whichcreatesanacidicenvironmentneartheelectrode.Asecondpossiblereactioninvolves corrosionoftheelectrode M M n+ +ne )Tj /T1_0 10.909 Tf 175.204 -4.504 Td ((1) Atthecathode, H 2 O+e )Tj /T1_3 10.909 Tf 10.115 -4.504 Td (! 1 2 H 2 +OH )Tj /T1_0 10.909 Tf 154.507 -4.504 Td ((1) creatingalocalbasicenvironment.Thewaterisdrivenfromtheanodetothecathode,creatinga tendencytodryouttheregionclosetotheanode.Therefore,thewatertravelsinthesamedirectionasconventionalcurrent. 6 ThesolutionpHisexpectedtoincreaserapidlytoapproximately 11or12atthecathodeandhydrogenwillbegenerated,asshownbyreaction(1).Cationsare driventothecathodebytheelectricpotentialgradientandcanbereduced(theinverseofreaction(1)).ApHgradientwillbegeneratedacrossthesoilasanoverallresultoftheelectrode reactions. Electrophoresisandelectro-osmosismethodsofdewateringphosphateclayshavebeen exploredsincethe1940s. 710 Althoughthebasictechnologywasconsideredtohavepromise itwasnotconsideredcommerciallyfeasible.Signicantprogresshassincebeenmadeinour understandingoftheunderlyingelectrokineticandelectro-osmoticprocessesandinthepractical implementationofthesetechniques.Forexample,recently,electrokineticmethodshavebeen demonstratedfordewateringofsoil, 11 minetailings, 4,12 sludge, 13 andclay, 1416 includingFlorida phosphateclay. 17 Theliteraturesuggestselectrokineticmethodscansignicantlydecreasethe 16

PAGE 17

timerequiredtodewaterclay.Therefore,furtherresearchinthisareaiswarrantedduetothe advancesthathavebeenmadeinthelasttwodecades. Thisworkispresentedinninechapters.Thebackgroundofthephosphateindustryis discussedinChapter2.Thephosphateproductionrequirementswhichdrivethephosphate industryaresummarized.ThelocationsofphosphateminesinthestateofFloridaarepresented andthemineralogyofthephosphatereservesarediscussed.Theminingprocessandproductionofphosphatearedescribedaswellasthewasteclaysettlingareasthatexistasaresultof phosphateproduction. Thisworkismotivatedduetosettlingandconsolidationissuesthatarisefromclaysettling areas;therefore,Chapter3providesdetailonthecharacteristicsofclayswhichincludestheir molecularstructures.Thesefeaturesareusedtoexplainwhyclayshavesuchpoorsettling characteristics.However,discussionalsoinvolveshowthesefeaturesmayaidsettlingor consolidationinthepresenceofanelectriceld. ElectrochemicalengineeringandelectrokineticphenomenaareintroducedinChapter4. Thefundamentalsofelectrokineticphenomenaareintroducedtoprovideaframeworkforhow separationsofclaysuspensionsoccur.Polarizationcurvesandimpedancespectroscopyare introducedastechniquesusedtocharacterizeelectrochemicalcells.Experimentalresultsinthe literaturearepresentedinChapter5.Aqualitativeassessmentoftheliteraturewasperformed whichdiscussesthefeasibilityofwaterremovalfromclays.Alsodiscussedareitemsabsent withintheliteraturethatprovidethebasisforthecontributionsofthiswork. TheexperimentalapproachtothisworkisdescribedinChapter6.Ex-situstudieswere performedontheclaysuspensionswhichwererepresentativeofsamplesusedinbench-top experimentspresentedwithinthiswork.Particlesizestudieswereperformedtocharacterizethe sizedistributionoftheparticlesinsuspension.X-Raydiffractionwasperformedtoidentifythe typesofclayspresentintheclaysuspensions.Suchanalysiswaswarrantedastheresponse toelectrokineticscanvaryduetothedifferentsurfacepropertiesofoneclaymineraltoanother. Zetapotentialmeasurementswerealsoperformedtohelpquantifytheamountofchargeon theclayparticlesinsuspension.Thischapteralsointroducesthebench-topapparatus(or electrochemicalcell)andtheinstrumentationusedtocontroltheapparatus. 17

PAGE 18

Thebench-topexperimentsthatwereperformedusinganelectriceldtoremovewater fromclayslurriesareintroducedinChapter7.Theexperimentalresultsaregivenanddiscussed withinthischapter.Constitutiverelationshipsweredeveloped,whichrelatethechangeinclay solidsweightpercentasafunctionofboththeoperatingtimeandtheelectriceld.Electrochemicaltechniquessuchasimpedancespectroscopyandpolarizationcurveswerealsoimplemented tocharacterizefurthertheexperimentalwork. Thesimulationworkperformedtoassessdifferentelectrodecongurationsinasimulated one-square-mileclaysettlingareaispresentedinChapter8.Relationshipsfoundintheexperimentalworkwereusedtomakeassumptionsregardingthedevelopmentandresultsofthe simulations.Theserelationshipsandassumptionssupportedtheprojectionsmadeforelectric powerrequirementsaswellastheenergyrequirementsofdifferentelectrodecongurations.The simulationswereperformedusingCP 3 D,amathematicalmodeldevelopedattheUniversityof Florida. TheconclusionsmadefromtheexperimentalandsimulationworkarereviewedinChapter 9.Thisdiscussionisfollowedbyproposedworkforthenextstepsofthisproject.Possible dewateringscale-updesignsforfutureimplementationareproposedandanewbench-top experimentaldesignisintroduced. 18

PAGE 19

CHAPTER2 THEPHOSPHATEINDUSTRY Thephosphateindustrywasstartedinthemidtolate19thcentury. 18 Theindustrywas developedduetophosphatebeinginvaluableasakeyingredientinfertilizersforfoodproducing crops.Italsocontributestotheproductionofotheritemssuchassoftdrinks,lightbulbs,vitamins,shavingcream,toothpaste,anditcanbeusedasafeedadditiveforlivestock.However, inthestateofFlorida,about90%ofthephosphateminedisusedasfertilizerforfoodproducing crops. PhosphateoredepositsarefoundallovertheworldincountriessuchastheUnitedStates, China,MoroccoandRussia. 19 PhosphateisalsominedinCanada,Australia,andJordan. Overall,signicantphosphatedepositsexistonvecontinentsandthereareover30countries thatminephosphate.However,about70%oftheworld'stotalproductioncomesfromtheUnited States,MoroccoandRussia. 20 InthestateofFlorida,thephosphateindustryisthethirdlargest industrybehindtourismandagriculture.In1982,itwasreportedthatthephosphateindustry injected1.5billiondollarsintotheFloridastateeconomyannually. 1 InCentralFloridaalone, therearecurrently4,000employeesworkinginthephosphateindustry. Phosphorusisconsideredasoneofthethreemostessentialplantnutrients. 21 Itisa nutrientthatissuppliedthroughthesoilalongwithnitrogenandpotassium.Phosphateisthe mainsourceofphosphorusfororganicfertilizersanditisanon-renewableresource. 1 Itis requiredforalllivingcellsinplantsandanimalsasitallowscellstostoreenergyfromsunlight andfood. 19 Plantgrowthisenhancedbysoilsthathavebeentreatedwithphosphatefertilizers. Oncethephosphaterockhasbeenminedandextractedthroughbeneciation,itisground intoanegrainsizeandmixedwithsulfuricacid.Thisprocessproducesphosphoricacidand calciumsulfate.Thecalciumsulfate,alsoknownasgypsum,canberemovedthroughltration. Theresultingphosphoricacidisthenreactedwithammoniatoproducediammoniumphosphate andmonoammoniumphosphate.Triplesuperphosphateisalsocreatedbyaddingphosphoric acidwiththephosphaterock.Thesethreeproductsarecombinedtogetherinmostfertilizers andtheyexistaswater-solublegranulestoallowforabsorptionbythesoilandplants.The conversionofinsolublephosphaterocktosolublephosphatefertilizerwasrstcommercially 19

PAGE 20

achievedinEnglandin1842. 22 InGermany,triplesuperphosphatewasrstcommercially producedin1870;however,itdidnotbecomeanimportantfertilizeringredientuntil1950. 22 2.1PhosphateMineralogy Worldwideitisestimatedthattherearebetween15and70billionmetrictonsofphosphate reserves. 20 InthesoutheasternUnitedStates,therearephosphatereservesthatexistinNorth Carolina,SouthCarolina,Georgiaandaspreviouslydiscussed,Florida.ThedepositsinFlorida havebeenthemostproductiveasonebilliontonsofphosphatewereproducedinthestateof Floridafrom1888through1990. 2 In1982itwasreportedthatreservesinthestateofFlorida produced80%ofthephosphateusedintheUnitedStatesand35percentworldwide. 1 Tenyears later,in1992,anotherreportcreditedthestateofFloridawithproducing85%ofthephosphate usedintheUnitedStatesand33%worldwide. 23 Itappearsthatthesepercentageshavebeen maintainedthroughpresentday.However,theresearchdirectoroftheFloridaInstituteof PhosphateResearch(FIPR)wasquotedaspredictingthatthephosphatereserveswillbe depletedinthestateofFloridabytheyear2040. 19 ThephosphatedepositsinFloridaaredominatedbythemineralscarbonate-uorapatite (alsoknownasapatiteorfrancolite),quartz,dolomite,anddifferentclayminerals. 2 Accordingto Bloomquist, 1 onesampletakenfromaspecicreserveinFloridaalsoincludedmineralssuchas wavelite,limoniteandfeldspar.Carbonate-uorapatiteisconsideredastheprimaryphosphate bearingmineral.Dolomiteiscalciummagnesiumcarbonateorcanbewrittenas CaMg(CO 3 ) 2 Quartzhasthemolecularformula SiO 2 anditisthesecondmostabundantmineralintheearth's crust.Themostabundantmineralintheearth'scrustisfeldspar,whichismadeupofacomplex groupofsilicates.ThemineralwaveliteincludestheelementPhosphorus,whilethemineral limonitedoesnot. ThelayoutandlocationsoftheFloridaphosphatereservesarepresentedinFigures 2-1 and 2-2.BothguresillustratetheentirestateofFlorida,buteachfocusesontheregionin CentralFloridawhichrepresentsthelocationwheremostofthephosphateminingoccurs. AmorespecicdepictionofphosphatereservesispresentedinFigure 2-1 whileadditional regionaldescriptionsinthestateofFloridaareidentiedinFigure 2-2 aswellasspecicmining locations. 20

PAGE 21

Figure2-1.Regions(takenfromBloomquist 1 )ofFloridaphosphatedepositsareidentied.The insetpresentsageneralviewofthemakeupoftheCentralFloridadepositswith varyingdepth. 21

PAGE 22

Figure2-2.Landstructuralregions(takenfromKauwenbergh etal. 2 afterUSGSbulletin1914)in Floridaareindicated.Theinsetpresentsthelocationsofseveralphosphateminesin theCentralFloridaregion.Eachminelocationismarkedbytheicon b X 22

PAGE 23

PhosphateinthesoutheasternUnitedStatesoriginatesfromrocksthatwereformedduring theMioceneage. 2 TheMioceneagecanbedescribedasthetimeperiodbetweenroughly5 to23millionyearsago.Muchofthephosphatedepositswereformedinthecoastalwaters surroundingFlorida.Thedepositshavealsobeenfoundtooriginatefromthefossilsofanimal remains.TherstdiscoveryofphosphateinFloridaoccurredin1881inthePeaceRiver,which isnearbythetownofFortMeade.ThediscoverywasmadebyJ.FrancisLeBaronoftheU.S. ArmyCorpsofEngineers. 24 2.2PhosphateBeneciation Phosphateexistsinamatrixofsand,clayandphosphatethatliesunderneaththeground. Thematrixismadeupofapproximatelyone-thirdsand,one-thirdclayandone-thirdphosphate. 23,24 Abovethematrixexistsanoverburdenwhichismadeofprimarilysandandclay. Theoverburdenliesfromthesurfaceofthegroundtoadepthofapproximately25feet.Directly beneaththeoverburdenliesthephosphatematrix.Thephosphatematrixusuallyextendsfrom adepthof25to50feet.Anextremelylargedraglineexcavationsystemisusedtoremovethe overburdenthatliesontopofthephosphatematrix.Theuseofdraglinesforphosphateexcavationbeganasearlyasthe1920s. 25 Thedraglinestodaycanreachsizesover3,500tonswith areachlargerthan300feet.Aphotographofacurrentlyoperateddraglineminingphosphate ispresentedinFigure2-3.Today'sdraglines,suchasthatpicturedinFigure2-3,candigupto 15acresamonth. 19 Suchlargeamountsareabletobeexcavatedduetothehugebucketthat extendsfromthedraglinedisplacinglargeamountsofthematrix.Aphotographofthebucket extendingfromthedraglinepicturedinFigure2-3ispresentedinFigure2-4.Thesebucketsare solargethatatleast25peoplecouldcomfortablystandnexttoeachotherwithinthedimensions ofthebucket. Uponcompletionoftheremovaloftheoverburden,thedraglinecontinuestoexcavatethe phosphatematrixtoanareathatismoreshallow.Thisshallowerarea,termedasapit,iswhere thematrixiscontactedwithhighpressurewaterwhichdissolvesthematrixintoaslurry.Itis advantageousforthematrixtobeinaslurryformsuchthatitcanbepumpedthroughpipes awayfromtheminingareaandtowardsthephosphatebeneciationplant.Atthebeneciation plantthephosphate,clayandsandareseparatedfromeachother.Thisisdonethroughaseries ofwashersandcyclones.Thewasherstakeadvantageofthepebblesizeofthephosphateto 23

PAGE 24

Figure2-3.Photographofelectricallyoperateddraglineshownminingthephosphatematrix (photographbyMarkOrazem,usedwithpermission). Figure2-4.Photographofbucketattachedtothedraglineusedtodisplacethephosphatematrix (photographbyMarkOrazem,usedwithpermission). 24

PAGE 25

aidinitsseparation.Theresidualwatertiedupwiththephosphatepebblesiseasilyremoved throughaseriesofdewateringtanks.Thephosphatepebbleswhendriedarereadytobe manufacturedintowater-solublefertilizer.Theseparatedsandisreturnedtothemineareato aidinrellingtheoriginalminecuttoexpeditelandreclamation. 22,26 Theleftoverclaysfromthe beneciationprocessarethensenttolargeclaysettlingareas,alsoknownasCSAs.Theclays existasa2-6solidsweightpercentslurrywhentheyarriveattheCSAs.Typically,theslurry senttotheCSAcontainsparticlesnerthan150micronsindiameter. 1,2325,27,28 Thisslurry includes20to30%oftheexcavatedphosphatenotincludingthequartzandsandparticlesalso presentintheslurry. 23 Ithasnotbeendeterminedwhetherimprovementsinthebeneciation processcouldbemadetoavoidsuchlossesofphosphate.Onesourcestatedthatapproximately onethirdoftheoriginal P 2 O 5 inthephosphatematrixisusuallylosttotheclayslurries, 29 while anotherstatedthat12to15%oftheavailablephosphateislosttothewasteslurry. 30 Theclay settlingareasandtheprimaryissuesregardingthemarediscussedingreaterdetailinSection 2.3. 2.3ClaySettlingAreas(CSAs) Disposalofthephosphaticclayshasbeenoneofthephosphateindustry'smostimportant challengesforseveraldecadescausingissueswithregardtowaterconsumption,waterpollution, andlandreclamation. 18 Issuesassociatedwithphosphaticclaydisposalhavestimulated researchsincethe1950s. 26,31 Theclaysarestoredinlargeabove-groundman-madedikes whichtypicallyreach40feetabovegroundlevel.Thesestoragepondsaretermedasclay settlingareas(CSAs).ThedepthfromthesurfaceoftheclaystothebottomofaCSAisusually between30to40feet.Claysettlingareasoccupyapproximately100squaremilesoflandin CentralFloridaasaresultofphosphatemining.EachindividualCSAistypicallydesigned tohaveanareaofone-square-mile.OnereportcitedthatFloridaphosphateminessend 20,000to30,000gallonsperminutetoclaysettlingareas; 32 whileotherreportshavelistedthat 100,000tonsperdayaregeneratedinFlorida. 28,33 Eitherway,itisclearthatthevolumeofclay suspensionsgeneratedfarexceedthevoidspacecreatedbyminingwhichleadstomajorland storageissues. 34 Tocompoundthisproblem,thepoorsettlingcharacteristicsoftheclaysmeans thatitcantake10yearsorlongertoreach25%solidsbymass(orweight). 30 25

PAGE 26

Figure2-5.Illustrationofclaysettlingareashowingwatermigrationtowardsonecornerfor recyclingtothebeneciationplant(imageaccessedfromMosaicwebsite). TheCSAsareconsideredtobeofmajorenvironmentalconcerntoenvironmentalgroups andnearbyresidents.From1942to1971thereweretwentyrecordeddamfailuresthatsent claysspillingintonearbyriversandstreams. 18 Thishasledtotheneedtoconsolidatetheclays morequicklyandavoidbuildingmoreCSAs.However,theclaysettlingareashavebeenknown toservethreemainfunctions.Theseincludetherecyclingofwatertothebeneciationplant, useasareservoirsystemtoconservegroundwater,andultimateconsolidationofwasteclays. 24 Typically,thediluteclayslurryat2to6%solidsenterstheclaysettlingareaatoneendofthe pondwhileclearwaterformsattheotherendofthepondasillustratedinFigure 2-5.Thisis furtherdemonstratedinlivephotographsofclaysettlingareasinCentralFloridapresentedin Figures 2-6 and 2-7.Theseparationpicturedaidsinrecyclingwatertothebeneciationplant. Intheearlytomid1970s,thestateofFloridabeganenforcingregulationsthatminedlands bereclaimed. 18, 22, 24 Generally,minedlandscanbereclaimedtoformhousingdevelopments, parks,orangegroves,andgolfcourses.However,claysettlingareasarenotactuallyreclaimed. Thesettlingareasareusuallyleasedforcattlegrazingoncetheyreachanacceptablesolids content(i.e.,40%solids).Theabovegroundstructureshavepooraestheticfeaturesaccording tonearbyresidents.Therefore,thegoalistoavoidbuildingmoreclaysettlingareasinorderto allowmoreoftheminedlandstobereclaimed. 26

PAGE 27

A B Figure2-6.Photographsofclaysettlingareaslookingdownfromraisedembankment;A)one endofthepondwhereclearwaterhassurfacedtothetop;andB)theotherendof thepondwhereclayhasnotsettledaswell(photographsbyMarkOrazem,usedwith permission). 27

PAGE 28

A B Figure2-7.Photographsofclaysettlingareasshowingtheinteriorslopedembankment;A)one cornerofthepondwhereclearwaterhassurfacedtothetop;andB)anothercorner ofthepondwhereclayhasnotsettledaswell(photographsbyMarkOrazem,used withpermission). 28

PAGE 29

CHAPTER3 STUDYOFCLAYS Phosphaticclaysconsistofapproximatelyone-thirdphosphate,one-thirdsand,andonethirdclay.However,theclaysfractiontendstodominatetheoveralldewateringbehaviorofthe slurry.Thisisduetotheclaysrepresentingtheneparticlefractionwhichcausepoorsettling characteristicsandwhenconsolidated,havepoorpermeabilityproperties.However,havinga nerparticlesizealsoallowsforapotentiallyfavorableresponsetotheapplicationofanelectric eld. Thepoorsettlingcharacteristicsarepartiallyduetosomeclaymineralsbeinglayered whichmeanstheirinternalsurfaceareaishugewhilealsohavingaverysmallapparentdensity. Metallurgistrefertophosphaticclaysasslimeswhicharelooselydenedasmaterialthatis toonetodewaterbybeneciationmethods. 21 Typically,98%oftheparticlescontainedinthe phosphaticclaysuspensionsarelessthan100micronsindiameter.Theaveragespecicgravity ofthedryparticlesofthephosphaticclaysis2.7witharangebetween2.6to2.9.Theyconsist ofclaymineralssuchassmectite(ormontmorillonite),palygorskiteandkaoliniteaswellas non-clayssuchascarbonate-uorapatite,quartz,waveliteanddolomite. 3.1ClayPropertiesandStructures Claymineralsareformedbychemicalweatheringofrocksintocrystallineparticlesthat typicallyhaveadiameterof2micronsorless. 3 Forexample,kaoliniteisformedfromwaterand carbondioxidebreakingdownfelspar. 3 Thegeneralshapeofmostclayparticlesisaplate-or at-likeshapeleadingtoalargeratioofsurfaceareatomass. 3 Thishighspecicsurfacearea leadstohinderedsettlingofclays.However,italsocausesthesurfacepropertiesoftheclaysto dominatewhichmaybefavorableforenhancedsedimentationduetoanappliedelectriceld. Mostclaymineralsaremadeofacombinationofsilicaandaluminasheets. 3 Thesilica sheetismadeofsilicatetrahedronsandthealuminasheetismadeofaluminaoctahedrons whicharepresentedinFigure3-1.Differentcombinationsofthesilicaandaluminasheetsform differentclayminerals.Forexample,kaoliniteismadeofastackofsheetsalternatingbetween silicaandalumina.Smectite,ormontmorillonite,iscomposedofanaluminasheetsandwiched inbetweentwosilicasheets.Thesocalled'sandwiches'thenstackconsecutivelytogetherbut withwatermoleculesoccupyingthespacebetweenthem.Theamountofwateradsorbedmay 29

PAGE 30

Figure3-1.Molecularrepresentations(reproducedfromCraig 3 )ofthebasicstructuresofclay minerals.Themolecularstructurescombinetoformthedepictedsheetstructures. increasecausingthesmectiteparticlestoswellwhichincreasestheirspecicsurfaceareaand furtherhinderssettling. 3.2ChargeonClays Typically,clayparticlesimmersedinwaterhaveanegativesurfacecharge.Thenegative surfacechargecanbeduetoseveralfactors.Onecauseforthiscanbeduetopreferential specicadsorptionofanionsatthesurfaceoftheclaysinsteadofcations.Anothercauseis duetoisomorphicsubstitution,orlatticesubstitution, 35 whichoccurswhenasilicaoraluminium atom(showninFigure 3-1)isreplacedbyanatomoflowervalency.Inotherwords,thepositive chargeonthecentralatominatetrahedron,forexample,isreducedleadingtoaresidual negativecharge.Anothercauseofthenegativechargeisbondbreakingatthesurfaceofthe clayparticles.Forexample,theoctahedroncontainshydroxylgroupswhichcandissociate leavingnegativelychargedsurfacegroups.Thenegativechargeontheclaysarenaturally balancedbypositivechargesorcationsintheadjacentwatersuchthattheoverallclay-water interfaceisneutral.Thesecationspredominateinaregionapproximately1to10nanometers fromthesurfaceoftheclay.Thisoverallinterfaceistermedasthedoublelayerwhichrepresents theclay-water(orsolid-solution)interfacewithanetnegativechargeonthesolidsideandanet 30

PAGE 31

positivechargeonthesolutionside.Beyondapproximately10nanometersfromtheclay-water interfacethewaterbecomesneutralastheconcentrationofcationsdecreaseswithincreasing distancefromtheinterface.Thisinterfacecanbemostsimplymodelledasacapacitor. Thecharacteristicsofthedoublelayerdeterminewhetheragivenparticlewillrespond favorablytoanelectriceld.Themagnitudeofchargeisusuallyexpressedinsteadintermsof thezetapotential.Thecharge q isdirectlyproportionaltothezetapotential as q = )Tj /T1_1 10.909 Tf 9.68 7.38 Td ( (3) where isthepermittivityand istheDebyelengthwhichrepresentsthethicknessofthediffuse partofthedoublelayer.Theelectrochemistryofthisrelationshipwillbediscussedinmore detailinSection4.3.Whilethechargeontheclayparticlesaidssedimentationwhenanelectric eldisapplied,thereisacounterforceastheinteractionoflikechargesbetweenclayparticles enhancesdispersion.Additionally,thereareshort-rangeattractionsbetweenparticlesdueto vanderWaalsforceswhichdecreasewithincreasingdistancebetweenparticles. 3 However, whennoelectriceldisapplied,therepulsionduetothelikechargesontheparticlesaswellas theirlargespecicsurfaceareadominateinkeepingtheclayssuspendedinwaterandhindering settling. 3.3TypesofClays Threetypesofclaysareknowntooccurwithinphosphaticclaysuspensions.Theseare knownassmectite,kaolinite,andpalygorskite.Thefollowingsectionsdiscusseachofthese claysingreaterdetail. 3.3.1SmectiteClay Smectiteclayisoftenreferredtoasmontmorillonite.Itincludesacombinationofsilica tetrahedral(SiO 4 )sheetsandaluminaoctahedral(Al(OH) 6 )sheets. 3,3638 Swellingofsmectites easilyoccursaswaterisadsorbedbetweenthestackedarrangementsofsilicaandalumina sheets.Thebondingbetweenthestackedsheetsisweakleadingtosmallerparticlesizes. 3,39 Magnesiumandironareknowntosubstituteforaluminiuminthealuminaoctahedronsina processknownasisomorphicsubstitution.Aluminiummayalsosubstituteforsilicainthesilica tetrahedrons.Smectiteshaveaveryhighdegreeofisomorphicsubstitutionincomparisonto otherclayminerals. 40 31

PAGE 32

3.3.2KaoliniteClay Kaoliniteclayismadeupofalternatingconsecutivesheetsofaluminaandsilica.An averagekaoliniteparticlecouldconsistofoveronehundredstacksofcombinedaluminaand silicasheetsleadingtolargerparticlesizesthansmectites. 3 Kaoliniteparticleshaveaplatelikeshapewithanegativesurfacechargeonitsfacesandapositivesurfacechargeonits edges. 4143 Kaolinitedoesnotadsorbwatereasilyanditsstabilityisduetohydrogenbonding betweensilicaandaluminasheets. 37 Thisincreasedstabilityalsolimitsisomorphicsubstitution asitdoesnotoccurasreadilyinkaolinite. 3 3.3.3PalygorskiteClay Palygorskiteisaclaymineralwhosemake-upismorecomplexthansmectitesorkaolinites. Palygorskiteisnotsimplycomposedofstacksofdifferentstructuralcombinationsofsilicaand aluminasheets.Insteaditismadeofchain-lengthlikestructuresleadingtolongerandthinner particles.Thepalygorskiteclayparticlesaredescribedasrod-shapedparticlescomposedof length-wiseberswhichvaryfrom0.01to5micronsinlength. 44 Grimhasclassiedpalygorskite asachain-structuretypeofclaymineralwithhornblende-likechainsofsilicatetrahedronslinked togetherbyoctahedralgroupsofoxygensandhydroxylscontainingaluminumandmagnesium atoms. 39 Thechargeonpalygorskiteismuchsmallerthanotherclaysandithasahigher surfaceareatomassratio.Itsminimalchargecausesittonotrespondfavorablytoanelectric eldanditshighsurfaceareatomassratiodoesnotallowittosettlewellnaturallyeither. Anothercommonlyusednameusedforpalygorskiteisattapulgite. 32

PAGE 33

CHAPTER4 ELECTROCHEMICALENGINEERINGANDELECTROKINETICTHEORY Electrochemicalengineeringisintroducedasadisciplineneededtomeetthegrowingenergydemandsoftheworld.Thisdiscussionincludestheintroductionofdifferentelectrochemical techniquessuchaspolarizationcurvesandimpedancespectroscopy.Electrokinetictheoryis introduced,whichisasubsetofelectrochemicalengineering.Thefundamentalpropertiesthat driveelectrokineticphenomenaarediscussedinthischaptertoprovideaframeworkforhow separationsofclaysuspensionsoccur. 4.1ElectrochemicalEngineeringTechniques ElectrochemicalEngineeringisadisciplinewithgreatpotentialfordemandduetothegrowingenergyneedsoftheearth.Electrochemicalsystemscanbeorganizedintotwocategories. Therstisasystemwhichusesspontaneouselectrochemicalreactionstoproduceelectrical energy.Theseareoftenreferredtoasgalvaniccells.Examplesofthisarebatteriesandfuel cells.Theothercategoryofelectrochemicalsystemsarethosewhichneedenergytodriveelectrochemicalreactionstoproduceaspecicproduct.Thisworkinvolvestheuseofanelectrolytic cell.Anotherapplicationinvolvingelectrolyticcellsisthechlor-alkiliindustry,whichisintroduced byPrentice, 45 yetdescribedinmuchgreaterdetailbyPletcherandWalsh. 46 Thisprocessuses electricalenergytodriveseveraldifferentelectrochemicalreactionstoproducechlorine,hydrogen,andsodiumhydroxide. 46 Theadvantageofelectrochemicalengineeringisthatitcanbe usedtoreducetheenergyrequirementneededtodrivethesereactionsforagivenamountof productproduced.ElectrochemicalImpedanceSpectroscopy(EIS)isapowerfultoolthatcanbe usedtoaidintheefforttoreducetheenergyrequirementneededtodrivethesereactions.The contributionsofEISarehighlightedbyitsabilitytodeterminedifferentelectrochemicalkinetic parameters. 47 EISandpolarizationcurvestypicallycanaidinthecharacterizationofelectrochemicalexperimentsandmayhelptodescribethereactionsorreactionmechanismsoccurring. DetailedusesofEISanditsmathematicalfoundationaregivenbyOrazemandTribollet. 48 Other excellentsourcesontheconceptsofelectrochemistryaregivenbyBardandFaulkner, 49 and BockrisandReddy. 50,51 However,Newmangivesthemostextensivedetailonthemathematics ofelectrochemicalengineering. 52 33

PAGE 34

4.2MethodstoOptimizeElectrochemicalCells Polarizationcurvesandelectrochemicalimpedancespectroscopy(EIS)aretwomeasurementtoolsthatcanbeusedtoobtainmoreinformationaboutagivenelectrochemicalsystem. Polarizationcurvescanprovideopen-circuitpotentials,whichrepresentthepotentialwhere thereisnonetcurrentowinthecell.Polarizationcurvescanalsobeusedtoidentifythemasstransfer-limitedcurrentforagivensystem,whichrepresentsthemaximumcurrentobtainable underagivensetofoperatingconditions.EISprovidesthemagnitudeofanelectrochemical cell'sresistanceandallowsidenticationofthedominatingresistancesinthecell.Bytaking impedancedataatdifferentpointsalongthepolarizationcurve,therolesofdifferentkinetic, Ohmic,andmass-transferphenomenacanbedistinguished. Thevoltagelossesforagivenelectrochemicalcellcanbeinvestigatedinordertocreatean optimalcelldesign.Theselossescanbeduetokinetic,Ohmic,andmass-transferlimitations thatareknowntooccurinelectrochemicalcells.TheycanbeidentiedthroughuseofEIS. Oncethelimitationsinanelectrochemicalcellaredeterminedfromimpedancedata,different methodscanbeusedtoreducetheoperatingvoltageofthecell.Forexample,thekinetic limitationsareassociatedwiththerateofthereactionthatoccursonanelectrode'ssurface. Thiscanbeimprovedbymethodsaddingimprovedcatalysts,increasingelectrodesize,and increasingoperatingtemperatures.Ohmiclimitationsarebasedontheelectrolyte'sresistance totheowofcurrentthroughsolution.Anexampleofamethodtodecreasethisresistanceisto increasetheconcentrationofthesupportingelectrolyte.Thesupportingelectrolyteisadissolved substancesuchasasaltthatwillreducetheresistanceofthesolutionwithoutparticipatingin electrochemicalreactionsasitremainsinert.Mass-transferlimitationsinvolvethedifcultyin deliveringreactantstoanelectrode'ssurface.Forexample,atthemass-transfer-limitedcurrent, theconcentrationofthelimitingreactingspeciesattheelectrode'ssurfaceisequaltozero.Two methodstoimprovethislimitationaretostirthesolutionandtoincreasetheconcentrationofthe reactingspeciesinthesolution. Theoverallimpedanceforagivencellcanbemeasuredbyusingtheanodeasapseudoreferenceelectrode.Inthiscase,noreferenceelectrodeswouldbeusedfortheimpedance measurement.However,useofreferenceelectrodesareoftenadvantageous,especiallyinmembranecells.AdditionalEISmeasurementsinvolvingreferenceelectrodescanbeperformedto 34

PAGE 35

identifythethedominantresistancespresentinanelectrochemicalcellbyisolatingtheircontributions.Suchmeasurementsprovidethebasisforidentifyingissuesoccurringinelectrochemical cells. 4.3PrinciplesofElectrokinetics Electrokineticphenomenaarisefromtherelationshipandinteractionbetweentheelectric eldandthediffusedoublelayeratasolid-solutioninterface. 53 Theexistenceofthediffusedoublelayerinthepresenceofanelectriceldleadstoahydrodynamicow. 5,54 Thisrepresentsthe drivingforceforelectrophoresisandelectro-osmosis,whichinvolveenhancedsedimentationand consolidation,respectively.Inordertodiscussthekeyprincipleswhichdescribeelectrokinetic phenomenaphysicallyandmathematically,itisnecessarytoreviewrsttheconceptsofthe electricalordiffusedoublelayer. 4.3.1ElectricalDoubleLayer Adoublelayernaturallyoccurswhenasolution(orelectrolyte)isincontactwithasolid. Asionsarepresentinsolution,somehaveagreaterafnityforadsorptiontothesolid'ssurface thanothers.Generally,anionsprefertobeclosertothesolidsurfacethancations.Therefore, aseparationofchargearisesasanionsadsorbtothesurfaceformingthecompactpartofthe doublelayer.Theionsthatarespecicallyadsorbeddonotmovefreelybecausetheyarebound tothesurface.Inordertobalancethenegativechargeadsorbedtothesurface,adiffuseregion ofchargeinthesolutionadjacenttotheadsorbedionsforms,andthisistermedthediffusepart ofthedoublelayerorsimplyasthediffuselayer. 55 Thediffuselayerisnotelectricallyneutralas, inthiscase,itishasahigherconcentrationofcationsthananions.Thecationsinthediffuse layerarebalancedbetweentheelectrostaticforcepullingcationstowardtheoppositelycharged surfaceandtheirtendencytodiffuseawayfromthesurfaceduetotheconcentrationgradient fromthediffuselayerintothebulksolution. 56 Inthediffuselayertheionsareablefreelymove astheyarenotboundtothesurface.Thethicknessofthediffuselayeristypicallygivenas1 to10nanometers; 5 althoughanothersourcehasreportedthethicknesstobeaslargeas100 nanometers. 57 ThephysicaldescriptionoftheelectricaldoublelayerasdescribedinthissectionisillustratedinFigure4-1.Beyondthediffusedoublelayer,thesolutioniselectricallyneutral.The 35

PAGE 36

lengthorthicknessofthediffusedoublelayerisgivenbytheDebyelength, = v u u u t RT F 2 n P i=1 z 2 i c i;1 (4) where R isthegasconstant(8.314J/molK), T isthetemperature(indegreesKelvin), z i is thechargeonionicspecies i, F isFaraday'sconstant(96,487C/equiv)and c i;1 isthebulk concentrationofionicspecies i .Thebulkconcentrationrepresentstheconcentrationinsolution awayfromtheinterfacewherethesolutioniselectricallyneutralandnoconcentrationgradients arepresent.TheinnerHelmholtzplane(IHP)andtheouterHelmholtzplane(OHP)areboth indicatedinFigure 4-1.TheIHP(alsoreferredtoastheSternsurface 58 )representstheplane intersectingthecentersofionsspecicallyadsorbedtothesolidsurface.TheOHPrepresents theplaneofclosestapproachtothesolidsurfaceforcentersoffreelymovingions. Ifthesoliditselfbecomescharged,thenaredistributionofchargecanoccurinthedouble layer.Theentireinterfacemustbeelectricallyneutral;therefore,thechargeonthesolid q 1 and thetotalchargeoftheallspecicallyadsorbedions q 2 mustbalancethechargeinthediffuse partofthedoublelayer q 3 as q 1 +q 2 +q 3 =0 (4) Anothertreatmentofthetermscaninvolvecombiningthechargeonthesolid q 1 withthetotal chargeofthespecicallyadsorbedions q 2 as q 1 +q 2 =q (4) Thisisadvantageouswhenconsideringparticlesincolloidalsuspensions.Asaparticlemoves, itcarrieswithitthechargeofthespecicallyadsorbedionswhichareboundtoitssurface. Therefore,thechargecanbecombinedasgivenbyequation(4). Asthechargeonthesolidsurfacearises,aredistributionofchargeinthediffusedouble layerwilloccurtobalancethetotalchargeoftheinterfacegivenby q+q 3 =0 (4) Changesinthebulkconcentrationwillalsoaffectthedistributionofchargewithinthediffuse doublelayer.Equation(4-1)indicatesthatasthebulkconcentration c i;1 increases,thethickness 36

PAGE 37

Figure4-1.SchematicofelectricaldoublelayerindicatingtheinnerHelmholtzplane(IHP),the outerHelmholtzplane(IHP),andthethicknessofthediffusepartofthedoublelayer .Thepositivecirclesindicatecationsandthenegativecirclesindicateanions. 37

PAGE 38

Figure4-2.Schematicillustratingthedirectionoftheelectriceld,thediffusedoublelayer,and thevelocityprolethatarises. ofdoublelayer decreases.Thisaffectsthechargeinthediffusepartofthedoublelayeras q 3 = )Tj /T1_1 10.909 Tf 9.68 7.38 Td ( (4) Bychangingthechargeinthediffusedoublelayer,thechargeonthesolid q mustalsochange forequation(4)tohold. 4.3.2ApplicationofElectricField Whenanelectriceldisapplied,aforceactsonthefreelymovingionsinthediffusepart ofthedoublelayer.Cationsrespondtomoveinthedirectionoftheelectriceldwhileanions moveinthedirectionoppositeoftheelectriceld.Ifthediffuselayerispositivelycharged(i.e., containsmorecationsthananions),thenthesolutioninthediffusedoublelayerwillbegin movinginthedirectionoftheelectriceld. 59 Ifthesolidistreatedasaatsurface,thenthe directionoftheelectriceldcanbedescribedastangenttothesurfaceaspresentedinFigure 4-2.ApartialdifferentialequationisdevelopedbycouplingtheNavier-Stokesequationwith Poisson'sequationwhichrelatesthevelocityofthesolution v withthepotentialinsolution b and theelectriceld E x as @v @y = @ b @y E x (4) where representstheviscosityofthesolutionand y isthepositionintothesolutionwhichis normaltothesolidsurface. 5 Thepotentialandvelocitygradientsareformedduetothepresence 38

PAGE 39

ofaboundarylayeratthesolid-solutioninterface.TheboundaryconditionsattheIHPare denedas v =0and b =b 0 .Theboundaryconditionsoutsidethediffusedoublelayerand towardsthebulksolutionaredenedas v = v x and b = b 1 .Fromintegrationandsolvingforthe constantsusingtheboundaryconditions,equation(4)issimpliedto ()Tj /T1_1 10.909 Tf (v x )= (b 0 )Tj /T1_2 10.909 Tf 10.909 0 Td (b 1 )E x (4) Thezetapotential isdenedasthepotentialdropthroughthediffusepartofthedoublelayer as =b 0 )Tj /T1_2 10.909 Tf 10.909 0 Td (b1 (4) Bysubstitutingequation(4)intoequation(4),thevelocityofwater v x justoutsidethediffuse doublelayercanbeobtainedas v x = )Tj /T1_1 10.909 Tf 9.681 7.38 Td [(E x (4) whichyieldsanexpressionrelatingthevelocityofthewaterasafunctionoftheelectriceld.The velocity v x canalsobedenedintermsofthechargeinthediffuselayer q 3 orthechargeonthe particle q bysubstitutingequation(4)intoequation(4)togive v x = qE x (4) Notethat v x representsthevelocityofsolutionwithrespecttotheparticleorsolidsurface. Furtherinformationontheprinciplesofelectrokineticscanbefoundintextbooks.Newman andThomas-Alyea 5 giveanextensivesectiononelectrokineticprincipleswhichincludes themathematicalderivationofkeyphenomenasuchastherelationshipoftheelectriceld withtheelectricdoublelayer.HiemenzandRajagopalan 58 alsoincludeadetailedsectionon electrokineticphenomenaandonthestrongroleofthedoublelayer.OtherrelatedtextsbyVan Olphen 56 andKruyt 60 primarilyfocusoncolloidalsystemsyethavekeyinformationpertainingto doublelayertheoryanditspertinencetoelectrokinetics. 39

PAGE 40

CHAPTER5 LITERATUREREVIEW Areviewoftheliteratureispresentedrangingfromsmall-scaletolarge-scaleapplicationsof electrokineticdewatering.Theapplicationspresentedareforremovalofwaterfromphosphatic claysaswellasothermaterials.Othermethodsforwaterremovalarealsopresented.Anoverall assessmentoftheaccomplishmentswithintheliteratureisdiscussed.Thisassessmentalso introducesitemsthatareabsentintheliteraturewhichmotivatethecontributionsforthiswork. 5.1ElectrokineticDewatering Numerousdewateringdesignsarepresentedintheliterature.Thefollowingsectionsinclude descriptionsofelectrokineticbench-topcellsaswellaslarge-scaleapplications.Althoughthe focusisonthedewateringofclays,muchcanbelearnedfromtheelectrokineticdewatering ofothermaterials.Someexamplesofelectrokineticdewateringapplicationsotherthanclays includewaterremovalofoilysludges, 13 harbordredges, 61 contaminatedriversediment, 62 waste sludge, 63 andactivatedsludge. 64 5.1.1Small-scaleBench-topCells Severalbench-topcellsarepresentedintheliterature.Alargeportionofthebench-topcells studiedarecylindricalinshape,verticallyoriented,andmadeofPlexiglasoracrylic.Buckland etal. 62 developedaPlexiglascylindricalcellforapplicationofanelectriceldtoremovewater fromcontaminatedriversediment.Thecellhadaninnerdiameterof20cmandaheightof100 cm.Electrodeswereplacedparalleltoeachotherwiththecathodeatthetopandtheanodeat thebottom.Theparallelarrangementoftheelectrodeswasusedtoensureauniformelectric eldalongthelengthofthecolumn.Referenceelectrodeportswereinstalledalongthelength ofcelltomeasuretheelectriceld.Samplingportswerealsoinstalledwhichtappedintothecell suchthatthesolidscontentcouldbedeterminedatvariouslocationswhiletheexperimentwas stilloperating.AsimilarlydesignedcellwaspresentedbyShangandLo, 17 Lo, 65 andShang,. 14 Thisbench-topcellhadthesamefeaturesasthecellreportedbyBuckland etal. 62 exceptthat itwassmaller,witha9cmdiameteranda20cmheight,itdidnothavesamplingports,andit wasdesignedspecicallyforwaterremovalfromphosphaticclaysuspensions.Shang 17 reported anincreaseinsolidsweightpercent(wt%)ofphosphateclayfrom12.7to29wt%injust30 40

PAGE 41

hours.Shang 17 reportedadewateringenergyrequirementof7.6kW-h/percubicmeterofclay suspension. HoandChen 66 alsodevelopedaverticallyorientedcylindricalcellwhichwasusedfor removalofwaterfromclaysuspensions.Thiscellfeaturedarotationalanodeplacedatthetop ofthecell.Byincreasingtherotationsperminute(RPMs)therateofdewateringincreasedas thehighernumberofrotationsforcedthecakedclaystofalloffoftheanode.Disadvantages ofthisdesignwerethattheelectrodegapcouldnotbeadjustedandtherotatinganodecould notoperateproperlyifthesolidsweightpercentbecametoolarge.HoandChen 66 reportedan increaseinsolidscontentfrom9.1to24.6wt%injust4hours. Reddy etal. 61 alsoproducedaverticallyorientedacryliccell.Thiscellincludedaweighton thetopelectrodetomaintainelectricalcontactwiththesedimentbeingdewatered.Theuseof theweightwasalsodesignedtoensureuniformconsolidation.Reddyetal. 61 reportedarangeof dewateringenergyrequirementsbetween280to1022W-h/percubicmeterofsedimenttreated. Jin etal. 67 developedacellhorizontallyorientedtoremovewaterfromafermentation broth.Thiscellwasrectangularinshape.Oneadvantageofthistypeofcellisthatitallowedfor efcientventingofthegasesgeneratedattheelectrodes.OtherdesignsbyMaini etal. 68 and Paczkowska 15 alsousedhorizontallyorientedbench-topcellsforelectrokineticremovalofwater. 5.1.2Large-scaleFieldApplications Severaloutdoortestshavebeenperformedtoremovewaterthroughapplicationofan electriceldonwasteclaysorminetailings.Someoftheseimplementationsinvolvedbuilding abovegroundstoragetankstoperformtestsandotherseitherbuiltexperimentalclayponds orappliedelectrokineticinstrumentationtopre-existingclayponds.Fourie etal. 4, 12 builtand operatedanabovegroundtanktodewaterminetailings.Thedesigninvolvedverticallyoriented electrodesinsertedintothetailings.Suchanarrangementofelectrodesisexpectedtobemore feasiblebecauseitmaybedifcultorimpossibletoinserthorizontallyorientedelectrodesat thebottomofanalreadyexistingclaysettlingarea(CSA).Verticallyorientedelectrodescould bemucheasiertoinstallastheycanbethrust(i.e.,asastake)downwardandimplantedinto suchsettlingponds.FromtherawnumbersreportedbyFourie etal., 4 adewateringenergy requirementof1.25W-h/kgofwaterremovedwasdetermined.Thisenergyrequirementwas baseduponanincreaseinsolidscontentfrom39to57wt%. 41

PAGE 42

McLean 69 reportedalarge-scaleclaydewateringexperimentwhichwasperformedon settlingpondsspecicallydesignedandbuiltforsuchtests.Thisinvolvedtheuseofdraglinesto preparetrencheswhichweresubsequentlylledwithdiluteclaysuspensions.Suchexperimental preparationwouldbebenecialforinstallationofhorizontallyorientedelectrodesatthebottom ofthepitsbeforetheyarelledwithsuspensions(orslurries).Large-scaledesignsreportedin theliteraturemaybebenecialwhenconsideringissuesthatmayarisefromanyscale-updesign basedonbench-topresults. 5.2OtherDewateringMethods TheTennesseeValleyAuthorityinvestigatednumerousmethodstoenhancewaterremoval ofslurriesthatfeedoralreadyexistinclaysettlingareas. 21 Techniquesthatwerelistedinclude occulants,ltration,hightemperaturedrying,electrophoresis,centrifugalapplications,magneticallyinducedcurrent,controlledstirringandothers.TheFloridaInstituteofPhosphateResearch (FIPR)haspublishednumerousdewateringmethodsthatdonotincludeelectrokineticeldapplications.Pittmannreportedandsummarizedseveralmethodssuchasafreeze-thawtechnique, useofmovablescreens,crustdevelopment,andasand-claysandwhichprocess. 25 5.2.1Flocculation Flocculationcanbedenedinvariouswaysasitinvolvesseveraldifferenttypesofmechanisms.Inthemostgeneralsense,occulationisdenedastheformationofocsorakesthat occurascolloidalparticlescomeoutofsuspension.Inphosphaticclaysuspensions,occulants areusedtoincreasethesettlingratewhichaidstherecyclingofwaterbacktothephosphate beneciationplant. 70 Thedilutesuspensionfeddirectlyfromthebeneciationplantistreated withocculantwhichincreasesthesolidscontentfrom2to10weightpercentwithinminutes, possiblyevenseconds. Theocculantsaremadeofpolymericmaterials. 71 Thepolymerusedischargedand termedaseitheranionicorcationic.Itsassociatedchargeattractsandattachestosuspended clayparticlesformingcondensedbundlesoftheparticles.Someocculantsreportedinthe literatureincludepolyethyleneoxide 72 andpolyacrylamide. 1,70 Polyacrylamidehasworkedwith greatsuccess,butisnotwidelyusedduetoitshighexpense.Althoughmostocculants(i.e., polyacrylamide)usedareorganic,inorganicocculantsalsoexist. 1 Inorganicocculantsactually 42

PAGE 43

hadalargersalesvolumeinthe1970sbutwereovertakenshortlyafteroncesyntheticorganic polymersarrivedonthemarket. 1 Rahman 70 describestheocculationmechanismintwostageswiththerststageinvolving theadsorptionofthepolymerreagentontoasingleclayparticle'ssurface.Thesecondstageinvolvestheaggregateocsthataresubsequentlyformed.Thepolymermoleculesaredescribed aslongstring-likechainswhichabsorbtoseverallocationsonanindividualparticle.Slackexists inthestring-likeocculantasitattachesfromlocationtoanotherontheparticle'ssurface.Loops thenformwhichextendoutwardfromthesurface.Theseloopscollideandattachwithotherclay particlesformingabridgefromoneparticletoanother. 70 Asthecollisionscontinuetooccurthe aggregateocsform.Astheaggregatesform,thespecicsurfaceareaoftheclaysdecrease andgravitationalforcesdominateoveranyrepellingsurfacechargesoftheclayswhichpromotes theclaystocommencesettling. 1 Anothermechanismwhichcanbetermedasocculationinvolvescompressingthesize ofthedouble-layerthatexistsattheclay-solutioninterface.Thepresenceofthedouble-layer onallclayparticlesleadstoarepulsiveforceduetothelikechargesontheclayparticles. Byincreasingtheconcentrationofionsintheelectrolyte,thethicknessofthedoublelayer decreasesasdescribedbyequation(4).Thecompressionofthedoublelayerleadsto areductionintherepulsiveforcesbetweenparticles. 1 Thisallowstheparticlestobeginto aggregate.Othermechanismsalsoexistwhichinvolvedifferentadsorptiontechniquesspecicto certaintypesofocculants(orpolymers). 5.2.2Large-scaleFieldApplication TheFloridaInstituteofPhosphateResearch(FIPR)haspublishedalargenumberofreports onthedewateringofclaysthatdonotincludetheuseofelectrokinetics.Carrier 73 reporteda eldtestincludingtheuseofaplastic,prefabricatedgeodrainsystemthatwasinstalledona pre-existingclaysettlingareatodecreasetheoverallreclamationperiodbyoneortwoyears. Thegeodraincoverwasplacedsuchthatitlaidonanembankmentthatslopeddownintothe CSA.AstheCSAwaslled,theslurrywouldcontactthegeodraincover,whichadsorbedwater fromtheslurrydecreasingthesolidscontentofthatenteringtheCSA.Small-scaletestshad previouslybeenperformedwhichindicatedthatthegeodrainmaterialwouldbeeffectiveona large-scale.Theadsorbedwaterwasreleasedfromthegeodrainbyapipelineconnectedtothe 43

PAGE 44

geodrainmaterial.Thistestultimatelyfailedbecausethepipelinewasnotinstalledproperlyas itwasincapableofwithstandingthechangesinheightleveloftheCSAovertime.However,the shearstrengthoftheclaysnearthegeodrainwas30%higherthanatareaswherethegeodrain wasnotused.Therefore,therecommendationofthisprojectwastorepeattheprojectwithan improvedinstallationofthedrainagepipes. 5.3AdditionalElectrokineticApplication Cellularparticleshavebeenpreviouslytreatedwithelectrokineticphenomenainorder toseparate,manipulate,oranalyzesuchparticles. 74 Typically,suchcellularparticlesareone micronindiameterorlarger.However,Hughes 74 hasproposedapplyingACelectrokineticsto particlesonthenanometer-scale.Thistypeofstudyinvolvesdielectrophoresis,whichisthe applicationofanonuniformelectriceldwhichinducesadipoleonparticlesthataresubjectto theelectriceld. 14,17,75,76 Thenonuniformelectriceldcoincideswithanalternatingcurrent (AC).Itwasinitiallybelievedthatparticlesonthenano-meterscalewouldbecontrolledbyBrownianmotioninsteadofthedielectrophoreticforce. 74 However,microfabricatedelectrodeswere usedwhichprovidedcontroloversuchsmallparticles. 77 Advancementofsuchanapplication couldhavelargeeffectsinexpandingordevelopingnanotechnologyinitsattemptstobecome mainstream. 5.4ElectrokineticParameters Experimentalissuesreportedwithintheliteraturewhichinvolveapplicationofanelectric eldtoremovewaterfromwasteclayorothertypesofsuspensionsarediscussedhere. Theobjectivesofthisworkcenterontheapplicationofanelectriceldtodewaterclay suspensions.However,applicationofanelectriceldtoremovewaterfromothermaterials isalsorelevantintermsofstudyingelectrokineticphenomena.Severalissuessuchasthe selectionofelectrodes,pHofsuspension,zetapotentialofparticles,particlesize,evolution ofgases,andcurrentpausingcangreatlyeffectthepowerrequirementofanelectrokinetic dewateringprocess.Thereportedenergyrequirementsintheliteraturerangefrom0.6to880 kW-hr/drymetricton. 4 Mostearlystudiesusedvoltagedropsaslargeas70V.Theassociated largecurrentsresultindryingofthesoilaroundtheanode,whichcauseslargeparasiticpotential drops.ThelargecurrentsalsocreatesignicantgradientsinpH,whichcanchangetheclay propertiesandcanincreaseparasiticlosses.IssuesthatariseduetochangesinpHare 44

PAGE 45

discussedinmoredetailinSection5.4.2.TheeffectsofchangesinpHandparasiticpotential dropscanbemitigatedbyreversingpolarityorbypausingthecurrenttoallownaturaldiffusionto relaxtheconcentrationproles.Usingintermittentcurrentorcurrentpausingisdiscussedfurther inSection5.4.3.Operatingatasmallerelectriceldcanalsoreduceissuesassociatedwith corrosionofelectrodesandparasiticvoltagedrops.Asmallerelectriceldshouldalsoreduce thepowerrequiredperunitofclayprocessed. 5.4.1SelectionandDesignofElectrodes Theselectionanddesignofelectrodesisoftendependentonthecelldesignortheobjectivesoftheexperiment.Ifitisdesiredtolterorremovewaterthroughtheelectrodes,then meshorporouselectrodesshouldbebenecial.Theseelectrodeswouldalsobeusefulifthere isconcernwithventilationofgasesevolvedduetoelectrochemicalreactionsoccurringatthe electrodesurfaces.Someelectrodescanbemachinedinorderensurethattheelectrodesare porousenoughsuchthatgasesdonotgettrapped. 62 Iftheelectrodesareorientedvertically, thengasesevolvedshouldnotgettrappedunderneaththeelectrodesastheyoftenmaywith horizontallyorientedelectrodes.Therefore,verticallyorientedelectrodesmaynotneedtobe porousforgasventilation. Corrosionisalsoabigissueindeterminingselectionofelectrodes.Duetothelargevoltagesappliedinearlyelectrokineticstudies,metalelectrodescorrodedandneededreplacement onaregularbasis.Alternativeelectrodessuchasthedimensionallystableanodesusedforcathodicprotectionmayprovideacost-effectivealternative. 6 Fouriehasreportedexcellentresults fordewateringofminetailingsbyusingelectrokineticgeosyntheticelectrodes,whichcomprise achargedporousfeltclothwrappedaroundacarbon-dopedplasticmesh. 4,11,12 Theelectrokineticgeosyntheticelectrodesalsoappeartoresolvetheproblemswithelectrodestabilityor degradation.Graphiteelectrodeshavealsobeenusedtomitigateissuesinvolvingcorrosion. 62 5.4.2pHandZetaPotential ChangesinpHwilloccurduetoelectrochemicalreactionsproducinghydrogenionsatthe anodeandhydroxideionsatthecathodewhicharegeneratedfromreactions(1)and(1). SuchchangesinpHaresignicantduetoitseffectonthechargeofclayparticles.Thelevel ofpHcanalsoaffectthestabilityofelectrodesascorrosionordegradationofelectrodescan occuratlowpH.Oftenintheliterature,insteadofquantifyingthechargeonclaysintermsof 45

PAGE 46

Coulombs,thezetapotentialisusedtocharacterizeclayparticleswhichareusuallyontheorder of10to50milli-Volts(mV).Severalstudiesintheliteraturemeasurethezetapotentialasa functionofthepH.Thezetapotentialiscriticaltoelectrokineticprocessesasgivenbyequation (4). SeveralstudieshaveinvolvedpHandzetapotentialsandhowtheyaffectvariouselectrokineticprocesses.Bjelopavlic etal. 78 reportsthatchangesinpHcanaffectthesurfacecharge ofclayparticlesduetothespecicadsorptionofionsontoparticles.AsthepHincreases,the hydroxylgroupsaddedtothesuspensionareknowntoadsorbtothesurfaceoftheparticles changingtheircharge.Thechargeontheparticle q isdirectlyproportionaltothezetapotential asgivenbyequation(4).OthereffectsofpHchangescaninvolvefurthercontaminationof claysuspensions.Cabrera-Guzman 79 reportsthatatlowpHvalues,hydrogenionscanpromote ionexchangeeffectswhichcanreleasemetalcontaminantsthatwerebondedtotheclayparticles.Thiscouldleadtofurtherissuesintermsofmanagingthehazardsassociatedwithland storageoftheclays. Dixit etal. 80 measuredthemobilityofclayparticlesasafunctionofpH.Theseresultsfound thataspHincreased,themobilityincreased.Thiswasattributedprimarilytodissociationof hydroxylgroupsattheedgesoftheclayparticleswhichisknowntooccuratlargerpHvalues. Thebondbetweentheoxygenatomandthehydrogenatombreaksleavingamorenegative chargeontheclayparticleasthepositivelychargedhydrogeniondiffusesaway.Asthecharge ontheclaybecomesmorenegative,theabsolutevalueofthezetapotentialincreasesleading toimprovedmobilityatlargerpHvalues.Sucharelationshipwasveriedexperimentallyby Gopalakrishnan etal. 81 asthezetapotentialsweremeasuredasafunctionofpHfortwotypes ofclayslurries.TheabsolutevalueofthezetapotentialincreasedwithincreasingpHforboth clays. 5.4.3UseofIntermittentCurrent Intermittentcurrentorvoltage(orcurrent)pausingisoftenemployedinelectrokineticcells torehydratedriedareaswhichcanreduceparasiticvoltagedropsandrelaxpHorconcentration gradients.Reducingareaswhereparasiticvoltagedropsoccurcanaidinreducingtheresistancetotheowofcurrentthroughagivenelectrochemicalcell.ShangandLo 17 implemented 46

PAGE 47

theuseofanintermittentcurrentintervaltodeterminethemostefcienton-offcycleforelectrokineticwaterremovalofclays.Anon-offcycleof15minutesonand5minutesoffwasreported asthemostefcientforremovalofwater.AnotherstudywasperformedbyGopalakrishnan et al. 81 todetermineanoptimalon-offtimeforcurrentpausing.Theirresultsindicatedanoptimal on-offtimeof30secondsonand0.1secondsoff.Thediscrepancybetweentheoptimumratios ofthetwopaperscouldbeduetothedifferenceintimescalesusedforeachratio.Another discrepancycouldbeduetodifferencesintheclaysusedfromonestudytotheother. 5.4.4ParticleSize Particlesizehasbeenreportedtohaveaneffectontherateofsedimentationofclayswhich iscontactwithanappliedelectriceld.Shang's 14 experimentalresultscomparedtwotypesof claysagainsteachotherwhichhaddifferentparticlesize.Oneclaywastermedasthebrown clayandtheotherwastermedasagreyclay.Thebrownclayhadaneraverageparticlesize andachievedbetterdewateringresultsfromtheapplicationoftheelectriceld.However,Shang alsoperformedcontroltestswithgravityonly,wherethegreyclayachievedbettersettlingresults. Therefore,theseresultsindicatedthatnerparticlesizeadverselyaffectsthesedimentation behaviorwhennoelectriceldisappliedbutenhancessedimentationwhenundertheforceof anelectriceld.Reasonsforthiscanbeattributedtosurfacepropertiesoftheparticles.Forner sizeparticles,thesurfacepropertiesareexpectedtodominatewhichallowthemtorespondto anelectriceld.However,whennoelectriceldisapplied,thesesurfacepropertiesmayaidin dispersingtheparticlesduetolikechargesontheclayparticlescausingtheirrepulsionofeach other. 5.5AssessmentoftheLiterature Theliteraturehaspresentedmanyexamplesofelectrokineticdewateringofclaysandother materials.Theseincludesmall-andlarge-scalestudieswhichconsistentlydemonstratethatthe applicationofanelectriceldcanbeusedtoremoveasignicantamountofwaterfromclays andnorothermaterials.However,theresultsofbench-topexperimentspresentedintheliterature havenotbeenusedtodevelopparametersforlarge-scaledesignofawaterremovalsystem. Constitutiverelationshipslinkingtheelectriceldwithoperatingtimeandtheamountofwater removedhavenotyetbeenpresentedwithintheliterature.Suchrelationshipsareneededin combinationwithamathematicalmodeltoprojecttheenergyrequirementsforalarge-scale 47

PAGE 48

waterremovalsystemtoassessitsfeasibilityintermsoflarge-scaletreatmentofclaysettling areas. 48

PAGE 49

CHAPTER6 EXPERIMENTALAPPROACH Ex-situcharacterizationofclaysamplesaswellasthebench-topexperimentalset-upare presentedinthischapter.Theex-situstudiesincludedX-Raydiffraction,particlesizedistribution measurementsandzetapotentialmeasurements.Theex-situmeasurementswereperformed toverifypropertiesoftheclaywhichcouldaffecttheresultsandreproducibilityofthebench-top experiments.Theintroductionofthebench-topexperimentalset-upincludesthedescriptionof theinstrumentationusedandthedesignofthebench-topcell. 6.1Ex-situAnalysisofClaySuspensions Characterizationofclaysuspensionswasdesiredtoverifyitsphysicalpropertieswhich mayaffecttheabilityofanelectriceldtoremovewaterfromtheclays.Thisalsoallowedfor determinationofwhethervariationsexistedbetweenthevariousclaysuspensionsamples providedbyMosaicFertilizerLLC. 6.1.1X-RayDiffraction X-RayDiffraction(XRD)patternsweregeneratedattheMajorAnalyticalInstrumentation Center(MAIC)attheUniversityofFloridausingtheXRDAPD3720instrument.XRDwas performedonphosphaticclaysamplesinordertodeterminetheidentityoftheclayminerals containedwithintheslurries.Ithasbeenreportedthatmineralogicalvariationsdoexistthroughoutclaysettlingareas. 18 Thesemineralogicalvariationscouldcausesignicantchangesinthe electrokineticdewateringresultsonphosphaticclays.Forexample,asamplethathasahigh contentofpalygorskitemaynotrespondwelltoelectrokinetictreatmentduetopalygorskite's smallsurfacecharge. 7,61,66,82 Sedimentationofpalygorskiteisalsodifcultduetoitslowdensityandsmallparticlesize.Bromwellreportedthatastheconcentrationofpalygorskiteinclay slurriesincreases,therateofsettlingdecreases. 18 Whileitislikelythattheclaysamplesfrom Mosaicdonotcontainlargeamountsofpalygorskite,thepossibilityoflargerquantitiespresents theneedforacompositionstudysuchasXRD. X-raydiffractionwasperformedonselectedsamplesprovidedbyMosaic.Somesamples hadbeentreatedwithMosaic'sproprietaryocculantwhileothersampleshadnot.Oneofthe mostcriticalissuesforgeneratingausefulXRDdiffractionpatternistopreparethesample foranalysisproperly.Aproperlydriedformoftheclaysmustbeprepared.Usinganovento 49

PAGE 50

drysamplesrepresentsanimproperdryingtechnique.Inovenornaturaldrying,asthewater evaporates,thesurfacetensionofthewatermoleculespushestheclayparticlestogether.This leavestheclaysamplesinacohesivephasethatcannotbeexaminedproperlybyXRD.The properpreparationmethodiscalledthePeeltechniquewhichinvolvesthepreparationofan orientedaggregatemountoftheclayontoaMilliporelterusingavacuumpump.Thisforcesthe claymineralparticlestolieat,whichenhancestheabilityoftheXRDbeamtoproduceuseful results. AnX-raydiffractionpatterngeneratedfromaocculatedclaysampleispresentedinFigure 6-1withthey-axisrepresentingintensitycountsinarbitraryunitswhilethex-axisistheangleof theX-raydiffractionbeamindegrees.ThepeaksinFigure6-1 wereidentiedusingpublished diffractionpatterndatarepresentingthelocationsforeachmineraltype.Thepresenceof smectite,palygorskite,andquartzwasidentied.Palygorskiteexistedintraceamountswhichis favorableforelectrokineticexperiments. AnX-raydiffractionpatternonasamplethatwasnottreatedwithocculantshowedno signicantdifferencefromFigure 6-1.Infact,severalothersamples(withorwithoutocculant) yieldedsimilarXRDresultsanditwasconcludedthateachsamplecontainednosignicant variationofclayminerals.Therefore,allsamplesprovidedbyMosaicareexpectedtobehave similarlyinthepresenceofanelectriceld.Thisprovidedcondenceinthereproducibilityof theelectrokineticexperimentspresentedwithinthiswork.Methodstodeterminetheidentityof phosphatebearingmineralswerenotperformedastheXRDsamplepreparationexcludedthe largersizedparticlesthattypicallycontainphosphatebearingminerals.Thelargerparticles wereexcludedbecausetheylimittheabilitytoprepareagoodmountandalsoclaymineralsare expectedtoprovidetheprimarylimitationstonaturalsettlingaswellastotheeffectivenessof electrokineticmethods.ForfurtherinformationregardingthestudyofXRDanditsuseonclays, Cullity 83 providesin-depthdescriptionandanalysisonXRDwhileGrim 39 discusseshowXRD canbeusedtostudyandidentifythemineralogyofclaysamples. 6.1.2ParticleSizeDistribution TheinstrumentationusedtodetermineparticlesizedistributionwasprovidedbytheDepartmentofGeologicalSciencesattheUniversityofFlorida.Resultsindicatedthatapproximately 90%ofparticlesfromthephosphaticclaysuspensionswerenerthan60micronsindiameter 50

PAGE 51

Figure6-1.X-raydiffractionpatterngeneratedfromanorientedaggregatemountofaphosphatic claysample.Peaksareidentiedwiththeirassociatedclayorsandmineral. andapproximately65%werenerthan2micronsindiameter.Theseresultsweregenerated usingasedigraphwhichmeasuressettlingvelocities.UsingStoke'sLaw,thesettlingvelocities wereconvertedintocharacteristicdiametersorsizes.Theseresultswerethesameforboth occulatedandunocculatedclaysamples. Particlesizedistributionmeasurementsandcalculationswerealsoperformedusing opticalmethodsintermsoftherealandimaginaryrefractiveindicesofclayparticles.The distributionsweredeterminedintermsofnumberdistributionandvolumedistribution.However, thismethodisonlyaccurateforhomogenoussuspensions,whichdoesnotincludephosphatic clays.Therefore,theseresultswerenotusefulandarenotpresentedinthiswork. 6.1.3ZetaPotentialMeasurements InstrumentationusedtoperformzetapotentialmeasurementswasprovidedbyParticle andEngineeringResearchCenterattheUniversityofFlorida.Theresultsweregenerated bythePALSZetaPotentialAnalyzer.Theinstrumentationexperimentallydeterminedthe averagezetapotentialofsuspensionscontainingaocculantandseparatelydeterminedthe averagezetapotentialforsuspensionsnotcontainingaocculant.Thedifferencebetweenthe averagedzetapotentialbetweenocculatedandunocculatedsampleswasnotsignicant.The 51

PAGE 52

occulatedclaysuspensionhadanaveragezetapotentialof-20.1mVandtheunocculated claysuspensionhadanaveragevalueof-19.9mV.Thisresultindicatesthatbothtypesofclays shouldrespondequallytoanequivalentelectriceldasthevelocityofwaterisafunctionofthe zetapotentialasgiveninequation(4). 6.2ElectrokineticStudies Bench-topexperimentswereperformedtoremovewaterfromclaysuspensions.The instrumentationusedtoperformexperimentsandtheexperimentalapparatususedarepresented.Informationregardingtheclaysamplesusedinexperimentsaredescribedaswellasthe methodsusedforperformingexperiments. 6.2.1ExperimentalInstrumentation Theinstrumentationusedforpotentiostaticorgalvanostaticmodulationinvolvedtheuseofa potentiostat(i.e.,Solartron1286ElectrochemicalInterfaceorEG&GPrincetonAppliedResearch (PAR)Potentiostat/GalvanostatModel273A).Thepotentiostatwasusedtoapplyaconstant potentialoraconstantcurrent.Thepotentiostatalsohadthecapabilityofgeneratingpolarization curves.ThepotentiostatwasoperatedthroughuseofCorrWaresoftware(ScribnerAssociates, Inc.).Inordertogenerateelectrochemicalimpedancespectroscopydata,animpedancefrequencyresponseanalyzer(i.e.,Solartron1250FrequencyResponseAnalyzerortheSolartron SI1260Impedance/Gain-PhaseAnalyzer)worksincombinationwithapotentiostat.Thistypeof measurementinvolvedtheuseofZ-Plotsoftware(ScribnerAssociates,Inc.). 6.2.2Bench-topCell AcylindricalPlexiglascellwasdesignedandconstructedforbench-topexperiments.The advantageofusingacylindricalcellisthatitsgeometryprovidesauniformpotentialdropalong thelengthofthecylinder.Thedimensionsofthecellwere30cminlengthwithaninnerdiameter of9cm.ThecellwasmodiedtoincludeAg/AgClreferenceelectrodessuchthatthevoltage dropthroughtheclayslurrycouldbemeasured.Thisallowedfordeterminationoftheelectric eldforceactingontheslurryasthereferenceelectrodesmeasuredonlytheOhmicpotential drop.Thiswasrequiredbecausethepotentialdifferencebetweenthecathodeandanode (i.e.,cellpotential)doesnotonlyincludethecontributionoftheOhmicpotentialdrop,butalso includesthepotentialdropduetoreactionkineticlimitationsattheelectrodes.Aschematic ofthecellispresentedinFigure6-2.Thedesignwasadaptedfromthedescriptionofthe 52

PAGE 53

experimentalcellpresentedbyShang, 14 ShangandLo, 17 andLo. 65 Allexperimentsbeganwith theelectrodesplaced19cmor20cmapart.Initially,onlytworeferenceelectrodeswereinstalled tomeasurethevoltagegradient.Theywerexedtobe10cmapart.Afteraninitialmatrixof experiments,thecellwasfurthermodiedtoincludeatotaloffourreferenceelectrodes.This allowedforadditionalmeasurementsforsituationswheretheelectrodegapwasreducedandthe originaltworeferenceelectrodesdidnotbothphysicallyremainincontactwiththeconsolidated clay.Suchascenariowouldoccurwhendecantingthesupernatantwateraftertheexperiment hadoperatedforsomelengthoftime. Forallexperiments,thecellwasorientedverticallywiththecathodeplacedatthetopofthe cellandtheanodeplacedatthebottomofthecell.Theelectrodeswereconguredinthisway suchthattheelectriceldwouldworkwithgravityandnotagainstit.Underthisconguration, theelectriceldappliedadownwardforceonthenegativelychargedclayparticlestowardsthe positivelychargedanodewhilesimultaneouslyattractingthewaterupwardstothecathode. Althoughnaturalsettlingduetogravityisnotexpectedtoachievebeyond15weightpercent solidsslurryovershortandintermediatetimes,itwasstilldesiredtohaveanelectriceldthat operatedconsistentlywithgravity.Meshelectrodeswereusedforboththecathodeandanode. Bothweredimensionallystableanodesmadeoftitaniumwitharutheniumoxidecoating.They weremachinedandttoworkwithmovableplungerswithinthecellsuchthattheelectrodegap couldbeshortenedoradjustedassupernatantwaterwasremoved. Allbench-topdewateringexperimentsincludedanon-offcycleof30minutesonand2 minutesoff.Thiscyclewasincludedtoallowconcentrationandhydraulicgradientstorelaxto reducetheresistancetotheowofelectriccurrentthroughthecell.Experimentsoperatedat longertimeswerecontrolledtooperatefor12hoursonandthen12hoursoffeachday.During the12hoursofon-timetheexperimentcontinuedtomaintaintheon-offcycleof30minuteson and2minutesoff.Uponcompletionofeachdewateringexperiment,thecontentswithinthecell wereremovedandimmediatelyweighed.Thetreatedclaysampleswerethenplacedinanoven todrycompletelyandwerethenre-weighedwhendriedtodeterminethesolidsweightpercent uponcompletionofeachexperiment. 53

PAGE 54

Figure6-2.Schematicofbench-topcellwithlabeledlocationsoftheelectrodesandthe temperatureandvoltagemeasurements.Darkershadedareawithinthecell representswhereclayslurryisloadedforexperiments. 6.2.3ExperimentalOperation Oneoftheinitialobjectivesofthebench-topexperimentswastostudythefundamental propertiesoftheclayandwaterandtheirassociatedresponsetoanappliedelectriceld.The small-scaleelectrokineticcellwasnotintendedasaproto-typemodelforlarge-scaledewatering. However,theexperimentswereexpectedtoaidindeterminingthefeasibilityofalargescale designforimplementationinanalreadyexistingclaysettlingarea(CSA).Allexperimentswere performedusingthephosphaticclaysuspensionsamplesprovidedbytheprojectsponsor, MosaicFertilizerLLC.Thesamplesprovidedvariedfrom2to12%solidsbymass(orweight). Samplesthatinitiallyexistedinthe2to4weightpercentsolidsrangehadnotyetbeentreated withaproprietaryocculantwhileothersamplesinitially8to12weightpercentsolidshadbeen treatedwiththeocculant.Additionalelectrokineticexperimentsperformedwithoutocculant showedthattheuseofaocculantdidnotaffecttheelectrokineticresults. Thesolidscontentoftheclaysuspensionswasdeterminedbyweighingtheslurrysample andthenweighingitagainafterthesamplewasdriedinanovensuchthatallmoisturewas removed.15weightpercentsolidswasthehighestcontentachievedthroughnaturalsettlingin 54

PAGE 55

Figure6-3.Schematicillustratingconnectionsanduseofresistorsbetweenpotentiostatand bench-topcell.Resistorsareincludedtoamplifytheappliedpotentialtothecell. combinationwithpre-treatmentofocculant.FromthecurrentthickeningoperationatMosaic, thesolidscontentquicklyreaches8to12weightpercentsolids.Therefore,theexperimentalcell wastypicallyloadedwithasolidsslurrycontentwithinthe8to12weightpercentrange. Initially,allexperimentswereoperatedunderpotentiostaticmodulationandtheelectric currentwasmeasuredasaresponsevariable.However,issuesdevelopedwhenattemptingto operateatsmallcellpotentialsintherangeof1.5to2Volts.Undersuchoperatingconditions theelectrodesdidnotremainpolarizedandtheresultingelectriceldwaseffectivelyequalto zero.Suchbehaviorwaslikelyduetokineticlimitationsdominatingthepotentialdropthecell wasattemptingtoapply.Inordertoremedythisissue,thepotentiostatwasswitchedtooperate undergalvanostaticmodulation.Aconstantelectriccurrentwasappliedwiththecellpotential measuredasaresponsevariable.Theappliedcurrentwasoperatedaslowas2to4milliamps (mA).Thematrixofexperimentspresentedinthisworkwasoperatedineitherpotentiostatic orgalvanostaticmodulation.Therewasnoevidencethatthetypeofmodulationaffectedthe dewateringresultsaslongasitprovidedastableelectriceld. Onevariationbetweengalvanostaticandpotentiostaticmodulationoccurredwhenadjusting thedistancebetweenthecathodeandanode.Ifoperatinginpotentiostaticmode,reducingthe electrodegaprequiredthereductionofthecellpotentialinordertomaintainthesameelectric eld.Forgalvanostaticmode,thisdistancehadnoeffectontheelectriceldaslongasthe conductivityofthesuspensiondidnotchange.Anotherexperimentalvariationforpotentiostatic modulationoccurredwhenattemptingtoapplyacellpotentialbeyond10V.TheCorrWare 55

PAGE 56

softwareprohibitsappliedpotentialsbeyond10V.However,thepotentiostatoperatedincludeda 100Vpowersupply.Aschematicofapotentiostatwithworking,referenceandcounterelectrode leadsispresentedinFigure6-3.Theworkingelectrodewasconnectedtothecathodeandthe counterelectrodewasconnectedtoanode.Thereferenceelectrodelead,forthepurposesof thebench-topexperiments,wasconnectedtotheanodesuchthattheanodeactsasapseudoreferenceelectrode.However,ifthecounterandreferenceleadsarenotshortedtogetherand insteadhavealargeresistorbetweenthemaswellasaresistorbetweentheworkingand reference(asshowninFigure6-3)thenacellpotentialbeyond10Vcanbegenerated.The potentiostatcontrolsthepotentialbetweentheworkingleadandthereferencelead;however,the workingelectrodedrawselectriccurrentfromthecounterelectrode.Largerresistorsareused toavoidasignicantleakagecurrent.Thecellpotentialcouldbemultipliedbyafactorof10if theresistancebetweentheworkingandreferenceleadsis10%oftheresistancebetweenthe workingandcounterleads.Therefore,uptoa100Vcellpotentialcanbereachedbyapplying just10VbetweentheworkingandreferenceundersuchconditionspresentedinFigure6-3.If itisdesiredtodoublethecellpotentialthenbothresistorsinFigure 6-3 wouldneedtobeequal. Forgalvanostaticmodulation,theabsolutecurrentappliediscontrolledbetweenthecathodeand anodeinsuchawaythatthereisnomultipliercapability.However,thisdidnotcauseanyissues undergalvanostaticmodebecausethebench-topexperimentalwaysoperatedwellbelowthe software'smaximumset-pointforelectriccurrent. 56

PAGE 57

CHAPTER7 EXPERIMENTALRESULTSANDDISCUSSION Thedevelopmentandresultsofthebench-topexperimentsarediscussedinthischapter. Thebench-topexperimentsinvolvetheapplicationofanelectriceldtoremovewaterfromclay suspensions.Amatrixofexperimentswereperformedvaryingoperatingtimeandtheelectric eld.Constitutiveequationsweredevelopedandpresentedhere,whichgivethechangeinsolids weightpercentoftheclayasafunctionofoperatingtimeandtheelectriceld.Electrochemical techniquessuchasimpedancespectroscopyandpolarizationcurveswereusedtocharacterize aspectsofthebench-topexperiments. 7.1ProofofConceptwithBench-topCell ExperimentalresultsdemonstratingaproofofconceptarepresentedinFigure7-1foran experimentoperatedfor9hoursatacellpotentialof80V.Figure 7-1A representsthePlexiglas cellloadedwiththeclaysuspensionbeforetheelectriceldisapplied.Thecompositionofthe claysuspensionwasinitially9weightpercent(wt%)solids.Uponcompletionoftheexperiment, theexistenceofathickenedclaymassformedwhichispresentedinFigure 7-1B.Supernatant waterthatformedtowardsthetopofthecellwasremovedbeforethephotographillustratedin Figure 7-1B wastaken.TheremainingthickenedclayinFigure 7-1B hadasolidscontentof33.5 wt%.Intheabsenceofanelectriceld,itnormallytakes25yearsfortheclaysuspensionto reachasolidscontentof40wt%.Fromthisexperimentalone,theapplicationofanelectriceld yieldedasolidscontentapproaching35wt%afterjust9hoursofcelloperation.Therefore,the applicationofanelectriceldisdemonstratedasamethodthatcanenhanceremovalofwater fromclayandaproofofconceptisestablished. 84 Notethatthesolidscontentoftheslurryin unitsofwt%isdenedwithinthisworkas w c = m c m c + m w 100% (7) with m c representingthemassofdryclayand m w representingthemassofwaterpresent. 7.2ConstitutiveRelationshipatShortTimes Amatrixofbench-topexperimentswereperformedvaryingoperatingtimeandtheelectric eld.Arelationshipwasdesiredtopredictthetimerequiredtoremovewaterasafunctionof theelectriceld.Uponcompletionofeachexperiment,thechangeinsolidsweightpercent 57

PAGE 58

A B Figure7-1.Photographsofthebench-topPlexiglasshown:A)Beforeexperimentwasstarted thecellwasloadedinitiallywitha9wt%solidssuspension,andB)afterthe experimentendedthecellwasoccupiedbyathickenedlumpofclayconsistingof 33.5wt%solids. wasmeasured.TheresultsoftheentirematrixofexperimentsarepresentedinFigure 7-2. Noevidentrelationshipwasfoundrelatingthechangeinsolidscontentwithoperatingtime basedupontheresultspresentedinFigure 7-2.However,byscalingthechangeinsolidsweight percentbytheappliedelectriceld(inV/cm),thedatasuperimposetoformalinearrelationship aspresentedinFigure 7-3.AtrendlinewasincludedtotthedatapresentedinFigure 7-3.The equationofthelinewasusedtosuggestarelationshipbetweenthechangeinsolidsweight percent w c andtheelectriceld E (inV/cm)as w c E =0 :72t (7) with t representingtheoperatingtime(inhours).Equation(7)suggeststhatapredictioncan bemadeforthetimerequiredtoachieveacertainsolidsweightpercentatagivenelectriceld. Thisindicatesthatthetimerequiredtoachieveadesiredseparationisinverselyrelatedtothe electriceldapplied.Therefore,thesizeoftheelectriceldcorrespondstoaspecicsolids weightpercentoftheclaymeaningthatanonuniformelectriceldshouldresultinnonuniform waterremoval.Whilethebench-topcellwasdesignedforauniformelectriceld,thisresultcould berelevanttoalarge-scaledewateringsystemwheretheelectriceldwilllikelybenonuniform insomelocations.Therelationshipgivenasequation(7)doesnotapplyforoperatingtimes 58

PAGE 59

Figure7-2.Resultsofbench-topdewateringexperimentsillustratingthechangeinsolidsweight percentasafunctionofoperatingtimewithappliedpotentialasaparameter.The datamarked1.5Vwerecontrolledgalvanostaticallytoyieldaconstantelectriceld. beyondtherangeofvaluespresentedinFigure7-3andisdescribedasavalidrelationshiponly atshorttimes. 7.3UniformityofWaterRemoval ThematrixofexperimentspresentedinSection 7.2 wereanalyzedingreaterdetailto examinethelevelofuniformdewateringwithinthecell.Samplingatvariouslocationswithin thecellwasperformedtodeterminetheextentofdewateringbyposition.Theresultsare presentedinFigures 7-4 7-8 whichgivethesolidsweightpercentasafunctionoftime.Each plotrepresentsaspecicappliedcellpotential.Theextentofdewateringwasfoundtobe non-uniformforallresultspresentedinFigures 7-4 7-8. AsshowninFigures 7-4 7-8,highersolidscontentwasachievedbyoperatingatlarger potentialsandforlongertimes.Foranappliedcellpotentialof10V,thehighestsolidsweight percentwaslocatedneartheanodeandthenexthighestsolidscontentwasadjacenttothe clay-supernatantinterface(zonesCandD)aspresentedinFigure 7-4.Thecenteroftheclay suspensionintermsofheight(zoneB)wasstilltheconsistencyoftheoriginalsuspensionat shorttimes.Forlargerpotentials(i.e.,seeFigure 7-7))themiddleregionbehaveddifferently 59

PAGE 60

Figure7-3.SolidsweightpercentresultsfromFigure7-2 arescaledbytheappliedelectriceld E .Atrendlinettothedatasuggestsalinearrelationshipwhichisgivenas equation(7). suchthatasolidcoredevelopedcontainingthehighestsolidsweightpercent.Thisbehaviorwas previouslypresentedinFigure 7-1B.Thisexperimentillustratedtheeffectsofgasevolutionat theanode.Thelargerateofgasevolutionforcedthesuspensiontobecomedilutearoundthe perimeterofthethickenedclaymasslocatedinthecenter.Additionalexperimentsidentied thatlowercurrentdensitieswhichdecreasedtherateofgasevolutionhelpedalleviatesuch nonuniformbehavior. Amuchlongerexperimentat20voltswasoperatedtodeterminethedegreeofnonuniform behavioroftheclayoverlongertimes.TheresultspresentedinFigure 7-8 indicatethatatlonger timestheclaydewateringbecomesmoreuniform.Physicalobservationsalsofoundthatthe transparencyofthesupernatantwaterwasmuchimprovedwhichisattributedtoallowingthe celltositovernightfor12hourswithoutanyappliedpotential.Thisbehaviorisindicatedin Figure 7-9 (onleft).Infact,thesupernatantwaterfromthisexperimentismuchclearerwhen comparedtothesupernatantwaterformedbyonlygravitationalsettlinginaonelitercylinder picturedinFigure 7-9 (onright).TheexperimentphotographedinFigure 7-9 wasalsovery 60

PAGE 61

Figure7-4.Cellpotentialof10Vwasappliedforthedurationoftheexperiment.Theinset representsaschematicoftheverticallyorientedcellwithzoneDatthetopofthecell representingthelocationofthesupernatantwaterseparatedfromthebulkclay. Figure7-5.Cellpotentialof20Vwasappliedforthedurationoftheexperiment.Theinset representsaschematicoftheverticallyorientedcellwithzoneDatthetopofthecell representingthelocationofthesupernatantwaterseparatedfromthebulkclay. 61

PAGE 62

Figure7-6.Cellpotentialof40Vwasappliedforthedurationoftheexperiment.Theinset representsaschematicoftheverticallyorientedcellwithzoneDatthetopofthecell representingthelocationofthesupernatantwaterseparatedfromthebulkclay. Figure7-7.Cellpotentialof80Vwasappliedforthedurationoftheexperiment.Theinset representsaschematicoftheverticallyorientedcellwithzoneDatthetopofthecell representingthelocationofthesupernatantwaterseparatedfromthebulkclay. 62

PAGE 63

Figure7-8.Cellpotentialof20Vwasappliedforthedurationoftheexperiment.Thedatain Figure7-5arereproducedhereinadditiontoanexperimentoperatingfor48hours. promisingintermsoftheamountofwaterremovedfromtheconsolidatedclaysportionofthecell aspresentedinFigure7-8. Otherobservationsfromthe20Vexperimentatlongtimesindicatedelectrokinetictreatment actuallyre-suspendssomeclayparticlesduetogasevolutionattheanode.Thiswasindicated byhavingclearersupernatantwateraftertheexperimenthadbeenturnedoffovernightfor12 hoursthanaftertheexperimentwasoperatedduringthedayfor12hours.Thissuggeststhat possiblefuturedesignsmayneedabetterpathwayforgasbubblestoventinsteadofupwards throughthecellcolumn.Thisalsosuggeststhatthesupernatantwatershouldberemovedafter sometimetoavoidre-suspendingclayparticlesintothewater.Fortheexperimentspresented inthissection,thegapbetweentheelectrodeswasmaintainedatthesamedistancethroughout theentireexperimentstoavoidchangingthepotentialgradient(inV/cm)inthecell.Inorderto maintaintheseconditionsthesupernatantwaterwasnotremovedallowingtheelectrodesto remaininelectricalcontactwiththeclayslurrywithoutadjustments. Theresultsofthissectiondemonstratetrendswhichindicatethatthesolidsweightpercent increasedasthecell(orapplied)potentialincreased.Theresultsalsoindicatedthesolids weightpercentoftheclaysincreasedwithincreasedoperatingtime.However,thepurpose ofpresentingtheresultsintermsofFigures7-4-7-8 wastostudytheuniformityoftheclay 63

PAGE 64

Figure7-9.Photographofcellafterthecompletionofa4day,20Vexperimentandits comparisontothesettlingofacontrol(onright)sampleintheabsenceofanelectric eld.Thecellpotentialwasappliedfor12hourseachday.Thearrowindicatesthe distancebetweenelectrodes.Aclearlayerofwaterisindicatedabovetheclay. suspensions.Theexperimentsrevealedthatlowercellpotentialsorcurrentdensitiescould improvetheuniformityofclaydewateringbydecreasingtherateofgasevolution.Longer operatingtimeswerefoundtoimprovetheuniformityofdewateringasillustratedbyFigure 7-8.Thismaybemeaningfulforalarge-scalewaterremovalsystemthatwouldbeexpectedto requiremanydaysofoperationtotreatanareaaslargeasaone-square-mileclaysettlingarea. 7.4ConstitutiveRelationshipatLongTimes Bench-topexperimentswereperformedatlongeroperatingtimestodeterminethevalidity ofequation(7)beyondtherangeofparameterspresentedinSection7.2.Uponcompletion ofanewsetofexperiments,theassociateddatapointswereaddedtothedatainFigure 7-3 togiveFigure 7-10.Theexperimentsperformedatlongeroperatingtimesshowedthat,even whensupernatantwaterwasperiodicallyremoved,alimitingvalueforsolidsweightpercentwas reached.Thevalueofthemaximumsolidscontentachievablewasfoundtobeafunctionofthe appliedelectriceldasthesolidsmaximumincreasedwithincreasingelectriceldstrength.The 64

PAGE 65

Figure7-10.Additionaldataatlongertimesandlargersolidsweightpercentareincludedwith thedatapresentedinFigure7-3.Thehorizontaldashedlinesrepresentthe suggestedplateaureachedatthreedifferentelectriceldsizes. datapointsthatconvergetowardsthehorizontallydashedlinesinFigure7-10 wereconcluded tohavereachedtheirmaximumsolidsweightpercent.Forsuchdata,themaximumchangein solidsweightpercentwasgivenasafunctionoftheelectriceldinFigure 7-11.Theequation ofthelinewasusedtosuggestarelationshipbetweenthemaximumchangeinsolidsweight percentandtheelectriceldas w c =7:1Log 10 (E )+16:5 (7) Aninterpolationmodelequationwasdevelopedtoincorporatethelinearbehavioratshorttimes (i.e.,equation(7))givingaconstitutiverelationshipforshortandlongtimesas w c =[(0 :72tE ) )Tj /T1_3 7.97 Tf (n +(7:1Log 10 (E )+16:5) )Tj /T1_3 7.97 Tf (n ] )Tj /T1_4 7.97 Tf (1=n (7) where n isadimensionlessparameterthatcontrolsthetransitionfromshort-timetolong-time behavior, E hasunitsofV/cmand t hasunitsofhours.Thecomparisonofequation(7)for n = 5totheexperimentaldataforthreedifferentappliedelectriceldsarepresentedinFigures 7-12 and 7-13B.Theresultispresentedintheformof w c =E inFigure 7-12 andas w c inFigure 65

PAGE 66

Figure7-11.Maximumchangeinsolidscontentplottedasafunctionoftheelectriceld.The relationshipdevelopedfromthelineartrendlineisgivenbyequation(7). 7-13.Theagreementwithexperimentaldatasuggeststhatthechangeinsolidsweightpercent canbedeterminedataknownelectriceldandforalargerangeofoperatingtimes. Theinterpolationmodelresultsusingequation(7)wereplottedforarangeofelectric eldsaspresentedinFigures7-14and7-15.Theresultsindicatethatatlargerelectriceldsthe transitionfromtheshort-timebehaviortolong-timebehavioroccurredmuchearlier.Atverysmall electricelds,thistransitionappearstotakemuchlonger,anditisdifculttodeterminewhenthe transitionoccursbaseduponthetimescalespresentedinFigures 7-14 and 7-15. Theinterpolationmodelgivenasequation(7)wasselectedafterconsiderationofother simplermodels.Themodelsconsideredwere w c = w max t k + t (7) and w c = w max (1 )Tj /T1_2 10.909 Tf 10.909 0 Td (exp()Tj /T1_2 10.909 Tf [(kt )) (7) where w max isgivenas w max =7:1Log 10 (E )+16:5 (7) 66

PAGE 67

Figure7-12.DatafromFigure7-10withtheconstitutiverelationshipforlong-times(eq.(7))t tothedataforthreedifferentelectriceldsizes. A B Figure7-13.DataincludedinbothFigures 7-10 and 7-12 arepresentedwithoutscalingbythe electriceld.InB),theconstitutiverelationshipforlongtimes(eq.(7))isttothe dataatthreeelectriceldsizes. 67

PAGE 68

Figure7-14.Constitutiverelationshipforlongtimes(eq.(7))plottedforseveralelectriceld sizes. Figure7-15.Constitutiverelationshipforlongtimes(eq.(7))plottedforseveralelectriceld sizeswiththechangeinsolidsweightpercentscaledbytheelectriceld. 68

PAGE 69

Equation(7)istheexpressionforthemaximumsolidscontentgiveninequation(74).The parameter k isdimensionlessandisadjustedseparatelytotthedataforbothequations(7 5)and(7),and t isoperatingtimewithunitsofhours.Thesemodelswerettothedata presentedinFigure7-10.Thetprovidedbyequations(7)and7werelesssatisfactory thanthetprovidedbyequation(7).Thebehaviorofequations(7)and(7)werenonlinearatshorttimesandfailedtoaccuratelytransitionfromshort-timetolong-timebehavior.As equation(7)wasfoundtoprovidethebestttothedata,itisintroducedwithinthisworkas themodelthatbestcharacterizesthebehaviorofthebench-topclaydewateringcell. 7.5EnergyUsageModel Theenergyrequirement(inW-h/kgwaterremoved)forwaterremovalwasdeterminedfor thebench-topexperimentspresentedinprevioussectionswithinthischapter.Thevalueswere calculatedbymultiplyingthecellpotential V cell bytheproductoftheoperatingcurrent I and operatingtime t andthendividingbythemassofwaterremovedduringagivenexperiment. Experimentalrequirementsweredeterminedfromagroupofbench-topexperimentsandare presentedinFigure7-16.Allvaluesliewithintherangeofenergyrequirementsreportedin theliterature.FromtherawnumbersreportedbyFourie etal., 4 anenergyrequirementof 1.25W-h/kgofwaterremovedwascalculated,whileLarue etal. 16 reported700W-h/kgof waterremoved.AccordingtotheexperimentalresultspresentedinFigure7-16,theenergy requirementincreaseswithincreasingelectriceldstrength.Thistrendprovidesinsightforthe managementofoperatingcostsforalarge-scalewaterremovalsystemastheexperimental resultsindicatethatoperatingatsmallerelectriceldswouldyieldsmallerenergycosts. TheexperimentalenergyrequirementspresentedinFigure 7-16 weremodeledbasedupon fundamentalelectrochemicalengineeringequationsincombinationwiththeconstitutiverelationshipforlongtimes(givenasequation(7)).Theresultsofthemodelarecomparedtothe experimentaldatainFigure 7-17.Favorableagreementisfound,verifyingtheelectrochemical parametersusedforthemodelandtheuseofequation(7). TheenergyrequirementsoftheexperimentalresultsfromFigures 7-16 and 7-17 were calculatedbymultiplyingthecellpotential V cell bytheproductoftheoperatingcurrent I and operatingtime t andthendividingbythemassofwaterremovedduringagivenbench-top 69

PAGE 70

Figure7-16.Energyrequiredpermassofwaterremovedisgivenasafunctionoftheelectric eld. experiment.AsindicatedfromFigures 7-16 and 7-17,theenergyrequirementswereadversely affectedastheelectriceldincreased. ThedevelopmentofthemodelpresentedinFigure 7-17 includedthecalculationofthe cellpotential V cell asafunctionofoperatingcurrentdensity i .Therelationshipbetweenthecell potential V cell andtheoperatingcurrentdensity i isgivenas V cell =1:23+ a )Tj /T1_1 10.909 Tf 10.909 0 Td ( c )Tj /T1_1 10.909 Tf 10.909 0 Td (iR e (7) where a representstheanodicoverpotential, c representsthecathodicoverpotentialand R e representstheOhmicresistanceoftheclaysuspension.Thevalueof1.23(Volts)isthe minimumpotentialrequiredtopromotethehydrolysisofwater.Theoperatingcurrent I ,or currentdensity i,wascontrolledasaninputwiththecellpotential V cell calculatedasaresponse variable.Thevaluesof a c ,and R e werethencalculatedinordertodetermine V cell atagiven operatingcurrentdensity i. 70

PAGE 71

Figure7-17.DatapresentedinFigure7-16 comparedwithamodeledenergyrequirementcurve. Energyrequiredpermassofwaterremovedisgivenasafunctionoftheelectric eld. Tocalculate a ,theButler-Volmerequation i = i o (exp( a b aa ) )Tj /T1_1 10.909 Tf 10.909 0 Td (exp()Tj /T1_1 10.909 Tf ( a b ac )) (7) wasusedwhere b aa = F RT (7) and b ac = (1 )Tj /T1_1 10.909 Tf 10.909 0 Td ( )F RT (7) with beingthedimensionlesssymmetryfactor.TheButler-Volmerequation(givenbyequation (7))iswrittenforareversiblereactionwithbothananodicandcathodiccontribution.Atlarge andpositiveoverpotentials,theanodiccontributiondominatesandthecathodiccontributioncan beneglectedsimplifyingequation(7)to i = i o exp( a b ca ) (7) Conversely,whenoperatingatlargeandnegativeoverpotentials,thecathodiccontribution dominatesandtheanodiccontributioncanbeneglected.Thisallowsequation(79)tobe 71

PAGE 72

simpliedto i = )Tj /T1_1 10.909 Tf (i o exp()Tj /T1_1 10.909 Tf ( a b cc ) (7) However,inordertosolvefor a ,aniterativemethodwasperformedastheButler-Volmer equation(givenbyequation(7))isrearrangedtogive a;n+1 = ln( i i o + exp()Tj /T1_1 10.909 Tf ( a;n b ac )) b aa (7) where a;n representstheinitialguessfortheanodicoverpotential.Theinitialguesswas approximatedbythelinearformoftheButler-Volmerequation i = i o (b aa + b ac ) a (7) whichisrearrangedto a;n = i i o (b aa + b ac ) (7) ThelinearformoftheButler-Volmerequationisvalidonlyatverysmalloverpotentialswhere neitheranodicorcathodiccontributionscanbecompletelyneglected.Aspreadsheetprogram wasusedtoperformtheiterations.Eachcalculationof a;n+1 wasusedasthesubsequentinput valuefor a;n whichwascontinuouslypluggedbackintoequation(7)untiltheresultingvalue of a;n+1 nolongerandthesolutionconverged. Asimilariterativemethodwasusedtodeterminethecathodicoverpotential c .TheButlerVolmerequation i = i o (exp( c b ca ) )Tj /T1_1 10.909 Tf 10.909 0 Td (exp()Tj /T1_1 10.909 Tf ( c b cc )) (7) wasagainusedwhere b ca = F RT (7) and b cc = (1 )Tj /T1_1 10.909 Tf 10.909 0 Td ( )F RT (7) TheButler-Volmerequationisrearrangedas c;n+1 = )Tj /T1_1 10.909 Tf [(ln()Tj /T1_4 7.97 Tf 34.505 4.296 Td (i i o + exp( c;n b ca )) b cc (7) 72

PAGE 73

where c;n representstheinitialguessused.Theinitialguess,wasagainapproximatedbythe linearformoftheButler-Volmerequation i = i o (b aa + b ac ) c (7) whichisrearrangedto c;n = i i o (b aa + b ac ) (7) Inordertocalculate V cell ,theOhmicresistanceoftheelectrolyte R e mustalsobecalculated. ThemethodstocalculatetheOhmicresistance R e experimentallyaredescribedinSection 7.6. Upondeterminationof V cell foragivensetofoperatingconditions,theenergyrequirement (inW-h/kgofwaterremoved)wascalculatedasafunctionofoperatingtime t forarangeof electriceldvalues.Theobjectiveofthissectionwastomodeltheexperimentaldatapresented inFigure 7-16 basedupontheuseoffundamentalelectrochemicalengineeringrelationshipsand theconstitutiverelationshipforlongtimes(givenasequation(7)).Therefore,theoperating currentdensity i wascontrolledtoyieldthesamerangeofelectriceldvaluesforcomparison withtheexperimentalresultspresentedinFigure 7-16.Inordertomodeltheenergyrequirement,theamountofwaterremovedwascalculatedasafunctionoftimeforagivenelectriceld usingtheconstitutiverelationshipforlongtimes(givenasequation(7)).Asequation(7) incorporatestheoperatingtime,theenergyrequirementcanbeplottedasafunctionofoperating timeataspeciedelectriceld.Thiswasperformedseparatelyatdifferentelectriceldvaluesto developthemodeledenergyrequirementcurvepresentedinFigure 7-17.Thevaluescalculated todeveloptheenergyrequirementcurvearepresentedinTable 7-1.Theparametersusedfor theButler-Volmerequationandtheexpressionforthecellpotential V cell aregiveninTable 7-2. Notethattheelectrolyteresistance R e andtheelectrolyteresistivity e arerelatedas R e = A c d (7) where A c representsthecross-sectionalareaofthecell,whichcorrespondstothenominalarea ofboththeanodeandcathode,and d representsthedistancebetweentheelectrodes. Anexampleoftheenergyrequirementgivenasafunctionofoperatingtimeataspecied electriceldispresentedinFigure7-18.ThemodelinFigure7-17 wascalculatedforthesame 73

PAGE 74

Table7-1.ResultsofcalculationsusedtomodeltheenergyrequirementspresentedinFigure 7-17. E /Vcm )Tj /T1_4 7.97 Tf (1 V cell I /W waterremoved/kg t/h energyrequirement/Whkg )Tj /T1_4 7.97 Tf (1 0.08 0.01 0.34 60 1.7 0.52 0.24 0.34 9 6.4 1.05 0.95 0.54 9 15.7 2.54 5.34 0.79 9 60.6 4.54 16.90 0.87 9 174.8 Table7-2.Valuesofparametersandvariablesusedinenergyusagemodel. parameter numericalvalue / ncm 1500 b aa ,b ac ,b ca ,b cc /V )Tj /T1_4 7.97 Tf (1 10 0.5 i o /mAcm )Tj /T1_4 7.97 Tf (2 1.0 A c /cm 2 63.6 d /cm 20 operatingtimeastheexperimentaldata.However,aterminaloperatingtimewasidentiedin Figure7-18wherethedatatransitionsfromnonlinearbehaviortolinearbehavior.Oncethe linearbehaviordominates,thesolidsweightpercenthasreacheditsmaximumbasedupon theconstitutiverelationshipforlongtimes(givenasequation(7)).Therefore,asindicated inFigure7-18B,thetransitionfromnonlineartolinearbehaviorcanbedeterminedwhichis denedastheterminaloperatingtimetoachievemaximumremovalofwater.Thisbehavior wasobservedatarangeofelectriceldvalues,witheachyieldingadifferentterminaloperating time.Attheterminaloperatingtime,thecorrespondingenergyrequirementcanbeidentied fromthedataaspresentedinFigure7-18.Themodelwascalculatedseparatelybaseduponthe terminaloperatingtimeidentiedateachelectriceld.ThisresultispresentedinFigure7-19. BothmodelsarecomparedtogetherwiththeexperimentaldataaspresentedinFigure7-20.The discrepanciesinthetwomodelsareconcludedtobebasedupontheuseofdifferentoperating timesasthiswastheonlyparameterchanged.Thedifferenceinoperatingtimesareincludedin Figure7-21.Themodeledenergyrequirementsarefoundtovaryasafunctionofoperatingtime aspresentedinFigure7-20.Forfutureenergyusagemodels,itissuggestedthattheoperating timeshouldbethesameasthatofexperimentstoobtainfavorableagreement. Theagreementofthemodelwiththeexperimentaldatasuggeststhattheconstitutive relationshipsforlongtimes(givenasequation(7))incombinationwithelectrochemical 74

PAGE 75

A B Figure7-18.Energyrequirementisgivenasafunctionofoperatingtime.InB),thetransition fromlineartononlinearbehaviorisindicated. relationshipscanbeusedtoeffectivelyprojectenergyrequirements.Thisfurthervalidatesthe abilityoftheconstitutiverelationship(givenasequation(7))tocharacterizethewaterremoval fromclayusingthebench-topexperiment. 7.6ElectrochemicalCharacterization AmanuallygeneratedpolarizationcurveispresentedinFigure 7-22.Eachdatapointwas generatedindividuallyfromseparateexperimentswithinthematrixdiscussedinSection 7.3. Aftereachpotentialwasapplied,thecurrentwasrecordedafterreachingsteadystate.Insome casespolarizationcurvescanbeusedtoidentifyelectrochemicalreactionsormechanisms withinthecell;however,fortheworkpresentedinthissectiontheobjectivewastodetermine whetherthedominantresistanceinthecellwascontrolledbykinetic,Ohmic,ormasstransfer limitations.ThelinearbehaviorpresentedinFigure 7-22 indicatesthatthedominantresistance inthebench-topexperimentalcellwastheOhmicresistance.Thissuggeststhatforoperating conditionsoftheexperimentaldatapresentedinFigure 7-22 thatnoadverseissuesoccurred regardingtheelectrodesusedfortheexperiments. Anelectrochemicalimpedancespectroscopy(EIS)scanwasalsoperformedonthebenchtopexperiment.TheimpedancescanispresentedinFigure 7-23.Theverticaldashedline representsthehighfrequencyasymptote.ThehighfrequencyasymptotegivestheOhmic resistancewhichis152.5Ohms.Are-scaledviewofFigure 7-23 ispresentedinFigure 7-24 whichincludesthefrequencyrangeusedfortheimpedancescan.Theimpedanceresultsalso suggestthattheOhmicresistanceisthedominantresistanceofthecellasthemagnitudeofthe 75

PAGE 76

Figure7-19.Energyrequirementasafunctionoftheelectriceldforthemaximumseparation achievableattheappliedelectriceld E .Valuesweredeterminedusingthe terminaloperatingtimeindicatedbythemethodpresentedinFigure7-18A. Figure7-20.DatapresentedinFigure7-17 areincludedforcomparisontothemodelbased upontheterminaloperatingtime. 76

PAGE 77

Figure7-21.Twosetsofoperatingtimesusedforthemodelaregivenasafunctionofthe electriceld. impedanceisverysmallcomparedtotheabsolutevalueoftheOhmicresistance.Therefore,the qualitativeobservationsfromtheimpedancescanwereconsistentwiththatfromthepolarization curve. TheOhmicresistanceofthecellisusedtodeterminetheconductivityofclaysuspension. Theclaysuspensionwasapproximately10%solidsbymass(orweight)whenbothpolarization andimpedanceplotsweregenerated.ByincludingthecalculatedOhmicresistance( R e )aswell asthebench-topcelldimensions,theconductivity()canbecalculatedas = d R e A c (7) where d isdistancebetweenelectrodes(in cm)and A c isthecrosssectionalareaofthecell(in cm 2 ).Fromequation(7)theconductivitywascalculatedas700microSiemenspercentimeter (S=cm).TheconductivitywasalsocalculatedfortheexactsameclaysampleusingaDC (directcurrent)measurement.TheOhmicresistancewascalculatedbytakingthepotential dropmeasuredbetweentwoAg/AgClreferenceelectrodeswithinthecellanddividingitbythe currentowingthroughthecell.Fromthismethod,theconductivitywascalculatedas682 S=cm usingequation(7)with d representingthedistancebetweenthetworeferenceelectrodes. 77

PAGE 78

Figure7-22.Polarizationcurvegeneratedfrombench-topcellloadedwithclaysuspension.Each datapointwasindividuallymeasuredfromseparateexperimentsforeachapplied potential. Figure7-23.Impedancescangeneratedatopencircuitpotentialonbench-topcellloadedwith claysuspension. Therefore,nosignicantdifferenceinthecalculatedconductivitywasfound,regardlessofthe differenttechniquesused.Thisfurtherindicatesthateithermethodshouldbevalidforcalculating conductivity.Thesevaluesarecomparedtoreportedconductivitiesofdifferentclaysuspensions fromtheliteratureaspresentedinTable7-3. Thepolarizationcurveandtheimpedancescanpresentedinthissectionwereusedto characterizefurtherthebench-topexperimentspresentedinthischapter.Bothmethodswere usedtodeterminethedominatingresistancewithinthebench-topcell.Bothmethodsindicated thattheOhmicresistancewasthedominatingresistanceofthecell.Thismeantthatanykinetic limitationsduetoelectrochemicalreactionsoccurringattheelectrodesurfaceswasnegligible 78

PAGE 79

Figure7-24.Re-scaledperspectiveofimpedancescanpresentedinFigure7-23withtheaxisfor therealimpedancesubtractedbytheOhmicresistance R e Table7-3.Conductivitiesofclaysuspensionsfromthisworkarecomparedwiththosereportedin theliterature.EISandDCwerethemethodsusedtoexperimentallydeterminethe conductivityofclaysuspensionsusedwithinthiswork. Source Claytype conductivity/ Scm )Tj /T1_5 7.97 Tf (1 EIS Mosaicphosphaticclay 701 DC Mosaicphosphaticclay 682 McLean 69 Floridaphosphaticclay 700 Shang 14 Ontario,Canadanaturalsoilclay 410 Shang etal. 85 Ottawa,Canadanaturalsoilclay 256-2520 incomparisontotheOhmicresistanceofthecell.Theconductivityoftheclaysuspensionwas determinedfrompotentiostaticexperimentsandcomparedtotheconductivityfoundfromthe impedanceresults.Verygoodagreementwasfoundfortheconductivitiescalculatedfromboth methodsaspresentedandcomparedwithotherresultsintheliteratureinTable 7-3. 79

PAGE 80

CHAPTER8 SIMULATIONSFORLARGE-SCALEDEWATERINGSYSTEMS Simulationswereperformedtoassessdifferentelectrodecongurationsinasimulated one-square-mileclaysettlingarea(CSA).CP 3 D,amathematicalmodeldevelopedatthe UniversityofFlorida,wasusedtoperformthesimulations.Thebackgroundforthemathematical frameworkofthemodelispresentedwhichisbaseduponitsprimaryapplicationinvolving cathodicprotectionofburiedpipelines.Theapplicationofthemodelcanbeeasilyttosimulate cathodesandanodesinaclaysettlingpond.Therelationshipsdiscoveredintheexperimental workfromChapter7wereusedtomakeassumptionsregardingthedevelopmentandresultsof thesimulations. 8.1IntroductiontoCP 3 DandApplicationforCathodicProtection CP 3 DisacomputerprogramdevelopedbyProfessorMarkOrazem'selectrochemical engineeringresearchgrouptomodelcathodicprotectionofburiedpipelines. 86,87 Itallowsfor thecreationofavisualizedthree-dimensionalcathodicprotectionsystemofburiedstructures. Thisprogramwasdevelopedasatooltoimprovetheabilitytoassesspipelineconditions. Thereareseveralparametersthatcanbeusedandvariedinthecalculationsperformedby themathematicalmodel.Someoftheseincludecoatingaw(orholiday)size,soilresistivity, cathodicprotectionlevel,coatingcondition,depthofcover,andpipelinethickness.Inorderto replicateabove-groundsurveytechniqueswithinCP 3 D,asoilsurfaceiscreatedandutilized withintheprogram.Thesoilsurfaceismadeupofnodeswhereon-andoff-potentialsare calculatedbythemodelateachnode'sexactlocation.Theon-potentialrepresentsthepotential measuredorcalculatedwhenthecathodicprotection(CP)currentisturnedonorenergized. Theoff-potentialrepresentsthepotentialwhentheCPcurrentisoff.Inordertostudythedetails ofinterest,soilsurfacesaretypicallyplacedabovetheanodeandalsoabovethecoatingaw locatedonthepipeline.Thesoilsurfacesarelocatedinproximityoftheanodeandcoatingaw becausetheselocationsrepresentareaswhereausefuldistributionofon-andoff-potentials arefound.Figure8-1isanimageofthethreedimensionalenvironmentofCP 3 Dshowingthe arrangementofthesoilsurfaceinrespecttothepipelineandthecoatingaw.Thesoilsurface mustbelimitedinsizetoavoidputtingastrainontheresourcesoftheprogram.However,in areal-lifeeldsurvey,measurementsmayincludetheentirelengthofthepipelineinorder 80

PAGE 81

Figure8-1.CP 3 Dimageshowingthephysicalorientationofthesoilsurfacewithrespecttothe pipeline.Thedarkerareaonthepipelinerepresentsthecoatingaworholiday. toinspectitsoverallintegrity.Sincetheareaofinterest(i.e.,locationofthecoatingaw)is speciedwithinCP 3 D,asoilsurfacethatcoverstheentirepipelineisnotnecessary.Amatrix ofsimulationswereperformedinanotherstudyusingCP 3 Dtodevelopdesignequationswhich incorporatepipelineandsoilparametersintheassessmentofsurveyindicationstopredict coatingawsize. 88 Thisstudyhasmajorimplicationsonhowpipelinesurveydatashouldbe interpretedtoassesspipelineintegrity. 8.1.1MathematicalDevelopment ThegoverningequationsfortheCP 3 Dmodelworkasabasisforallcalculationsmadeby theprogram.Forprotectionofundergroundpipelines,themodelaccountsforthecurrentow throughthesoil,thepipeline(orcathode),andthroughthecircuitrybacktotheanode.There aretwodifferentdomainswhicharegovernedseparatelyinthemodel.Oneistheouterdomain whichisrepresentedbythesoil.Theotheristheinnerdomainwhichrepresentsthepipeline,the anode,andtheelectricalwiringthatconnectsthem. Therstofthegoverningequationsfortheouterdomainisthematerialbalanceofasolute species @c i @t = )Tj /T1_5 10.909 Tf ((r N i )+ R i (8) 81

PAGE 82

where c i istheconcentrationofaspecies i, N i isthenetuxvectorforspecies i,and R i representstherateofgenerationofspecies i duetohomogeneousreactions.Homogeneous reactionsaredenedasthereactionswhichoccurintheelectrolyteandnotattheelectrode surfaces,whichinthissystemareeithertheanodeorthepipeline.Equation(8)mustbe coupledwiththeequationofelectroneutralityinordertoaccountfortheconcentrationsand potentialsthatarepresentinthesoil.Theequationforelectroneutralityisgivenas X i z i c i =0 (8) where z i representsthechargeassociatedwithspecies i.Foradiluteelectrolyte,theuxofa givenspeciescanbegivenbasedonitscontributionsfromconvection,diffusion,andmigration as N i = vc i )Tj /T1_1 10.909 Tf 10.909 0 Td (D i rc i )Tj /T1_1 10.909 Tf 10.91 0 Td (z i u i Fc i rb (8) where v istheuidvelocity, D i isthediffusioncoefcientforspecies i u i isthemobility, F is Faraday'sconstantand b representsthedistributionofpotentialinthedomain.Thediffusion coefcientisrelatedtothemobilitybytheNernst-Einsteinequationas D i = RTu i (8) where R isthegasconstantand T isthetemperature.Theequationforcurrentdensityisbased onthecontributionofthemovementofeachionicspeciesandisgivenas i = F X i z i N i (8) Iftheconcentrationofionsintheelectrolyteareuniformandsteadystateisassumedthe equationforcurrentdensitycanbewrittenasOhm'slaw.Therefore, i = )Tj /T1_1 10.909 Tf (rb (8) where representstheconductivityoftheelectrolyte.Theconductivityisgivenby = F 2 X i z 2 i u i c i (8) 82

PAGE 83

anditisuniformbecausetheconcentrationisuniform.Duetouniformconcentration,the potentialisgovernedbyLaplace'sequationwhichisgivenas r 2 b=0 (8) Laplace'sequationcanbederivedbyrstmultiplyingequation(8)by z i F andsummingover allionicspecieswhichgives @ @t F X i z i c i = r F X i z i N i + F X i z i R i (8) Fromelectroneutralityandfromtheassumptionthat R i iszerobecauseitrepresentsreactionsin thebulk,equation(8)reducesto r i =0 (8) BysubstitutingOhm'slawintoequation(8)andbytheassumptionofconstantconductivity, Laplace'sequationisobtainedasequation(8). Sinceitisassumedthattherearenoconcentrationgradientsinthesoilorelectrolyte,the concentrationgradientsduetoreactionsatthesurfaceoftheanodeandpipelinearetreatedso thattheylieinathinlayeradjacenttotheelectrodesurfaces. 86 Theconcentrationgradientsin thisthinlayerareincorporatedintotheboundaryconditionwhichisbasedonelectrochemical reactions. Fortheinnerdomain,therearealsosomeassumptionsthatmustbemade.Forexample, thismodeltreatsthepotentialthroughthepipelineasnonuniform.Previously,pipelinesof shorterlengthshaveneglectedthepotentialdropalongthepipelinesteel. 89 However,ithas beenproventhatforlongpipelinesthispotentialdropcannotbeneglected. 9092 Laplace'sequationalsogovernstheowofcurrentthroughthepipesteel,anode,andthe connectingwires.Itisgivenas r (rV )=0 (8) where istheconductivityofeitherthepipelinesteel,theanode,ortheelectricalwiresand V is thedropinpotentialofthemetalfromauniformvalue.Theconductivityoftheinnerdomainis notconstantbecauseitisnotthesamefortheanode,pipeline,orthecopperconnectingwires. 83

PAGE 84

Laplace'sequationcanbesimpliedas V = IR = I L A (8) whichaccountsforthepotentialdropalongthecopperconnectingwires.Forthisequation, R representstheresistanceofthewire, representstheelectricalresistivityofthewire, L isthe length,and A isthecross-sectionalareaoftheconnectingwire. Thetwodomainsarecoupledthroughboundaryconditions.Theboundaryconditions developarelationshipbetweenthelocalvaluesofpotentialandthecurrentdensityonthemetal surface. 86 Thisrelationshipvariesdependingonwhetherthetypeofsurfaceisthebaremetal pipeline,thecoatedpipeline,asacricialanode,oranimpressedcurrentanode. 8.1.2BareSteel Foranon-coatedpipeline,baresteelisexposedtothesoil.Baresteelcanalsobeexposed inplaceswhereacoatedpipelinehasscratchesorawsonit.Therearethreedifferentelectrochemicalreactionsthatcantakeplaceatthesurfaceofthebaresteel.Thesereactionsinclude theoxidationofiron,reductionofoxygen,andhydrogenevolution.Hydrogenevolutioncanoccur ifthemetalispolarizedtoverynegativepotentials.Thecurrentcontributionofeachofthese reactionscanbeincludedintherelationshipbetweenlocalcurrentdensity i,thepotentialofthe steel V ,andthepotentialofthesoilnexttothesteel b.Thefollowingequation 93,94 represents theboundaryconditionofthebaremetalpipelineandtheadjacentsoilandisgivenas i =10 ( V )Tj /T1_4 5.978 Tf (b)Tj /T1_5 5.978 Tf (E Fe ) f Fe )Tj /T1_1 10.909 Tf 10.909 0 Td (( 1 i lim;O 2 +10 ( V )Tj /T1_4 5.978 Tf (b)Tj /T1_5 5.978 Tf (E O 2 ) f O 2 ) )Tj /T1_10 7.97 Tf (1 )Tj /T1_1 10.909 Tf 10.909 0 Td (10 )Tj /T1_4 5.978 Tf (( V )Tj /T1_4 5.978 Tf (b)Tj /T1_5 5.978 Tf (E H 2 ) f H 2 (8) Theterm E Fe representstheequilibriumpotentialfortheoxidationofironandthistermiswritten similarlyforthereactionsofoxygenreductionandhydrogenevolution.The f termisgivenfor eachreactionanditrepresentsthetafelslopeofthecorrespondingreaction.The i lim;O 2 term representsthemasstransferlimitingcurrentdensityofoxygenreductionatthemetalsurface. Therefore,thecurrentcontributionofoxygenreductioncannotbelargerthanthevalueof i lim;O 2 8.1.3CoatedSteel Forcoatedpipelines,treatmentofthereactionsatthepipeline-soilinterfacemustbe differentthanthatofbaresteel.Thepurposeofthecoatingistoprovideresistanceforthe transportofreducingspeciestothemetalsurface.ItalsoreducestheamountofCPcurrentthat 84

PAGE 85

isneededtoprotectagivenpipeline.Therearetwomaintypesofcoatingbehaviorandthese arebothmodeleddifferently.Onemodelofcoatingbehavioriswherethetransportofspecies isuniformthroughthecoating.Theelectrochemicalreactionstakeplaceoncethetransported speciesreachthecoating-metalinterface.Thesereactionsaredrivenbythedifferencein potential V ofthemetalandthepotential b in justunderneaththecoatingbutstillabovethe metalorsteel. Theothertypeofmodelinvolvesthepresenceofporeswhichallowforthetransportof solutespeciestotakeplace.Ithasbeenshownthattheporestructurewillexpandafterithas beencontactedwithwaterandthattheconductivityofthecoatingincreaseswithtimeafterits exposuretowater. 95 Theresistivityofthecoatingwithporesisafunctionofthenumberofpores perunitarea.Theelectrochemicalreactionsalsotakeplaceatthecoating-metalinterfacefor thismodelandthesereactionsarealsodrivenbythepotentialdifferencebetweenVand b in Ithasbeenshownthatthenon-porouscoatedsteelformsadiffusionbarrierwhenputin aqueousenvironments.Asthewaterisadsorbedbythecoating,thesteelcanbepolarized slightlyevenifthecoatingisdisbonded. 9699 Equation(8)canbemodiedbasedoneither modeloftransportthroughthecoating. 100,101 Thecurrentdensitycanbewrittenasafunctionof thepotentialdropthroughthecoatingas i = b )Tj /T1_2 10.909 Tf 10.909 0 Td (b in (8) where b isthepotentialofthesoiladjacenttothecoating, istheresistivityofthecoating, and isthethicknessofthecoating.Bywritingthecurrentdensityintermsofelectrochemical reactionsitisalsogivenas i = A pore A [10 ( V )Tj /T1_5 5.978 Tf (b in )Tj /T1_6 5.978 Tf (E Fe ) f Fe )Tj /T1_2 10.909 Tf 10.909 0 Td (( 1 (1 )Tj /T1_1 10.909 Tf 10.909 0 Td ( blk )i lim;O 2 +10 ( V )Tj /T1_5 5.978 Tf (b in )Tj /T1_6 5.978 Tf (E O 2 ) f O 2 ) )Tj /T1_9 7.97 Tf (1 )Tj /T1_2 10.909 Tf 10.909 0 Td (10 )Tj /T1_5 5.978 Tf 7.782 6.35 Td (( V )Tj /T1_5 5.978 Tf (b in )Tj /T1_6 5.978 Tf (E H 2 ) f O 2 ] (8) where A pore A istheeffectivesurfaceareaavailableforreactionstooccurand block isthereductionofthetransportofoxygenthroughthediffusionbarrier.Inordertodeterminethevaluesfor thecurrentdensity(i )and b in ,bothequation(8)andequation(8)aresolvedsimultaneouslybytheNewton-Ralphsonmethod. 85

PAGE 86

8.1.4SacricialandImpressedCurrentAnodes Thereactionsatthesurfaceofasacricialanodetypicallyincludeoxygenreductionandthe corrosionoftheanode.Thecurrentdensityexpressionistreatedsimilarlyasthatofthebare steelexceptthatthehydrogenevolutionreactionisneglected.Theexpressionisgivenas i = i lim;O 2 (10 V )Tj /T1_5 5.978 Tf (b)Tj /T1_6 5.978 Tf (E corr f anode )Tj /T1_3 10.909 Tf 10.909 0 Td (1) (8) where E corr isthecorrosionpotentialattheanodeand f anode isthecorrespondingTafelslopefor theanodiccorrosionreaction. Thecurrentdensitymodelequationforimpressedcurrentanodesissimilartothatofthe galvanicorsacricialanode.Theonlydifferenceistheinclusionoftherectierpotentialsetting. Thisequationisgivenas i = i lim;O 2 (10 V )Tj /T1_5 5.978 Tf (b)Tj /T1_5 5.978 Tf (V rect )Tj /T1_6 5.978 Tf (E corr f anode )Tj /T1_3 10.909 Tf 10.909 0 Td (1) (8) Thisequationmustbemodiediftherearechlorideionspresentinthesoil. 8.2ApplicationforClayDewatering CP 3 Dwasusedtodevelopdifferentelectrodecongurationstosimulatethedesignofa largescalewaterremovalsysteminaclaysettlingarea.Simulationsspeciedallelectrodes asconstantpotentialsurfaces.Thepipelineswithinthesoftwareprogramareconsidered,and moreappropriatelyreferredto,ascathodesinthissection.Thegoverningequationthroughthe electrolyteorclayslurryisLaplace'sequationgivenasequation(8).InsolvingforLaplace's equation,oneofthekeyassumptionsisthattherearenoconcentrationgradients.Therefore, sincetheconcentrationisuniform,theconductivitymustalsobeuniform. 8.2.1Large-scaleSimulations Twodifferentelectrodecongurationswereexplored.Therstcongurationincluded cylindricalelectrodesthatwere6inchesindiameterandplaced3feetapart.Theelectrodes werehorizontallyorientedandonemilelongtobeconsistentwithaone-square-mileclay settlingarea.Theywereequallyspacedfortheentirelengthofonemile.Anillustrationis giveninFigure8-2showingcylindricalcathodesplacednearthesurfaceofthesimulatedclay settlingarea(CSA)andcylindricalanodesplacednearthebottom.ThedepthoftheCSAis simulatedas40feet.Theelectrodecongurationyieldedauniformelectriceldasshownbythe 86

PAGE 87

Figure8-2.Representationofalargeclaysettlingareawithonemilelongcylindricalelectrodes spacedequallyalongtopandbottomsurfaces.Thezoomedinportionrepresentsa cross-sectionofthecylindricalelectrodes.Althoughonlyrowsofthreecylindersare shown,thesimulationwasscaledforrowsthatextendtoadistanceofonemile. Figure8-3.IllustrationofhorizontallyorientedelectrodecongurationpresentedinFigure8-2 withthecalculatedpotentialdistributionfromCP 3 Dpresentedasafalse-colorimage. 87

PAGE 88

A B Figure8-4.Verticallyorientedelectrodesarepresented:A)CP 3 Dimageofverticallyoriented electrodeconguration;andB)photographofageosyntheticelectrodecoveredwith alterclothrequiredasaseparatorforremovalofwaterusingaverticallyoriented cathode(takenfromFourie etal., 4 Copyright c r 2008NRCCanadaoritslicensors andreproducedwithpermission). potentialdistributioncalculatedbyCP 3 DpresentedinFigure8-3.Theuniformchangeincolor withrespecttothepositionbetweenthetoprowofcathodesandthebottomrowofanodesis representativeofauniformpotentialdistribution.Thisbehaviorwasduetotheclosespacingof theelectrodesandbyhavingonlycathodesintherownearthesurfaceandonlyanodesinthe rowatthebottom. ThesecondcongurationexploredisrepresentedinFigure8-4A.Forthissimulation,the electrodeswereorientedvertically,withcathodesandanodesplaceddiagonallyfromone another.Theaverageelectriceldwascalculated,andtheappliedpotentialwasadjustedsuch thattheaverageelectriceldwasthesameforthetwodifferentcongurations(i.e.,horizontal andvertical).ThevisualizationinFigure8-4Arepresentsaunitcellthatisscaledupbyadding additionalelectrodestocoveranentiresquaremile.Aseparatororlterclothisrequiredfor verticallyorientedcathodessuchthattheclayislteredfromthewaterenteringtheinner diameterofthecathodeaspicturedinFigure 8-4B.Theelectriceldcorrespondingtothe potentialdistributionshowninFigure 8-4A ispresentedinFigure 8-5.Thelocationnearthe centerwheretheelectriceldisatrepresentsaregionwheretheelectriceldtendstoward 88

PAGE 89

A B Figure8-5.IllustrationoftheelectriceldcalculatedfromthesimulationpresentedinFigure 8-4A:A)threedimensionalrepresentation,andB)twodimensionalrepresentation. zero.Thepeaksshownatthecornersrepresentareaswheretheelectriceldismuchstronger. Theexperimentalworkdemonstratedarelationshipbetweenthesizeoftheelectriceldandthe solidscontent(orsolidsweightpercent)foragivenoperatingtime.Therefore,thenonuniform electriceldwouldhavenonuniformdryingwithlocationslocatedinproximityoftheelectrodes (i.e.,thefourcornersinFigure8-4)havingastrongchangeinsolidscontentandthelocations awayfromtheelectrodes(i.e.,atthecenterinFigure8-4)maintainingtheconsistencyofthe originalcontentofthesuspension. TimeandpowerrequirementsarepresentedinTables 8-1 and 8-2 basedonsimulation resultsforthetwodifferentelectrodecongurations.Thesimulationresultforthevertical congurationreportedinTable 8-1 wasrepresentedinFigure 8-4.Acomparisonwithabenchtopexperimentalresultisalsoincludedineachtable.Forthebench-topexperimentalresults presentedinTables 8-1 and 8-2,theelectriceldandenergyrequirementvalueswerereported directlyfromtheexperiment.Thetimerequiredwasdeterminedusingequation(74)presented inSection 7.4.Thepowerandenergycalculationsfortheseexperimentalresultswerebasedon ahypotheticalscale-uptoaone-square-mileclaysettlingarea.Thesevaluesdonotrepresent thebench-topexperimentalcell'sactualoperatingpowerandenergyrequirements.Thevalues inTable 8-1 arebasedonasolidscontentincreasefrom10to25wt%foraone-square-mileclay settlingareaandTable 8-2 isbasedonanincreasefrom10to20wt%.Theenergyrequirement 89

PAGE 90

Table8-1.Resultsofpowerandenergycalculationsfordewateringofsimulatedone-square-mile claysettlingarea.Solidsweightpercentwasincreasedfrom10to25wt%. Conguration HorizontalSimulation VerticalSimulation Bench-topExperiment AverageElectricField/Vcm )Tj /T1_3 7.97 Tf (1 1.2 1.2 1.1 TimeRequired/days 0.8 1.3 1.0 PowerRequired/kW 4.4E+7 3.5E+7 1.2E+7 Energy/kWh 8.5E+8 1.1E+9 2.8E+8 EnergyRequirement/Whkg )Tj /T1_3 7.97 Tf (1 43.6 53.7 13.9 Table8-2.Resultsofpowerandenergycalculationsfordewateringofsimulatedone-square-mile claysettlingarea.Solidsweightpercentwasincreasedfrom10to20wt%. Conguration HorizontalSimulation VerticalSimulation Bench-topExperiment AverageElectricField/Vcm )Tj /T1_3 7.97 Tf (1 0.1 0.1 0.08 TimeRequired/days 8.8 12.9 9.6 PowerRequired/kW 3.5E+5 2.8E+5 1.0E+5 Energy/kWh 7.4E+7 8.8E+7 2.4E+5 EnergyRequirement/Whkg )Tj /T1_3 7.97 Tf (1 4.4 5.2 1.5 results(inW-h/kgwaterremoved)presentedinTables8-1and8-2areincludedalongwith additionalbench-topexperimentalenergyrequirementsinFigure 8-6.Alldataliebetweenthe twodifferentenergyrequirementsreportedintheliterature.Fromtherawdatareportedby Fourie etal., 4 anenergyrequirementof1.25W-h/kgofwaterremovedwascalculated;whereas, Larue etal. 16 reportedanenergyrequirementof700W-h/kgofwaterremoved. 8.2.2EconomicImplications Powerandenergyrequirementswerecalculatedforelectrokineticdewateringofaonesquare-mileclaysettlingareainSection 8.2.1.Thevalueswereachievedthroughuseofthe mathematicalmodel,CP 3 D,incombinationwiththeconstitutiverelationshipsdevelopedto characterizetheexperimentalwork.Thepowerandenergyrequirementsforachievingasolids contentupto25wt%foraone-square-mileclaysettlingareawereaslargeas 10 7 kWand 10 9 kW-h,respectively.Suchlargepowerandenergyrequirementssuggestthattheremediation ofanentirelarge-scaleclaysettlingareamayrequireamodiedapproachforimplementation. Apowerrequirementof 10 7 kWwouldexceedthecapabilitiesofmostpowerplants,which typicallydonotgenerateabove1,000MW.Inordertoavoidsuchlargepowerrequirements, thevolumecouldbereducedbytreatingonlyaportionofaclaysettlingareaforagiventime. SuchamodicationispresentedinFigure 8-7.Thismodicationwouldinvolveisolatingaportion oftheCSAandpumpingtheslurryfromotherlocationswithintheCSAintotheisolatedarea 90

PAGE 91

Figure8-6.Energyrequirementforwaterremovalisgivenasafunctionoftheelectriceld. ExperimentaldatapresentedinFigure7-16areincludedwithbothsimulationand additionalexperimentaldata. forelectrokinetictreatment.Thiswouldallowthegapbetweenelectrodestobesmallerand theelectriceldtobelargerwithreducedpowerrequirements.Oncetheareahasreachedan acceptablesolidscontent,thistreatmentcouldbemovedtoanothersectionoftheCSA.The penaltyforsuchamodicationwouldbethatthetimerequiredtodewatertheentireCSAwould beincreased.However,suchamodicationwouldonlybemadeifthebenetsofdecreased powerconsumptionoutweighedcostsassociatedwithincreasedoperatingtime. Anotheralternativecouldbetomodifyanexistingin-lineprocessdewateringunitsuchasa thickenerpresentedinFigure8-8.Forthistypeofimplementation,thebottomofthethickener couldserveasananodeandtherotatingvanecouldserveasacathode.Theappliedelectric eldwouldenhancetheseparationofsolidstypicallyachievedthroughthethickener.This typeofimplementationcouldbeattractivebecauseitwouldinvolvemodicationstothenewly designedthickeneroperationon-siteatMosaic.Theoperationwouldbecontinuous,witha solids-richstreamleavingfromthebottomofthethickenerandadecantedstreamleavingfrom theoverow.OtherthickenerschematicscanbefoundinGeankopolis, 102 McCabe etal., 103 PerryandChilton, 104 andKos. 105 91

PAGE 92

Figure8-7.Atopviewoftwodifferentone-square-mileclaysettlingareas.Themodieddesign ispicturedontherightdemonstratingthetreatmentofanisolatedsectionofthe CSA. Figure8-8.Schematicrepresentationofthickeneradaptedforelectrokineticallyenhanced separation. 92

PAGE 93

CHAPTER9 CONCLUSIONSANDFUTUREWORK Thisimportanceofthisworkoriginatesfromtheworld'sgrowingdemandforphosphate. Phosphateisusedprimarilyasfertilizertoenhancetheyieldfromfoodcrops.Asaresultof phosphatemining,largeamountsofdilutewasteclaysareformedwhicharestoredinman-made enclosedretentionpondsknownasclaysettlingareas(CSAs).Over100squaremilesofthe CSAsexistinCentralFloridacausingamajorlandstorageissue.Duetothepropertiesofthe clays,itcantakeasmuchas25yearsfortheCSAstosettleanddrytoappropriatelevels.This workinvolvesapplicationofanelectriceldtoincreasetherateofwaterremovaltoreducethe timerequiredtoconsolidatetheoccupiedlandarea.Thischapterreviewstheconclusionsmade fromthebench-topexperimentsandsimulationsperformedfromthiswork.Thisdiscussion leadstotheproposednextstepsofthisproject.Anewbench-topcellisproposedforfuture experimentsandapplicationsforscale-updemonstrationsarediscussed.Whilemuchworkhas beenpreviouslyreportedintheliteratureonelectrokineticdewateringofclaysandothermethods forwaterremoval,theliteraturehasnotyetincludeddevelopmentofparametersaimedtowards large-scaledesign.Theobjectiveofthisworkwastousesmall-scaleelectrokineticexperiments todevelopparametersthatcanbeusedforlarge-scaledesign.Thedevelopmentofsimulations hasbeenincludedtoaidinlarge-scaleprojectionsintermsofpowerandenergyrequirements. Thischapterproposesupcomingobjectivesforthisprojectwhichbuildfromtheresultsand ndingspresentedinthiswork. 9.1Conclusions Abench-topPlexiglascellwasusedforapplicationofanelectriceldtodewaterwaste clayslurrieswhichariseduetophosphatemininginCentralFlorida.Thewasteclaysarestored inlargesettlingpondsinCentralFloridawhichcover100squaremilesofland.Thesesettling ponds,alsoknownasclaysettlingareas(CSAs),takeasmuchas25yearstoachieveanacceptablesolidscontentof40wt%.Aproof-of-conceptbench-topexperimentwasperformedata cellpotentialof80V.Afterjust9hours,thesolidscontentapproached35wt%.Therefore,from thebench-topcell,whattypicallytakes25yearsthroughnaturalprocesses,wasinsteaddemonstratedtotakeonly9hoursfromapplicationofanelectriceld.Thisexperimentdemonstrated 93

PAGE 94

conclusivelythatsubstantialwaterremovalfromphosphaticclaysisachievableinthepresence ofanelectriceld. Amatrixofexperimentswereperformedfordifferentoperatingtimesandelectricelds. Trendsweredemonstratedasthechangeinsolidsweightpercentincreasedwithincreasing cellpotentialandoperationtime.Theexperimentalresultswereusedtodevelopconstitutive relationshipsforbothshortandlongoperatingtimestopredictthesolidscontentoftheclaysfor agivenoperatingtime.Amaximumsolidscontentwasreachedatlongoperationtimes.This maximumwasastrongfunctionoftheelectriceld.Forexample,anelectriceldof3V/cm wouldgiveamaximumsolidscontentof30wt%whileanelectriceldof0.1V/cmwouldallow only20wt%tobereached.Therefore,alargerelectriceldisneededtoreachanacceptable solidscontent. Simulationswereperformedtocalculatepowerandenergyrequirementsforwaterremoval oflarge-scaleCSAs.Thecalculationsincludedtheuseoftheconstitutiveequationsdeveloped fromtheexperimentalwork.Thepowerandenergyrequirementswerefoundtobeextremely large.Forexample,toachieveasolidscontentof25wt%thepowerrequiredwasontheorder of 10 7 kWandtheenergyrequiredwasontheorderof 10 9 kW-h.Thisresultsuggeststhat alternativedesignsshouldbeconsideredwhichscaledowntheamountofvolumetreatedata giventime.Thiswouldreducethenumberofelectrodesneededandthepowerrequiredtodrive theseparationofthewaterfromtheclays.Thedistancebetweenelectrodeswouldbesmaller allowingalargerelectriceldandcorrespondinglargermaximumsolidscontenttobeachieved. Energyrequirementsfromthebench-topexperimentswerecalculatedintermsofkW-h perkgofwaterremoved.Theresultsindicatedthatenergyrequirementsweresmallestat smallerelectricelds.Experimentalenergyrequirementsweremodeledseparatelyonthebasis offundamentalelectrochemicalrelationshipsandequation(7).Thiscomparisonshowed favorableagreementandveriedtheuseoftheconstitutiverelationshipforlongtimes(givenas equation(7))tocharacterizetheexperimentaldata. 9.2FutureWork Theresultsofthepresentworkareusedtoproposeadditionalstepsforthecontinuationof thisproject.Collaborationswithengineersfromtheprojectsponsor,MosaicFertilizerLLC,are proposedtoinvolvediscussionstodevelopdesignsforelectrokineticdewateringimplementation. 94

PAGE 95

Figure9-1.Schematicrepresentationofbench-scalesettlingbasinadaptedforelectrokinetically enhancedseparation. Discussionsareproposedtoincludeassessmentofwhetherdewateringshouldtakeplacewithin claysettlingareasoraspartofanin-lineprocessdewateringunit. BasedupontheeconomicassessmentpresentedinSection 8.2.2,furthermodelingand experimentalvericationisproposedformodieddewateringdesignsforclaysettlingareas aswellasdifferentin-lineprocessdewateringunits.Thethickener,presentedinFigure 8-8 wouldbebestsuitedformodelvericationtoexploreitsabilitytoforelectrokineticallyenhanced dewateringincombinationwithitsexistingfeatures.Experimentalvericationofsuchadesign wouldbeproposedafterthedevelopmentofsuchamodel. Otherworkproposedinvolvesasemi-batchoperationaspresentedinFigure 9-1.This designwouldinvolveacontinuousinletfeedstreamofadilutesuspensionofphosphaticclays withacontinuousoutletstreamofclearsupernatantwater.Theinletfeedstreamwouldoriginate directlyfromthephosphatebeneciationplantwhilethecontinuousoutletstreamofclear supernatantwatermaybeusedforrecyclebacktothebeneciationplant.Thesolidsremoved fromtheincomingsuspensionwouldbeaccumulatedwithinthebasin.Thistypeofdesigncould beretrottedtoanexistingclaysettlingareaforremediationoritcouldbeusedasanin-line processoperation.ItmayalsobeeffectivefortreatmentofanisolatedsectionofaCSAas presentedinFigure 8-7.Suchadesigncouldbestudiedthroughsimulations,butwouldrst involvetheoperationofabench-topexperiment.Theresultsofthebench-topexperimentswould giveinsighttowhichtypeofimplementationsuchadesignwouldbebest.Theoperationofthe bench-topcellisproposedtoexploretherelationshipsamongresidencetime,appliedelectric 95

PAGE 96

eld,andsolidscontentoftheefuentsteam.Turbiditymeasurementswillbeimplemented toallowreal-timemonitoringofefuentsolidscontent.Uponcompletionofthebench-top experiments,itisproposedthatascale-updemonstrationbebuiltandoperatedon-siteat Mosaicwhichwouldtestthescalabilityoftheelectrokineticdesignrelationshipsestablishedfrom thiswork. 96

PAGE 97

REFERENCES [1]D.Bloomquist,Centrifugemodelingoflargestrainconsolidationphenomenainphosphatic clayretentionponds,Ph.D.thesis,UniversityofFlorida(1982). [2]S.J.V.Kauwenbergh,J.B.Cathcart,G.H.McClellan,Mineralogyandalterationofthe phosphatedepositsofFlorida,U.S.GeologicalSurveyBulletin1914(1990)1. [3]R.F.Craig,SoilMechanics,6thEdition,E&FNSpon,AnImprintofChapman&Hall, London,UK,1974. [4]A.B.Fourie,D.G.Johns,C.J.F.P.Jones,Dewateringofminetailingsusingelectrokinetic geosynthetics,Can.Geotech.J.44(2007)160. [5]J.S.Newman,K.E.Thomas-Alyea,ElectrochemicalSystems,3rdEdition,JohnWiley& Sons,Hoboken,NewJersey,2004. [6]J.Morgan,CathodicProtection,2ndEdition,NACEInternational,Houston,TX,1993. [7]S.Speil,M.R.Thompson,Electrophoreticdewateringofclay,Trans.Electrochem.Soc.81 (1942)119. [8]M.H.Stanczyk,I.L.Feld,Electro-dewateringtestsofFloridaphosphaterockslimes,Bu. MinesRI6451. [9]O.Terichow,A.May,ElectrophoresisandcoagulationstudiesofsomeFloridaphosphate slimes,Bu.MinesRI7816. [10]H.J.Kelly,H.M.Harris,Electricaldewateringofdiluteclayslurries,Bu.MinesRI6479. [11]S.Glendinning,C.J.F.P.Jones,R.C.Pugh,Reinforcedsoilusingcohesivelland electrokineticgeosynthetics,Int.J.Geomech.5(2005)138. [12]A.Fourie,Harnessingthepower:opportunitiesforelectrokineticdewateringofmine tailings,GeotechnicalNews(June2006). [13]L.Yang,G.Nakhla,A.Bassi,Electro-kineticdewateringofoilysludges,J.Hazard.Mater. B125(2005)130. [14]J.Q.Shang,Electrokineticdewateringofclayslurriesasengineeredsoilcovers,Can. Geotech.J.34(1997)78. [15]B.Paczkowska,Electroosmoticintroductionofmethacrylatepolycationstodehydrate clayeysoil,Can.Geotech.J.42(2005)780. [16]O.Larue,R.Wakeman,E.Tarleton,E.Vorobiev,Pressureelectroosmoticdewateringwith continuousremovalofelectrolysisproducts,Chem.Eng.Sci.61(2006)4732. [17]J.Q.Shang,K.Y.Lo,Electrokineticdewateringofaphosphateclay,J.Hazard.Mater.55 (1997)117. [18]L.G.Bromwell,Physico-chemicalpropertiesofFloridaphosphaticclays,Tech.rep., FloridaInstitueofPhosphateResearch(1982). [19]C.Barnett,Mineeld,FloridaTrend:TheMagazineofFloridaBusiness(May2008). 97

PAGE 98

[20]G.N.Anastassakis,Phosphatesprocessing(chapter2.4),TAILSAFE:TailingsManagementFacilities-AStateoftheArtReport,editors:T.Meggyes,K.E.RoehlandD. Dixon-Hardy(2005). [21]P.M.Tyler,W.H.Waggaman,Phosphaticslimeapotentialmineralasset,Industr.Eng. Chem.46(1954)1049. [22]H.W.Long,D.P.Orne,Regionalstudyoflanduse,planning,andreclamation,Tech.rep., FloridaInstituteofPhosphateResearch(1990). [23]B.M.Moudgil,Enhancedrecoveryofcoarseparticlesduringphosphateotation,Tech. rep.,FloridaInstituteofPhosphateResearch(1992). [24]S.R.Weaver,Evaluationoftheuseofgeotextilesforcappingphosphaticwasteclay ponds,Master'sthesis,UniversityofFlorida(1985). [25]W.E.Pittman,J.W.Sweeney,State-of-the-artofphosphaticclaydewateringtechnologyanddisposaltechniques(intwoparts):1.Areviewofphosphaticclaydewatering research,Tech.rep.,FloridaInstituteofPhosphateResearch(1983). [26]H.El-Shall,P.Zhang,Processfordewateringandutilizationofminingwastes,Miner.Eng. 17(2004)269. [27]W.R.Reigner,C.Winkler,Reclaimedphosphateclaysettlingareainvestigation:Hydrologicmodelclaibrationandultimateclayelevationprediction,Tech.rep.,FloridaInstituteof PhosphateResearch(2001). [28]P.Zhang,RecoveryofphosphatefromFloridaphosphaticclays,Tech.rep.,Florida InstituteofPhosphateResearch(2001). [29]P.Zhang,M.Bogan,RecoveryofphosphatefromFloridabeneciationslimesI.Reidentifyingtheproblem,Miner.Eng.8(1995)523. [30]S.A.McClimans,Centrifugalmodelevaluationoftheconsolidationbehaviorof sand/phosphaticclaymixes,Master'sthesis,UniversityofFlorida(1984). [31]H.E.El-Shall,Developmentandevaluationofarapidclay-dewateringprocessasa reclamationtechnique,Tech.rep.,FloridaInstituteofPhosphateResearch(1996). [32]B.Scheiner,A.G.Smelley,D.Brooks,Large-scaledewateringofphosphaticclaywaste fromCentralFlorida,Bu.Mines8611(1981)1. [33]D.Znidarcic,A.N.Abu-Hejleh,T.Fairbanks,A.Robertson,Consolidationcharacteristics determinationforphosphaticclays,Volume1:Seepageinducedconsolidationtest equipmentdescriptionandusersmanual,Tech.rep.,FloridaInstituteofPhosphate Research(1994). [34]W.D.Carrier,Wasteclaydisposalinminecuts,Tech.rep.,FloridaInstituteofPhosphate Research(1982). [35]K.Fa,V.K.Paruchuri,S.C.Brown,B.M.Moudgilc,J.D.Millera,Thesignicanceofelectrokineticcharacterizationforinterpretinginterfacialphenomenaatplanar,macroscopic interfaces,Phys.Chem.Chem.Phys.7(2005)678. 98

PAGE 99

[36]A.Aysen,SoilMechanics,A.A.BalkemaPublishers,Lisse,TheNetherlands,2002. [37]P.L.Berry,D.Reid,IntroductiontoSoilMechanics,McGraw-HillBookCompany,London, UK,1987. [38]V.Mets,K.Amram,J.Ganor,Stoichiometryofsmectitedissolutionreaction,Geochim. Cosm.A.69(2005)1755. [39]R.E.Grim,ClayMineralogy,McGraw-HillBookCompany,1968. [40]A.McFarlane,K.Bremmell,J.Addai-Mensah,Improveddewateringbehaviorofclay mineralsdispersionsviainterfacialchemistryandparticleinteractionsoptimization,J. ColloidInterf.Sci.293(2006)116. [41]K.Ma,A.C.Pierre,Claysediment-structureformationinaqueouskaolinitesuspensions, ClaysClayMiner.47(1999)522. [42]K.Ma,A.C.Pierre,EffectofinteractionbetweenclayparticlesandFe3+ionsoncolloidal propertiesofkaolinitesuspensions,ClaysClayMiner.45(1997)733. [43]K.Ma,A.C.Pierre,SedmentationbehaviorofanekaoliniteinthepresenceoffreshFe electrolyte,ClaysClayMiner.40(1992)586. [44]A.K.Helmy,S.G.deBussetti,E.A.Ferreiro,Thesurfaceenergyofpalygorskite,Powder Technol.171(2007)126. [45]G.Prentice,ElectrochemicalEngineeringPrinciples,PrenticeHall,EnglewoodCliffs,NJ, 1991. [46]D.Pletcher,F.C.Walsch,IndustrialElectrochemistry,2ndEdition,ChapmanandHall, 1990. [47]D.Jones,PrinciplesandPreventionofCorrosion,2ndEdition,Prentice-Hall,UpperSaddle River,NJ,1996. [48]M.E.Orazem,B.Tribollet,ElectrochemicalImpedanceSpectroscopy,JohnWiley&Sons, Hoboken,NJ,2008. [49]A.J.Bard,L.R.Faulkner,ElectrochemicalMethods:FundamentalsandApplications, Wiley,NewYork,1980. [50]J.O.M.Bockris,A.K.N.Reddy,ModernElectrochemistry:Ionics,2ndEdition,Vol.1, PlenumPub,NewYork,1998. [51]J.O.M.Bockris,A.K.N.Reddy,ModernElectrochemistry:Electrodics,2ndEdition,Vol.2, PlenumPub,NewYork,2000. [52]J.Newman,K.E.Thomas-Alyea,ElectrochemicalSystems,3rdEdition,JohnWiley& Sons,Inc.,Hoboken,NJ,2004. [53]A.J.Rutgers,M.D.Smet,ElectrochemicalConstants,NationalBureauofStandards Circular524,1953,Ch.25.Electrokineticresearchesincapillarysystemsandincolloidal solutions,pp.263. 99

PAGE 100

[54]J.S.Newman,ElectrochemicalSystems,2ndEdition,PrenticeHall,EnglewoodCliffs,NJ, 1991. [55]A.Klinkenberg,J.L.vanderMinne,ElectrostaticsinthePetroleumIndustry,Elsevier, Amsterdam,1958. [56]H.V.Olphen,AnIntroductiontoClayColloidChemistry,2ndEdition,JohnWiley&Sons, 1963. [57]J.G.Sunderland,Electrokineticdewateringandthickening.1.Introductionandhistorical reviewofelectrokineticapplications,J.App.Electrochem.17(1987)889. [58]P.C.Hiemenz,R.Rajagopalan,PrinciplesofColloidandSurfaceChemistry,3rdEdition, MarcelDekker,Inc.,NewYork,NewYork,1997. [59]D.H.Gray,J.K.Mitchell,Fundamentalaspectsofelectro-osmosisinsoils,J.SoilMech. Found.Div.,ASCE93(6)(1967)178. [60]H.R.Kruyt,ColloidScience,ElsevierPublishingCompany,NewYork,NewYork,1952. [61]K.R.Reddy,A.Urbanek,A.P.Khodadoust,Electroosmoticdewateringofdredged sediments:bench-scaleinvestigation,J.Environ.Man.78(2006)200. [62]D.G.Buckland,J.Q.Shang,E.Mohamedelhassan,Electrokineticsedimentationof contaminatedWellandRiversediment,Can.Geotech.J.37(2000)735747. [63]M.H.M.Raats,A.J.G.vanDiemen,J.Lav en,H.N.Stein,Fullscaleelectrokinetic dewateringofwastesludge,ColloidsSurf.A:Physicochem.Eng.Aspects210(2002) 231. [64]H.Saveyn,G.Pauwels,R.Timmerman,P.V.derMeeren,Effectofpolyelectrolyteconditioningontheenhanceddewateringofactivatedsludgebyapplicationofanelectriceld duringtheexpressionphase,WaterRes.39(2005)3012. [65]K.Lo,Electro-kineticdewateringofFloridaphosphateclays,Tech.rep.,FloridaInstituteof PhosphateResearch(1995). [66]M.Y.Ho,G.Chen,Enhancedelectro-osmoticdewateringofneparticlesuspensionusing arotatinganode,Industr.Eng.Chem.Res.40(2001)1859. [67]W.Jin,Z.Liu,F.Ding,Brothdewateringinahorizontalelectriceld,Separ.Sci.Technol. 38(2003)767. [68]G.Maini,A.K.Sharman,C.J.Knowles,G.Sunderland,S.A.Jackman,Electrokinetic remediationofmetalsandorganicsfromhistoricallycontaminatedsoil,J.Chem.Technol. Biotechnol.75(2000)657. [69]J.E.McLean,ECOPACeldtestinFloridaphosphateslimepond,Tech.rep.,Monsanto Enviro-Chem(1986). [70]M.K.A.-E.Rahman,Dewateringofphosphaticclaywastebyocculation,Chem.Eng. Technol.23(2000)457. 100

PAGE 101

[71]S.Mathur,P.Singh,B.Moudgil,Advancesinselectiveocculationtechnologyforsolidsolidseparations,Int.J.Miner.Proc.58(2000)201. [72]A.G.Smelley,I.Feld,FlocculationdewateringofFloridaphosphaticclaywastes,Bu. Mines8349(1979)1. [73]W.D.Carrier,RapidclaydewateringphaseII:Field-scaletests,Tech.rep.,FloridaInstitute ofPhosphateResearch(2001). [74]M.P.Hughes,ACelectrokinetics:Applicationsfornanotechnology,Nanotechnol.11(2000) 124. [75]J.Q.Shang,W.A.Dunlap,Improvementofsoftclaysbyhigh-voltageelectrokinetics, JournalofGeotechnicalEngineering122(4)(1996)274. [76]H.A.Pohl,Dielectrophoresis,CambrigeUniversityPress,Cambridge,UK,1978. [77]M.Washizu,O.Kiurosawa,ElectrostaticmanipulationofDNAinmicrofabricatedstructures,IEEETrans.Ind.Appl.26(6). [78]M.Bjelopavlic,P.K.Singh,H.El-Shall,B.M.Moudgil,Roleofsurfacemoleculararchitectureandenergeticsofhydrogenbondingsitesinadsorptionofpolymersandsurfactants, J.ColloidInterf.Sci.226(2000)159. [79]D.Cabrera-Guzman,Theuseofelectrokineticsforhazardouswastesiteremediation,J. AirWasteManag.Assoc.40(1990)1670. [80]S.Dixit,R.Gombeer,J.D'hoore,Theelectrophoreticmobilityofnaturalclaysandtheir potentialmobilitywithinthepedon,Geoderma13(1975)325. [81]S.Gopalakrishnan,A.Mujumdar,M.Weber,Optimaloff-timeininterruptedelectroosmotic dewatering,Separ.Technol.6(1996)197. [82]A.Neaman,A.Singer,Theeffectsofpalygorskiteonchemicalandphysico-chemical propertiesofsoils:Areview,Geoderma123(2004)297. [83]B.D.Cullity,ElementsofX-rayDiffraction,3rdEdition,PrenticeHall,UpperSaddleRiver, NewJersey,2001. [84]J.P.McKinney,M.E.Orazem,Electrokineticmethodsfordewateringofphosphaticclay slurries,ECSTrans.19:26(2009)35. [85]J.Q.Shang,K.Y.Lo,I.Inculet,Polarizationandconductionofclay-water-electrolyte systems,J.Geotech.Eng.121(1995)243. [86]D.P.Riemer,M.E.Orazem,Modelingcoatingawswithnon-linearpolarizationcurvesfor longpipelines,in:R.A.Adey(Ed.),CorrosionandBoundaryElementMethods,Vol.12of AdvancesinBoundaryElements,WITPress,Southampton,UK,2005,pp.225. [87]D.P.Riemer,Modelingcathodicprotectionforpipelinenetworks,Ph.D.thesis,Universityof Florida(2000). [88]J.P.McKinney,Evaluationofabove-groundpotentialmeasurementsforassessingpipeline integrity,Master'sthesis,UniversityofFlorida(2006). 101

PAGE 102

[89]M.E.Orazem,J.M.Esteban,K.J.Kennelley,R.M.Degerstedt,Mathematicalmodelsfor cathodicprotectionofanundergroundpipelinewithcoatingholidays:Part1-theoretical development,Corros.53(4)(1997)264. [90]J.Morgan,CathodicProtection,2ndEdition,NACEInternational,Houston,TX,1993. [91]J.Wagner,CathodicProtectionDesignI,NACEInternational,Houston,TX,1994. [92]A.W.Peabody,ControlofPipelineCorrosion,NACEInternational,Houston,TX,1978. [93]J.F.Yan,S.N.R.Pakalapati,T.V.Nguyen,R.E.White,R.B.Grifn,Mathematical modelingofcathodicprotectionusingtheboundaryelementmethodwithanonlinear polarizationcurve,J.Electrochem.Soc.139(7)(1992)1932. [94]K.J.Kennelley,L.Bone,M.E.Orazem,Currentandpotentialdistributiononacoated pipelinewithholidaysparti-modelandexperimentalverication,Corros.49(3)(1999) 199. [95]C.Coras,N.Pebere,C.Lacabanne,Characterizationofathinprotectivecoatingon galvanizedsteelbyelectrochemicalimpedancespectroscopyandathermostimulated currentmethod,Corros.Sci.41(8)(1999)1539. [96]I.Thompson,D.Campbell,InterpretingNyquistresponsesfromdefectivecoatingsonsteel substrates,Corros.Sci.36(1)(1994)187. [97]D.Diakow,G.V.Boven,M.Wilmott,Polarizationunderdisbondedcoatings:Conventional andpulsedcathodicprotectioncompared,Mater.Perf.37(1998)17. [98]T.R.Jack,Externalcorrosionoflinepipe-asummaryofresearchactivities,Mater.Perf. 35(1996)18. [99]J.A.Beavers,N.G.Thompson,Corrosionbeneathdisbondedpipeline,Mater.Perf.36 (1997)13. [100]D.Riemer,M.Orazem,Developmentofmathematicalmodelsofcathodicprotectionof multiplepipelinesinarightofway,in:Proceedingsofthe1998InternationalGasResearch Conference,GRI,Chicago,1998,p.117,paperTSO-19. [101]D.P.Riemer,M.E.Orazem,Cathodicprotectionofmultiplepipelineswithcoatingholidays, in:M.E.Orazem(Ed.),ProcedingsoftheNACE99TopicalResearchSymposium: CathodicProtection:ModelingandExperiment,NACEInternational,Houston,TX,1999, pp.65. [102]C.J.Geankoplis,TransportProcessesandUnitOperations,3rdEdition,PrenticeHall, EnglewoodCliffs,NewJersey,1993. [103]W.L.McCabe,J.C.Smith,P.Harriot,UnitOperationsofChemicalEngineering,5th Edition,McGraw-Hill,Inc.,1993. [104]R.H.Perry,C.H.Chilton,ChemicalEngineer'sHandbook,5thEdition,McGraw-HillBook Company,1973. [105]P.Kos,Gravitythickeningofsludges,Ph.D.thesis,UniversityofMassachusetts(1978). 102

PAGE 103

BIOGRAPHICALSKETCH PatrickgraduatedfromtheUniversityofSouthCarolinawithaBachelorofSciencedegree inchemicalengineeringinMayof2004.WhileattheUniversityofSouthCarolina,Patrickdevelopedaninterestinelectrochemistrybyworkingonanundergraduateresearchprojectinvolving thecharacterizationofadirectmethanolfuelcellandbytakinganundergraduatecoursein corrosionengineering.HeenteredgraduateschoolinAugustof2004attheUniversityofFlorida andjoinedProfessorMarkOrazem'sresearchgroupwhichspecializesinelectrochemicalengineering.InMayof2006,PatrickcompletedaMasterofSciencedegree.Patrick'smaster'sthesis involvedsimulationsforcathodicprotectionofburiedpipelinestoimproveprioritizationofsurvey indicationsthatlocatecorrosioncausingcoatingdefects.Inthesummerof2006,Patrickwasthe NACE/CCTechnologiesSummerInternshipRecipient.Fromthisaward,Patrickwassponsored byNACE(NationalAssociationofCorrosionEngineers)tointernatDNVColumbus(formerly CCTechnologies).ThisinternshipallowedPatricktoparticipateinpipelinesurveyswhichhe hadpreviouslysimulatedinhismaster'sresearch.Inthefallof2006,Patrickreturnedtothe UniversityofFloridawherehebeganhisdoctoralresearch.Patrickhasrecentlyreceivedand acceptedajobofferfromIntel,locatedinHillsboro,Oregon.HisemploymentatIntelwillbeginin Julyof2010. 103