Eindhoven University of Technology MASTER Physical ... · The collected coal combustion fly ashes...
Transcript of Eindhoven University of Technology MASTER Physical ... · The collected coal combustion fly ashes...
Eindhoven University of Technology
MASTER
Physical-chemical upgrading and use of bio-energy fly ashes as building material in theconcrete industry
Doudart de la Grée, G.C.H.
Award date:2012
Link to publication
DisclaimerThis document contains a student thesis (bachelor's or master's), as authored by a student at Eindhoven University of Technology. Studenttheses are made available in the TU/e repository upon obtaining the required degree. The grade received is not published on the documentas presented in the repository. The required complexity or quality of research of student theses may vary by program, and the requiredminimum study period may vary in duration.
General rightsCopyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright ownersand it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.
• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain
EINDHOVENUNIVERSITYOFTECHNOLOGY
GraduationProjectPhysical‐chemicalupgradinganduseofbio‐energyflyashesasbuildingmaterialinthe
concreteindustry
ing.GuillaumeDoudartdelaGrée
27/6/2012
ing.GuillaumeDoudartdelaGrée page2
GraduationProject
Physical‐chemicalupgradinganduseofbio‐energyflyashesasbuildingmaterialinthe
concreteindustry
ing.GuillaumeDoudartdelaGrée
27/6/2012Author ing.G.C.H(Guillaume)DoudartdelaGréeAddress: Heereindsestraat5
5741RBBeekenDonkTheNetherlands
Studentnumber: S090076E‐mail [email protected] University: EindhovenUniversityofTechnologyAddress: DenDolech2
Postbus5135600MBEindhovenTheNetherlands
Tel: 31(0)40‐2479111Fax: 31(0)40‐2456087URL: http://www.tue.nl Supervisors: Prof.dr.ir.H.J.H.(Jos)Brouwers Dipl.Eng.M.V.A.(Miruna)Marinescu– FloreaURL: http://josbrouwers.bwk.tue.nl Internship: VanGansewinkelMineralsAddress: Loswalweg50
3199LGMaasvlakteRotterdamPostbus10163180AARozenburgTheNetherlands
Tel: +31181363099Fax: +31181362812URL: http://www.vangansewinkel‐minerals.nl/ Supervisor: ir.A.(Arno)Keulen
GraduationProject Preface
ing.GuillaumeDoudartdelaGrée page3
Preface"I would like to thankmy supervisors, startingwith: Prof. Dr. ir Jos Brouwers, for his inspiringlecturesandtheopportunitytotakepartinthisinnovativeproject.Dipl.Eng.MirunaFlorea,forhervaluableadviceandgenerousshareofknowledge,butmostlyforherenthusiasmthatmotivatedmeduringtheproject,theencouragementtouseandexploremyabilitiesandexpandmyboundaries.ir. Arno Keulen for supporting and motivating me in every practical way, sharing his practicalexperienceandhelpingmetocollectrelevantdataformyresearch.SpecialthankstoPeterCappon,for supporting and helpingmewhen I was working in the laboratory and had some new crazyideas.Togetherwewereinnovativeandsharedplentyofideasandbuiltdifferentexperimentalset‐ups. When looking back it was a significant learning process, during which I gained moreenthusiasmformaterialsciencethenIcouldeverimagine.Thereisasaying:Timeflieswhenyou’rehavingfun,wellformethatwasdefinitelytrue”.GuillaumeDoudartdelaGrée27‐6‐2012
GraduationProject Summary
ing.GuillaumeDoudartdelaGrée page4
SummaryTheaimofthisresearchwastocombinefourdifferentbio‐energyflyashesfrombio‐energypowerplantsinconcretemixtures.Therolesofthistypeofflyashinaconcretemixtureasabinder(partlyor totally replacingcement)or fillerwere investigated.First,physical andchemicalpropertiesofthe different fly ashes are determined followedby the determination of unwanted contaminantsandneededtreatmentmethods.Second,testsareperformedtoindicateanyimprovementandiftheuseof bio‐energy fly ash is promising.Todrawconclusions, theobtaineddata is comparedwithreferencecoalcombustionflyashtermedPKVASMZ.
Itwasfoundthatfromtheinvestigatedbio‐energyflyashes,thebio‐energyflyashesdeliveredbyHVC‐Alkmaar have self cementitious and pozzolanic properties and Twence‐Hengelo haspozzolanic properties. Physical and chemical properties like particle size distribution and oxidecompositionoftheflyashesvaryenormousamongeachotherandarealsonotreallycomparablewith PKVA SMZ. The same applies for the amount of unwanted contaminants like chlorides andcarbon.Thisall canhoweverbe relatedby theused fuel for the incinerationplantsand theusedburningprocesses.Byusingdifferenttreatmenttechniqueslikecrushing,thermallyandwatertreatingbio‐energyflyash,itistriedtoupgradebio‐energyflyashtoamaterialcomparablewithPKVASMZ.From the results in can be concluded that the thermal treatment and water treatment methodreduce unwanted contaminants and not only make the material not hazardous, but they alsoincreasethepotentialofthematerialascementreplacement.Aftertreatmentandcrushingthebio‐energyflyashperformssimilarlyascoalcombustionflyash.
GraduationProject TableofContents
ing.GuillaumeDoudartdelaGrée page5
TableofContentsPreface..........................................................................................................................................................................................3 Summary.....................................................................................................................................................................................4 TableofContents.....................................................................................................................................................................5 1. Introduction.....................................................................................................................................................................8 1.1 Problemstatementbio‐energyflyash........................................................................................................8 1.2 Problemdefinition...............................................................................................................................................8 1.3 Objectivesandresearchquestions...............................................................................................................9 1.4 Hypothesis...............................................................................................................................................................9 1.5 Generalapproach.................................................................................................................................................9 1.6 Structureofthereport.......................................................................................................................................9
2. Flyashoriginandtypes............................................................................................................................................10 2.1 Bio‐powerplantinstallations........................................................................................................................10 2.1.1 History...........................................................................................................................................................11 2.1.2 Fuel.................................................................................................................................................................12 2.1.3 Process..........................................................................................................................................................13
2.2 Theby‐productflyashingeneral................................................................................................................17 2.2.1 Coalcombustionflyash.........................................................................................................................17 2.2.2 Bio‐energyflyash.....................................................................................................................................19
3. Concrete...........................................................................................................................................................................20 3.1 Composition.........................................................................................................................................................20 3.1.1 Water.............................................................................................................................................................20 3.1.2 Cement..........................................................................................................................................................20 3.1.3 Aggregates...................................................................................................................................................20 3.1.4 Fillerandlegislation................................................................................................................................21
3.2 Propertiesofconcreteandinfluenceofflyash......................................................................................21 3.2.1 Freshconcrete...........................................................................................................................................21 3.2.2 Hardeningprocess...................................................................................................................................22 3.2.3 Hardenedconcrete...................................................................................................................................22
4. Bio‐energyflyashproperties.................................................................................................................................24 4.2 Mineralogicalproperties.................................................................................................................................24 4.2.1 Flyashesusedinthisstudy..................................................................................................................24
4.3 ParticleMorphology..........................................................................................................................................25 4.4 Physicalproperties............................................................................................................................................26 4.4.1 Density..........................................................................................................................................................26 4.4.2 Colour............................................................................................................................................................26 4.4.3 Carboncontent..........................................................................................................................................27 4.4.4 Fineness,PSDandSSA............................................................................................................................28 4.4.5 Moisturecontent.......................................................................................................................................31
4.5 Flyashpropertiesandlegislation...............................................................................................................31 4.5.1 Leachingvalue...........................................................................................................................................32 4.5.2 Flyashcharacteristics............................................................................................................................33 4.5.3 Landfill..........................................................................................................................................................33 4.5.4 Buildingmaterial......................................................................................................................................34 4.5.5 Regulationforflyash..............................................................................................................................36
5. Cement..............................................................................................................................................................................37 5.1 Composition.........................................................................................................................................................37 5.2 Thehydrationproducts...................................................................................................................................39
GraduationProject TableofContents
ing.GuillaumeDoudartdelaGrée page6
5.3 Hydrationprocess(Barron,2010)..............................................................................................................41 5.4 Hydrationproductsduringtime..................................................................................................................42
6. HydrationofPortlandcementwithFlyash......................................................................................................44 6.1 Tricalciumsilicateandflyash.......................................................................................................................44 6.2 Dicalciumsilicateandflyash........................................................................................................................45 6.3 EffectFlyashonthehydrationofC3AandC4AF...................................................................................45 6.4 Effectofflyashoncementhydration........................................................................................................46 6.4.1 C‐S‐Hcontent..............................................................................................................................................46
7. Effectsofcontaminantsoncementhydration.................................................................................................47 7.1 Carboncontent/lossonignitiontest.........................................................................................................47 7.2 Chloridecontent.................................................................................................................................................48
8. Treatmentofbio‐energyflyash............................................................................................................................50 8.1 Treatmentmethods...........................................................................................................................................50 8.1.1 Sieving...........................................................................................................................................................51 8.1.2 Thermaltreatment&Air‐filtering.....................................................................................................51 8.1.3 Metallicaluminiumremoval................................................................................................................52 8.1.4 Washing........................................................................................................................................................52 8.1.5 Seperation/Grinding...............................................................................................................................53
8.2 Treatmentevaluation.......................................................................................................................................53 8.2.1 Carboncontent.........................................................................................................................................53 8.2.2 Chloridecontentinwater.....................................................................................................................53 8.2.3 Chloridecontentofthesolidmaterial.............................................................................................53
9. Results..............................................................................................................................................................................55 9.1 Chlorideremoval(partone)..........................................................................................................................55 9.1.1 Twenceboilerflyash(B1)....................................................................................................................55 9.1.2 Twencecycloneflyash(B2)................................................................................................................57 9.1.3 Validationmeasurement.......................................................................................................................59
9.2 Chlorideremoval(parttwo)..........................................................................................................................59 9.2.1 Twenceboilerflyash..............................................................................................................................59 9.2.2 Twencecycloneflyash...........................................................................................................................60 9.2.3 HVCcycloneflyash..................................................................................................................................61 9.2.4 HVCFilterflyash.....................................................................................................................................62
9.3 Treatmentdataparticlesizedistribution................................................................................................62 9.3.1 Twenceboilerflyash..............................................................................................................................62 9.3.2 TwencecycloneflyashandHVCcycloneflyash.........................................................................63 9.3.3 ComparisonPSDandSSAbetweenoriginalandtreatedbio‐energyflyash...................64
9.4 Strengthdevelopment......................................................................................................................................64 9.5 Sideeffects............................................................................................................................................................69 9.5.1 Watertreatment............................................................................................................................................69 9.5.2 Thermaltreatment........................................................................................................................................69 9.5.3 Strengthresults..............................................................................................................................................70 9.5.4 Scanningelectronmicroscopy(SEM)...................................................................................................71 9.5.5 X‐raydiffractionpattern(XRD)...............................................................................................................72
10. Discussionsandconclusions..............................................................................................................................75 10.1 Relationsanddifferences................................................................................................................................75 10.2 Treatmentmethodsandreliability.............................................................................................................75 10.2.1 Carbonremoval.........................................................................................................................................75 10.2.2 Watertreatment.......................................................................................................................................76 10.2.3 Methodofmeasuringchloridecontent...........................................................................................76 10.2.4 Grinding........................................................................................................................................................77
GraduationProject TableofContents
ing.GuillaumeDoudartdelaGrée page7
10.2 Strengthresults...................................................................................................................................................77 10.3 Recommendations.............................................................................................................................................79
References................................................................................................................................................................................80
GraduationProject 1.Introduction
ing.GuillaumeDoudartdelaGrée page8
1. Introduction
1.1 Problemstatementbio‐energyflyashWorldwideincreasedconcernoftheCO2emissionsanddependencyfromfossilfuelsleadstoanincreasinguseofrenewableenergysourcesinordertoreplacetheuseoffossilfuelsandtodecreasethegreenhouseemissions.Oneofthoserenewableenergysourcesis100%biomass that can be used as a replacement of coal in power plants. These so called bio‐powerplantsusewastewoodasfuelintheircombustionroomtoproduceheat.Thatenergyisthenconsumedinsteamturbinestogenerateelectricityforownplant’sneedorsupplytothe electricity network of surrounding districts. This central method of electricitygenerationwasalreadyappliedduringtheindustrialrevolution.During that period of time, pulverized coal was used as fossil fuel and there was noawarenessoftheby‐productsexposedtotheenvironmentbytheexhaustgases.However,thoseby‐productsofCO2 emissionsandair‐polluting fly ashes increasedhealthproblemsnear the factories. When these problems were recognized, better installations wereconstructedtocollect flyashesusingcycloneseparators,electrostaticprecipitatorsorbaghousestofiltertheexhaustgases.Alsothecombustionroomprocessesimprovedresultingindevelopmentoflessby‐products.Thecollectedcoalcombustionflyashesweredisposedat landfill sites.However, those responsible for disposal of fly ash are constantly seekingpotential ash utilization options because of expensive costs and increasing production.Nowadays, flyashcanbeusedasgroundstabilizationunderroadsorapplied inconcretemixturesduetoitsphysicalandchemicalproperties.Flyashisafinematerialthatcanbeusedasafillerinconcretemixturesandbecauseofitspozzolanicactivityitcanbeusedasabinder and partly replace cement and therefore reduce the use of natural resources.However,flyashhasalsodisadvantagesbecauseofitscontaminantslikeheavymetalsandchlorides that can decrease the hydration degree of cement and reduce the strength ofconcretestructures.Theincreaseofusingbiomassforelectricitygenerationinpowerplantsleadstoproductionofflyashesdifferentfromthe‘oldfashion’coalcombustionflyash.Theseflyashescreatedfromburningbiomasshavedifferent characteristicsandproperties incomparison tocoalcombustionflyash.
1.2 ProblemdefinitionWithintheNetherlands,thereistheneedtoincreasetheknowledgeabouttheutilizationofbio‐energy fly ashes in concretemixtures. So far the reuseof this typeof fly ashes is notstartedbecauseofthefollowingreasons:
1. Bio‐energy flyash isa chemicallyandphysicallyvariableproductwhichmakes itscombinationwith cement questionable. Particle size distribution, loss on ignition,density, specific surface area, leaching as well as its pozzolanic/cementitiouspropertiesneedtobetestedtoconfirmitssuitability;
2. Bio‐energy fly ash contains contaminants like lead, zinc and chromium and largeamountsofchloridesthatmayhavenegativeinfluencesonthehydrationofcement;
3. As replacement of cement, its slow pozzolanic activity influences the hydrationprocess,consequentlyloweringthepropertiesofconcreteatearlyages;
GraduationProject 1.Introduction
ing.GuillaumeDoudartdelaGrée page9
4. Thereisnoworldwideagreementonthechemicalandphysicalinfluencesofflyashincombinationwithconcretemixtures,whichmakesitsapplicationstilluncommon;
5. So far there is no treatment method for contaminated fly ash to remove thecontaminantsandincreaseitpozzolanicactivitythatcouldmakeitsutilizationmorecost‐efficientandsustainable;
1.3 ObjectivesandresearchquestionsTheaimofthisresearchistocombinebio‐energyflyashfrombio‐energypowerplantsinconcretemixtures.Theroleofthistypeofflyashinaconcretemixtureasabinder(partlyortotally replacing cement) or filler will be investigated. The aim is to develop a moresustainable and cost‐efficient concrete that givesby‐products a second life andwith that,reducestheCO2emissionsofthecementproductionindustry.Inordertodoso,bio‐energyflyashesneedtobetreatedtoremoveunwantedparticlessothat itcanbeutilizedintheconcreteindustry.Thefollowingobjectivescanbedefinedinaccordancewiththeproblemdefinition:
1. Determinethephysicalandchemicalpropertiesofthedifferentflyashes;2. Remove unwanted contaminants in the bio‐energy fly ashes by treatment to
increasethereactivityinconcretemixtures.
1.4 HypothesisBymeansofabroadliteratestudy,thefollowinghypothesesareformulated:
1. After the treatmentmethoddescribed inParagraph8.1, theproperties of thebio‐energyflyashusedinconcretemixturesandmentionedinChapter4areimproved;
2. After washing and optimization the fly ashes will fulfil at least one of therequirementsforuseasabuildingmaterialinshapedornon‐shapedformasdefinedbytheBuildingMaterialDecree,Paragraph4.5.4.
1.5 GeneralapproachToanswerthedescribedobjectives,thisresearchisstructuredintotwoparts.Thefirstpart(Chapters 1‐7) is the theory based on literature study to obtain information, insight anddatathatcanbeusedtosetupthepracticallaboratorytestandtobetteranalyzetheresults.Thesecondpart(Chapters8‐11)consistsofthepracticalexperiments.Thispartdescribestheperformedlaboratorytests.Basedontheexperimentalresults,furtheroptimizationandtestsareperformedtomeetthestatedrequirementsandobjectives.
1.6 StructureofthereportThis report consists of twoparts. The first part is the theory that describes origin of thedifferent fly ashes, utilization, properties and legislation, basedon a literature study.Thesecondpartwilldescribe thepracticalpartwith thedifferent laboratory testsperformed,results and further optimization processes, ending by a discussion, conclusions andrecommendations.
GraduationProject 2.Flyashoriginandtypes
ing.GuillaumeDoudartdelaGrée page10
2. FlyashoriginandtypesInthisresearchdifferenttypesofflyashesareinvestigated.Eachpowerplanthasitsowntechnology of the bio‐energy fly ash generation. However, one thing they all have incommon is that the generated bio‐energy fly ashes cannot be reused because of itsconcentrationofunburnedcomponentsandharmfulsubstances(metalsandsalts).Togetmore insight about fly ash in general, background information about the origin of thematerial andproduction isvery important. In thischapter thebackground informationoftheflyashesisprovided.
2.1 Bio‐powerplantinstallationsThe types of bio‐energy fly ash used in this study were collected from the cyclone andelectrostaticprecipitatorsoftwodifferentpowerplantsintheNetherlands.Thereasonforthisapproachisthatflyashesgeneratedinpowerplantsareinherentlyvariablematerialsbecauseofseveralfactors.Amongthosearethetypeandmineralogicalcompositionofthefuel, degreeofpulverization, typeof furnaceandoxidationconditions including fuel ratioandthemannerinwhichflyashiscollected,handledandstoredbeforeuse.Sincenotwoutilitiesorplantsmayhaveallofthesefactorsincommon,flyashfromvariouspower plants is likely to be different. The following types of bio‐power plant fly ash areexamined:cycloneflyash(A1)andfilterflyash(A2)fromHVC‐Alkmaar,andboilerflyash(B1)andcycloneflyash(B2)fromTwence‐Hengelo.Usingseveraltypesofflyash,amoregeneralapproachoftreatmentandapplicationmightbefound.In HVC‐Alkmaar, 170.000 ton waste wood (dry biomass) is incinerated every year,comparedto140.000tonofwastewood incineratedbyTwence‐Hengelo.The incineratedwasteoftheHVC‐Alkmaaristhesameasthecontentof20960trucksthatdeliverthewaste,and reduces the CO2 emission by 100.000 ton per year, compared to fossil fuelwhich isequaltotheannualemissionof500.000cars.Thebio‐powerplantofHVC‐Alkmaardelivers25 MW electricity, which is equal to the electricity use of 60.000 households and theproduced heat can be used for 48.000 houses (HVC, 2009). An overview of the fuelincinerationandby‐productsfortheyears2008and2009isprovidedinTable1.
Table1:Fuelincinerationandby‐products(DWMA,2010)
Year InstallationIncineratedfuel[ton] Totalwaste
Bottomash[%]
Flyash[%]
RGR‐salts[%]
2008HVC‐
Alkmaar171.829 8624 54 25 21
2008 Twence‐Hengelo
137.440 12587 66 20 14
2009HVC‐
Alkmaar 176.000 6485 29 50 22
2009 Twence‐Hengelo
147.894 15766 66 22 11
GraduationProject 2.Flyashoriginandtypes
ing.GuillaumeDoudartdelaGrée page11
2.1.1 HistoryThe bio‐power plants described in this study are built recently. Twence‐Hengelo is inoperation since 2007 and a year laterHVC‐Alkmaar started to operate. The reason for adevelopmentofbio‐powerplantsistheincreasedconcernoftheCO2emissionsworldwide.
IntheKyotoprotocol,article3theclimatechange,orsocalledgreenhouseeffect,isoneofthereasonstoreducetheCO2worldwidebyatleast5%comparedtoreferenceyear1990intheperiod2008‐2012(UnitedNations,1998).Thismeansareductionof50MtonCO2 fortheNetherlands. The reduction is doneby reducing25MtonCO2within theNetherlandsand 25Mton CO2 outside the Netherlands because it is possible to reduce the obligationthroughactionsinforeigncountries.Accordingtothe‘UitvoeringsnotaKlimaatbleid1en2”thereductionneeds toberealizedcompared to thereferenceyear1990 (TweedeKamer,1999).In themidterm evaluation the reduction of 25Mton CO2 within the Netherlandswill bereducedto20MtonCO2withintheperiodof2008‐2012.InadditiontotheKyotoProtocol,theDutchgovernmenthasalreadyshapedapolicy ‘Derde‐energienota’ forthepromotionofrenewableenergy(Wijers,1995).Itwasstatedthat10%ofenergyconsumptionin2020shouldcomefromrenewableenergy.Furthermore, theEuropeanguideline(HetEuropeesParlementendeRaad,2001)ispromotingrenewablepower.Thedirectivewastoreachatarget of 9% renewable electricity by 2010. In 2009 this goal was reached and did notdeclinein2010.Accordingtothe“Landelijkafvalbeheersplan,2009’(VROM,2010)theaimof theDutchgovernment is tobeoneof theenvironmentally friendlyandenergyefficientcountriesinEuropeandeffectivelyusecertainexistingenergycontentinwaste.Themacroeconomicsurvey‘biobasedeconomy’performedintheNetherlandsonbehalfoftheEnergyTransitionPlatformforGreenMaterialsconcludesthatlargescaleapplicationofbiomass might have a huge environmental benefit and longer‐term positive economicimpactof5and8billioneurosperannum.Biomass is considered a sustainable and renewable resource that can replace fossil fuelslikecoalandgas.Beingapartof theCO2cycle, it reduces theCO2‐emissionbecauseof its'carbon neutral' origin. Thus, CO2 released in converting biomass into energy does notcontribute to the increaseof thegreenhouseeffect.Therefore,biomasscomplieswith theobjective of the government for the reduction of the CO2 emission and increasedindependency from fossil fuels. Furthermore, using waste as fuel, instead of using it aslandfill, is more favourable for the environment because of preventing the emission ofmethane from landfills. The landfill generated gasmethane is a greenhouse gaswhich isabout 20 times more harmful than CO2. Therefore, the amount of the land filling withbiodegradablewasteandhence,thelandfillgasemissionsarebeingreducedinrecentyears.Between1990 and 2006 the annual emission of landfillmethane has decreased bymorethan300Ktons,from572Ktonmethanein1990to257.6Ktonin2006.Thiscorrespondstoabout6MtonCO2inequivalent.The Copernicus Institute (Utrecht University), in cooperation with the LEI (WageningenUR),conductedastudyontheeconomiceffectsofbiomass.Thesurveyprovidessomelong‐term macroeconomic scenarios for the use of biomass for bio‐fuels, chemistry andelectricity generation (Banse et al., 2009). The scenarios aredesigned for situationswithhighandlowamountofbiomassimportedtotheNetherlandsfromEuropeancountriesandforsituationswithhighandlowlevelsoftechnologicaldevelopment.Thestudyencourages
GraduationProject 2.Flyashoriginandtypes
ing.GuillaumeDoudartdelaGrée page12
theministrytopursuethescenarioofhightechnologydevelopmentsinlarge‐scaleimportof biomass. The predicted effects of such a scenario by 2030 are: an additional annualturnover of 5 and 8 billion euros, 25%of the fossil fuels are being replaced by biomass,reducinggreenhousegasemissionsbyabout25%.
Everyyearthepotentialofbio‐powerplantsis increasedduetothereasonsstatedabove.AccordingtotherecentEnergyReport2011(CBS,2011),theNetherlandsislessdependenton fossil fuels and gradually switching to renewable energy. Figure 1 illustrates thebreakdownbysourceofthetotalrenewableenergyproductionbetween1990and2010.
Figure1:Renewableelectricity1991‐2010*Preliminaryresults.(CBS,2011)
2.1.2 FuelThe fuel used in the bio‐power plant of HVC‐Alkmaar and Twence‐Hengelo is a wastemainly consistingofwood.Thewoodenwaste canbedivided into threedifferent classes,namely A‐wood, B‐ and C‐wood. A description and example for each class of wood areprovided in Table 2. Not all these types of timber are suitable as fuel for the bio‐powerplantsbecauseoftheircomponents.Usuallythewoodwasteusedinthebio‐powerplantsconsists mainly of B‐wood. Other compositions, such as beam grass, cocoa husks andresidualproducts,nuts,shellsandkernelsarealsousedasfuel.AccordingtoVROM(2010)"theminimumstandardformanufacturingandprocessingofA‐andB‐woodisusefulapplication".Tobeuseful,theapplicationofthewoodenwasteshouldsatisfycertainrequirements:
By combustion, more energy is generated and recovered and then used in thecombustionprocess;aportionofthegeneratedenergymustactuallybeused,eitherimmediatelyintheformofheat,orafterconversionintheformofelectricity(C‐228/00,paragraph42);
Mostofthewastemustbeconsumedduringtheoperationandmostoftheenergygeneratedmustberecoveredandused(C‐228/00,paragraph43).Becausemostofthe waste must be incinerated, the waste has to consist of at least 50% organicmatter;
The installation that isusingthe fuelshouldnothave the function/statusofwasteincineration.
0.00
2.00
4.00
6.00
8.00
10.00
Percentage total energy production
[%]
Year
Other biomass combustion
For co‐firing biomass in power plants
Municipal solid waste incineration
Solar power
Biogas
Wind energy
Hydropower
GraduationProject 2.Flyashoriginandtypes
ing.GuillaumeDoudartdelaGrée page13
Thebio‐powerplantsystemmeetstheserequirements,becausemoreenergyisrecoveredthanwhatisusedforproduction.Recoverytakesplaceintheformofpartlyelectricityandpartlyheat.Inaddition,theusedbiomassconsistsofmorethan50%organicmaterial.Thesystem runs on A‐wood when wood biomass in the form of wood chips is no longeravailable.Thismeansthatthepowerplantisnotprimarilydesignedfortheincinerationofwaste, but to produce electricity and heat. This is very important because otherwise theclassAandBwoodcouldnotbeusedaccordingtotherequirementsdescribedin(VROM,2010).Sinceno fossil fuel is involved, theelectricity that isgenerated fromthebio‐powerplantsobtainsthelabelof100%greenenergy.
Table2:Classificationofwoodenwasteanddescription(VROM,2010)Class Description Example
Class‐A Unpaintedanduntreatedwood ‐ Beams‐ Staircomponents‐ Rafters‐ Battens‐ Pruning‐ Palletwood
Class‐B Woodproductsnotmentioned inclassesAandC,includingpainted,varnishedandgluedwood.
‐ Hard‐board‐ Soft‐board‐ Chipboard‐ Woodfibreboard‐ Pressedwood‐ Furniture(exceptrattan)‐ Paintedwood‐ Doorsandframes(withoutglass
andaluminium)‐ Notimpregnatedwood‐ DemolitionWood‐ Plywood
Class‐C Impregnatedwoodandpreservedwood(CCandCCA‐wood);CCAwoodalsocontainscopper,chromiumandarsenic,CCWoodcontainscopperandchromium,butnotarsenic.Timberwithotheragents(fungicides,insecticides,boron‐containingcompounds,quaternaryammoniumcompounds)inordertoprolonglife.
‐ GardenFencing/fenceparts‐ Sleepers‐ MeadowPoles‐ Greenwood(woodusedin
playgrounds)
2.1.3 ProcessThis section presents processes of wooden waste treatment for the HVC‐Alkmaar andTwence‐Hengelopowerplants.Becausethesebio‐powerplantscanhavedifferentprocessstages, an explanation is provided for the insight into the process of bio‐energy fly ashextraction from waste products. Figure 2 presents the overall process that can vary foreverypowerplantbutingeneraltheprincipleworksasfollows:Thewastetravelsfromthecombustionroomtotheboiler.Theoutputisaboilerflyashandastreamofairwithsmallparticles.Thissteamtravelstoanelectrostaticfilterwhereflyashiscollectedandthentoafibrefilterthatcollectsfurthersolidresidue.Afterthat,differentsteps of washing take place, additional water vapour with the clean gas is added andreleasedintheatmosphere.
GraduationProject 2.Flyashoriginandtypes
ing.GuillaumeDoudartdelaGrée page14
Figure2:exampleoftheprocessesinbio‐powerplantwithgrilloven(AfvalEnergieBedrijf,2006)
2.1.3.1SupplyWastewoodisdeliveredbyship,trainortruckandgoestothewarehouse.Sometimes,thewoodwaste suppliedby trainsandships firstneeds tobeunpackedbefore itgoes to thewastebunker.Thiscanbeperformedautomaticallybyspecialunpackingmachinery.
2.1.3.2WastebunkerIn the warehouse, fully automatic cranes equipped with a hydraulic polyp grab mix thesuppliedwaste in the bunker. This is an important step because the supply comes fromdifferent factories that process different products and in order to minimize this effect,mixingiscrucial.Afterwards,thewoodisdeliveredfromthebunkertothefunnels,wherethewasteslidesduetogravityintothecombustionchamber.
2.1.3.3CombustionchamberTherearedifferent typesof combustion chambersavailable for the incinerationofwaste.Thetwomainlyusedtypesarethegrilloven(furnace)andthefluidizedbedincinerators.
ThegrillovenusedattheTwence‐Hengelobio‐energypowerplantThe grill oven consists of the following devices: moving tiles for the transport of wastematerials;combustionzones;awaterbasinandanairsuctionsystem.The transport tilescanshiftandtumbleunderanangleovereachotherandthusmovethewastecomingfromthe funnels over the surface of the grill. Thewaste then undergoes various stages of thecombustionprocess,likedrying,degassing,andfinallyburningundertemperaturesaround850degreesCelsius.AnoverviewofthegrillisgiveninFigure3.Afteracombustiontimeofaroundonehour,thesolidcombustionresiduesleftonthegrate(bottomash)fallinabasinfilledwithwater.The liquid level in thebasin is regulated.Evaporatedwater is removedwiththesuctionsystemfor thecombustionair.Besidescoolingdownthebottomash, thebasinisusedtomaintainthepressureinthecombustionchamber(waterseal).Atthesametime,extraairiscomingthroughtheairsupplybelowthegrillstocoolthemdownandtoprovideoxygenforcombustion.Theamountofair,wastedosageandgrillcontroldevicearecontrolledbyanautomaticprocesscontrolsystem(automaticfiring).Therestoftheairisused as combustionair,which is functioningas a recirculationairblownover thegrill inorder to achieve proper turbulence and hence a better post‐combustion, improving the
GraduationProject 2.Flyashoriginandtypes
ing.GuillaumeDoudartdelaGrée page15
efficiency of the system. In somedesigns, the first part of these grills is cooledbywater,which improves the durability of the system and facilitates the combustion because thesuppliedairnowonlyfulfilstheroleofsupplyingoxygenandreducestheNOxemission.
Figure3:exampleofgrilloven(AfvalEnergieBedrijf,2006).Arrowsindicatehotandcoldairinlets
Inthedesignofthegrilloventhefocusshouldbeonthefollowingpoints:
Sufficient distribution capabilities of thewaste on the grill, so the covering of thewasteonthegrillshouldbeashomogeneousaspossible;
Thorough mixing of the waste on the grill, so that combustion takes place ashomogenouslyaspossible;
Equalandadjustableairdistributionacrossthegridzone,accordingly; Limitation of the primary combustion air and flue gases entraining dust and ash
particlestoreduceflyashproduction; Propercontroloptions for changing conditionsso thatanoptimal combustioncan
beachieved.
In general, a sufficiently high temperature (above 850 degrees Celsius), the presence ofoxygen(residualcontentofatleast6%inthefluegases),sufficientstandtime(atleasttwoseconds) and thoroughmixing of the flue gases should provide a proper burning of thewood.
FluidizedbedincineratorusedattheHVC‐Alkmaarbio‐energypowerplantIngeneral,thebiomassisfedtothefluidizedbed,whichcontainsalargeamountofsand(aninert, non‐combustible material). From the bottom combustion air is blown through thefluidized bed at high speed. The high volume of air passing the fluidized bed createsturbulence that ensures the complete combustionof the fuel particles.Also, byprovidingprimary and secondary combustion air for a staged combustion and recirculation, theformationof the fluegasNOx isreduced.Therefore, thecombustion in the furnacecanbecharacterizedasa"lowNOx"process.Acycloneremovesthesolidandunburnedparticlesfromthefluegasesandcarriesitbacktothebed.Therearetwomaintypesoffluidizedbeds.Inthefirstone,thevelocityischosensothatthesandandthefuel justperformabubblingmotion.Thiscanbecalledastationaryfluidized
Waterbasin
GraduationProject 2.Flyashoriginandtypes
ing.GuillaumeDoudartdelaGrée page16
bedorabubblingfluidizedbed(BFB).Inthesecondtype,thespeedoftheairflowisfurtherincreased creating flows that are carrying sand and fuel. Such an installation is called acirculating fluidizedbed (CFB).This system isused atHVC‐Alkmaar. Compared to aBFB,CFBhastheadvantagethatbythegreaterturbulencetheheattransferwillbehigher,thatmeans a lower flue stream resulting in a highly efficient system. Figure 4 illustrates theprincipleofafluidizedbedcombustionroom.ThedisadvantagesoftheCFBarethehigheruse of electric power due to the need for an increased airflow and the higher dustconcentrationinthefluegas.Mostofthedust,however, issimplyseparatedfromthefluegasinthecyclone.Unburnedparticlesfromthefluegasesaregoingbacktothecombustionroom. That process is controlled by the cyclone. After this, there is another cyclone thatcapturesred‐hotashparticlesandashparticlesgreaterthan10microns.Theflyashthatisremovedbythecycloneisstoredinclosedflyashsilos.
Figure4:principleofacirculatingfluidizedbedcombustionroom
2.1.3.4BoilerAfterthecombustionroomthefluegasespassfirstthroughthethreesections(theradiantsection)andthenthefourthsection(theconvectionpartoftheboiler),wheretheexistingheatisrecovered.Inthefirstthreesections,thegasesreleaseheatthroughradiationtothewatertubesresultinginasteamproduction.Thecoolingofthegasesoccurswithoutdirectcontactbetweenthegasstreamandtheheatedsurfacesof theboiler, inordertopreventcorrosion. The amount of steam generatedhas the conditions of 90bar and 500degreesCelsius.Duringthepassageoffluegasthroughtheboilertheradiationdecreasesandmoreconvectivetransfer takesplace.Duringthisprocessdirectcontactbetween fluegasesandpipesisrequired.Duringalltheseprocessesthetemperatureisreducingandpartoftheashinthegasisfallingdown.Thissocalledbottomashisthencollectedandstoredinstoragetanks.
2.1.3.5PowergenerationFromtheprocessdescribedabovehotwaterandsteamareproduced.Thesteamcanflowthroughaturbinetoproduceelectricity.Thehotwatercanbeusedfordistrictheating, inwhichcasetransfertakesplacethroughaheatexchanger.Also,thehotwatercangotoanevaporatortobeconvertedintosteam.Thissteamcanthenbetransportedtotheturbine.
GraduationProject 2.Flyashoriginandtypes
ing.GuillaumeDoudartdelaGrée page17
2.1.3.6FluegascleaningFor theseparationof flyash fromtheboileroutlet,anelectrostatic filter is installed.Thisfilterandthecombinationofthetwo‐wayashhandlingsystemoftheboilermakeitpossibleto clean flyash that canbe reused.For the removalofdust a fine fibre filter isused.Drypowder(CaOorunslakedlime,orNaHCO3orsodiumbicarbonate) is injectedontheclothfilter to absorb fine particles and powdered limestone is added to prevent fire andexplosions.Thenextcleaningstepistoremovetheacidiccomponentsandammoniafromthefluegases.Thisisdonebyawetcleaningprocesswithwater.
So,ingeneral,flyashesinthegrillovenarecollectedinthefollowingway:intheboiler“bigparticles”, intheelectrostaticfilterandintheclothfilter,“smallparticles”.Inthefluidizedbed incinerator, fly ash is collected as follows: in the cyclone, “big particles” and in theelectrostaticfilter,“smallerparticles”.
2.2 Theby‐productflyashingeneralSince ancient times, Romans, Chinese and Indians have used volcanic ashes and othersimilar natural andman‐madematerials to produce cementingmaterials bymixingwithlime,volcanicashesandpulverizedburnedbricks.Inthiswaytheyproducedcementitiousmortarsfortheconstructionofancientmonumentsthatarestillinexistencetoday(Joshi&Lohtia,1997).Theterm‘flyash’appearedinliteraturein1937.However,from1914,dataabout the use of finely pulverized powdered coal (fly ash) as a pozzolan in concrete hadbeenalreadypublished(Joshi&Lohtia,1997).Flyashcanbedescribedasafinematerialprecipitatedfromthestackgasesofburningsolidfuels. In Europe the ashwas always referred to as pulverized fuel ash, but in the UnitedStatesthisashwastermedflyashbecauseitescapedwiththefluegassesofcoalfuelpowerplantsand“flew”intotheatmosphere.Therearemanytypesofflyashproducedfromdifferentinstallations:
Coalcombustionflyash,from1882,designThomasEdison(NETL,2011); Bio‐energyflyash,from1971,(Dpcleantech,2011); Paper‐sludgeflyash,from1990,(CDEM,2011); Municipal SolidWaste Incineration (Destructor) fly ash, from 1874, design Albert
Fryer(Herbert,2011).Inthisparagraphthefirsttwotypesofflyasharediscussed.
2.2.1 CoalcombustionflyashDuring the industrial revolution, the production of coal combustion fly ash started. Theelectricpowerplantscreatedelectricity fromsteamproduction.Thesteamwasproducedby burning 63microns sized pulverized coal in a 1600degrees Celsius fired combustionroom(Hendriksetal.,1999)ThecreationofcoalcombustionflyashisillustratedinFigure5,itstartsfromthemomentwhenthemineralportionofcoalisheatedabovethemeltingpoint.Fromthatpoint,smalldrops are formed and when they are cooling down the formation of spherical fly ashparticlesbegins.Whenthetemperatureoftheparticlesisnotabovethemeltingpoint,lesssphericalflyashparticlesareformedandmorefusedparticlesaredetectedasillustratedintheleftpartofFigure6.Thisleadstolesssmallparticlesof10micronsandmoreporousflyash particles as illustrated on the right part of Figure 6 (CUR, 1992). In general, theproductionofallflyashesissimilartotheformationofcoalcombustionflyash.
GraduationProject 2.Flyashoriginandtypes
ing.GuillaumeDoudartdelaGrée page18
Figure5:formationofcoalcombustionflyash
Figure6:Left,“ideal”flyashandfusedparticles.Right,lesspherical:finecoalparticles
Later,whenairpollutionproblemswererecognized,better installationswereconstructedtocollectflyashusingcycloneseparators,electrostaticprecipitatorsorbaghousefiltersasdescribedinParagraph2.2.Dependingonthecollectionsystem,varyingfrommechanicaltoelectricprecipitatorsorbaghousesandfabricfilters,about85to99.9%oftheashfromthefluegasesisretrievedintheformofflyash.Flyashaccountsfor75to85%ofthetotalcoalash and the remainder is collected as bottom ash or boiler slag. The current annualproductionofcoal combustion flyash isabout500million tonsand forms75‐80%of thetotalashproductionworldwide.Only3to57%ofthisproductionisusedworldwide(Joshi,1979).Therest is landfilled,butthis isnoteconomicallydesirablebecauseofhigh landfillcosts,andenvironmentalriskssuchasleachingtotheground,therebycreatingwaterandalso air pollution. However, the government in the Netherlands aimed to reuse theproducedcoalcombustionflyashfromthebeginning.Thisgoalwasachievedin1988,whenfrom the annual production of 712,400 tons fly ash, 98% was reused. This percentagerepresents thehighestamount in theworld(CUR,1992). In2007,814.717tonsof flyashwereproduced:506.139 tonsof flyashwereusedascement fillerandconcretemixturesand88.054tonofflyashwereusedintheproductionofpozzolaniccements.Intotal, thisforms73%offlyashproduction(Vliegasunie,2008).Oneofthecementsiscalled“Portlandflyashcement”thatisclassifiedamongothertypesasdescribedinFigure7.
GraduationProject 2.Flyashoriginandtypes
ing.GuillaumeDoudartdelaGrée page19
Figure7:PortlandcementtypesconformingEN‐197‐1:Composition,specificationsandconformitycriteriafor
commoncements
2.2.2 Bio‐energyflyashTheincreaseofusingbiomass forelectricitygeneration inpowerplants leadsto flyashesdifferentfromthe‘oldfashion’coalcombustionflyash.Thebio‐energyflyashcreatedfromburning biomass has different characteristics and properties in comparison to coalcombustion fly ashes because of different combustion input and therefore, still need toprove their utilization in concrete mixtures. Their characteristics and properties will befurtherdescribedinChapter4.
GraduationProject 3.Concrete
ing.GuillaumeDoudartdelaGrée page20
3. ConcreteConcretehasbeenusedasabuildingmaterialforages.Itconsistsofrawmaterialssuchascement, aggregates like sand and gravel, water, additives and fillers. The properties ofconcrete can be characterized during three different phases: immediately after mixing“fresh concrete”, during the hardening process “young concrete” and the “hardenedconcrete”.Thesephasescanbedistinguishedfromeachotherbyvaluesofproperties(e.g.:density,humidity,strength)thathaveaninfluenceonthequalityoftheproducedconcrete.
3.1 CompositionConcreteiscreatedwhenwaterismixedwithcementandaggregates,additivesandfillers.Withthecorrectproportion,voidsinthemixarefilledwithsandandcementcombinesallthegrainsintooneoverallstructurewhenithydrateswithwater.
3.1.1 WaterWhenalargeamountofwaterisaddedtothemix,theexcessivewaterdoesnottakepartinthecementhydrationandsealsthecapillaryporestructure.Thisleadstoreducedstrengthandvulnerabilityofconcretetofreezing,thusreducingitsdurability.Incaseofdeficiencyofthewater, the cement cannot be completelyhydrated resulting again in the strength anddurabilitylossoftheconcreteproduct.
3.1.2 CementA large amount of cement would increase the strength development in concrete and itsdurability.However,cementproductiongeneratesCO2andisexpensivebecauseitconsistsofrawmaterialslikecalcite,clay,ironoxideandquartzneededtobefirstlyexcavatedandthen heated. Therefore it is important to use cement as little as possible (nomore thanneeded).Paragraph5.1describesthecompositionindetail.
3.1.3 AggregatesTheaggregatesconsistof:
Gravel8‐32mm; Finegravel4‐8mm; Coarsesand0‐4mm; Finesand0‐1mm; Stonepowder0‐0,125mm.
An ideal mix of aggregates would consist of an amount of small particles that would fitperfectly in the voids created by the bigger particles. Such a “perfect” mix however, isdifficult to produce and is not necessary because pore space is needed for the hydrationproducts. Still, fillersmostly finer than0.125mmareused for concretemixtures,by thatincreasing the consistency of cement. The reason for that is the enlarging of the particlesurface and thewater retention capacity. In the presence ofwater, all the interfaces aredrawntogether.Theparticlesseemtostickmoretogetherwhenfinerparticlesarepresentresulting intheincreasedconsistency.Appliedaggregatesandfillershavedirect influenceon the density of concrete. Depending on density, concrete can be divided into threecategories:
Lightconcrete,density2000kg/m3orlower;
GraduationProject 3.Concrete
ing.GuillaumeDoudartdelaGrée page21
Normalconcrete,densitybetween2000–2600kg/m3; Heavyconcrete,density2600kg/m3orlarger.
3.1.4 FillerandlegislationAccording toNEN‐EN‐206‐1 there are two types of fillers. The first type represents inertparticlesfunctioningasafillerwithoutanychemicalreactioninconcrete.Thesecondtypecharacterizesreactivefillersthathavepozzolanicorhydraulicproperties.Flyashisrelatedtothesecondcategory.Thepozzolanicfillersformreactionproductswiththeavailablecalciumhydroxide(Ca(OH)2orCH)thatiscreatedbyreactionofcementwithwater. Thehydraulic properties of the filler are developedby reactionwith the availablecementmineralsandformationofadenserporestructure,thusincreasingthestrengthanddurability(Joshi&Lohtia,1997).Becauseofthis,thesecondcategorycanfunctionnotonlyasafillerbutalsoasabinder.Thebindingfunctionofafillerisfurtherincorporatedintotheso‐called k‐value in the NEN‐EN‐206‐1. This k‐value is so far only available for the “old‐fashioned”coalcombustionflyashincombinationwithCEMI32.5N,CEMI32.5R,CEMI42.5 R, CEM III/A and CEM III/B. All other types should be implemented through acomparative study (equivalent concrete performance concept) to determine the binderfunction (see 5.2.5.2 and 5.2.5.3 of NEN‐EN‐206‐1). The k‐value is important because itdeterminestheamountofflyashthatcanbeseenasbinder(maximum1/3oftotalcement),andthereforereducestheminimumamountofcementrequiredbyNEN‐EN‐206‐1,whichiscost‐andCO2reducing.
3.2 Propertiesofconcreteandinfluenceofflyash
3.2.1 FreshconcreteConcrete isacontinuously transformingmaterial. Itstarts fromamixvarying from liquidformtoearth‐moist form,dependingon the functionandworkabilityof theconcretemix.The definition of workability is the total properties of concrete that are important for acorrecthandlingoftheconcretemixinthecircumstances.Forexample:
During a construction process with in situ concrete structures, the concrete should bepumpedtotheconcretestructureandequallydistributedalongthesurface,fillingallofthecorners and gaps. A lowpumping abilitymeans that the concrete cannot travel from thetruck to the concrete structure when long distances or heights should be over passed(Figure8).
Figure8:Importanceofthepumpingabilityandtraveldistanceofaconcretemixture
GraduationProject 3.Concrete
ing.GuillaumeDoudartdelaGrée page22
A lowworkabilitymeanspoordistribution of concretewhile a highworkabilitymakes itmore efficient. However, workability is not the only important parameter; the in‐situconcrete shouldnot segregateand theconsistencyshouldbehigh.Normally, forconcretewithahighplasticityandalowconsistencymostoftheheavygravelsmaysegregatetothebottom, thus resulting in a poormixture andpossibly affecting the concrete constructionnegatively,while a low plasticity and high consistencywill decrease its ability to equallydistributealongthesurface.Fly ashes canhave an effect on theworkability of the concretemixtureswhenused as afiller,becausetheirusemostlyincreasesthespecificsurfacearea(dependingontheparticlesizedistribution)resultinginahigherwaterdemand.Ontheotherhand,flyashusedasapartial replacement of cement can decrease the water demand, because less cement isneededsolesswaterisneededtohydrate.Ontheotherhand,itwillstillneedsomewaterifitpossessespozzolanicproperties,butlessthancementhydration(Wangetal.,2012).Flyashescanalso increase thewaterdemandbecauseof its limeandcoalcontent,whichadsorb water. Finally, fly ash can also increase the workability because of the availablecenospheres and plerospheres that increases the fluidity (rolling effect) of the mortars.Thereforeitseffectonthewaterdemandisstronglytypedependent.Fly ash influences the following aspects ofworkability:mix ability, transportability, flowbehaviour, compact ability, stability, pumping ability, finishing and green strength. Theseimportantaspectswillbeconsideredduringthepreparingoftherecipes.
3.2.2 HardeningprocessThehardeningprocessofcementpasteisanimportantphasethatdeterminestheconcretequalities.Inthisphaseconcreteneedstobesealedinordertoreducethewaterevaporation.In addition, during the hydration of cement, heat is released while increasing thetemperaturewithina largemassof concrete, since theheat isnotquicklydissipated.Theoccurred difference in temperatures can cause internal stresses which may result incracking. By using bio‐energy fly ash in concrete, less hydration heat is released at thebeginningbecauselesscementisadded.Bio‐energyflyashstartstohydratelaterduetotheslowerpozzolanicreaction.Therefore,thehydrationheatisspreadoutmoreovertime.Thisresults in less cracking but also in a decrease of early strength. Increasing the reactivity,resultinginafasthardeningcouldresultinanincreaseofearlystrength.
3.2.3 HardenedconcreteAlmostallpropertiesofconcretearelargelydeterminedbyitscomposition.Thepropertiesofconcreteare:compressivestrength,elasticdeformation(intermsoftheelasticmodulus),permeability,porosity,wear resistance,density, relativehumidity, frost resistance, colourandeventheappearance.Tocharacterizethepropertiesofconcretethefollowingfeaturescanbeused:
Cement:typeandamount; Aggregate:type,particleshape,particlesizedistributionandquantity; Additives:additives,fillers,dyesandfibres.
Thecompressivestrengthandsustainabilityarepropertiesofconcretethatareprescribedforeachproductandprojectbydesigninstitutes.
GraduationProject 3.Concrete
ing.GuillaumeDoudartdelaGrée page23
Thecompressivestrengthisimportantforthebearingcapacityoftheconstruction,whichisinfluencedbythestrengthoftheaggregateusedandbythechoiceofthewater‐cementratioandcementstrengthclass.Furthermore,thecompositionofconcretewillbedeterminedbyitsapplicationandrelatedworkabilityasmentionedinParagraph3.2.1.Foraconcretewallamore liquid concretemix is needed,while for a concrete brick an earth‐moist concreteconsistencyisneeded.Thereareseveralreasonstouseflyashes intheconcreteproduction.Coalcombustionflyashcanprovidecostsavings, improvedworkability,bettersurfacefinishing, lowerheatofhydration, improved long term or ultimate strength, reduced permeability, improvedsulphateresistanceandreinforcementcorrosionprevention.Ontheotherside,usingcoalcombustionenergyflyashinconcreteresultsindelayedstrengthgain,increaseddemandofair entraining agent with increasing carbon content in the fly ash, and slightly reducedresistance to scalingdue to saltsused forde‐icingon concrete roads. So far it isbelievedthatthesameprincipleappliestobio‐energyflyash.
GraduationProject 4.Bio‐energyflyashproperties
ing.GuillaumeDoudartdelaGrée page24
4. Bio‐energyflyashpropertiesFly ash can be used in concrete as a filler or partial replacement of cement due to itsparticle‐size distribution and pozzolanic properties as described in Paragraph 3.1.4.Pozzolansaredefinedas“siliciousandaluminousmaterials”whichinthemselvespossesslittleornocementitiousvalue.Howeverwill,infinelydividedformandinthepresenceofmoisture,chemically react with calcium hydroxide at ordinary temperatures to form compoundspossessing cementitiousproperties. Some fly asheswith self cementitious properties reactimmediatelywithwater(Joshi&Lohtia,1997).Themineralogicalcomposition,crystallineandnon‐crystallinephases,particlemorphologyas well physical characteristics define largely the pozzolanic reactivity of fly ash. Thephysical characteristics of fly ashwhich affect concrete performance are loss on ignition,fineness, moisture content and specific gravity, and pozzolanic activity. These are theprincipalparametersforpredictingtheperformanceofflyashinconcrete.
4.2 MineralogicalpropertiesFlyash consistsof aglassphaseanda crystallinephase.Thesephases canbe recognizedusing theX‐raypowderdiffractionmethod. If structuresare recognised, it is a crystallinephase. If themethoddoesnotdetect structures, it isaglassphase.Theglassphase is thereactive phase (pozzolan) consisting of alumina‐silicates. Firstly, Portland cement inconcreteneedstohydratewithwaterresultinginaformationofcalciumhydroxide(lime).Then it reacts with the alumino‐silicates presented in fly ash with the creation ofcementitiouscompoundspossessingadhesiveproperties.Incontrasttopozzolanicflyashes(ClassF according toAmerican standard testmethod (ASTM)C618), self cementitious flyashes (Class C according to ASTM C618) are able to hydrate almost in the sameway asPortlandcementdoes.Thedegreeofselfhardeninggenerallyvarieswiththecalciumoxidecontentoftheflyash.
4.2.1 FlyashesusedinthisstudyInthisstudy,sixflyasheswillbeusedasfollows:
Thefirstoneisacycloneflyash,providedbyHVCandtermedA1; ThesecondoneisafilterflyashprovidedbyHVCandtermedA2; Thethirdoneisaboilerflyash,providedbyTwenceandtermedB1; Thefourthoneisacycloneflyash,providedbyTwenceandtermedB2; Thefifthoneisacommercialtype,knownas:PKVASMZ(ClassF)andisusedasa
referenceflyashcomingfromthecoalcombustionprocessandistermedR.
Table3presentsthemost importantoxidespresent intheflyashes;thisisusedforoxideengineering.Inthismethodtheoxidesofdifferentflyashesarecomparedwithareference.This way an expectation can be made of how the material will react; the smaller thedifference,thebetterthecomposition.Thecalciumoxideamountislowerinbio‐energyflyash than in cement but higher than in PKVA SMZ (R). On the other hand the amount ofsilicate ismuchhigher inmostof thebio‐energy flyashescomparedtocementbut lowerthaninthereferenceflyash.ThiscombinationwillresultinalowC/Sratiowhichcanhavebothnegativeandpositiveeffectsonthefinalproduct,dependingontheratio.AddingsomegypsumwillincreasetheCaOconcentrationanddecreasetheC/Sratio.Ingeneralthebio‐energy flyasheshavemore similarities from thepointofviewofoxide compositionwithcementthanwiththereferenceflyash.
GraduationProject 4.Bio‐energyflyashproperties
ing.GuillaumeDoudartdelaGrée page25
Table3:OxidecompositionofdifferentflyashtypescomparedtoreferenceflyashtypeobtainedbyXRF
OxideHVCcycloneflyash(A1)
[%]
HVCfilterflyash(A2)[%]
Twenceboilerfly
ash(B1)[%]
Twencecycloneflyash(B2)[%]
CEMI42.5N[%]
PKVASMZ(R)[%]
MgO 2.7 2.1 2.7 3.2 1.6 1.9Al2O3 7.6 3.7 6.4 5.5 5.8 22.3SiO2 22.4 8.5 43.3 39.8 18.8 54.8SO3 12.3 12.6 6.5 8.3 4.5 1.4CaO 30.5 48.9 23.4 22.9 62.0 4.4Fe2O3 5.2 2.5 3.9 3.7 3.6 8.4
4.3 ParticleMorphologyWhileexaminingcoalcombustionflyashwithascanningelectron‐microscope(SEM)itcanbe seen that it consists of small particles that are typically spherical and fused particles(Paragraph 2.2.1). These spherical particles can be very useful in concrete because offunctioning as a lubricant between the irregularly shaped cement particles (CUR, 1992).Therearehoweverdifferentsphericalparticles;themostcommonaredescribedbelow:
Cenospheres, small sphericalparticles that arehollow,owing to an entrapmentofgasesbythemoltenphaseinthecourseofburning,representingabout20%oftotalflyash;
Plerospheres, the spherical hollow particles that contain entrapment of particlesinsteadofgases.
Due to the low density of cenospheres, they are valuable for low density concreteproduction. Figure9presents reference fly ashunder anOlympus SZX9microscopewithmagnificationof150xandsoftwarepackageAnalisySIS3.2.Fortheseparationofcenospheresandplerospheresfromtheflyashdropsofthewaterareused. The left picture illustrates reference fly ash under water and right picture on thesurface of the water. On the surface of the water large amount of cenospheres andplerospherescanbeclearlyseen.
ComparingFigure9withFigure10whichpresentsTwencecyclone flyash(B2) indicatesthatthebio‐energyflyashismuchcoarserandhasalmostnofloatingspheresbutinsteadfloating carbonparticles. Also theparticle size distribution is far greater than that of thereferenceflyashwhichcanindicatealowerpozzolanicactivity.
Figure9:Left,PKVASMZflyash(R)underwaterindicatingsmallspheresandminerals.Right,PKVASMZflyash(R)floatingonwaterindicatinglowdensityspheres
GraduationProject 4.Bio‐energyflyashproperties
ing.GuillaumeDoudartdelaGrée page26
Figure10:Left,Twencecycloneflyashunderwater(B2),indicatingcoarsematerials.Right,Twencecyclonefly
ash(B2)floatingonwaterindicatinglowdensityspheresandcarbonparticles
4.4 Physicalproperties
4.4.1 DensityFlyashcanbedistinguishedbyitsdensity.Tomeasurehedensity,theflyashisfirstdriedintheovenatatemperatureof105degreesCelsiusaccordingtoNEN‐12880for14hoursinorder to remove adsorbed water. Table 4 presents the different densities of fly ashmeasuredwithaMicromeriticsAccuPyc II1340gaspycnometerunderconditionsof20.6degreesCelsiuswiththreepurgesandacellvolumeof108.23cm3andexpansionvolume74.003cm3.Theobtaineddensitiesarequitehighcomparingtocoalcombustionflyashesthatarearound2.1‐2.4g/cm3butstilllowerthancement(±3.1g/cm3).ThelowerdensityofPKVASMZisprobablyduetothecenospheresandplerospheresasdescribedinParagraph4.3.
Table4:Densityofflyashfromthebio‐energypowerplants
HVCcycloneflyash(A1)
HVCfilterflyash(A2)
Twenceboilerflyash(B1)
Twencecycloneflyash(B2)
PKVASMZ(R)
Density[g/cm3] 2.73 2.59 2.65 2.68 2.36
4.4.2 ColourBio‐energy fly ash can alsobe classifiedby colour. This quality is important for aestheticreasonsbutcanalsobeusedtodistinguishparticlesthatarecontainingalargeamountofiron oxide and coal particles. These particles are dark, blackish in colour. Changes in theconcentration of these particles can affect the colour as can be observed from the nextsection“carboncontent”.Figure11presentsfourdifferentbio‐energyflyashesinvestigatedinthisresearch.TheflyashessuppliedbyTwencearemuchdarkerthantheonesfromHVC,whichindicatesahighercarboncontent.TheHVCflyashesaremorestickyandconsistingof clumped particles. The Twence cyclone fly ash has a consistency more like finedistributedpowder (particle size<0.125mm)and theTwenceboiler flyash– likemixoffinedistributedpowderandcoaldust.Foraesthetic reasons theamountof flyash that isused can have influence on the colour output of the concrete. In the pictures belowonlyunburned coal particles (black) can be distinguished from the samples, other unburnedparticleshave a similar colour as the rest of the sample and are thereforehard todetectvisually.
GraduationProject 4.Bio‐energyflyashproperties
ing.GuillaumeDoudartdelaGrée page27
4.4.3 CarboncontentThecarboncontentsofthefourbio‐energyflyashesarepresentedinTable5.TheTwenceboilerflyashhasalargeamountofunburnedcoalasalsoillustratedinFigure11andTable6.Thiscanbeduetothecombustioninstallationofthegrilloveninsteadofafluidizedbedoven,wherealessefficientcombustiontakesplace.ThefactthattheTwencecycloneflyashandHVCasheshaveacomparableamountoflossonignition(LOI),andisnotillustratedinFigure11, isprobablybecause theLOIofTwence ismostly carbon;while in theHVCashconsistofotherelements.AccordingtoNEN‐EN‐450theLOIisallowedsmallerthan5%.
Table5:LOIforthefourdifferentflyashesobtainedbyX‐rayfluorescence(XRF)from2011
Bio‐powerplant HVCcycloneflyash(A1)
HVCfilterflyash(A2)
Twenceboilerflyash(B1)
Twencecycloneflyash(B2)
PKVASMZ(R)
LOI2011masspercentage
[%]0.5 0.5 18.7 1.2 3.9
AsillustratedinFigure11,theTwenceflyashescontainblackcoalparticles.Toinvestigatetheamount,theflyashesarevisuallyobserved.Firstly,flyashesaresievedfrom500to125micronstoremovethe largeunburnedcoalparticlesandto investigatetheiramount.It isfoundthatTwenceboilerflyash(B1)consistsof13.1%coarsecarbonparticlesandthatthecarbonparticlesalsoremaininthelowersievediametersasillustratedinFigure12(visualanalysis)andpresentedinTable6(measured).However,theTwencecycloneflyash(B2)hasonly0.7%coarsecarbonparticles in the 500 micron sieve diameter and the carbon particles are almost entirelyfiltered out using the 125micron sieve diameter as illustrated in Figure 13 (visual) andpresentedinTable7(measured).Remainingflyash<125micronspresents41.5%oftotalamountofflyashwhichmeansthatitismuchmoresuitableforapplicationinconcretethantheB1basedonfinerparticlesthatarereactingfasterandhaveless/noncarboncontent.
Table6:Thesievediameter,masspercentageandcarboncontentforTwenceboilerflyash(B1)Sievediameter[µm] Masspercentage[%] Volumepercentage[%]
500 13.1 14.09300 19.2 19.06250 8.3 8.21200 14.9 14.78125 24.9 24.77<125 19.5 19.09Total 100 100
Figure11:Picturesof thefourdifferentflyashes(2010)WhereA1(=HVCcycloneflyash),A2=(HVCfilterflyash),B1(=Twenceboilerflyash),B2(=Twencecycloneflyash)
GraduationProject 4.Bio‐energyflyashproperties
ing.GuillaumeDoudartdelaGrée page28
Table7:Thesievediameter,masspercentageandcarboncontentofTwencecycloneflyash(B2)Sievediameter[µm] Masspercentage[%] Volumepercentage[%]
500 0.7 1.34300 5.7 21.45250 6.2 20.30200 11.3 19.56125 34.5 18.31<125 41.5 19.03Total 100 100
4.4.4 Fineness,PSDandSSAA larger amount of particles smaller than 45microns can result in a better packing andmorepozzolanicactivity.Morepozzolanicactivityisduetoreactionkinetics:finerparticlesreact faster, and is not due to the fact that finer particles consist ofmore spherical glassparticlesandthereforemorepozzolanicactivity(Fraay,1987).Alsotheworkabilityof theconcretemixwillbehigherbecausesmallerparticlesresultinahigherHägermannflowsize(CUR, 1992) as described in Paragraph 3.2. For other fly ash types a large percentage of
Figure13:SievedTwencecycloneflyash(B2)from500to<125µm
Figure12:SievedTwenceboilerflyash(B1) ashfrom500to<125µm
GraduationProject 4.Bio‐energyflyashproperties
ing.GuillaumeDoudartdelaGrée page29
particleslargerthanthe45micronshasbeenreportedtohaveanegativeeffectonthe28dayand90daystrengthsofnormallycuredPortlandcementflyashmortars(Mehta,1984;Diamond,1985).Fineness is one of the primary physical characteristics of fly ash that relates to itspozzolanic activity (Joshi, 1970).When examining fly ash for its particle sizedistribution(PSD)theNEN‐EN‐450setsthelimitof40%forthemaximumamountofflyashretainedonthe45micronmeshsieveonwetsieving,asaqualitycontrolmeasure.Figure14illustrates(PSD) for fivedifferent typesof flyashmeasuredwithaMastersizer2000using the laserdiffractionmethod.Figure15presentsthepercentageofparticlesthatarepassingthrougha sieve subdivided from 100% (everything passes) to 0% (nothing passes). The X‐axispresents theparticlesizedistributionwith increasingdimensions from left toright,whilethe Y‐axis represents a volume percentage from the total sample increasing from thebottomtotop.ThisdataissummarizedinTable8.
Figure14:Particlesizedistribution(PSD)ofthesixflyashes
Figure15:Cumulativefinervolumeoftheflyashes.Thelinesatsizeof45micronsandvolumeof60%indicate
thebreakpointbetweengoodandlesspositivequalityofflyashasdescribedbyNEN‐EN‐450
0
1
2
3
4
5
6
7
8
9
0 1 10 100 1000
Volume [%]
Particle Size distribution [μm]
HVC Cyclone ash (A1)
HVC filter ash (A2)
Twence Boiler ash (B1)
Twence Cyclone ash (B2)
PKVA SMZ (R)
0
10
20
30
40
50
60
70
80
90
100
0 1 10 100 1000
Cumulative finer vo
lume [%
]
Particle Size distribution [μm]
HVC Cyclone ash (A1)
HVC filter ash (A2)
Twence Boiler ash (B1)
Twence Cyclone ash (B2)
PKVA SMZ (R)
GraduationProject 4.Bio‐energyflyashproperties
ing.GuillaumeDoudartdelaGrée page30
Table8:Cumulativefinervolumetableofdifferentflyashes
HVCcycloneflyash(A1)
HVCfilterflyash(A2)
Twenceboilerash(B1)
Twencecycloneash
(B2)PKVASMZ(R)
d(min.)[µm] 0.7 0.7 1.7 2.2 0.6d(0.10)[µm] 10 3.8 18 8.7 5.0d(0.50)[µm] 38 13 110 110 25d(0.60)[µm] 47 19 138 130 30d(0.80)[µm] 75 40 202 187 106d(0.90)[µm] 110 69 255 235 120d(max.)[µm] 631 209 479 417 832
TheflyashesfromTwenceexaminedwiththeMastersizeraresievedtoamaximumparticlesizeof250microns.However,fromtheresultspresentedinFigure15,about10%hasstillalargersize.Thisisduetothefactthatsomeparticlesareneedle‐shapedsoiftheyarelyinghorizontallyonthesievetheywillbestopped,butfallingverticallytheywillslipthroughthesieve.As the graph illustrates, the fly ash of HVC is finer than the fly ash of Twence. This canrelatedtotheburningprocesswereHVCreducesthewasteforthecombustionroomintosmall fragments. As expected, the HVC filter fly ash (A2) has a larger amount of smallparticlesthanHVCcyclone(A1).Thiscanbeexplainedbythefactthatgasseswithflyashescoming from the combustion room are first collected with the cyclone (HVC) or boiler(Twence), collecting the bigger particles and after that the bag filter (HVC) or cyclone(Twence) collect the remaining fly ashes. However, the Twence boiler has finer particlesthentheA1whichingeneralwasnotexpected;thiscouldbemainlycarbondust(uselessingredientsofflyash).ComparingtheresultwithPKVASMZ(R),A1isabitcoarserwhileA2isabitfiner.Probablymixing these two fly ashes will create a fly ash that is almost identical on that of thereferenceflyash.TohaveaclearviewoftheparticlesScanningElectronMycroscopy(SEM)isperformed.The(SEM)usesafocusedbeamofhigh‐energyelectronstogenerateavarietyofsignalsatthesurfaceofsolidspecimens.Thesignalsthatderivefromelectron‐sampleinteractionsrevealinformation about the sample including external morphology (texture), chemicalcomposition,andcrystallinestructureandorientationofmaterialsmakingup thesample.InthisresearchaPhilipsXL30ESEM‐FEG,equippedwithGSE,SE,BSEdetectors,EDXSEMisusedtoobservetheexternalmorphologyoftheflyashes.InFigure16theimagesarepresentedofHVCcyclone(A1)situatedontheleft,inthemiddleTwenceboiler(B1)andontherightHVCfilter(A2).FromthisimagesitcanbeconcludedthatA1hasverysmallparticlescomparedtoB1,which isalsoconcluded fromthePSDinFigure14 andFigure15.However, as thePSDs indicates there is an increaseofparticlessmaller than 10micronswhen comparing HVC asheswith B1. According to (CUR, 1992)regardingcoal‐combustion flyashan increaseofparticlessmaller than10micronswouldindicate a higher amount of spherical particles. For the bio‐energy fly ashes this is notobserved indicatingagain thatbio‐energy flyash isofanentirelydifferentnature.This is
GraduationProject 4.Bio‐energyflyashproperties
ing.GuillaumeDoudartdelaGrée page31
even enhanced by the picture of A2 illustrating irregularly‐shaped particles even forparticlessmallerthan10microns.
Figure16:SEMimages.Left,HVCcycloneflyash(A1).MiddleTwenceboiler(B1).Right,HVCfilterflyash(A2)Forusingflyashasafiller,flyashshouldcontainsmallparticlesofsphericalformsinsteadofthefusedformsasmentionedinParagraph2.4.1.SofarA2fulfilsthisrequirementsandtherequirementsofNEN‐EN‐450formaximumretainedflyashonthe45micronsieveasdescribedinParagraph4.5.5.Thefinenessofflyashisalsodefinedbyaspecificsurfacearea(SSA)perunitofmass.Ifthesurfaceareaisverylargeaconsiderableamountofsmallparticlesforminglargeactiveareaareavailable.However,theeffectofincreaseinspecificsurfaceareabeyond6000cm2/gisreportedtobeinsignificant(Joshi&Marsh,1986).Table9presentsthespecificsurfaceareaoftheflyashesmeasuredwiththeMastersizer2000Ver.5.60.Thismethod,however,doesnotaccountthesurfaceassociatedwiththeshapeoftheparticles.AccordingtothismethodthespecificsurfaceareaofTwenceflyashesismuchsmallercomparedtothereferenceorHVCflyashes.
Table9:Specificsurfacearea(SSA)offlyashes
Bio‐powerplantHVCcycloneflyash(A1)[cm2/g]
HVCfilterflyash(A2)[cm2/g]
Twenceboilerflyash
(B1)[cm2/g]
Twencecycloneflyash(B2)[cm2/g]
PKVASMZ(R)[cm2/g]
Specificsurfacearea 1160 2690 382 559 2090
4.4.5 MoisturecontentIt is important that the ashes consist of less than 3% moisture to prevent caking andpacking,especiallyifflyasheshaveselfcementitiousproperties(Joshi&Lohtia,1997).Thewater can reactwith theseparticles forminghydrationproducts,which can in turn reactwith the remaining fly ashes. The fly ash is then less cementitious and pozzolanic activewhenusedinconcretemixtures.
4.5 FlyashpropertiesandlegislationFlyashes fromdifferentpowerplantshavevariableamountsof eachconstituentelementdue to the applied combustion process and fuel type. The fly ash is seen as awaste andshouldnotbehazardous inorder tobe landfilledandthebestoptionwhenpossible is toutilizeitinabuildingmaterialproduction.As described in Paragraph 2.1.2, the fuel used in the bio‐energy power plants mainlyconsistsofB‐woodandcanbecomposedofmanydifferentproducts (e.g.window frames
A1 B1 A2
GraduationProject 4.Bio‐energyflyashproperties
ing.GuillaumeDoudartdelaGrée page32
madefromtimber).However,theseproductsallhavetheirownmanufacturingprocessandfinishing (e.g. paint or impregnation). This has influence on the quantities of differentelementsinwoodandeventuallyinflyash.Thewoodiscomingfromtreesgrowingindifferentforests.Intheseforestsitgrowswhileaccumulatingsubstancesfromthesoil.Somespecieslikealder,birch,poplarorwillowcanabsorbtoxicmetalslikeAl,Cd,Cu,Fe,Mn,Mo,NienZn(PortofAntwerp,2000‐2006).Thisabilitycanevenbeusedtoimmobilizetoxicmetalsinground(termed:Phytoremediation)(Glimmerveen,1996).Thesoilcanforinstancebepollutedbytheexistinggroundwater.Farfromtheforeststhegroundwatercanbepollutedbyhumanandnaturalactivitiessuchasindustriesandcars,volcanoesandmines.Theoriginofthesecontaminantscanbeinitiatedfarbackintime.Thus, it is not obvious if fly ashes from bio‐energy combustion are less polluted than flyashes from coal combustion power plants. In the current chapter the fly ashes arecharacterized and the legislation providing hazardous limits of contaminant products isdescribed.Itisrequiredforlandfillorreuseofflyashtotakeoutthepotentiallydetrimentalsubstancesinordertoavoidanyharmfulinfluenceontheenvironmentandthehealthofthepeople.
4.5.1 LeachingvalueTo qualify the fly ash, a leaching test is performed. The leaching values for elements aredeterminedbyNEN‐5773usingacolumntest.Inthistesttheliquidflowsthroughthesolidsample. The amount of that liquid should be at least two times the volume of the solidsample, but less than ten times. The emission values stated in the Landfill Ban Decree(VROM, 2010) are defined by a relative amount of liquid to solid (L/S) equal to ten.However,thisamountishardtoobtainforpowdersamples.InthatcaseaL/Sequaltotwocanbeappliedandfinallyextrapolatedtoten(VROM,2007):
∗∗
∗ (1)
Where:EL/S=10:thecumulativeleachingofabuildingmaterial,soilorsludgeataratioofliquidand
solid(L/S)often;EL/S=y: thecumulativeleachingofabuildingmaterial,soilorsludgewithaL/Svalueequal
toy,whichisequalorhigherthantwo,butlowerorequaltoten;K: material‐dependent constant representing ameasure of the rate of leaching.The
valuesaregiveninTable10.Table10:K‐unitperelement
Element K Element K‐unitAntimony 0.04 Nickel 0.25Arsenic 0.01 Selenium 0.16Barium 0.17 Tin 0.10Cadmium 0.32 Vanadium 0.04Chrome 0.25 Zink 0.28Cobalt 0.13 Bromide 0.51Copper 0.27 Chloride 0.65Lead 0.18 Fluoride 0.26
Molybdenum 0.38 Sulphate 0.33
GraduationProject 4.Bio‐energyflyashproperties
ing.GuillaumeDoudartdelaGrée page33
4.5.2 FlyashcharacteristicsTable11presentsthemaximumandminimumvaluesofelementsfordifferentbio‐energypower plant fly ashes. These values are obtained with X‐ray fluorescence (XRF) andreportedaccording tostandard leaching testdescribed inAccreditationProgramBuildingMaterials Decree AP04 (Accreditatiecollege Bodembeheer, 2008). The samples for theconcentrationmeasurementsarecollectedtwiceperyearatthecorrespondingpowerplant.It canbe seen that foreachplant the concentrationof the components isvarying.This isbecauseofthedifferentfueltypes,thecombustionprocessandthecollectingpointofeachplant,asmentionedinChapter2.Thecontentofcertainelementsinsamplesisreportedinmilligramsofelementperkilogramofdrymatter(mg/kgds).
Table11:Elementalcompositionofdifferentflyashes,valuesobtainedbyXRFElement(symbol)
HVCcycloneflyash(A1)[mg/kgds]
HVCfilterflyash(A2)[mg/kgds]
Twenceboilerflyash(B1)[mg/kgds]
Twencecycloneflyash(B2)[mg/kg
ds]Aluminium(Al) 17000 32000 9900‐13000 13000‐30000Antimony(Sb) 110 67 28‐49 65‐300Arsenic(As) 94 66 16‐34 37‐160Barium(Ba) 160 150 80‐770 130‐780Bromide(Br‐) 420 130 <15‐59 15‐150Cadmium(Cd) 22 14 3.6‐10 11‐28Calcium(Ca) 270000 140000 40000‐96000 67000‐190000Chloride(Cl‐) 83000 23000 990‐5600 5300‐27000Chromium(Cr) 270 240 88‐130 130‐280Fluoride(F‐) 2.1 5 <2‐6.8 6.9‐100Potassium(K) 16000 12000 9500‐20000 9900‐20000Cobalt(Co) 20 21 13‐19 16‐23Copper(Cu) 890 830 210‐380 210‐620Mercury(Hg) 3.5 0.81 <0.05‐0.09 0.14‐2.5Lead(Pb) 3800 2400 370‐1100 970‐4200
Magnesium(Mg) 8500 9700 6600‐16000 10000‐16000Manganese(Mn) 1300 1200 590‐920 1100‐2600Molybdenum(Mo) 11 8.8 4.6‐5.8 7.1‐13
Sodium(Na) 11000 12000 4100‐6300 4500‐8100Nickel(Ni) 37 52 27‐42 37‐66
Selenium(Se) 4 4 <4‐<4 <4‐9.4Silicon(Si) 440 540 410‐4600 390‐1600
Strontium(Sr) 470 500 180‐340 320‐740Sulphate(SO42−) 6100 7200 9800‐15000 11000‐16000
Tin(Sn) 34 45 18‐38 33‐97Vanadium(V) 30 39 23‐27 39‐45Tungsten(W) 16 29 13‐35 29‐62Zinc(Zn) 3000 4600 1100‐3300 2700‐10000
pH 12.4 12.6 12.6‐12.9 12.6‐12.8
4.5.3 LandfillTheflyashcanbeseenasawasteorbuildingproduct.Bothproductsneedtofulfilcertainrequirements.TheLandfillBanDecree contains requirements that classifywaste streamsintoinert,non‐hazardousandhazardous.Ifthesampledoesnotfulfiltheserequirements,itcannot be used even for landfill, before it undergoes a certain treatment. The leachingvaluesofthedifferentflyashesfromthebio‐powerplantcomparedtotheclassifyingvaluesaccordingtothelandfilldecreearepresentedinTable12.Itcanbeseenthatalmostalltheleachedquantitiesofelementsintheflyasharenon‐hazardous.Onlythechloride,leadand
GraduationProject 4.Bio‐energyflyashproperties
ing.GuillaumeDoudartdelaGrée page34
sulphatecontentsarefarabovethelimitthattheycannotgotolandfillbeforehavingapre‐treatmenttoremoveunwantedelements.Thecleaningprocesscanbechemicaltreatment,immobilizationorwithwashingtechniques,aswillbedescribedinChapter7.Table12:ClassificationofInert,Non‐hazardous,HazardousandNo‐landfillelementsintheflyashes(L/S=10)
*Materialsthatfulfilthiscategoryfirstneedapre‐treatmentbeforedisposedonlandfill
4.5.4 BuildingmaterialIf the fly ash is intended for application in building production it should also fulfil otherrequirements stated in the Building Material Decree (BMD) (VROM, 2007). This decreedividesthebuildingmaterialsintodifferentcategories:
Shapedbuildingmaterials; Non‐shapedbuildingmaterialswithoutinsulation‐management‐control(Dutch:IBC‐
measures; Non‐shaped building materials that need IBC‐measures, this is an IBC‐building
material.(Thisistopreventleachingintotheenvironment).To distinguish if a building material belongs in the “shaped” group, the followingrequirementsshouldbesatisfied:
Thesmallestelementinthematerialshouldhaveavolumeofatleast50cm3; Thematerialneedstoinasolidshape.
Inotherwords, thebuildingmaterialsshouldconsistof largeshapedvolumesandshouldnotshowabrasion(wearing).
Elements
Inert
[mg/K
gds]
Non‐hazard
ous
[mg/K
gds]
Hazard
ous
[mg/K
gds]
No‐lan
dfill*
[mg/K
gds]
HVCcyclon
eflyash
(A1)
[mg/K
gds]
HVCfilterflyash
(A2)
[mg/K
gds]
Twenceb
oilerflyash
(B1)
[mg/K
gds]
Twencecyclon
eflyash
(B2)
[mg/K
gds]
(Sb) <0.06 0.06‐0.7 0.7‐5 5< 0.33 0.33 0.33 0.33(As) <0.5 0.5‐2 2‐25 25< 0.96 0.96 0.48 0.48
(Ba) <20 20‐100 100‐300
300< 16.73 88.88 1.30 2.07
(Cd) <0.04 0.04‐1 1‐5 5< 0.04 0.04 0.02 0.02
(Cl‐) <800 800‐15000
15000‐25000 25000< 21000 80000 5764 12628
(Cr) <0.5 0.5‐10 10‐70 70< 0.41 0.41 3.27 0.23
(F‐) <10 10‐150 150‐500
500< 9.30 8.09 4.57 17.35
(Cu) <2 2‐50 50‐100 100< 0.55 1.7 0.22 0.22(Hg) <0.01 0.01‐0.2 0.2‐2 2< 0.01 0.02 0.00 0.00(Pb) <0.5 0.5‐10 10‐50 50< 0.86 1.52 0.28 30.37(Mo) <0.5 0.5‐10 10‐30 30< 1.18 0.87 3.67 1.84(Ni) <0.4 0.4‐10 10‐40 40< 0.41 0.1 0.23 3.03(Se) <0.1 0.1‐0.5 0.5‐7 7< 0.43 0.21 0.23 1.22
(SO42−) <1000 1000‐20000
20000‐50000
50000< 15000 15000 25915 16545
(Zn) <4 4‐50 50‐200 200< 13.84 11 3.94 11.39
GraduationProject 4.Bio‐energyflyashproperties
ing.GuillaumeDoudartdelaGrée page35
Thematerialsthatdonotfulfiltheserequirementsareautomaticallyreferredtothe“non‐shaped” group. An example is ashes and granulates. If the materials cannot be appliedwithoutIBCmeasures,theybelongtothenon‐shapedIBCbuildingmaterials.TherequirementsoftheBMDencompassmaximumcompositionand leachingvalues.Theconsidered elements are those which are mainly available in buildingmaterials and caninfluencethesoilquality.Fly ash is a non‐shaped buildingmaterial because it is in powder form.When fly ash isappliedintoconcreteblocksitwillbecomeashapedbuildingmaterial.Thisconcreteblockwillneedtofulfiltherequirementsthatareassociatedwithshapedbuildingmaterials.InTable13theconsideredflyashesarecomparedwiththerequirementsforanon‐shapedbuildingmaterialdeterminedforL/S=2byBMD(VROM,2007).Thegreycolourhighlightsvaluesthataresufficientforusinginanon‐shapedbuildingmaterial.Thedarkgreycolourillustrates the elements that are above the limit. All these materials need to undergotreatmenttoremovethedetrimentalsubstances.However,ifitisimpossibletofulfilallthenon‐shaped buildingmaterial requirements it can be used in a shaped buildingmaterial,part of the elements will then be immobilized and therefore not hazardous substancesanymore(Eijk,2001).Ingeneraltherearethreesortsofmaterials:
Theonesthatfulfilallthisrequirementsandcanbeusedwithoutanytreatment; Theonesthatdonotfulfilallthisrequirementsandneedtobetreated; Theonesthatdonotfulfilalltheserequirementsbutcanbeputinashapedbuilding
material where the elements will be immobilized and fulfil the requirements forshapedmaterials.
Table13:Non‐shapedbuildingmaterialrequirementscomparedwiththeflyashvalues(L/S=2)
Elements Value(BMD)
A1(HVCF) A2(HVCC) B1(TB) B2(TC)
(Sb) 0.16 0.17 0.17 0.078 0.1(As) 0.9 0.48 0.48 0.1 0.73(Ba) 22 48.21 9.07 0.46 0.01(Cd) 0.04 0.02 0.02 0.01 0.1(Cr) 0.63 0.23 0.23 1.4 0.1(Co) 0.54 0.32 0.32 0.1 0.1(Cu) 0.9 0.96 0.31 0.1 0.001(Hg) 0.02 0.01 0.00 0.001 11(Pb) 2.3 0.83 0.47 0.1 1(Mo) 1 0.51 0.70 2 1.3(Ni) 0.44 0.23 0.23 0.1 0.42(Se) 0.15 0.11 0.23 0.078 0.2(Sn) 0.4 0.35 0.35 0.2 0.1(V) 1.8 0.43 0.43 0.1 5.2(Zn) 4.5 0.99 7.89 1.8 80(Br‐) 20 559.59 310.88 0.46 9200(Cl‐) 616 108432.54 28823.84 4200 7.6(F‐) 55 4.57 5.25 2 8300(SO42‐) 1.730 1993.42 2990.13 13000 0.1
GraduationProject 4.Bio‐energyflyashproperties
ing.GuillaumeDoudartdelaGrée page36
4.5.5 RegulationforflyashThe NEN‐EN‐450 describes legislations for coal combustion fly ash concerning fly ashproperties like particle size distribution and chemical composition. It is stated that thepozzolanicactivityof flyash isdeterminedbythecontentofSiO2(sand)andAl2O3(clay),andthatthereactiveformofSiO2shouldbeatleast25%(m/m).
Additionalrequirementsare:
Lossofignition ≤5%(m/m); Chloridecontent ≤0.10%(m/m)=1000(mg/kg); Sulphatecontent ≤3.0%(m/m)=30000mg/kg; Freecalciumoxide ≤1.0%(m/m)=10000mg/kg.
Physicalrequirementsareassessedasfollows.Atfirst,themaximumsievedresidueonthe45 micron sieve is ≤ 40% (m/m). Secondly, the strength loss by replacing 25% of thecementbyflyashinconcretemixturesshouldbelessthan25%after28daysand15%after90days.
GraduationProject 5.Cement
ing.GuillaumeDoudartdelaGrée page37
5. Cement
5.1 CompositionCementisusedasabinderinconcrete.Cementneedstomeetrequirementsthataredefinedin European standards such as NEN‐EN‐197‐1 and additional standards (NEN‐EN‐197‐4andNEN‐EN‐14216).RawmaterialslikeCaCO3(calciteintheformoflimestone),2SiO2.Al2O3(clay,shale),Fe2O3(iron oxide) SiO2 (quartz in the form of silica) are used in the production of Portlandcement.Therawmaterialsaregroundandmixedincertainproportionsandburnedat1450degreesCelsius formingparticlesknownas clinker.After that, the clinker is cooleddownandground toa finepowder. InFigure17 theprocessand formationof clinkerphases isillustrated.Finally,somegypsumisaddedtoformPortlandcement.The cement industry has its own notation for the common oxides. Table 14 provides adescriptionoftheoxidesandusednotations.
Figure17:Schematicdiagramshowingthevariationsintypecontentsofphasesduringtheformationof
Portlandcementclinker(Taylor,1997)
Table14:OxidesthatarefoundinPortlandcementwithcementchemistrynotationOxide SymbolCaO CSiO2 SAl2O3 AFe2O3 FSO3 SH2O HCO2 CMgO M
Theoxidesinteractwitheachotherformingseriesofmorecomplexcompounds.Table15describesfivemineralcompoundsthatusuallyareregardedasthemajorconstituentsofcementandwhichundergohydration.
GraduationProject 5.Cement
ing.GuillaumeDoudartdelaGrée page38
Table15:maincompoundsofPortlandcementCompounds ActualFormula Name MineralPhase
C3S 3CaO•SiO2 Tricalciumsilicate AliteC2S 2CaO•SiO2 Dicalciumsilicate BeliteC3A 3CaO•Al2O3 Tricalciumaluminate CeliteC4AF 4CaO•Al2O3•Fe2O3 Tetracalciumaluminoferrite FerriteCS CaSO4(•2H2O) Calciumsulphate Gypsum
The potential composition of Portland cement is based on the work of R.H. Bogue andothers(Neville,2004)referredtoas ‘Boguecomposition’ thatdescribesthepercentageofmaincompoundsincementby:x 3.043x (2)x 2.650x 1.692x x 4.072x 7.600x 6.718x 1.430x 2.852x x 2.867x 0.754x x 1.701x whereC is computed as the totalCaOminus the free lime (estimatedas0.7% fromENCImeasurementsin2012).ThegeneralcompositionofthemaincompoundsofcementisillustratedinTable16.
Table16:CompositionlimitsofPortlandcement(Neville,2004)Oxide Content[%]CaO 60‐67SiO2 17‐25Al2O3 3‐8Fe2O3 0.5‐6.0SO3 2.0‐3.5MgO 0.5‐4.0Alkalis 0.3‐1.2
AsummaryofthepatternofformationandhydrationofcementisgiveninFigure18.
Figure18:Patternofformationandhydrationofcement(Taylor,1997)
Chemically, the hydration of Portland cement consists of a series of reactions betweenindividual clinker minerals, calcium sulphate and water, which proceed both
Component elements
O Si Ca Al Fe
+ impurities (K,Na, Mg etc.)
Component oxides
CaO SIO2 Al2O3 Fe2O3 SO3
Cement minerals
C3S C2S C3A C4AF
Portland Cements
(various types of Portland cement)
Hydration products
C‐S‐H gel Ca(OH)2 AFm AFt
CS
GraduationProject 5.Cement
ing.GuillaumeDoudartdelaGrée page39
simultaneouslyandsuccessivelyatdifferentratesand influenceeachother.The followingfactorsdeterminethekineticsofthehydrationprocess:
Thephasecompositionoftheclinkerandthetypeandquantityofforeignionsincorporatedinthecrystallinelatticesoftheindividualclinkerminerals;
Theprocessinghistoryoftheclinker,includingtheheatingrate,maximumburningtemperatureandcoolingrate;
Thequantityandformofcalciumsulphatepresentinthecement; Thefinenessofthecement; Thetechnologyemployedforcomminution(processinwhichsolidmaterialsare
reducedinsize,bycrushing,grindingandotherprocesses)ofthecement; Thewater/cementratioofthemix; Curingconditions(airorwatercuring); Thehydrationtemperature; Thepresenceofchemicaladmixturesinthemix.
Thehydrationprogressdependson: Therateofdissolutionoftheinvolvedphases; Therateofnucleationandcrystalgrowthofthehydratestobeformed; Therateofdiffusionofwateranddissolvedionsthroughthehydratedmaterial
alreadyformed.
5.2 ThehydrationproductsHydration occurs in different forms and there is a divergence of opinions betweenresearchersaboutthechemicalequations.Thisdependsontheconditions,cementtype,ordifferent analyze techniques, temperatures etc. Furthermore the complete hydrationprocessisnotclearlyknown,alotofdifferentresultsarefound.Belowashortdescriptionisgiven for the main hydration process as in reality much more will happen because ofdifferentcompositionsinthecementcompoundsandinteractionbetweeneachother.Twoprimarymechanismsoccurwhenhydrationhappens.
Through solution involves dissolution of anhydrous compounds to their ionicconstituents, formation of hydrates in solution, and eventual precipitation due totheirlowsolubility;
Topochemicalorsolid‐statehydration‐reactionstakeplacedirectlyatthesurfaceoftheanhydrouscementcompoundswithoutgoingintosolution.
Whenwater isaddedtocement, thedissolutionofcementgrainsoccurs.Thisresults inagrowing ionic concentration in “water”which is now a solution. The ionic concentrationforms compounds in the solution and after reaching their saturation point, compoundsprecipitateoutashydrationproducts(solids).TodescribethehydrationofsilicatesandaluminatesareviewofBrouwers(2005)isused,whichcompared60typesofcementtocreateageneralmodelpresentedinTable17.
Table17:Mainhydrationproductsofcementat100%RHName Equation Mainhydrationproduct
Tricalciumsilicate C3S+4.5HC1.7SH3.2+1.3CH C‐S‐Hgel,CHDicalciumsilicate C2S+3.5HC1.7SH3.2+0.3CH C‐S‐Hgel,CH
Tricalciumaluminate C3A+CS+14HC4ASH14 SO4‐AFmTricalciumaluminate C3A+CH+21HC4AH22 Hydroxi‐AFmTricalciumaluminate C3A+3CS +36HC6AS3H36 AFt(Ettringite)
Tetracalciumaluminoferrite C4AF+2C3S+22HC6AFS2H18+4CH C‐A‐F‐S‐Hhydrogarnet,CHTetracalciumaluminoferrite C4AF+2C2S+20HC6AFS2H18+2CH C‐A‐F‐S‐Hhydrogarnet,CH
GraduationProject 5.Cement
ing.GuillaumeDoudartdelaGrée page40
In this model, the probability of C‐A‐F‐S‐H hydrogarnet and CH formation is included,formedfromthereactionbetweenferritephasesandcalciumsilicates.Thisbecauseferritehasbeenfoundinhydrogarnetsandtheresultingcalculatedproductscouldbeformedbythe hydration of Tetracalcium alumina ferrite and calcium silicates in cement, which isaccordingtotheauthormorelikelythanthereactionofC4AFwithwaterand/orCS.
5.2.1 TricalciumsilicateOver60%bymassofmostcementscomprisesthetricalciumsilicate(C3S).Thismineralisthemostimportantoneasitcontrolsthesettingandhardeningofcement.Incombinationwithwater,C3Shydrates into twohydrationproducts,CalciumSilicateHydroxide (C‐S‐H)and calcium hydroxide (known as CH or Ca(OH)2). During the hydration, CH forms withwater calcium and hydroxide. During the acceleration period see also Paragraph 5.1.2,Ca(OH)2 concentration in the liquid phase attains a maximum, at this time it begins todecline,crystallinecalciumhydroxidestartstoprecipitateasdescribedby:C3S+H2OC‐S‐H+Ca(OH)2 (3)ThehydrationofC3Scanbeacceleratedbytheadditionofanhydrite,gypsumorfineinertcalciumcarbonate.
5.2.2 DicalciumsilicateOver 20%bymass ofmost cements comprises dicalcium silicate (C2S). The hydration ofdicalciumsilicateisassociated(sameastricalcium)withthereleaseofcalciumandsilicateions into the liquid phase (water) and is similar to that of C3S even though the wholeprocess progresses more slowly. However, (for example) the hydration amount issignificantlowerthanthatofC3Swiththesamedegreeofhydration.
5.2.3 TricalciumaluminateTheother10%ofhydrationproductsarealuminates(comingfromC3A)intheformofmainproductsofhydroxy‐AFm,monosulfo‐aluminateandettringite.Despite their littleamountthese products are crucial for cement because the reactions with them are affecting thehydrationofthecalciumsilicatesphases.ComparingtoC3S,thehydrationofC3Aisveryfast.However, a very rapid hydration of C3A prevents cement to develop sufficient strength.Thereforegypsumintheformofcalciumsulphatedihydrate(CaSO42H2O)isaddedtoslowdownthehydrationspedofC3A.TherearetwomainhydrationproducttypesgeneratedbythereactionofC3Awithwaterinthepresenceofcalciumsulphates,AFmandAFtdescribedinequationfour.C3A+3CaSO4+36H2O→C6AS3H36 (4)C3A+CaSO4+14H2O→C4ASH14ThemostcommonlyconsideredAFmcompoundismonosulfate,C4ASH14.TheAFtphaseisrepresented by ettringite, C6AS3H36. Ettringite crystallizes in the form of well‐developedprismaticoracicularcrystals.Monosulfatecrystallizesintheformofthinhexagonalplates.
GraduationProject 5.Cement
ing.GuillaumeDoudartdelaGrée page41
5.2.4 TetracalciumaluminoferriteTheremaining10%ofthecompositionofCEMIisC4AF.Thereactivityoftheferritephasemay vary in awide range anddependon theAl/Fe ratio.As soon as there is insufficientamountofgypsumtoconvertalloftheC4AFtoettringite,silicatehydrationissloweddown.ThereactionmechanismoftheferritephaseimpliesthatinthepresenceofferritelessC‐S‐HandmoreCHareformedthanwithoutferrite(Brouwers,2005).ThroughthehydrationofC4AFahydrogarnetwillbeformed,asexplainedinParagraph5.2.
5.3 Hydrationprocess(Barron,2010)ThehydrationofPortlandcementmaterialsisexothermic.Thisvariationofhydrationheatcan bemonitored using a calorimeter,which helps to visualize the heat evolution curve.ThiscurvecanbedividedinfourdifferentstagespresentedinFigure19.
Figure19:Rateofheatevolution(Mindess&Young,1981)
1. “Pre‐inductionperiod”(lastingafewminutes)
Immediatelywhencementgets in contactwithwater, a rapiddissolutionof ionic speciesintotheliquidphaseandtheformationofhydratephasesstarts.Alkalisulphatespresentinthecementdissolvecompletelywithinseconds,contributingK+,Na+andSO42‐ions.Calciumsulphates (gypsum)dissolvesuntil saturation, contributingCa2+ andadditional SO42‐ ions.C3SdissolvesandalayerofC‐S‐Hphaseprecipitatesonthecementparticlesurfacewhicheventually slows down the hydration of C3S and C2S. As the C/S ratio of the producedhydrate is lower than that of C3S, the formation of C‐S‐H phase is associated with anincreaseoftheCa2+andOH‐concentrationintheliquidphase.Atthesametime,silicateionsenteralsotheliquidphase,althoughtheirconcentrationremainsverylow.ThefractionofC3Shydratedinthepre‐inductionperiodremainslow(2to10%).C3Adissolvesandreactswith Ca2+ and SO42‐ ions present in the liquid phase (mostly from gypsum), yieldingettringite (AFt) thatalsoprecipitateson thecementparticlesurfaceandeventually slowsdownthehydrationofC3SandC2S.C3Ahydratesinthepre‐inductionperiodaround5‐25%.The concentration of Al3+ in the liquid phase remains very low. C2S reacts in the pre‐inductionperiod,yieldingaC‐S‐HphaseandcontributingtotheCa2+andOH‐concentrationintheliquidphase.The early fast hydration reaction appears to be slowed down due to the deposition of alayer of hydration products at the cement grain surface. In thisway a barrier is formedbetween the non‐hydrated material and the bulk solution, causing a rise in theconcentration of dissolved ions in the liquid phase in immediate contact with the non‐
GraduationProject 5.Cement
ing.GuillaumeDoudartdelaGrée page42
hydrated material to values approaching the theoretical solubility of the anhydrouscompound.ThemainproductsinthisstageareettringiteandCH(Hewlett,2004).
2. “Inductionperiod”(orDormantstage)Afterashortperiodoffasthydration,theoverallhydrationrateslowsdownforaperiodofa few hours. In this period the hydration of all clinker minerals progress very slowly.Different authors describe different theories about the induction period. Taylor (1997)describesthefollowingtheories:“Impermeablehydratelayertheory”,“Electricdoublelayertheory”, “Nucleation of Ca(OH)2 theory”, “andNucleation of C‐S‐H theory”. In general the“Impermeablehydrate layertheory”assumesthecreationofabarrier layer thatpreventsthereactionofnonreactedcementbecauseof thepreventionofwaterpenetration to theinorganicoxide in the cementgrains. In theendof theDormant stage, cracksare formedanddissolutionofthelayerensuresthathydrationaccelerates.Thistheoryofbarrierlayerformationisworldwideexcepted(Taylor,1997).
3. “Accelerationperiod”In this period the progress of hydration accelerates again and is controlled by thenucleation and growth of the resultant hydration products. The rate of C3S hydrationacceleratesandthe'second‐stageC‐S‐H'startstobeformed.AnoticeablehydrationofC2Sstarts.Becauseofthecalciumhydroxideintheliquidphase,theconcentrationofCa2+intheliquid phase gradually declines. The calcium sulphate becomes completely dissolved andtheconcentrationofS042‐ inthe liquidphasestartstodecline,duetotheformationof theAFtphase.Inthisperiod30%ofthecementishydratedformingCHandC‐S‐HduetoC3Swhichislargelyavailableincement.
4. “Decelerationperiod”(Post‐acceleration)In this period the hydration rate slows down gradually by the slow penetration of H+throughtheC‐S‐HtotheanhydrousCaOandSiO2andthepenetrationofCa2+and(SiO4)4‐totheOH‐ionsleftinthesolution,astheamountofstillnon‐reactedmaterialdeclinesandtherateofthehydrationprocessbecomesdiffusion‐controlled.Duringthisfinalstagehydrationslowly continues, hardening the cement paste. The contribution of C2S to this processincreaseswithtimeand,asaconsequence,therateatwhichadditionalcalciumhydroxideare formed declines. After the supply of calcium sulphate becomes exhausted, the S042‐concentration in the liquidphasedeclines.Asaconsequence, theAFtphasethathasbeenformedintheearlierstagesofhydrationstartstoreactinathrough‐solutionreactionwithadditional C3A, yielding monosulfate. At sufficiently high initial water/cement ratios thehydrationprocessprogressesuntilalloftheoriginalcementbecomesconsumed.However,the residue of larger cement particles may persist even in mature pastes. At lowwater/cementratios thereactionmaystop inthepresenceofsignificantamountsofnon‐reactedmaterial, due to the lack of sufficient amountsofwaterneeded for thehydrationprocess.
5.4 HydrationproductsduringtimeTheformedhydrationproductsfromthetimezerountil90daysispresentedinFigure20,inthefirstminutes,ettringiteiscreatedbyC3AdissolvingintheliquidandreactswithCa2+andSO42‐ionspresentintheliquidphasewhichareproducedbyC3SandC3Shydration.AtthesametimeCHproductsareformedbytheC3SandC2Shydration.
GraduationProject 5.Cement
ing.GuillaumeDoudartdelaGrée page43
AfterafewhourstheformationoftheC‐S‐Hphasefollowsthehydrationofaliteandbelite.Aftera fast initial formationof small amountsassociatedwith thehydrationofC3S in thepre‐inductionperiod,theamountofC‐S‐Hincreasesonlyslowlyintheinductionperiod.
From a few hours to one day, the formation of hydration products accelerates. Theformation of ettringite by C3A in the presence of calcium sulphate CaSO4 slows downbecauseoftheconsumptionofCaSO4.Becauseofthis,C3Awillreactwithettringiteresultinginmonosulfate.Whilemonosulfateformationisincreased,theettringitecontentdeclines.
Thoughthewholeprocesstheporosity(emptyspaces)isdecreasingbecausethehydrationproductsarefillingthevoidsintheconcrete.
Figure20:IntroducesdifferenthydrationproductsintimewhereC4AFandC3Aarethefirsthydrationproducts
andalmostcompletelyhydrateduringtime(Kurtis,2011)Typicalhydrationkineticsofpureclinkerminerals(C3Awithoutandwithaddedgypsum)hydration at ambient temperature are illustrated in Figure 21. During the first hours the“earlyage”strengthdevelopmentiscausedbyC3S.Duetoitsslowreaction,C2Scontributestothe‘longterm’strengthdevelopmentofcement.
Frac
tion
hyd
rate
d
Hydrationtime(h)
Time (log scale) days Progressofhydration(%)
Figure21:Typicalhydrationkineticsofpureclinkerminerals(Hewlett,2004)
GraduationProject 6.HydrationeffectofPortlandcementcombinedwithFlyash
ing.GuillaumeDoudartdelaGrée page44
6. HydrationofPortlandcementwithFlyashThepozzolanic activity of fly ash is due to the silicate andaluminatephases of fly ash incombination with the Ca(OH)2 that is formed during cement hydration which togetherproduce calcium silicates and aluminate hydrates (C3S, C2S and C3A) (Hewlett, 2004).However, the hydration and pozzolanic reactions do not occur independently. Water,soluble alkalis, sulphates, calcium and organics from the fly ash may affect the surfacereactionsandthenucleationandcrystallizationprocesses,especiallyintheearlystagesofcementhydration.Similarly,thepozzolanicreactionswilldependontheamountofcalcium,alkalis,sulphates,silicatesandaluminatesreleasedintotheliquidphasefromcementandflyash.
DifferentauthorsdescribevariousinfluencesofflyashincombinationwithPortlandcement(PC).Thiscanvaryupto:
Retardinginfluence(Diamond,1981); AccelerationofhydrationofPC(Costa&Massazza,1983).
Thisdifferenceismostlyduetothedifferentchemicalandphysicalpropertiesoftheusedflyash intheresearch.AlsothehydrationofC3Siscanbedelayedduetochanges inioniccomposition,which affects thedissolutionofC3S. If for examplemore gypsum ispresent,moreCa2+ispresentintheliquidphase;thiscanalsobehappeningwithflyashproducingCa2+:eventuallylessC3ShydratesbecausethereisnoneedforCa2+.IfalltheCa2+reactswithC3Aandalsogetsabsorbedbytheflyashparticles,thespeedofC3Shydrationwillincrease.Itmaybeclearthatitisnotsurewhichionicchangewilloccur.Belowaresomeconclusionsdescribed by different authors. It can be stated that there ismore known and explainedaboutthepositivereactionsthanaboutnegativereactionswhich,alsooccurbutarehardlyunderstood.
6.1 TricalciumsilicateandflyashItisfoundthatC3SreactsfasterwithflyashthanwithoutinbothPre‐inductionperiodandAccelerationperiod(Takemoto&Uchikawa,1980;Costa&Masazza,1981);.After24hoursC3Sishydratedfor55%inthepresenceofflyashcomparedto38%pureC3S.ThisisduetotheflyashparticlesthatcanabsorbCa+2andprovideadditionalsurfaceonwhichC‐S‐Hgelcan precipitate. In consequence, the dissolution of C3S because of increasing provision ofCa+2totheliquidandlesC‐S‐HgelformationnearC3SbecauseotherwiseitwoulddisturbthewaterpenetrationandthereforethehydrationofC3S.
Flyashwasalsofoundtoacceleratethepolymerizationofhydratedsilicates;about60%ofSiintheC‐S‐HofC3Sflyashpastewaspresentaspolymerscomparedwith40%forpureC3S paste (Wesche, 1991).When 4% gypsum is added, the hydration speed decreases atearly ages but increases at later ages. This is due to the increasing provision of Ca+2 andtherefore C3S reacting less, but when fly ash is also reacting (absorbing Ca2+) the Ca+2contentreducesandC3Sreactsfaster.Iftheamountofaddedgypsumisincreased(8%)thehydration decreases in all stages because a large amount of Ca2+ is present in the liquidphase.
GraduationProject 6.HydrationeffectofPortlandcementcombinedwithFlyash
ing.GuillaumeDoudartdelaGrée page45
Fly ash in combinationwithPC alsohas an influence on the C3S andC2S hydration ratio.After18monthsC3SisfullyhydratedintheC3S‐flyashmix,comparedtothepartlyhydratedinpureC3Spaste.TheoppositehappenedforC2Shydrationproducts.
ItisalsofoundthatthehydrationheatcurveofC3Sincombinationwithflyashischanged.Theheathydrationpeaksareretardedandarelesshigh.ThiscouldbeduetothedelayofnucleationandhydrationofCa(OH)2andC‐S‐Hby the solublealuminate species releasedfromflyash(Wesche,1991).Asignificantdelayofmorethan12hoursinthemaximumheatevolution peak in the presence of fly ashwas also found byOgawa et al. (1980). Fly ashdecreases theCa+2and silicate concentrations in the liquidphaseofhydratingC3S‐fly‐ashsignificantly(20and50%,respectively).Fly ash is expected to affect the composition of C‐S‐Hproduced in the hydration system.SmallamountsofAFtandAFmphaseswerefoundat7‐28daysandwithin3monthsalltheFe,AlandSO42‐suppliedbytheflyashwereincorporatedintheC‐S‐Hgel.
Thepozzolanicreactionofflyashhoweverstartslater.Theglassphaseparticlesneedtobecrackedbutthisdependsontheamountofalkali(pH‐value) intheconcretemix,whichisdependingontheamountofOH‐ionspresentintheliquid.ThelayerofCa+2andC‐S‐Hgelaroundtheflyashpreventreactionofflyashparticles.InalaterstagewhentheporewaterhasalowerCa+2content(becauseitwasconsumedforproductionofettringite)andthepHvalueishigher(resultingindissolutionoftheglassphase),lessC‐S‐Hiscreatedaroundflyashparticlesbecause the formationofhydrationproducts ismorenear thecementgrain.FlyashcanthereforestartreactingandconsumingCH.Alsomorehydrationproductsarecreated, resulting in an increasing temperature; a higher temperature will increase thereactionofflyash.
ItappearsthattheCHcontentisincreasedwhenflyashandPCarecombined.Thisisduetothe acceleration of the hydration of C3S. After the pozzolanic reaction of fly ash starts, itgraduallyconsumesCHandthisdecreasestheCHcontent(Tayloretal.,1985).
6.2 DicalciumsilicateandflyashFortheC2Shydrationproduct,differentopinionsareformulated.ItisfoundthattheC2Scanhavetwoaccelerationperiods,aroundfourand30days.Itisalsofoundthatithasnoeffectonthehydrationuntil28days.However,Sakaietal.(2005)foundthattheC2Shydrationisaccelerated up to 91 days and after that retarded. This is due to the formation of densehydrationproductsonthesurfaceofC2S.
6.3 EffectFlyashonthehydrationofC3AandC4AFFly ash is a more effective retarder of C3A and C4AF hydration compared to the sameamountofgypsum.SO42‐andCa+2dissolved fromflyashmaypartlyexplain theretardingeffectofflyash.Analyzingtheliquidphaseofthehydratingsystemshowedthatsaturationintermsofgypsumoccurredwithafewsecondsofwaterbeingaddedtoflyash.TheinteractionofflyashwithC3Aprobablyinvolvesthefollowingprocess(Wesche,1991):
Adsorptionofsulphateions,whichreducesitsactivedissolutionsites; Formationofettringiteatanearlyage,whichreducesavailablemigrationof
sulphates; Migrationofsulphateionsandstabilizationofhexagonalstructures.
GraduationProject 6.HydrationeffectofPortlandcementcombinedwithFlyash
ing.GuillaumeDoudartdelaGrée page46
Pozzolanicmaterialsincludingflyashacceleratenotonlytheformationofettringiteanditsconversiontomonosulfoaluminate,butalsothehydrationofC3Ainthepresenceofgypsum.Higher amounts of alkalis in the pozzolanic materials promoted the formation of cubichydrates. The formation of ettringite and its conversion to monosulfoaluminate wereretardedbyCa(OH)2.FlyashincreasedthedissolutionofCHandhencethehydrationofC3AbyprovidingsurfaceforettringiteprecipitationandCa2+absorption(Uchikawaetal.,1985).
6.4 EffectofflyashoncementhydrationLukas(1976)foundthatflyashincombinationwithPCincreasestheformationofCa(OH)2inpastesuptothreedaysofhydrationandattributedittotheacceleratedhydrationofC3Sincement.TheCa(OH)2contentdecreaseswithtime,indicatingthatithadbeenusedforthepozzolanicreactionofflyash.ItwasfoundthatthedegreeofhydrationofC3Sinacement‐flyashpastewashigherthaninpurecementpastesfromonedayonwards.Also,Ghose&Pratt (1981) reported a retardation of both C3A andC3S heat evolution peakmaxima forcementflyashpastes.By adding fly ash into themixture, the amount of hydration products in the concrete isincreasingaswellastheconcretestrength.However,thatcanonlyoccuraftercementstartstohydratebecauseoftheneedofCa(OH)2.Therefore,inthecaseofflyashusageasapartialreplacement for cement, theearly strength is reducedbut the long termstrengthmaybeequal to the one of plain cement, or even be higher. The rate of strength gain, however,dependsupon theproperties of fly ash and cement,mixproperties, aswell as the curingconditionsoftheflyashconcrete(Joshi,1979)
6.4.1 C‐S‐HcontentIn a combination of PC with fly ash, C‐S‐H is produced by the hydration of PC and thepozzolanic reaction of fly ashwith CH. Both C‐S‐H are different due to different reactionmechanisms (Killoh et al., 1989). The Ca/Si ratio for C‐S‐H in fly ash cement paste isexpectedtodifferfromthat incementpaste,sinceC‐S‐HisalsoformedbythereactionofCa(OH)2withaluminosilicatephaseoftheflyash.ThedecreaseofCa/Siratio(whichisnotalways a negative effect) of the inner hydrate can be attributed to an increase in the Sicontent.ItisfoundthattheCa/SiratioofC‐S‐HislowernearthepozzolangrainsofflyashascomparedtotheC‐S‐HneartoC3S(Ogawaetal.,1980).
Normallyatearlyages,therangeofCa/Siratioistypically1.2‐2.3forOPCpastes.At lateragesitrangesaround1.60‐1.85.C‐S‐HmaybeformedwithCa/Siratiosvaryingfrom0.8toabout 1.75 in CEM I pastes (Taylor, 1997). This author also describes that some studieshaveindicatedminimumCa/Siratiosof0.6‐0.7.AccordingtoseveralauthorstheCa/SiratioofC‐S‐Hcloseto flyashparticles is lowercomparedtoordinaryPCmix. (Rayment,1982;Tayloretal.,1985;Uchikawa,1986).AlowerCa/SiratiogivesamorestableC‐S‐Hgelwhichis a positive effect. The greater proportion of C‐S‐H gel in the hydrated fly‐ash cementresultsinlowpermeability,whichtogetherwiththereductionintheCa(OH)2contentoffersan explanation for the improved resistance to chemical attack, particularly by sulphates,observedforflyashopposedtoplaincementconcrete.However,becauseofallthisitisalsopossiblethatC‐S‐Hgelcandeveloplessbecauseoflowvoidspace,resultinginadecreaseofwaterpenetrationtothecementgrain,whicheventuallyreducesthestrengthofconcrete.
GraduationProject 7.0Effectofcontaminantsoncementhydration
ing.GuillaumeDoudartdelaGrée page47
7. EffectsofcontaminantsoncementhydrationThe use of bio‐energy fly ash in concretemixtures can not only have positive effects asdescribed in Chapter 6 and other previous chapters, but some elements can also createstructuraldamage.Tominimizethesenegativeeffectstheregulationofcoal‐combustionflyash (NEN‐EN‐450) has been developed. In this chapter the importance and reasons aredescribedforsomeoftheregulations.TheregulationsaredescribedinParagraph4.5.5andconsider:lossofignition,chloridecontent,sulphatecontentandfreecalciumoxidecontent.
7.1 Carboncontent/lossonignitiontestLossonignition(LOI)isdefinedasthemasslossthattakesplacewhenheatingupmaterialsto a temperature of 950 degrees Celsius according to NEN EN‐196‐2. In the first phase,water is evaporated, organic matter is combusted to ash and carbon dioxide at atemperaturebetween500and550degreesCelsius.ThisLOIcanbecalculatedusing:LOI550=((DW105–DW550)/DW105)*100 (5)whereLOI550representsLOIat550degreesCelsius(asapercentage),DW105representsthedry weight of the sample after drying (usually 12–24 h at ca. 105 degrees Celsius) andbeforecombustion,andDW550isthedryweightofthesampleafterheatingto550degreesCelsius (both in g). In a secondphase, carbondioxide is evolved from carbonate, leavingcalciumoxideandLOIiscalculatedas:LOI950=((DW550–DW950)/DW105)*100 (6)whereLOI950istheLOIat950degreesCelsius(asapercentage),DW950representsthedryweightofthesampleafterheatingto950degreesCelsius(Heirietal.,2001).TocalculatetheLOIforflyashthetotalmasslossofthedryweightupto950degreesCelsiusneedstobeconsidered.BecausetheLOIhasahighrelationwiththecarboncontentofbio‐energyflyash,alowLOIpracticallyinmostcasesmeanslowcarboncontent.Alowcarboncontentofbio‐energyflyash is importantbecausehigh carbon contentsmay lead to the followingnegative effects(vandenBerg,2006):
Coarse coal particles can formgalvanic elementswithpre‐stressed steel products.As a result, small electric circuits are created, which can cause corrosion of thereinforcement(pitting).Therefore,itisnotoftenusedincombinationwithpre‐loadand pre‐tensioned steel. For that reason, high carbon fly ash can only be used incombinationwithCEMI,becauseitconsistsofless/nocoalparticles,whileinCEMIIa part of the clinker is replaced by fly ash or blast furnace slag that contain coalparticles, 60% of which are smaller than 45 microns. In addition there is arequirementthattheamountofflyashparticleslargerthan212micronsshouldbelessorequalto3%(m/m)(NEN‐EN‐5950);
Theabsorptionofthecellulatingagentbythecarbonparticlescanpreventtheformationofairbubbles,whichcandecreasetheresistanceofconcretetofrostandde‐icingsaltsandhaveaneffectonthedurabilityofconcrete,
Carbonactsasfilleroftheactivepozzolanicmatterintheflyash,whichisnotdesirable;
GraduationProject 7.0Effectofcontaminantsoncementhydration
ing.GuillaumeDoudartdelaGrée page48
Carbonparticlesabsorbwaterandthusreducetheavailablewaterforthecementtohydrate.
7.2 ChloridecontentChlorides,especiallyCaCl2wereaddedinthepasttoconcretetoincreasethehydrationrate.Lateranegativeeffectwasfound‐thecorrosionofsteelreinforcement.Chloridesreducethedurability of reinforced concrete. Steel reinforcement in concrete is protected againstcorrosionby the “passive layer” on the steel surface. This passive layer is createddue tooxidation of the steel combined with a high alkaline environment. However, when asufficientamountofchloridesreachesthesteelreinforcement,corrosionstartsreducingthequalityandfunctionofthesteelreinforcement.Thebio‐energy flyashescontain largeamountsof chlorideswhichmakes itunsuitable touse inreinforcedconcrete.Mostchlorideshoweveraresolubleandcanberemovedusingwatertreatments.ThisapproachwillbefurtherdiscussedinChapter8.Chlorideisformedfromchlorine(halogen).Whenchlorine(Cl)gainsanelectronitformsananion chloride (Cl‐). From different studies including a complete review of Yudovich &Ketris(2005)itisknownthatcoalashcancontainlargeamounts(morethan1000ppm)ofCl‐containingphases.ThespeciesofCloccurringincoalasharesurprisinglyvaried.Amongthemarefound:
‐ Inorganic saltlike NaCl and other chlorides, Cl‐bearing silicates, carbonates,sulfides,aswellasdissolvedchloridesinporemoisture;
‐ Organic‐associated Cl seems to predominate in coal. Itmay consist of two types.Firstly,(minorsite(“true”Clorg)water‐insolubleCl‐organiccompounds,whereCliscovalent bonded with coal organic matter (Hodges et al., 1983; Saunders, 1980).Secondly,“semi‐organic”Cl,asanionCl‐,whichispartlyorfullywater‐solubleClthatissorbedontheporesurfaceofcoalorganicmatter.Suchformdoesnotenterthecoal organic macromolecule and may be exchangeable (Daybell & Pringle, 1958;Edgcombe, 1956; Gluskoter & Ruch, 1971; Saunders, 1980; Caswell et al., 1984;Pearce&Hill,1986;Oakeyetal.,1991;Huggins&Huffman,1991;Spears,inpress).
Furthermore itneeds tobenoted that small size coalparticlescancontainmorechlorinethanlargesizecoalparticles(Yudovich&Ketris,2005).
All this information from the review of Yudovich & Ketris (2005) indicates that coalparticles in fly ashes can contain large amounts of chlorine that can be represented bychlorides.InChapter8,furtherresearchisdoneonthisphenomenaandpossibleoccurringproblemswhentryingtowashoutthesechlorides.Looking closer with X‐ray diffraction (XRD) provides knowledge about the elementalcompositionofflyashes.Theanalysisareperformedusingcopperradiationwithastepsizeof0.02degreesand1‐secondcounttime.PatternprocessingwasdoneusingEVAsoftwareand the ICCDD pattern database. To find out which chlorides the fly ashes contain, asufficient amountof chlorides shouldbe available (content of >5%). For this reasonHVCfilterflyash(A1)isusedbecauseitcontainsapproximately9%chlorides.Figure 22 illustrates the XRD‐pattern of original and water treated A1, to see whichchlorides are present and if there is a relationwith soluble chlorides. It can clearly beenseenthatthechloridecontentismainlyconsistofKClbutalsotracesofCaCl2andNaClare
GraduationProject 7.0Effectofcontaminantsoncementhydration
ing.GuillaumeDoudartdelaGrée page49
present.After treatingwithwater,mostof thechloridesaredissolved inwater.However,the solubilityof theCl‐containingphasesaredifferent.The solubility inwateratambienttemperaturesofCaCl2(740g/l)ismuchhigherthanthatofNaCl(360g/l)orKCl(344g/l),whilethesolubilityofMgCl2(5.43g/l)isratherlow.
Figure22:XRDpatternofHVCfilterflyash(A1),indicatingdifferentCl‐containingphases
Figure23:XRDpatternofCarbonfromTwenceboilerflyash(B1)
GraduationProject 8.Treatmentofbio‐energyflyash
ing.GuillaumeDoudartdelaGrée page50
8. Treatmentofbio‐energyflyashThetreatmentofbio‐energyflyashesisdividedintofivedifferentsteps.Theseincludetheremovalofbigandsmall carbonparticles, chlorides,metallic aluminiumand fulfilling therequirements like particle size distribution and sulphates according to NEN‐EN‐450 andmakingtheflyashNon‐hazardousorevenInert(asdefinedbytheLandfillBanDecree).Theremovalofheavymetalswillnotbetestedafterthesesteps.However,afterthewatertreatment almost all leachable metals should be removed. Remaining metals will beenclosed into the cement matrix as described by Eijk (2001). Removal of metallicaluminiumwillbedescribedbecause internal swellingoccurs increasing theporosityanddecreasingstrength.
8.1 TreatmentmethodsThetreatmentsstepsusedinthisresearchdependonthebio‐energyflyashandmoreoverthecompoundswhichitconsistof.Ingeneralthefollowingtreatmentmethodologycanbeused:
GraduationProject 8.Treatmentofbio‐energyflyash
ing.GuillaumeDoudartdelaGrée page51
First,carbonparticlesareremoved,becauseofthenegativeeffectonthechlorideremoval.Secondly,theflyashiswashedusingwatertreatmentstoreducethesolublechloridesandaluminium content and in the end grinding can be performed to decrease the PSD andpossibleincreasethereactivityofthebio‐energyflyashes.
8.1.1 SievingThis treatment is needed for bio‐energy fly asheswith a high content of carbon. Using asieve of 500 microns, big carbon particles will be removed. In this way the LOI will bereduced, as well as the chloride content of the fly ash. The relation between removingcarbonandchlorideisalreadydiscussedinParagraph7.2.
8.1.2 Thermaltreatment&Air‐filteringThis treatment is needed for bio‐energy fly ashes with high content of small carbonparticles (<40µm). These fly ashes can be recognized because of their black appearance,even after Treatment 1. For this treatment, two options (2a and 2b) are available.Treatment2a is a thermal treatment.Treatment2b is a separationbyelectrostatic filters(bothforindustrialuse).However,inthelabthethermaltreatmentisperformedbyusingan oven and electrostatic filters are replaced by a shaking device with air suction aspresentedinFigures24and25.Thismeasureshouldgivesimilarendresultsastheabovedescribedindustrialprocesses.Thechoicewhichmethodtousedependsonwhethertherearephasespresent that can changewhen thermal treatment isused.Aphase change canresult in crystallisationwhichnegatively affects thepozzolanic reactivity of the fly ashes,andshouldthereforebeavoided.Phasechangesaredependingonthecontentoftheflyash,but can as an example consist of: formation of mullite, hematite and crystallisation ofmeliliteoroxidationofmagnetite.Ingeneralthereductionoftheglassphaseandincreaseof crystalline phase occur at temperatures above 700 degrees Celsius, resulting in adecreaseofthepozzolanicactivity(Fox,2005)
Figure24:Shakingdevicewithairsuction Figure25:Heatedflyashinovenat800°C
Ifthereisapossibilityofphasechange,Treatment2aisnotanoptionandTreatment2bcanbeapplied.Treatment2busesashakingmechanismwithairexhaust.Byshaking,thelightcoal particles are lifted up and removed using the air exhaust.Whenwell calibrated, theremoval of small light particles that are not coal should be limited. In Chapter 10, thedifference between these two methods and the removal of small particles is furtherdiscussed.
GraduationProject 8.Treatmentofbio‐energyflyash
ing.GuillaumeDoudartdelaGrée page52
8.1.3 MetallicaluminiumremovalThis treatment is only required when large amounts of metallic aluminium are present,whichincontactwithwaterwillformhydrogenaccordingto:2Al+6H2O2Al(OH)3+3H2 (11)If thishappens in themortars,an increase involume, resulting inan increaseofporosityanddecreaseofcompressiveandflexuralstrengthwilltakeplace.Toremovetheunwantedaluminium for industrial purpose, an eddy current separator can be used. In this way apowerful varying magnetic field separates the non‐ferrous metals from the rest of thematerial.Forlabuse,theflyashisstirredinwaterfor72hourswithaliquidtosolidratiooffour.Afterwards,theliquidisfiltered,flushedanddriedaccordingtothesteps2,3,4whicharedescribedinParagraph8.1.4.
8.1.4 WashingAfter this,water treatment isused toremovewatersoluble compoundsmostly chlorides.Forthis,fourstepsareperformedwhicharedescribedbelow:
1. Flyashincombinationwithdemineralisedwaterisshakeninbottlesusinga“StuartreciprocatingshakerSSL2”toremovesolublechloridesandmetalsasillustratedinFigure25;
2. Thewaterisseparatedfromtheflyashusing15‐30micronfiltersas illustratedinFigure26;
3. Theremaining flyashon the filter is flushedwithdemineralisedwater to removeremainingwaterwithsolubleminerals;
4. TheremainingflyashisdriedtoremovetheavailablewatercontentasillustratedinFigure27.
Figure26:Shakingprocess Figure27:Filtering Figure28:DriedflyashafterfilteringTheefficiencyof thewater treatment techniquedependson theparameters statedbelowandwillbediscussedlater.
Temperatureofwater Treatmenttime; pH‐valueofwater Amountofflushingwater; Water/solidratio; Filtersize. Oscillationspeed;
GraduationProject 8.Treatmentofbio‐energyflyash
ing.GuillaumeDoudartdelaGrée page53
8.1.5 Seperation/GrindingThislastisusedtocreatesmallparticlesusingaballmill(onlyforlabuse),increasingthereactivityandfillereffectoftheparticles.Thisresultsinanincreaseofpozzolanicactivityatearlycuringages,creatingcompacterconcretewithahigherdensityandhigherflexuralandcompressivestrength.
8.2 Treatmentevaluation
8.2.1 CarboncontentAfter Treatments 1 and 2 the carbon content can bemeasured using ThermogravimetricanalysiswithaNETZSCHSTA449F1,wheretheweightlossatdifferenttemperaturescanbelinkedtotheremovalofcertaincompoundsincludingcarbon.
8.2.2 ChloridecontentinwaterFromthewashedsample,3mlofusedwateristakenandafterfiltering,twomlaretakenusing a “Macro 25ml pipette”. The chloride content of the 2ml togetherwith additionaldemineralised water is then measured using titration with “Metrohm 785 DMP Titrino”with0.01MsolutionofsilvernitrateaspresentedinFigure29.
Figure29:Metrohm785DMPTitrino
8.2.3 ChloridecontentofthesolidmaterialFormeasuringthechloridecontentofsolidmaterials,twogramssolidmaterial(bio‐energyflyash),togetherwith37mlofdemineralisedwaterand3mlofnitricacidiscombinedinabottle. Themix is then stirred using amagnetic stirrer on a heating plate of 45 degreesCelsiusfor15minutes.Afterwards,thesolutionisfilteredandflusheduntilavolumeof100ml is obtained. From this, 2‐10ml, depending on the chloride content, can bemeasuredusing Metrohm 785 DMP Titrino with a 0.01 M solution of silver nitrate. The chloridecontentisgivenby:
∗ ∗ . ∗ .
∗ ∗ (12)
GraduationProject 8.Treatmentofbio‐energyflyash
ing.GuillaumeDoudartdelaGrée page54
wereP10istheamountinthepipetteinml(usedfordeterminationoftheconcentrationofchloride). B100 is the total volume of the solution (in ml). V is the titrated silver nitratesolution inmlandZ0.01 the concentrationof silvernitrate inmolesper liter (M).G2 is theused solid (bio‐energy fly ash) in grams for making the solution and Ct the chlorideconcentrationing/g.
GraduationProject 9.Results
ing.GuillaumeDoudartdelaGrée page55
9. ResultsThis chapter describes the treatment data together with the effects and obtained finalcompressive and flexural strengths. During the measurements, assumptions are made,whicharethenassessedduringthemeasurements.Thisdataprovidesincreasedknowledgeonhowtoperformthemeasurementsandexplanationsoftheobtainedresults.Itshouldbestatedthatatthebeginningmostofthedatadidnotmeettheexpectations.Becauseofthis,thefirstparagraphsprovideinformationaboutthemeasurementsandimprovementsthatweremade,followedbyasecondparagraphwithobtaineddata.
9.1 Chlorideremoval(partone)
9.1.1 Twenceboilerflyash(B1)InthisfirstexperimentwithTwenceboilerflyash(B1),differenttreatmentparametersandtheir influenceonflyashpropertiesare investigated. Inthiscasetheflyash isshakenforonehourwithdifferenttemperatures(20and60°C),twodifferentshakingspeeds(120and240rpm)andtwodifferentliquidtosolidratios(L/S=2,L/S=4).Thechlorideismeasuredby taking threemlwaterandmeasuring thechloridecontentby titration.TheresultsaregiveninTable18.Table18:RemovedchloridesfromTwenceboilerflyashwithdifferentparameters
Liquidtosolidratio
Original20°C,120rpm[mg/kg]
Increasedshakingspeed240rpm
[mg/kg]
Increasedtemperature60°C
[mg/kg]
Combination60°C,240rpm[mg/kg]
L/S=2 1957 2575 2091 2867Percentageimproved* 0% +11.8% +2.6% +17.4%
L/S=4 2043 2320 2320 2702Percentageimproved* 0% +6.0% +5.3% +12.6%
*Thispercentageiscomparedtotheoriginal20°C,120rpmvalues
The results in Table 18 present a large improvement by increasing the shaking speedcomparedtotheoriginal,especiallyforaliquidtosolidratiooftwo.Thelargerquantityofchlorides which are removable with an L/S = 2 can be explained by different theories.Firstly,becauseofgravity,particlesaremoreattachedtothebottomsurfaceandthewaterpressuredownwardsreducesthemovementoftheparticles,thesocalled“dampingeffect”.Secondly,whenwavesareclosetothebottomsurface,ahigherturbulentflowisrealized,the“mixingeffect”.Whentheflowisfurtherawayfromthebottomsurfacethewaveshavealmostnoinfluenceontheparticlesandreducetheefficiencyofdissolvingchlorides.The increase of temperature for L/S = 4 has a bitmore advantage than for L/S= 2. It isbelievedthatthisisduetotherapiddecreaseintemperatureofL/S=2andthereforehasashortereffect.Acombinationofincreasedtemperatureandliquidtosolidratioworkswellforremovingchlorideions,becausethepositiveeffectscumulate.Inordertomonitorthebehaviourofthebio‐energyflyashintime,thechloridecontentofB1ismeasuredduringonehourwithashakingspeedof120rpm,extractingevery15min3mlandadding3mldemineralised‐watertokeeptheliquidtosolidratioconstant.Foreachstep thecalculation iscorrected for theremovedcontentofchlorides ineachthreeml. In
GraduationProject 9.Results
ing.GuillaumeDoudartdelaGrée page56
ordertoobservedifferencesbetween2samples,thebottleswithaliquidtosolidratioequalto2and4aremeasuredtwicetocheckreliability(groupAandB).Resultsarepresentedinfigure30.
Figure30:ChloridesremovedfromTwenceboilerflyash(B1)during60minuteswithdifferentL/Sratios
FromFigure30itcanbeconcludedthatintimemorechloridesareremoved.However,themeasurementsarehardly repeatable.Toevaluate the reliabilityof theprevious results, asecondtestisperformed.Inthiscase,fivebottlesofTwenceboilerflyash(B1)arewashed(TBw). The used fly ash is coming from the same bucket of 15 kg. From this bucket fivesamplesof100gramsaretakenwhilemixing.The100gramisthenputintobottles(A‐E)which are shaken for one hour at a speed of 120 rpm and with L/S = 4 to test therepeatability. Furthermore, the chloride content in the used distilled water is measuredthreetimestoseeifthereisalargevariability.TheresultsarepresentedinTable19.
Table19:InvestigationoffivethesameTwenceboilerflyashsamples(A‐E),todeterminethechlorideremoval
L/S SampleTBw(A)[mg/kg]
SampleTBw(B)[mg/kg]
SampleTBw(C)[mg/kg]
SampleTBw(D)[mg/kg]
SampleTBw(E)[mg/kg]
4 1980‐1993 2153‐2172 1997‐2026 1852‐1905 2149‐2200Reduction* 37.9‐38.1% 41.2‐41.6% 38.2‐38.8% 35.4‐36.5% 41.1‐42.1%
*Percentageremovediscalculatedfromtheoriginal5226mg/kg
TheresultsinTable18showthatineverysamplethereisadeviationofresults.SampleDistheworstcasewithavariationof1.1%,thelowestdifferenceissampleAwithavariationof0.3%.ComparingthesamplesandtakingthelowestvalueofsampleDandthehighestvalueofsampleEresultsintheworstcasescenariowithdifferencebetweenthesamplesof6.7%.
InordertoinvestigatethequantityofchloridewhichisremovedfromTwenceboilerflyashbydistilledwater,thechloridecontentofsolidTwenceboilerflyash(TBs) isdetermined.This isdone for twopairsofmeasurements.The firstone isapureTwenceboiler flyashwithout any changes. The second one is a crushedTwence boiler fly ash. Samples 1 to 4presentthepureTwenceboilerflyashandSamples5to8–thecrushedTwenceboilerflyash. Sample 9 is sieved, crushed and represents everything below 500microns (87% ofmass),while sample 10 is sieved, crushed and represents everything above 500microns
25
35
45
55
65
15 30 45 60
Percentage
chloride rem
oved [%]
Time [min.]
A L/S 2
B L/S 2
A L/S 4
B L/S 4
GraduationProject 9.Results
ing.GuillaumeDoudartdelaGrée page57
(13%bymass).TheresultsarepresentedinTable20,wheresamples9and10togetherareassumedtoconstitute100%(100%mass).Table20:ChloridecontentofsolidTwenceboilerflyash
TBs1
[mg/kg]
TBs2
[mg/kg]
TBs3
[mg/kg]
TBs4
[mg/kg]
TBs5
[mg/kg]
TBs6
[mg/kg]
TBs7
[mg/kg]
TBs8
[mg/kg]
TBs9
[mg/kg]
TBs10
[mg/kg]2680.3 2545.4 3553.6 2751.3 3714 3634.3 3738.2 3771.9 2855 237051.3% 48.7% 68.0% 52.6% 71.1% 69.5% 71.5% 72.2% 54.6% 45.4%
Itcanbeseenthatsamples1‐4contaunlesschloridesthansamples5‐8,probablybecausenot all chloride ions from these samples were dissolved. Probably some chloride ionscontained incoalparticleswerenotdissolved inwater.This is incontrast to thesamples5‐8thatarecrushed,andthereforechlorideparticleswerereleasedandcouldbesolvedinwater. Sample10 ispurecarbon thatwason topof the500micronsieveand represents13%bymass. Even this low content bymass contains almost 50%of the chlorides. Thisexplains why samples 5‐8 have more chlorides than sample 9, because a small part ofcarbonwithchlorideswasstillpresent.Nevertheless,becauseof thesmallamountof twograms that is investigated, samples 5‐8 are not a realistic representation of the chloridecontentinTwenceboilerflyash.However,itisbelievedthatsample9togetherwithsample10giveagoodrepresentation.Moreover,thisvalueisclosetotheresultscomingfromtheXRFanalysis(990‐5600mg/kg).Theseresultsshowthatcoalparticleshaveabiginfluenceonthechloridecontentofaflyash,aschloridesattachandcombinewithcoalparticlesasdescribed earlier in Paragraph 7.2. These facts can explain the observed differencesbetweenthesamesamples.Forthatreason,furtherinvestigationisperformedonTwencecyclone fly ash due to less carbon particles in its composition and should result in lessfluctuationwhenmeasuringthechloridecontentindistilledwaterandpureflyashes.
9.1.2 Twencecycloneflyash(B2)ToinvestigatethequantityofremovedchloridefromTwencecycloneflyash(B2),at firstthe initial chloride content of plain B2 is determined. This is done for two pairs ofmeasurements.ThefirstoneisplainB2withoutanychanges.ThesecondoneiscrushedB2.Samples TC1‐TC3 present the plain B2 and samples TC4G‐TC6G the crushed B2. SampleTCArepresentstheaverageofsamplesTC4G‐TC6Gandistakenas100%chloridecontentbecauseofthesmalldifferencesbetweensamplesTC4G‐TC6G.TheresultsarepresentedinTable21.Table21:ChloridecontentofsolidTwencecycloneflyash
TC1[mg/kg]
TC2[mg/kg]
TC3[mg/kg]
TC4G[mg/kg]
TC5G[mg/kg]
TC6G[mg/kg]
TCA[mg/kg]
3887.3 4276.0 4020.4 4158 4090 4143 4130.894.1% 103.5% 97.3% 100.7% 99.0% 100.3% 100%
FromTable21itcanbestatedthatthedifferencebetweentheoriginalsampleandcrushedsampleisverysmall.Thisisprobablyduetosmallercontentofcarbonparticles.TheresultsareincontrasttoTwenceboilerflyash,wherealargedifferencewasobserved.Inordertomonitorthebehaviourofthebio‐energyflyashintime,thechloridecontentoffour samplesB2aremeasuredandpresented inFigure31.The four samplesaredividedintogroupAandB,bothmeasuredevery15minutesforonehourlongwithashakingspeedof120rpmandtwodifferentliquidtosolidratios(L/S),namely2and4.
GraduationProject 9.Results
ing.GuillaumeDoudartdelaGrée page58
Figure31:ChlorideremoveinTwencecycloneflyash(B2)during60minuteswithdifferentL/Sratios
It was expected that with increasing time more and more chlorides would be removed.However, Figure 31 illustrates that there are still differences between the same‐samplepairs,althoughtthisisalreadyconsiderablyreducedcomparedtoB2.Thereducedbutstillpresent inaccuracy canhavedifferent reasons.Oneof the first reasonswouldbe that thechloride content in fly ash is not homogenously distributed, resulting in a difference inchloride removal. However, it can also be that the 3 ml from every sample are notrepresentativeforthetotal400mlthat isused.Thiswouldexplainwhyat60minutesalltheresultsareincloseagreement,becauseatthistimethewashingstoppedand50mlofwaterwasextractedformeasuringthechloridecontent.Thefollowingmeasurementisperformedusingaliquidtosolidratioof2andtaking25mlsamples insteadof the threemlused toconfirmthis statement.Also theshakingspeed isincreasedto240rpmbecausethiswouldreducethetimetosolvethechloridesinwater,asmentionedinTable18.ResultsareillustratedinFigure32.
Figure32:ChlorideionsremovedinTwencecycloneflyash(B2)during60minuteswithL/S=2andshaking
speedof240rpm
TheresultspresentedinFigure32lookmorepromissing.Notonlythediscrepancybetweenthe same‐samples decreased, but close to 30 minutes more than 95% of chlorides are
60
70
80
90
100
110
15 30 45 60
Percentage
chloride removed [%]
Time [min.]
A L/S 2
B L/S 2
A L/S 4
B L/S 4
92
93
94
95
96
97
98
99
100
0 10 20 30 40 50 60 70
Percentage
chloride
removed [%]
Time [min.]
A L/S 2
B L/S 2
GraduationProject 9.Results
ing.GuillaumeDoudartdelaGrée page59
removed. The fact that there is still a bit of shifting in the results during time could beexplainedbysmallmeasurementerrors.Thedifferencebetweentheseresults is less than5% and is assumed to be trustworthy. Therefore, futher investigations for the other flyashesaremadeusingtheknowledge thata largesamplesizeof25mlofwater isneededandthatcarbonparticleshavea large influenceontheoutcomeofthewashingtechniquebecauseofpossiblechloridecontent.
9.1.3 ValidationmeasurementTotestthereliabilityofthepredictionofchloridesinsolidsamples,twodifferentsolutionsofNaClandCaCl2aremixedwithCaO‐SiO2indifferentratioswithatotalsamplesizeof10grams. Afterwards the chlorides aremeasured and results are analysedwhether there isanydiscrepancy.TheresultsaregiveninTable22.Table22:ChloridecontentofNaClandCaCl2solutionscomparedtomeasuredvalues
Cl‐[%] OriginalNaCl[g]
MeasuredNaCl[g]
Difference[%]
OriginalCaCl2[g]
MeasuredCaCl2[g]
Difference[%]
2 0.33 0.25 ‐40.0 0.41 0.18 ‐44.04 0.66 0.53 ‐36.4 0.83 0.47 ‐29.26 0.99 0.70 ‐44.4 1.24 0.60 ‐39.08 1.32 0.95 ‐43.3 1.66 1.02 ‐35.8
Ingeneraltheaboveresultspresentanerrorof40%inmeasuringthechloridecontent,butthis is strongly related to theusedchloride type. It canbe seen that thedifferenceof thehighly soluble CaCl2 is smaller than that of the less soluble NaCl. The reason for thesedifferences can so far only be speculated about. Perhaps the time for dissolving (five‐tenminutes) was too short, and that a longer mixing time is needed to dissolve all solublechlorides.Asimilartrendisinfactvisiblewhenmonitoringthedissolvingrateofchloridesduringtime.Fromthisittakes30‐60minutestohaveastableresult.
9.2 Chlorideremoval(parttwo)
9.2.1 TwenceboilerflyashTo ensure that themaximum chloride contentwill not exceed 1000mg/kg the followingtreatmentstepsarecarriedout:1) Thebio‐energyflyashissievedona500micronsieve,toremovelargecoalparticles;2) Thebio‐energyflyashis:
a. airfilteredremovingsmallcoalparticles;b. thermallytreatedat750°C,toremovesmallcoalparticles.
3) Bothsamplesdescribedundersteptwoarethenwashedusingthewatertreatment.TheresultsareillustratedinFigure33.
GraduationProject 9.Results
ing.GuillaumeDoudartdelaGrée page60
Figure33:ChlorideionsinTwenceboilerflyash(B1)beforeandafterdifferenttreatmentsteps
The first step already reduced the available chloride content with 45%. This is due toabsorbedchlorideandchlorineinthecarbonstructure.Washingthisflyashwillresultinatotaldecreaseof51%.Thisis6%lowerthannotwashing,fromwhichitcanbeconcludedthatwashing has almost no effect as long as small carbonparticles are still present. Thesoluble chlorides are somehow attached to the surface of the carbon particles andpreventedtodissolveinwater.RemovingthesmallcarbonparticlesusingTreatment2aor2b results in a reductionof75%and70%, respectively, compared to theoriginal flyash.Washingthisflyashesresultsinatotalreductionof93%and82%.Thisindicatesthatfirstremoving the coal particles increases the removal of chlorides. Secondly, air filteringremoves chlorides that are less soluble and with this decreases the remaining chloridecontent.
9.2.2 TwencecycloneflyashBecausethisflyashonlyhaslargecarbonparticlesthefollowingstepsareperformed:1) Thebio‐energyflyashissievedona500micronsieve,toremovelargecoalparticles;2) Thebio‐energyflyashiswashedusingthewatertreatment.ResultsareillustratedinFigure34.
0
1000
2000
3000
4000
5000
6000Chloride content mg/kg
Treatment steps
chloride content
GraduationProject 9.Results
ing.GuillaumeDoudartdelaGrée page61
Figure34:ChlorideionsinTwencecycloneflyash(B2)beforeandafterdifferenttreatmentsteps
The chlorides inB2ash areeasily soluble andafter treatment almost fulfil themaximumlimitof1000mg/kg.Whensievedwitha500micronsieve,thechloridecontentisreducedby12%.Aftersievingwitha500micronsieveandwatertreatmentonly9%ofthechloridecontent remains, compared to 29% when washed without sieving. This means that byremovingcarbontheefficiencyof thewashing technique increasesby8%,andasaresultthechloridecontentfulfilsthestatedrequirements.
9.2.3 HVCcycloneflyashThisflyashhasnocarbonissuesandthereforethewatertreatmentcouldimmediatelybeapplied to fulfil the chloride requirements presented in Figure 35. However, beforetreatmentalmostallthemetalaluminiumisremoved.Thisisdonebyputtingtheflyashinabottlewithwaterusingaliquidtosolidratiooffourandshakingitfor72hours.
Figure35:ChlorideionsinHVCcycloneflyash(A1)beforeandaftertreatment
Afterwashingthechloridecontentisreducedby96%andfulfilsthestatedrequirements.
0500100015002000250030003500400045005000
Chloride content mg/kg
Treatment steps
chloride content
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
20000
22000
Chloride content mg/kg
Treatment steps
chloride content
GraduationProject 9.Results
ing.GuillaumeDoudartdelaGrée page62
9.2.4 HVCFilterflyashThis fly ash has (like HVC cyclone fly ash) no carbon particles and therefore canimmediatelybewater treated. It also containsmetallicaluminium,butbecause thewatertreatmentwasnot successful further researchhasnot beenperformed.As it canbe seenfromFigure 36, the chloride content of this fly ash is 86 timesmore than allowed. Aftertreatment this amount is reduced by 80% but still it is too high to be used as cementreplacementinconcretestructures.
Figure36:ChlorideionsinHVCfilterflyash(A2)beforeandaftertreatment
9.3 TreatmentdataparticlesizedistributionThe effect of removing small coal particles by thermal treatment or air filtering on theparticle size distribution (PSD) will be discussed, as well as the effect of the watertreatment.Furthermore,crushedbio‐energyflyashsampleswhichfulfilstheNEN‐EN‐450andNEN‐EN‐5950requirementsasillustrated.
9.3.1 TwenceboilerflyashIn Figure 36 the effect of both thermal treatment (TBH) and air filtering (TBA) areillustrated, togetherwith theoriginalPSDofTwenceboiler flyash (B1).AlsocrushedTB(TBAG and TBHG) are present that fulfils the NEN‐EN‐450 andwill be used for strengthdevelopmentinthefollowingchapters.Itcanbeseenthat there isalmostnodifferencebetweenthethermalandtheair filteringtreatment.BothPSDsareshiftedtotheright,indicatingbiggerparticlesduetotheremovalof thesmallcoalparticles.Lookingatthegrindedflyash,65%issmallerthan45micronsand0.1%isbiggerthan212microns,fulfillingthestatedrequirements.Theeffectof thewater treatmenton thePSDof this flyash is illustrated inFigure37.Toobtain these results, the fly ash is first heated andground and afterwards separated intotwogroupswhereoneiswatertreatedforcomparison.Theeffectofthewatertreatmentisnegligible andonly the solublematerials that are removedprovide a small change in thePSD.
0
10000
20000
30000
40000
50000
60000
70000
80000
90000
100000
Chloride content mg/kg
Treatment steps
chloride content
GraduationProject 9.Results
ing.GuillaumeDoudartdelaGrée page63
Figure37:PSDTBbeforeandafterdifferenttreatmentsteps
Figure38:PSDTwenceboilerthermallytreated(TBH)beforeandafterwatertreatment
9.3.2 TwencecycloneflyashandHVCcycloneflyashBothTwence cyclone fly ash andHVC cyclone fly ash arewater treated and ground. Theresults are presented in Figure 39. Also these two bio‐energy fly ashes are fulfilling thestatedrequirementsaftersuccessfullybeenwatertreatedandground.
0
10
20
30
40
50
60
70
80
90
100
0.1 1.0 10.0 100.0 1000.0 10000.0
Cumulative
finer volume [%
]
Particle Size distribution [μm]
TB (B1)
TB <500 and heated (TBH)
TB <500 and air filtered (TBA)
TB <500, air filtered and grinded (TBAG)TB <500, heated and grinded (TBHG)
0
10
20
30
40
50
60
70
80
90
100
0.1 1.0 10.0 100.0 1000.0
Cumulative
finer volume [%]
Particle Size distribution [μm]
TBH
TBH (water treated)
GraduationProject 9.Results
ing.GuillaumeDoudartdelaGrée page64
Figure39:PSDofTwencecyclone(TC)andHVC(A1)beforeandaftertreatment
9.3.3 ComparisonPSDandSSAbetweenoriginalandtreatedbio‐energyflyashAlldataregardingparticlesizedistribution(PSD)andspecificsurfacearea(SSA)oforiginal(O)andtreatedbio‐energyflyashes(T)aredescribedinTable23.Thetreatedbio‐energyflyashisthefinalbio‐energyflyashafteralltreatmentsstepsincludingcrushing.Table21:PSDandSSAdataregardingoriginalandtreatedbio‐energyflyashes
HVCcyclone
(A1) Twenceboiler(B1)Twencecyclone
(B2)PKVASMZ(R)
(O) (T) (O) (TH)* (TA)* (O) (T) (O)d(min.)[µm] 0.7 0.7 1.7 0.7 0.8 2.2 0.7 0.63d(0.10)[µm] 10 7 18 4 5 31 6 5d(0.50)[µm] 36 34 110 31 31 110 34 21d(0.60)[µm] 47 46 138 40 40 130 42 30d(0.80)[µm] 75 80 202 63 63 187 68 63d(0.90)[µm] 126 113 257 83 82 235 90 106d(max.)[µm] 631 275 479 158 158 417 182 832SSA[cm2/g] 1160 1370 382 1980 1720 559 1590 2090
*TH=TwenceboilerthermaltreatedandTA=TwenceboilerairfilteredTable23indicatesthatstillaftercrushingthebio‐energyflyashesPKVASMZhaveasmallerPSDandalargerSSA.However,thereisalargeimprovementaftertreatment(forinstance,theB1d(0.90)decreasesfrom257to82).
9.4 StrengthdevelopmentTheBio‐energyflyashesareusedas5%,10%and20%cementreplacement.MortarsareproducedusingthestandardmethoddescribedinNEN‐EN‐196andarecuredunderwaterfor28days.The7daystrengthisnotmeasuredbecauseoftheslowreactivityofflyashingeneral.TheredlineindicatesPKVASMZandisusedasareference.
0
10
20
30
40
50
60
70
80
90
100
0.1 1.0 10.0 100.0 1000.0
Cumulative finer volume [%]
Particle Size distribution [μm]
HVC Cyclone ash (A1)
TC <500, washed and grinded (2)
Twence Cyclone ash (B2)
HVC cyclone washed
GraduationProject 9.Results
ing.GuillaumeDoudartdelaGrée page65
Firsttheinfluenceofthewatertreatmentisinvestigated,replacing5%and20%ofcement.Regarding flexural strength, the difference is negligible and in the order ofmagnitude of±0.5MPa.CompressivestrengthresultsarepresentedinFigure40andFigure41.
Figure40:Effectofwatertreatmentonthe28daysstrengthdevelopmentusing5%bio‐energyflyashascement
replacement(PKVASMZ(R)47.5MPa)
Figure41:Effectofwatertreatmentonthe28daysstrengthdevelopmentusing20%bio‐energyflyashas
cementreplacement(PKVASMZ(R)41MPa)FromFigure40andFigure41itcanbeconcludedthatingeneralthewatertreatmenthasapositiveeffectonthestrengthgenerationandfurthertreatmentcanbeperformed.Onlythe5%replacementwithTwencecycloneflyashseemstohaveanegativeeffect.Thishowevercanalsobedueattributedtotheaccuracyofthetest.
Twence cyclone ash (B2)
Twence boiler ash (B1)
HVC cyclone ash (A1)
Original 47.5 42.5 43.4
washed 45.7 46.7 49.6
20.0
25.0
30.0
35.0
40.0
45.0
50.0Compressive strength M
Pa]
Bio‐energy fly ash
Twence cyclone ash (B2)
Twence boiler ash (B1)
HVC cyclone ash (A1)
Original 25.6 24.4 25.6
washed 35.3 33.6 30.8
20.0
25.0
30.0
35.0
40.0
45.0
50.0
Compressive strength [MPa]
Bio‐energy fly ash
GraduationProject 9.Results
ing.GuillaumeDoudartdelaGrée page66
Figures42‐47illustratethe5%,10%and20%replacementofcementbythedifferentbio‐energyflyashes.Theresultsareseparatedinoriginalbio‐energyflyash,treatedbio‐energyfly ash (fulfilling the carbon and chloride requirements) and the afterwards ground bio‐energyflyashfulfillingtherequirementsforparticlesizedistribution.Figures42‐44showthemeasuredflexuralstrengthandFigures45‐47–thecompressivestrength.
Figure42:28daysflexuralstrengthdevelopmentusing5%bio‐energyflyashascementreplacementbefore
andafterdifferenttreatmentsteps(PKVASMZ(R)7.5MPa)
The28daysflexuralstrengthwith5%bio‐energyflyashascementreplacementseemstobe almost the same as PKVA SMZ (R) and even close to that of CEM I 42.5 R (8.1MPa).ThermallytreatedTwenceboilerflyashhasthehigheststrengthdevelopment,followedbyHVCcycloneandTwenceboilerair filtered flyash.Twencecyclone flyashseemstoreactnegativelyontreatmentandthestrengthisdecreasedby0.4‐0.6MPa.
Figure43:28daysflexuralstrengthdevelopmentusing10%bio‐energyflyashascementreplacementbefore
andafterdifferenttreatmentsteps(PKVASMZ(R)7.5MPa)
Twence cyclone ash
(B2)
Twence boiler ash air filtered
(B1A)
Twence boiler ash thermal treated (B1H)
HVC cyclone ash (A1)
Original 7.5 7.6 7.6 8.1
Crushed 6.7 7.6 7.9 7.7
Treated 7.1 7.4 7.6
0.01.02.03.04.05.06.07.08.09.0
Flexural strength [MPa]
Bio‐energy fly ash as 5% replacement
Twence cyclone ash
(B2)
Twence boiler ash air filtered
(B1A)
Twence boiler ash thermal treated (B1H)
HVC cyclone ash (A1)
Original 6.8 6.2 6.2 5.5
Crushed 7.1 7.8 7.5 7.2
Treated 6.6 6.7 7.1
0.01.02.03.04.05.06.07.08.09.0
Flexural stren
gth [MPa]
Bio‐energy fly ash as 10% replacement
GraduationProject 9.Results
ing.GuillaumeDoudartdelaGrée page67
Figure43presentsasmalldecreaseof28daysflexuralstrengthforalmostallbio‐energyflyashes (replacement level 10%) compared to the reference (R). However, the flexuralstrength of crushed particles is significantly increased compared to original flexuralstrengthandisevencomparablewiththereference(R).
Figure44:28daysflexuralstrengthdevelopmentusing20%bio‐energyflyashascementreplacementbeforeandafter
differenttreatmentsteps(PKVASMZ(R)7.2MPa)
Figure 44 presents a decrease of 28 days flexural strength for all bio‐energy fly ashes(replacementlevel20%)comparedtoPKVASMZ(R).Stillcrushedparticleshaveahigherflexuralstrengthforallbio‐energyflyashescomparedtothereferenceflexuralstrength.
Figure45:28dayscompressivestrengthdevelopmentusing5%bio‐energyflyashascementreplacement
beforeandafterdifferenttreatmentsteps(PKVASMZ(R)46.9MPa)The 28 days compressive strength development using 5% bio‐energy fly ash as cementreplacement (described in Figure 45) illustrates comparable and even higher strengthdevelopmentscomparedtothereference(R).EspeciallythetwotreatedTwenceboilerflyashesareperformingwell.TwencecycloneandTwenceboilerthermallytreatedflyashare
Twence cyclone ash
(B2)
Twence boiler ash air filtered
(B1A)
Twence boiler ash thermal treated (B1H)
HVC cyclone ash (A1)
Original 5.6 5.4 5.4 5.3
Crushed 6.6 6.9 6.7 6.1
Treated 6.0 6.0 5.7
0123456789
Flexural strength [MPa]
Bio‐energy fly ash as 20% replacement
Twence cyclone ash
(B2)
Twence boiler ash air
filtered (B1A)
Twence boiler ash thermal treated (B1H)
HVC cyclone ash (A1)
Original 47.5 42.5 42.5 43.4
Crushed 43.1 47.3 46.0 46.3
Treated 46.0 45.3 49.5
30.0
35.0
40.0
45.0
50.0
55.0
60.0
Compressive strength [MPa]
Bio‐energy fly ash as 5% replacement
GraduationProject 9.Results
ing.GuillaumeDoudartdelaGrée page68
examples that indicate that crushing not always increases the strength developmentcomparedtonotcrushed.Theaboveresultsalsoindicatethatthebio‐energyflyashhasabindingactivitybecausetheydevelopahigherstrengththanthereference(R).
Figure46:28dayscompressivestrengthdevelopmentusing10%bio‐energyflyashascementreplacement
beforeandafterdifferenttreatmentsteps(PKVASMZ(R)46.1MPa)
The10%replacement is atall fronts lower than thatofPKVASMZ.HVC cyclone showsahighcompressivestrengthdevelopmentcomparedtotheoriginal.
Figure47:28dayscompressivestrengthdevelopmentusing20%bio‐energyflyashascementreplacement
beforeandafterdifferenttreatmentsteps(PKVASMZ(R)41MPa)The Twence cyclone 28 days compressive strength with 20% replacement has similarresults compared to the 10% replacement regarding the treated sample. It seems thatcrushing somehowaffects the fly ash, creating lower strength results in all replacements.Furthermore,thestrengthresultsoftheTwenceflyashesarecomparablewiththestrengthresultsofPKVASMZ.
Twence cyclone ash
(B2)
Twence boiler ash air
filtered (B1A)
Twence boiler ash thermal treated (B1H)
HVC cyclone ash (A1)
Original 44.4 42.2 42.2 27.5
Crushed 44.8 45.8 43.6 43.8
Treated 45.7 40.7 42.7
25.0
30.0
35.0
40.0
45.0
50.0
55.0
60.0Compressive strength [MPa]
Bio‐energy fly ash as 10% replacement
Twence cyclone ash
(B2)
Twence boiler ash air
filtered (B1A)
Twence boiler ash thermal treated (B1H)
HVC cyclone ash (A1)
Original 25.6 24.4 24.4 25.6
Crushed 41.0 39.5 39.9 34.6
Treated 35.7 30.6 31.4
20.025.030.035.040.045.050.055.060.0
Compressive strength [MPa]
Bio‐energy fly ash as 20% replacement
GraduationProject 9.Results
ing.GuillaumeDoudartdelaGrée page69
9.5 SideeffectsThis paragraph summarizes different experiments that got attention during the researchbecauseof interestingphenomena. Itstartedbywater treatingthebio‐energy flyashestoinvestigatetheeffectiveness,followedbythermaltreatingallfourbio‐energyflyashesandremoval of metallic aluminium. Results are later investigated using scanning electronmicroscopyandX‐raypowderdiffraction.
9.5.1WatertreatmentThat thewater treatmentdidnot always result inwhatwashoped (which canbe clearlyseen in Figure48).The figure illustratesHVC filter fly ash (left) andHVC cyclone fly ash(right)afterwatertreatmentandseparationofliquidandsolidusingafilter.Theflyashesaredriedintheovenfor12hoursat50degreesCelsius.
Figure48:FixedHVCfilterflyash(left)andHVCcycloneflyash(right)duetocementitiouspropertiesThe figure illustrates that the fly ash already posses cementitious properties. The sameappliesforHVCcycloneflyash,whichis,however,easiertobreak(lowerstrength).
9.5.2ThermaltreatmentThethermaltreatmentalsocreatedsomeratherspectacularresultsaspresentedinFigure49.After treatmentat750degreesCelsiusHVC filter flyashcompletelyclompedtogetherinto small granulates. After crushing this thermally treated fly ash is used as a cementreplacement;lowstrengthresultswerefound,probablyduetophasechanges.
Figure49:ThermaltreatedHVCfilterflyashafteronehourinovenat750degreesCelsius
GraduationProject 9.Results
ing.GuillaumeDoudartdelaGrée page70
9.5.3StrengthresultsAtestwasperformedinordertoinvestigatethestrengthresultsofthermallytreatedbio‐energy fly ash, followedbyawater treatmentand crushing toparticles smaller than100microns.Furthermorea longercuring time is tested for thermally treatedandonlywatertreatedbio‐energyflyashes,toinvestigateifthereisanypozzolanicactivityafter28days,theresultsarepresentedinTable24.
Table22:Strengthdevelopmentafter28daysofthermaltreatedbio‐energyflyashes
Bio‐energyflyash 28daysflexuralstrength[MPa]
28dayscompressivestrength[MPa]
60dayscompressivestrength[MPa]
HVCcycloneflyash(A1)(20%replacement)
6.3 25.4 ‐
HVCfilterflyash(A2)(20%replacement)
5.5 35.5 36.2
Twencecycloneflyash(A1)(20%replacement)
7.1 40.3 ‐
Twenceboilerflyash(B2)(20%replacement)
7.4 40.6 ‐
Twenceboilerflyash(B2)watertreated
(40%replacement)3.7 18.7 23
Table24illustratesthatthethermaltreatmentdidnothaveapositiveeffectonthetensile,andespeciallynotonthecompressivestrength,ofHVCbio‐energyflyashes.Evenafter60daysnopozzolanic activity is found.This indicates thathydration ishinderedby thermaltreatment. The small increase of 0.7 MPa can be seen as a measurement error. Twenceboilerwatertreatedflyashwith40%replacementofcementillustratesanincreaseof23%,whichindicatesthatthebio‐energyflyashpossesslowpozzolanicreactions.Further investigationof theeffectof thewater treatmenton the strengthdevelopment isdonebyincreasingreplacement.TheresultsarepresentedinTable25.
Table23:28daysstrengthdevelopmentofwatertreatedbio‐energyflyasheswithdifferentcementreplacementpercentages
28dayscompressivestrength[MPa] 28dayscompressivestrength[MPa]
Replacement[%]
HVCcycloneflyash
original(A1)
HVCcycloneflyash
watertreated(A1)
HVCfilterflyash(A2)original
HVCfilterflyash(A2)watertreated
5% 43.4 49.6 46.7 43.920% 25.6 30.8 40.8 33.730% 26.8 21.4 33.0 14.940% 19.9 15.9 23.0 5.0
28dayscompressivestrength[MPa] 28dayscompressivestrength[MPa]
Replacement[%]
Twencecycloneflyash(B1)original
Twencecycloneflyash(B1)water
treated
Twenceboilerflyash(B2)original
Twenceboilerflyash(B2)water
treated5% 47.5 45.7 42.5 46.720% 25.6 35.3 24.4 33.630% 23.8 23.9 20.1 26.040% 13.7 17.5 16.1 18.7
GraduationProject 9.Results
ing.GuillaumeDoudartdelaGrée page71
ItwasfoundthatthewatertreatmenthadaneffectonHVCfilterflyash.Thisisduetothecementitious properties that it possesses. The effect on HVC cyclone fly ash was onlyobservedaboveareplacementof20%.Twenceashesare ingeneralnotaffected.Onlythe5%replacementofTwencecycloneflyashislowerthanexpected.
9.5.4Scanningelectronmicroscopy(SEM)HVCflyashesseemtohavecementitiousproperties.Stillstrengthresultsshowsometimespromisingresultsandsometimesdisappointingresults,mostlydependingonthetreatment.Tofurtherinvestigatetheeffectofthewatertreatment,1hourwatertreatedand72hourswatertreatedHVCcycloneflyashareinvestigatedandcomparedwith1hourand24hourswatertreatedTwenceboilerflyash.TheresultsarepresentedinFigure50andFigure51.
Figure50:HVCcycloneflyashafteronehourofwatertreatment(left)and72hoursofwatertreatment(right)
Figure51:Twenceboilerflyashafteronehourofwatertreatment(left)and24hoursofwatertreatment(right)Afterwatertreatmentofonehour,almostnocementitiouspropertiesarefoundinbothHVCcycloneandTwenceboilerflyash.However,after72hoursofwatertreatmentthereseems
GraduationProject 9.Results
ing.GuillaumeDoudartdelaGrée page72
tobehydrationproductsformedthatcompletelycoverthesurfaceofthebio‐energyflyashparticles.ThisresultisnotobservedwhenwatertreatingtheTwenceboilerflyash.ItseemsthatthethreedaysofHVCcyclonewatertreatmenttoremovemetallicaluminiumhasmoreeffect thanwouldbeexpected.Thisall seems tobecausedbyself‐cementitiousproperties.
9.5.5X‐raydiffractionpattern(XRD)Theeffectofdifferenttreatmentslikewaterandthermaltreatmentforbio‐energyflyashesare investigatedusingX‐raydiffraction.The results ofHVC filter fly ash arepresented inFigure52andtheresultsofHVCcycloneflyashinFigure53.Figure54presentstheXRD‐pattern of PKVA SMZ.TheXRD‐pattern of Twence boiler andTwence cyclone fly ash areillustratedinFigures55and56,respectively.
Figure52:XRDpatternoforiginal,watertreatedandthermallytreatedHVCfilterflyash
HVCfilterflyashcontainsphaseslikeanhydrite(CaSO4),portlandite(Ca(OH)2),lime(CaO),quartz (SiO2), chlorides (CaCl2 and NaCl) and calcite (CaCO3). After water treatment thestructure is similar and there is only an increase of Ca(OH)2 and a decrease of chlorides.WhenHVCfilterflyashisthermaltreatedat750degreesCelsiusthestructureiscompletelychanged.Portlanditeisreducedbecauseofevaporatedwaterandcalciteisreducedbecauseof releaseof carbondioxide (CO2)andbecauseof this free lime is increased.Anhydrite isalsocompletelyremoved.
GraduationProject 9.Results
ing.GuillaumeDoudartdelaGrée page73
Figure53:XRDpatternoforiginal,onehourwatertreatedandthreehourswatertreatedHVCcycloneflyash
HVC cyclone fly ash compared to HVC filter fly ash contains similar phases of anhydrite(CaSO4),quartz(SiO2),chlorides(CaCl2andNaCl)andcalcite(CaCO3),but lessportlandite(Ca(OH)2) and lime (CaO). After water treatment there is a decrease of chlorides andanhydrite.WhenHVC cyclone fly ash iswater treated for72hours the structure is still comparablewith the 1 hour water treated. Small changes are found in the available anhydrite andsilicate.Themainpeeksarestillthoseofquartz.
Figure54:XRDpatternofPKVASMZ(R)
PKVASMZmainlyconsistofquartz,freelime,mullite,magnesiumoxideandhematite(Fe2O3).
GraduationProject 9.Results
ing.GuillaumeDoudartdelaGrée page74
Figure55:XRDpatternoforiginal,watertreated,thermallytreatedandthermally+watertreatedTwenceboiler
flyashTwencecycloneandTwenceboilerhavenotonlysimilaroxidecompositionbutalsoX‐raydiffractionpatterns.Mainpeaksarequartzandotherformsarenotfound,whichmeansthattheyareinverylowamountwhencomparedtoquartz.
Figure56:XRDpatternoforiginal,watertreated,thermallytreatedandthermally+watertreatedTwencecycloneflyash
GraduationProject 10.Discussionsandconclusions
ing.GuillaumeDoudartdelaGrée page75
10. DiscussionsandconclusionsThissection isdivided intothreedifferentparts.The firstpart isabout thebio‐energy flyashes, comparingwithcoal combustion flyash.Thesecondpart isadiscussionabout theused treatmentmethods and reliability.The thirdpart is about theobtained final resultsregardingtreatmentandstrengthresults.
10.1 RelationsanddifferencesDuringthisresearch,workingwithdifferenttypesof flyash, itcanbestatedthatthewellknowncoal‐combustionflyashthathasbeenusedfordecadesiscompletelydifferentthenbio‐energy fly ash or any types of fly ash described in this report. The benefits of coalcombustionflyashlikeasmallparticlesizedistribution,andthepresenceofcensospheresand plerospheres, increasing the workability and creating a denser structure do notnaturally apply to bio‐energy fly ash. Studying the physical and chemical properties, bio‐energy fly ash has a larger particle size distribution with very little cenospheres andplerospheres.The largerparticlesizedistributionwouldnegativelyaffect thereactivityofbio‐energyflyashandwouldmakeit lesssuitable inconcretemixtures.Lesscenospheresandplerospheresareleadingtoamaterialwithhigherdensityandlessglassyphases.Mostofthebio‐energyflyashconsistsoffusedparticleswithunburnedcoalparticlesthatnegatively affect the workability. These fused particles as well as coal particles can berelated to the incomplete burningprocess, aprocesswhere the focal point is feeding thecombustionroom,keeping the fireandtemperaturesconstant, resulting inwastestreamsthatcontain incompletelyburnedparticles.Theseparticlesnegativelyaffecttheuseoftheflyashinconcretemixtures,resultinginaneedfordifferenttreatmentmethodsmakingthismaterialalreadymoreexpensive thancoalcombustion flyash.However,bynotusingthematerial in concrete mixtures it will be landfilled, which is not desirable because ofincreasingcosts.TheuseoftheNEN‐EN‐450asqualitymeasureforbio‐energyflyashisdebatablebecauseactually we are talking about two different materials that are called “fly ash”. Still therequirements regarding developed strength and chloride content can be applied becausethese are safety measures to ensure quality and have nothing to do with what type ofmaterialisused.Finally,thecontentofbio‐energyflyashislikeanytypeofwastecontinuouslydaily,weekly,monthly etc. depending on thewaste streams that are put in the combustion room. It isthereforethatcontinuouslythereisaneedofchloridetestingtoensurequality.
10.2 Treatmentmethodsandreliability
10.2.1CarbonremovalBecauseofthehighLOIofTwencebio‐energyflyashes,mainlyduetohighcarboncontentand the negative influence of carbon on the chloride removal of bio‐energy fly ash, thecarbon content needed to be reduced. After treatment the carbon content is significantlyreducedand fulfils theLOI requirements.Thereductionof thecarboncontentcausedtheremoval of chlorides present in the carbon structure. Furthermore, it increased theefficiencyof thewater treatment toremovechlorides.Whenthecarbonwasstillpresent,
GraduationProject 10.Discussionsandconclusions
ing.GuillaumeDoudartdelaGrée page76
the soluble chlorideswhere somehowattached to the surfaceof the carbonparticles andpreventedfromthedissolvinginwater.Duetothereductionofcarbon,thechloridescouldnowdissolveinwater.
10.2.2WatertreatmentOptimizationofthewatertreatmentmethodwasoneofthebiggestchallengesduringthisresearch,becausenoinformationwasavailableaboutwatertreatmentofbio‐energyflyashtoremovesolubleelements.Available informationaboutremovingsolubleelements frommunicipal solid waste was not applicable or incomplete. First results did not meet theexpectations.Afteralotoftrialanderrorandtryingtounderstandthechloridebehaviourof bio‐energy fly ashes in water, slowly reliable results came up. Still there is a lot ofuncertaintyaboutthewatertreatment.Theinfluenceofspeedandtimeandtemperatureisnotfullystudiedandcanhaveenormousimpactontheefficiencyofthetreatment.Until30‐60minutes the removal of chloride ionswas stable. For this research thiswas sufficientbecause when applying in an industrial environment, hours of water‐treatments areincreasing costs and decreasing the utilization. However, longer duration could removechlorideswith low solubilitywhichwouldmake themethodmore beneficial. During thetreatmentalsoothersolubleelementsaredissolvedinthewater,butthisquantityhasnotbeen fully investigated, and perhaps increasing the time will have a positive effect onremovingtheseelements.Thesameprincipleappliestotheshakingspeedortheincreaseoftemperature.However,italsoneedstobeconsideredthatthewatertreatmentisnotalwaysbeneficial,evenifitisneededtoreducethechloridecontent.Among the used bio‐energy fly ashes, there are bio‐energy fly ashes with cementitiousproperties,mainlyduetotheavailablefreelimeandanhydrite.Availablefreelimewill, incontact with water, form portlandite that will ensure a high pH, activating the fly ash.Reducingtheamountoffreelimebywashingwilldecreasetheactivationofflyashandwillresultinastrengthdecrease.Theavailableanhydritewill increaselongtermstrength,butafteralongwatertreatmenttheanhydriteisdisappeared.During the research it is found that the formation of portlandite takes time. Short timewashingofaroundanhourwouldhave lessnegativeeffect thana longdurationofwatertreatment.
Allinall,thisresearchshowsthatthemethodofwatertreatingbio‐energyflyashtoremovesolubleelements isworking,butstill there isaneed for furtheroptimizationto increasetheremovalofsolubleelementsandtoensurethereliabilityandefficiencyofthemethod.Sofar,thewatertreatmentusedfornoncementitiousflyashesindicatesapositiveeffectonthe strength development because of decreasing the concentration of contaminants thatotherwisewouldhavedisturbedthecementmatrix.
10.2.3MethodofmeasuringchloridecontentIn this research twomethods are used to test the chloride content of fly ash. Using dataprovidedbythefactory,thechloridecontentisknownbutcanstilldifferfromtheoriginalsample because it is not a homogenous material. To test the solid material, the methoddescribedinParagraph8.2.3isused,thatprescribestheuseofnitricacidtodissolvetypesCl‐containingphases.Testing the chloride contentofwashed solidbio‐energy fly ash and
GraduationProject 10.Discussionsandconclusions
ing.GuillaumeDoudartdelaGrée page77
comparingittotheobtainedchlorideconcentration(whichisobtainedwhenmeasuringthechloridecontentofwaterfromthewatertreatment),adifferenceof500kg/mgisfound.Inmostcasesthereis lesschlorideinthewater‐treatedflyashthanthenwouldbeexpectedwhenmeasuringthechloridecontentofthewater.Thereasonforthisisnotknown,butthemeasuredchloridecontentinwaterisafact.Thesamewasproventobethecasewhenmeasuringthechlorideconcentrationofthesolid‐flyash. From the validation study described in Paragraph 9.1.3 with known chlorideconcentration, indicates thatonly55‐70% ismeasuredwhile theusedchloridesarequitesoluble. It is therefore not exactly known if the stated limit of 1000 kg/mg is indeedguaranteed when measuring the solid samples, it is only certain that the chlorideconcentrationhasdroppedsignificantly.
10.2.4GrindingWhen grinding thematerials, they fulfil not only the requirements described inNEN‐EN‐450regardingparticlesizedistribution(PSD),butalsotherequirementsthatthemaximumpercentage larger than 212micron should not exceed 3%m/m (NEN‐EN‐5950), and theexpectedpozzolanicactivityandworkability.Suchasearlierdescribed,decreasingthePSDand increasing the specific surface area increases the reactivity of fly ashes and has apositiveeffecton the strengthdevelopmentduring the first28days. Itwasexpected thatthe strength results of crushed bio‐energy fly ashwould indicate higher strength resultsthan not crushed bio‐energy fly ash. However, from the obtained strength results thedecreaseinparticlesizedistributiononlyhasasignificanteffectwhenahighpercentageofbio‐energy flyash isused.Upon5%and10%replacement theeffect isbarelyvisibleandmayindicatethatbio‐energyflyashisnotassensitivetoadecreaseofparticlesizesasthenormalreferenceflyash.Moreover,thegrindinginordertoreducetheparticlesizesisanexpensivemethod.Althoughinthisresearchtherewasnoinvestigationwhytheinitialbio‐energy fly ash from the two factories (Twence and HVC) have different particle sizedistributions. Two facts are known, HVC Alkmaar crushes its waste wood beforeincinerationandithasaparticlesizedistributionof85%below100microns.Ifthiswouldbe the reason of a decrease in particle size distribution it could be preferable to firstdecrease the size of the waste that is incinerated in a more efficient way and therebydecreasingthesizeofthebio‐energyflyashes.Thesecondfactisthatbothfactorieshaveadifferentincinerationprocess,andbecauseofthehighcarboncontentofTwenceHengeloitcan be concluded that the burning process of this factory is not complete enough, andprobablyalsothereasonofbiggerparticles.
10.2 StrengthresultsReplacingcementwithflyashwill inmostcasesresultinalowerstrengthduringthefirst28days.Themore cement is replaced the less strength is developed.This is because thecementstructure isdisturbedbyreplacementby flyashparticles thatareslowlyreactiveandthereforehasfewerconnections.Whenreplacing20%ofcementwithflyash,andtheinitial100%cementhasstrengthof50MPa.Thestrengthshouldbeatleast40MPaifnot;theflyashnotonlydoesnotparticipatetothedevelopmentofstrength,butactuallyhinderscementhydration.Comparingthereplacementofcementwithoriginalnottreatedflyashandwashedflyashindicatesanincreaseofstrengthformostofthebio‐energyflyashes.OnlyTwencecyclone
GraduationProject 10.Discussionsandconclusions
ing.GuillaumeDoudartdelaGrée page78
flyashhaslowerstrengthresultsat5%replacement,andcementitiousbioenergyflyashessuchasHVCfilterflyashshowanegativeeffectonthestrengthdevelopmentafterwashing.In general HVC cyclone fly ash is less promising at high replacement levels compared toTwenceflyashes.Thereareseveralreasonsthatcanleadtolowerstrengthresults.IntheoriginalcontentofHVCcycloneflyashtherewasahighamountofmetallicaluminiumandchloridespresent. It isbelieved that the chlorideshavenoeffect on the28days strengthdevelopmentandonlyaffectthedurabilityoftheconcretewhenusedwithreinforcement.However, the available metallic aluminium, which in contact with water will createhydrogen,will increasethevolumeofthesample,resulting inan increaseofporosityanddecreaseofcompressiveandflexuralstrengths.Fromearliertrialstheremovalofmetallicaluminium by using water will take longer than 24 hours. From experiments it took 72hourstoremovetheavailablemetallicaluminium.Theremovedquantitywasonlyvisuallyobserved,showingtheformationofhydrogenuptothe72hours.Afterthisperiodoftimenoformationofhydrogenwasobserved.Thisislaterconfirmedwhennodeformationoftheprisms was detected when measuring strength results. The effect of 72 hours of watertreatmentisfurtherinvestigatedusingSEMandXRD.SEM images of the 72 hours treated HVC cyclone fly ash illustrate the formation ofhydrationproductsonthesurfaceoftheparticles.Thenatureofthesehydrationproductsislater investigated using X‐ray diffraction, but could not reveal any explanation of a cleardecreaseofanhydrite.Still,becauseofthesehydrationproducts,treatmentofHVCcycloneashaffectsthestrengthresultsandisthereasonthatHVCcycloneashfor20%replacementhasa5MPa lowerstrengthcomparedtoTwence flyashes. Ifmetallicaluminiumcouldberemovedbyaneddycurrentseparatorandonehourofwatertreatmentisnoteffectingtheflyash(asisbelievedduetoearlierresults),HVCcyclonecanstillhaveapotentialasfillerorasapartlyreplacementofcement,withthesamepropertiesasTwenceflyashes.ComparingthermallytreatedTwenceboilerflyashwithTwenceairfilteredboilerflyash,comparableresultsarefoundfor20%replacement.Alsothestrengthresultsofthecrushedsampleswithdifferentreplacementpercentagesof5%and10%aresimilar.ThetreatmentofTwencecycloneonlyhaseffectforhighreplacementpercentagesof20%.Thisisprobablyduetothefollowingreason:ingeneralasmallreplacementamountwillnothavemuch influenceonstrengthresults than largerreplacementamounts. If treatment isreallyeffectiveit isclearlyvisiblebyanincreaseinstrengthcomparedtotheoriginal,nottreated bio‐energy fly ash. For this sufficient amount of replacement is needed,which isclearlyvisiblefor20%replacementbecausethentheappliedbio‐energyflyashreallyplaysaroll.Observing the flexural and compressive strength results of bio‐energy fly ashes show amaterialthathassimilarpropertiestothoseofreferencePKVASMZ.Theeffectofcrushingtheflyashestoincreasethereactivityisnotalwaysthateffectiveaswasinitiallythought.It is believed that the obtained strength results can be improved because of two effectsduring thecreationof thesamples.First, thewaterdemand iskept thesamefordifferentreplacementfactorsbecauseitwasuncertainifthebio‐energyflyashbehavesthesameasPKVASMZ. From the strength results and thedelayed strength increase in time, it seemsthatbio‐energyflyashindeedpossessespozzolanicpropertiesandthatthewaterdemandthereforecouldbedecreased.Second,theusedjoltingmethoddescribedinNEN‐EN‐196‐1
GraduationProject 10.Discussionsandconclusions
ing.GuillaumeDoudartdelaGrée page79
is believed to be insufficient. After crushing, a high porosity was found by insufficientremovalofairbubbles.Itisthereforepreferredtouseavibrationtable.
Finally,normallythesphericalshapeoftheflyashparticlesproducesarollingeffectatthepointofaggregatecontact,reducingfrictionandimprovingthefluidityofthecementpaste.Itwasfoundthatfortreatedbio‐energyflyashthespreadofthecrushedandnotcrushedbio‐energyflyashwasgoodatallreplacementfactors,despitethecoarseparticles.Thisisan improvementbecausethespreadoftheoriginalbio‐energyflyashwaspoor(probablyduetotraceelementsandcontaminantsonthesurfaceoftheparticles).Thiswouldindicatetheneedforplasticizers.Becauseofthegoodflowabilityofthetreatedbio‐energyflyashes,plasticizersareneverused.
10.3 RecommendationsDuringtheresearchitwasnotpossibletoobtainallrequiredresultsanddiscussedsubjectsin the literature study. The points described below and subsequent investigations couldlead to better understanding and potential of the use of bio‐energy fly ash as cementreplacement.
Furtheranalysisthatcouldbeperformed:
LOIcontentoftreatedbio‐energyflyash; Chloridecontentoftreatedbio‐energyflyash(XRF‐data); Testing the efficiency of water treatment when removing known amounts of
chlorideswithsimilarpH‐valueaswhenwatertreatingbio‐energyflyashes. Leachingtestontreatedflyashestoconformthatthebio‐energyflyashesareinert
ornon‐hazardous; Leaching tests on created samples to test if they fulfil the requirements of non‐
shapedbuildingmaterials; AnalysisofchloridecontentofcoalparticlesusingSEM/XRF; PerformX‐raydiffractionmeasurementswith smaller step size (<0.02) and count
time(<1.0)formoreaccurateresults; X‐raydiffractiononbio‐energyflyashestoinvestigatethequantityofamorphousor
glassphases; X‐raydiffractiononthecreatedsamplestoinvestigatethehydrationproducts.
Investigationsthatcouldbeperformed:
UsingK‐factortodeterminetherequiredwatercementratio’s; Useofplasticiserstoseetheireffectonthestrengthdevelopment; Activationeffectbyaddingfreelime/otheractivators.
Graduationproject References
ing.GuillaumeDoudartdelaGrée page80
ReferencesAccreditatiecollege Bodembeheer, 2008.Accreditatieprogramma voor keuring van partijen grond,bouwstoffen en korrelvormige afvalstoffen Onderdeel : Algemeen AP04 ‐ A. Gouda: StichtingInfrastructuurKwaliteitsborgingBodembeheer(SIKB).
Advanced cement technologies, 2012. Metakaolin. [Online] Available at:http://www.metakaolin.com/metakaolin[Accessed3February2012].
Afval Energie Bedrijf, 2006. Méér waarde uit afval. Amsterdam: Energie Bedrijf gemeenteAmsterdam.
Banse,M.,Faaij,A.,Hoefnagels,R.&Dornburg,V.,2009.AnalysisoftheEconomicImpactofLarge‐ScaleDeployment ofBiomassResources for Energy andMaterials in theNetherlands. Santiago deChile:EnergieTransitiePlatform'GroeneGrondstoffen'.
Barron, A., 2010. Hydration of Portland Cement. [Online] (1.11) Available at:http://cnx.org/content/m16447/1.11/content_info[Accessed5November2011].
Brouwers,H.J.H.,2005.TheworkofPowersandBrownyardrevisited:Part2.CementandConcreteResearch,35,pp.1922‐36.
CBS,2011.Renewableelectricity1991‐2010.DenHaag:CentraalBureauvoordeStatistiek.
CDEMMineralsBV,2003.Top‐Crete.Productinformation.Geldermalsen:CDEM.
CDEM,2011.Companyoverview,OurHistory. [Online]Availableat:http://www.cdem.nl/node/13[Accessed25November2011].
Changling,H.,Osbaeck,B.&Makovicky,E.,1995.Pozzolanicreactionofsixprincipalclayminerals:Activiation reactivity assessment and technological effects. Cement and Concrete Research, 25,pp.1691‐702.
Costa, U. & Masazza, F., 1981. Natural pozzolanas and fly ashes.Material and Research Society:effectsofflyashincorporationincementandconcrete,16November.pp.134‐44.
Costa,U.&Massazza,F.,1983.Somepropertiesofpozzolaniccementscontainingflyashes.InACI,A.c.i.,ed.ProceedingsoftheCANMET/ACI first internationalconcferenceontheuseof flyash,silicafume,slagandothermineralby‐productsinconcretevol1.Detroit,1983.
CUR, 1992. Vliegas als vulstof in beton. Gouda: Civieltechnisch Centrum Uitvoering Research enRegelgeving.
Diamond,S.,1981.Effectsoftwodanishflyashesonalkalicontentsofporesolutionsofcement‐fly‐ashpastes.CementandConcreteResearch,11,pp.383‐94.
Diamond,S.,1985.SelectionandUseofFlyAshforHighwayConcrete.JointhighwayResearchProject.WestLafayette,Indiana:PurdueUniversity.
Doudart de la Grée, G.C.H., 2012. Physical‐chemical upgrading and use of bio‐energy fly ashes asbuilding material in the concrete industry. Graduation report for master degree. Eindhoven:EindhovenUniversityofTechnology.
Dpcleantech, 2011. Biomass and waste to energy solutions. [Online] Available at:http://www.dpcleantech.com/about‐dpcleantech/past‐a‐future[Accessed21October2011].
Graduationproject References
ing.GuillaumeDoudartdelaGrée page81
DWMA, 2010. Jaarverslag 2009. Monitioring reststoffen van de verbranding van afval, biomassazuiveringsslib[Monitoringresidues from incinerationofwaste,biomassandsewagesludge].Annualreport.Bunnik:LibertasDutchWasteManagmentAssociation.
Eijk, R.J.v., 2001. Hydration of cementmixtures containing contaminants. Twente: PrintPartnersIpskampBV.
Fox,J.M.,2005.Changesinflyashwiththermaltreatment.InSubmittedforconsideratoninthe2005WorldofCoalAsh.Cleveland,2005.
Fraay,A.,1987.Vliegasinmortelsenbeton,puzzolanewerking,sterkteenduurzaamheid,vol.1en2.
Ghose,A.&Pratt,P.L.,1981.Studiesof thehydrationreactionsandmicrostructureofcement‐fly‐ashpastes.InDiamond,S.,ed.Materialresearchsosciety‐in:Effectsofflyashincorporationincementandconcrete.Boston,1981.
Glimmerveen,L.,1996.Shouldtreesnowbemoreactivelyusedintherehabilitationofheavymetalcontaminatedsites?AspectsofAppliedBiology,44,pp.357‐61.
Heiri,O.,Lotter,A.F.&Lemcke,G.,2001.Losson ignitionasamethod forestimatingorganicandcarbonate content in sediments: reproducibility and comparability of results.Paleolimnology, 25,pp.101‐10.
Hendriks,C.F.,Koppen,A.E.&Nijkerk,A.A.,1999.DeBouwcyclus.Best:ÆneasBV.
Herbert, L., 2011. The chartered insitution of waste management. [Online] Available at:http://www.ciwm.co.uk/web/FILES/About_CIWM/100_yrs_London_and_SE_centre.pdf [Accessed30November2011].
HetEuropeesParlementendeRaad,2001.RICHTLIJN2001/77/EGbetreffendedebevorderingvanelektriciteitsopwekking uit hernieuwbare energiebronnen op de. Brussel: Publicatieblad van deEuropeseGemeenschappen.
Hewlett,P.C.,2004.Lea'sChemistryofCementandConcrete.4thed.ElsevierScience&TechnologyBooks.
HVC, 2009. Publicaties. [Online] Available at: http://www.hvcgroep.nl/over_hvc/publicaties[Accessed26September2011].
Joshi, R.C., 1970. Pozzolanic Reaction in Synthetic Fly ashes. Ames: PhD Dissertation, Iowa StateUniversity.
Joshi,R.C.,1979.SourcesofPozzolanicActivity inFlyAshes. InAcriticalReview.Proceedings,5thinternationalflyashUtilizationsymposium.Atlanta,Georgia,USA,1979.
Joshi, R.C. & Lohtia, R.P., 1997.Flyash in concrete :production,propertiesanduses. Amsterdam:Gordon&BreachScience.
Joshi, R.C. & Marsh, B.K., 1986. Some Physical, Chemical and Mineralogical Properties of SomeCanadianFlyAshes.MRSProceedings,86,pp.113‐25.
Killoh,D.C.,Parrott,L.J.&Patel,R.G.,1989.InfluenceofCuringatDifferentRelativeHumiditiesontheHydrationandPorosityofaPortland/FlyAshCementPaste,inFlyAsh,SilicaFume.InMalhotra,V.M.,ed.SlagandNaturalPozzolansinConcrete.Detroid,1989.
Kinuthia, J.M., Wild, S., Sabir, B.B. & Bai, J., 2000. Self‐compensating autogenous shrinkage inPortlandcement‐metakaolin‐flyashpastes.AdvancedCementResearch12,pp.35‐43.
Graduationproject References
ing.GuillaumeDoudartdelaGrée page82
Kurtis, K., 2011. Lecture, Portland Cement Hydration. [Online] Atlanta, Georgia Available at:http://people.ce.gatech.edu/~kk92/hyd07.pdf.
Lea,F.M.,1970.TheChemistryofCementandConcrete.3rded.London:EDWARDARNOLDPUBL.
Lukas,W.,1976.Matér.Constr.9.p.331.
Mehta,P.K.,1984.TestingandCorrelationofFlyAshPropertieswithRespectofPozzolanicBehavior.California:EPRI,PaloAlto.
Mindess,S.&Young,J.F.,1981.Concrete.N.J.:Prentice‐Hall:EnglewoodCliffs.
Murat,M.,1983.Hydrationreactionandhardeningofcalcinedclaysandrelatedminerals.CementandConcreteResearch,13,pp.259‐66.
NETL, 2011. History of Coaluse. [Online] Available at:http://www.netl.doe.gov/keyissues/historyofcoaluse.html[Accessed30November2011].
Neville,A.M.,2004.PropertiesofConcrete.4thed.England:PearsonEducationlimited.
Ogawa, K., Uchiwaka, H., Takemoto, K. & Yasul, I., 1980. Themechanism of the hydration in thesystemc3s‐pozzolana.Cementandocncreteresearch10,V,pp.683‐96.
Port of Antwerp, 2000‐2006. Research project landscapemounds, App 06. Thesis “Broekpolder”.[Online] Available at: http://www.landscapingwithsediments.be/html/en_download.html[Accessed5Oktober2011].
Rayment,P.L.,1982.Theeffectofpulverized‐fuelashonthec/smolar ratioandalkali contentofcalciumsilicatehydratesincement.CementandConcreteResearch,12,pp.133‐40.
Sakai,E.etal.,2005.HydrationofFlyAshCement.Cem.Concr.Res.,35(6),pp.1135‐40.
Services, P.I., 2011. Cenospheres. [Online] Available at:http://www.primaryinfo.com/projects/cenospheres.htm[Accessed2011].
Takemoto,K.&Uchikawa,H.,1980.Hydrationofpozzolaniccement. In7th internationalcongressonthechemistryofcement.Paris,France,1980.
Taylor,H.F.W.,1997.Cementchemistry.2nded.New‐York:ThomasTelford.
Taylor,H.F.W.,Mohan,K.&Moir,G.K.,1985.Analyticalstudyofpureandextendedportlandcementpastes.JournaloftheAmericanCermicSociety.
TweedeKamer,1999.UitvoeringsnotaKlimaatbeleid.’s‐Gravenhage:SduUitgevers.
Uchikawa,H.,1986.J.Res.OnadaCementCompanay38.
Uchikawa, H., Ogawa, K. & Uchida, S., 1985. Influence of character of clinker on the property ofcementpaste.CementandConcreteResearch,15,p.561.
UnitedNations,1998.KyotoprotocoltotheUnitedNationsframeworkconventiononclimatechange.Kyoto.
vandenBerg,P.,2006.Beton technologievooronderwijsenpraktijk.Betonvereniging/Cement&BetonCentrum.
Vliegasunie,2008.JaarverslagVliegasunie2007.p.28.
VROM,2007.Regelingbodemkwaliteit.Staatscourant,20December.p.67.
Graduationproject References
ing.GuillaumeDoudartdelaGrée page83
VROM, 2010. Landelijk afvalbeheerplan 2009‐2021. Naar een materiaalketenbeleid. Den Haag:RuimteenMilieu.MinisterievanVolkshuisvesting,RuimtelijkeOrdelingenMilieubeheer.
Wang,Q.,Feng,J.&Yan,P.,2012.Themicrostructureof4‐year‐oldhardenedcement‐flyashpaste.ConstructionandBuildingMaterials,29,pp.114‐19.
Wesche,K.,1991.FlyashinConcretePropertiesandPreformance.Rillemreport7.London:E&FNSPONTheInternationalUnionofTestingandResearchLaboratoriesforMaterialsandStructures.
Wijers,G.J.,1995.DerdeEnergieNota.’s‐Gravenhage:SduUitgeverij.
Wild, S., Khatib, J. & Roose, J.L., 1998. Chemical and autogenous shrinkage of Portland cement‐metakaolinpastes.AdvanceCementResearch,10,pp.109‐19.
Yudovich,Y.E.&Ketris,M.P.,2005.Chlorineincoal:Areview.InternationalJournalofCoalGeology,67,pp.127‐44.
Zhang, M.H. & Malhotra, V.M., 1995. Characteristics of a thermally activated alumino‐silicatepozzolanicmaterialanditsuseinconcrete.CementandConcreteResearch,25,pp.1713‐25.