铝土矿和红柱石经热处理后的耐火性的比较.docx
《铝土矿和红柱石经热处理后的耐火性的比较.docx》由会员分享,可在线阅读,更多相关《铝土矿和红柱石经热处理后的耐火性的比较.docx(19页珍藏版)》请在冰豆网上搜索。
铝土矿和红柱石经热处理后的耐火性的比较
Effectofthermaltreatmentondamagemechanicalbehaviourofrefractorycastables:
Comparisonbetweenbauxiteandandalusiteaggregates
M.GhassemiKakroudia,b,E.Yeugo-Fogaingc,C.Gaultb,M.Hugerb,
andT.Chotardb,
aDepartmentofCeramicEngineering,UniversityofTabriz,Tabriz51666-16741,IslamicRepublicofIran
bGrouped’EtudedesMatériauxHétérogènes(GEMH),ENSCI,Limoges,France
cMatériauxàFinalitésSpécifiques,EA3834,InstitutdesSciencesdel’IngénieurdeToulonetduVar,Av.G.Pompidou,BP56,83162LaValetteduVarCedex,France
Received15January2008;
revised20March2008;
accepted28March2008.
Availableonline6June2008.
Abstract
Duringservicelife,refractorycastablesaresubjectedtodifferentsolicitations.Theknowledgeoftheirdamagebehaviourundersuchsolicitationsishighlyneededforabetterunderstandingofmechanisms,whichinducethefinalruptureofstructuresmadewithrefractories.Sincethesematerialsareoftenusedaswallsofmetallurgicaltools,thermalgradientsinsuchstructurecanleadtoseveremechanicalstressesintheouterlayeroftherefractorypartwhichisataratherlowtemperature.
Thisstudydealswiththemechanicalpropertiesatroomtemperature(bytensiletest)oftworefractorycastablestreatedatdifferenttemperatures(110 °C,250 °C,500 °C,700 °C,900 °Cand1100 °C)inordertoreproducethethermalgradientinwallsofmetallurgicaltools.Tworefractorycastablesareconsidered:
anultra-lowcementcontentbauxite-basedmaterial(Bau-ULCC)andalowcementcontentandalusite-basedmaterial(And-LCC).
Keywords:
Tensiletest;Mechanicalproperties;Thermalexpansion;Refractories;Castables
ArticleOutline
1.Introduction
2.Materialsandexperimentalprocedures
2.1.Materialandsamplepreparation
2.2.Dilatometry
2.3.Ultrasonicmeasurements
2.4.Tensiletest
3.Resultsanddiscussion
3.1.Young'smodulusvaluesatroomtemperature
3.2.Microstructuralevolutionduringafirstthermalcycleupto1500 °C
3.3.Young'smodulusevolutionduringthermalcyclesatlowertemperatures
3.4.Mechanicalbehaviourintensionafterthermaltreatment
3.5.Sourceofdamageinthecastables
4.Conclusion
Acknowledgements
References
1.Introduction
Theuseofmonolithicrefractoriesinvariousindustries(metallurgical,cement,etc.)iscontinuouslyincreasingforthelast20years[1]and[2].Duringtheirservicelife,refractorycastablesaresubjectedtoseveresolicitations,especiallyfromathermomechanicalpointofviewandaredegradedbyacombinationofseveralmechanisms,mainlythermalshock,abrasion,corrosionandmechanicalimpact.Thebehaviourofthesematerialsfacetothosemechanismsisinfluencedbytheevolutionofmanyfactorssuchastheirchemicalcomposition,theirmicrostructureaswellastheirphasetransformation,whichoccurathightemperatureduringfiringprocess,and/orinservice[3]and[4].
Thephysicalpropertiesofarefractoryconcretearehighlytemperature-dependent.Thisisprimarilycausedbythecomplexhydrationanddehydrationreactionsofcalciumaluminatecement[5],[6]and[7].
Theelaborationofmonolithicscontainingcalciumaluminatecementscontainsseveralstepssuchasmixing,placingandconsolidation,curinganddryoutandfinallyuseinservice.Eachofthesestepswithinthecastableplacingchainareintimatelylinkedtotheinitialhydrationprocessofthecalciumaluminatecement(CAC)[8]and[9].
Therefractorycastablesgenerallypresentcomplexheterogeneousmicrostructureswhichcanprovidestronginternalstressesbythermalsource.Becauseofthemismatchbetweenthepropertiesofphasesandmainlybetweentheircoefficientsofthermalexpansion,theserviceconditionscanconsiderablyaffecttheirinitialmicrostructuralstateandthustheirthermomechanicalproperties[10]and[11].
Previousstudieshavealreadybeenperformedinthefieldofthehightemperaturebehaviourofrefractorycastables[3]and[12].Thispaperdealswithresultsofanexperimentalapproachdevelopedtocharacterisethemicrostructuralchangesanddamageprocesses,whichoccurinsuchmaterialsduringthefirstheating.
2.Materialsandexperimentalprocedures
2.1.Materialandsamplepreparation
Twocommercialcastablesareconsidered.Thefirstoneisalowcementandalusitecastable(And-LCC)madeofandalusiteaggregates,fumedsilica,aluminaandofacalciumaluminatecement.Thesecondisanultra-lowcementbauxitecastable(Bau-ULCC)madeofbauxiteaggregates,fumedsilica,aluminaandofthesamecement.Bothmaterialsarecharacterisedbythesamefumedsilicacontent.InBau-ULCC,thealuminacontentistwotimeshigherthaninAnd-LCC.Table1showsthechemicalcompositionsofthecastablessuppliedbythemanufacturer.Thehighdifferencebetweenthesilicacontentsofthetwomaterialsismainlyduetothehighsilicacontentinandalusiteaggregatescomparedtobauxiteones.Forbothcastables,themaximumaggregatesizeisabout5 mm.Thematerialswerecuredduring24 hat110 °C.Fig.1showspicturesofpolishedsectionsofcuredmaterials.Aftermachining,somesampleshavebeenfiredat250 °C,500 °C,700 °C,900 °Cand1100 °Cinordertosimulateseveralthermalhistoriesbeforecharacterisation.Thesetemperaturelevelshavebeenfixedaccordingtothetemperaturerangerepresentingtherefractorycastablesinspecificindustrialapplications.Firingthermalcyclesarecharacterisedby5 °C/minheatingandcoolingratesandbya5 hisothermaldwellatthemaximumfiringtemperature.
Table1.
Chemicalanalysisandcharacterisationsdataofthetworefractories
Castabletype
And-LCC
Bau-ULCC
Aggregatetype
Andalusite
Bauxite
Al2O3(wt.%)
58
85
SiO2(wt.%)
37.5
10
CaO(wt.%)
2.3
1.1
Fe2O3(wt.%)
0.9
1
Maximumaggregatesize(mm)
5
5
Waterrequirement(wt.%)
4.5–5.5
4.2–5.2
Openporosity(vol.%)
6
10
Apparentdensity(kg/m3)
2600
2970
Full-sizetable
ViewWithinArticle
Full-sizeimage(122K)
Fig.1. Microstructureofstudiedrefractories:
(a)And-LCC;(b)Bau-ULCC.
ViewWithinArticle
2.2.Dilatometry
Inordertostudythermalexpansionmismatcheffects,samplesofmatrixandofaggregateshavebeenprepared.Thesamplesofthematrixwerepreparedbycastingandthesamplesofaggregates(grainsizelessthan200 μm),wereshapedbypressing.Thethermalcycles(heating/cooling)werecarriedoutwithaslopeof5 °C/min.
Theinfluenceofthetemperatureonamaterialcauses,ingeneral,variationsofitsapparentvolume.Theknowledgeofthesevariationsmakesitpossibletocharacterisemanyphysicalphenomenawhichoccurwithinamaterialduringagivenheattreatment.DilatometrictestswerecarriedoutbyadilatometerADAMELDI.Thesamplesusuallyusedareofdimension10 mm × 5 mm × 5 mm.
2.3.Ultrasonicmeasurements
Anultrasonictechniquebasedonacontinuousinsitumeasurementofthevelocityoflongitudinallongbarmodewavesinthematerialhasbeenusedtomonitortheevolutionoftheelasticmodulusversustemperatureonbothmaterials[13]and[14].Fig.2isaschematicrepresentationoftheultrasonicdevice.Thedeterminationoftheultrasonicvelocityisbasedonthemeasurementoftheroundtriptimeτbetweentwosuccessiveechoesinthesample.Theultrasonicpulseistransmittedfromthetransducertothesamplethroughawave-guide.ThemeasurementofthetimeτbetweentwosuccessiveechoeswithinthesampleallowstocalculatethewavevelocityandthentoobtainthevalueoftheYoung'smodulusbyEus = ρ(2L/τ)2,whereLandρaresamplelengthanddensity,respectively.
Full-sizeimage(33K)
Fig.2. Experimentalset-upusedforYoung'smodulusmeasurementathightemperaturebylongbarmodeultrasonicpulsetechnique.
ViewWithinArticle
2.4.Tensiletest
TensiletestshavebeenperformedwithanINSTRON8862electro-mechanicaluniversaltestingmachineatroomtemperature.Fig.3presentsaschematicofthetensiletestdevice.Thestrainismeasuredbytwoextensometersequippedbysiliconcarbiderodswhichareplacedontwooppositefacesofthespecimen.Theextensometergaugelengthis25 mm.
Full-sizeimage(53K)
Fig.3. Schemeofthetensiletestdevice.
ViewWithinArticle
Thelowvaluesofthedisplacementatruptureexhibitedbythesematerials(3–5 μm)requiredagoodcontrolofthethermalstabilityoftheextensometers(±0.1 °C).Therefractorysamplesareconstitutedofacylindricalrod(18 mmindiameter)gluedtotwometallicparts.Thegeometryispreciselyadjustedthankstoafinalcylindricalmachiningstepofthetotalassembly.Thetensiletestsarecarriedoutuntilrupturewithaconstantdisplacementvelocityof0.04 mm/minwithintermediateunloadingatseverallevelofstress.ToaccuratelydeterminetheYoung'smodulusfromtensiletestresults(ET),theearlyslopeofthefirstloadingstepofthestress–straincurveshasbeenevaluated.Fig.4illustratesanexampleofstress–straincurveobtainedonAnd-LCCtreatedat110 °CandthemethodtodeterminetheYoung'smodulusattheveryfirsttime.
Full-sizeimage(45K)
Fig.4. Stress–straintensilecurvesofAnd-LCCatroomtemperatureaftertreatmentat110 °C.ZoomoftheinitialpartofthecurvesthatillustratestheearlydeterminationoftheYoung'smodulusET.
ViewWithinArticle
3.Resultsanddiscussion
3.1.Young'smodulusvaluesatroomtemperature
Foreachmethod,exceptfortheultrasonicone,atleastfivesamplesweretested.Table2presentstheresultsoftheYoung'smodulusmeasuredatroomtemperaturebythetwotechniques(ultrasonicmeasurementandtensiletest)onbothcastables.Theobtainedvaluesa