ebook img

Development of Alkaline-Activated Self-Leveling Hybrid Mortar Ash-Based Composites PDF

22 Pages·2017·8.25 MB·English
by  
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview Development of Alkaline-Activated Self-Leveling Hybrid Mortar Ash-Based Composites

materials Article Development of Alkaline-Activated Self-Leveling Hybrid Mortar Ash-Based Composites LuísUrbanoDurloTambaraJúnior ,MalikCheriafandJanaídeCavalcanteRocha* LaboratoryofWasteValorizationandSustainableMaterials(ValoRes),DepartmentofCivilEngineering,PPGEC, FederalUniversityofSantaCatarina(USFC),CampusTrindade,88040-900Florianópolis,SC,Brazil; [email protected](L.U.D.T.J.);[email protected](M.C.) * Correspondence:[email protected];Tel.:+55-4837-21-51-69 (cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1) (cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7) Received:1September2018;Accepted:22September2018;Published:26September2018 Abstract: Thisstudyinvestigatedthereactivitypropertiesofself-levelinghybridalkali-activated cements,suchasordinaryPortlandcement(OPC)anditsresidualprecursors,coalbottomash(BA), andricehuskash(RHA).DuetotherelativelylowreactivityofBA,binarymixeswereproducedwith OPCusingcontentsof2.5–30%inthetreatedBAsamples. Furthermore,ternarymixeswereprepared inproportionsof25%,50%,and75%withRHAasareplacementmaterialfortheOPC(mixwith 90%:10%BA:OPC).Forallofthemixesthespreadingbehaviorswerefixedtoobtainaself-levelling mortar, anddimensionalchanges, suchascurlingandshrinkage,wereperformed. Mortarswith 30%OPCreachedacompressivestrengthof33.5MPaandflexuralstrengthof7.53MPa. Ascanning electronmicroscope(SEM)andX-raypowderdiffraction(XRD)wereusedtoindicatetheformation ofN-A-S-Handa(N,C)-A-S-Hgel,similartothegelwithtraceofcalcium. Thebestperformancewas achievedwhenthebinarymixproduced10%OPC.AhybridmortarofOPS-BApresented10times lowersusceptibilitytocurlingthananOPCmortar. Theresultsshowedthatbothashesreducedthe shrinkageandcurlingphenomena. Keywords: hybridcements;self-levellingmortar;ashes;alkali-activation 1. Introduction Alkali-activatedmaterials(AAM)arewidelyknownasaneco-efficientcementduetoitslow CO emissions and reuse of waste residue, achieving physical and mechanical properties similar 2 to ordinary Portland cement (OPC) [1–4]. An aluminosilicate mineral and an alkaline solution are required for activation [5]. Industrial waste, as the fly ash (FA), blast-furnace slag (BFS), and bottom ash (BA) are commonly used as aluminosilicate minerals in alkaline activation [6–9]. The developmentofalkali-activatedsystemswiththeuseofashfromcoal-firedthermalplantshasproven tobepromising. However,theuseofBAisapotentialalternativetotheincreasingconsumptionof FAasasupplementaryadditiontoOPCclinker,orduringhighlevelsofproductionwhenmaterials becomeunderusedandblendedintopondsasBA. Some recent studies have demonstrated that BAs can be used as an alternative source of aluminosilicates in mixes for alkaline activation [6–15]. To ensure the sufficient reactivity of this precursor material, a larger amount of amorphous silica and relatively small particle size are required[14–16]. However,todevelopsagreatermechanicalstrengthAAM-baseddependsonash reactivity[6,11,12],andahigherrateofdissolutioninalkalinemedium[5]. Duetoalessamorphous content,glassyfilmcanshowattheparticlesurface,thushinderingtheinitialdissolution[15]. The reactionscouldbeimprovedbyashesmilling[14,16]inadditiontoachangeintheSiO /Al O molar 2 2 3 ratio[17]. IfthedissolutionandsolidificationprocessofcalciumisfasterthanSi/Al,thenacceleration hardens[17–19]. Materials2018,11,1829;doi:10.3390/ma11101829 www.mdpi.com/journal/materials Materials2018,11,1829 2of22 Rice husk ash (RHA) is generally produced by the controlled burning of rice husks [20]. Amorphous silica is the predominant component, but its content varies according to the ash treatment [21,22]. Their use in AAM is principally an alternative source of soluble silica [23–25]. Kusbiantoro et al. [26] observed that the presence of RHA in alkali-activated concretes, with replacement of up to 7% of ash, increased the compressive strength and delayed the dissolution andthepolycondensationofalumina,thusaffectingthegelationprocessandincreasestheSi/Almolar ratioandthepolymericandthepolysialatestructuresleadingtoanincreaseinstrength. Nonetheless, Nazari et al. [27] noticed that the binary mixtures of FA and RHA (containing 20–40%ofRHA)hadalargeramountofRHAreducedalumino-silicategelsphasesformation. Higher NaOH molarity accelerates dissolution, althought if it accelerates above 12 M then obstruction of polycondensation occurs. The authors related better mechanical performance with higher materialfineness. Xu and Van Deventer [28] claim that Al and Si dissolve quicker when used in NaOH when comparedtoKOH,resultinginamoreresistantstructure,duetoanentropyincreaseatthesystem. However,thereareimprovementsofalkaliactivationwhencombinedwithactivators(e.g.,hydroxide andsilicate). Nucleationturnsslowlywhenusedwithhydroxides. Whenusedwithsilicate,however, there is a higher presence of soluble silicon, which accelerates the nucleation process, acting as a “seeding”agent,risingthedissolutionofvitreousphaseoftheashes[29]. Palomoetal.[12]describedthreetypesofalkaline-activatedcements: Cementrichincalcium, wheretheprecursorhasCaOlevels>10%,inwhichthemainproductsofthereactionareC-A-S-H gels;cementwithlowcalciumcontent,inwhichthemainproductsofthereactionareN-A-S-Hgels; andthehybridcements,withlowlevelsofPortlandcementinconjunctionwiththealuminosilicate,in whichtheproductsofthereactionarecomplex,includingC-A-S-HandN-A-S-Hgels. Similarly,Garcia-Lodeiroetal.[30]establishedamodeldetailingahybridof70%FA–30%OPC, activated with NaOH and a water-glass. Based on the existence of an initial dissolution of the alumino-silicateandcalciuminthesolution,SiandAlinalumino-silicateandofCa-OandSi-Ointhe OPCbrokechains. Whenthesolutionbecomessaturated,aprecipitationofN-A-S-Handformation ofC-S-Hgelsoccur. Overtime,Sidissolves,transformingN-A-S-Hchainsfromtype1(Si/Al=1)to type2(Si/Al=2). Inaddition,thepresenceofCaandAlionsintheaqueousmediumthatreactwith C-S-HgeltoformaC-(A)-S-H(2D).AportionofCathatdoesnotreactbefore,interactswithN-A-S-H, forming(N,C)-A-S-Hgel[31]. Quetal.[32]studiedtheeffectofthecuringtemperatureonhybridcementscontainingFA,BFS, and30%OPC.TheyfoundthatthedominantreactionproductsformedwereC-(A)-S-HandC-A-S-H gels. Additionally,theyconcludedthatcuringtheproductsat85◦Cacceleratedthereaction,resulting in higher mechanical strength in the initial days of curing. Yip et al. [33] reported the coexistence ofC-S-HandN-A-S-Hgelsinhybridcements,sincethemediumisnottooalkaline. Furthermore, Angulo-Ramírezetal.[34]reportedthepresenceofmainlyC-S-HandC-A-S-Hgelsinhybridcements with20%OPCand80%BFS. Therearehighexpectationsassociatedwiththeuseofhybridcements,sincetheycanreplace theOPCcementsinmortarsandconcretes[35]. Nevertheless,fewstudieshavebeenconductedto characterizethepropertiesofhybridcementscontainingcoalBA.Thecuretimeandtemperatureaffect theformationofthestructureobtainedfromalkalineactivation,wherehighreactiontemperatures increasethemixesmechanicalstrength[36]. Thepotentialfortheapplicationofhybridcementto produceaself-levellingcompound(SLC)wasassessedinthisstudy.FormortarscomprisedofSLC,the propertiesrequiredare: Fluidity,lowviscosity,spreading,fastdrying,dimensionalstability,surface resistance, and durability [37]. In other words, a self-leveling material needs to present resistance to segregation, the capacity for fluidity, and self-levelling. For the characterization of the fluidity, monopointtests,suchasmini-slumptest,arewidelyused. Alkali-activatedproductsusuallypresent highviscosity,resultingfromtheuseofanalkalinesolution. Thismayhaveapositivecontributionin termsofavoidingthesegregationofself-levellingproducts. However,therearerecurringproblems, Materials2018,11,1829 3of22 includinghighshrinkageandcurling,especiallywhenappliedinmaterialswiththinthicknesses. The shrinkagephenomena,ontheotherhand,isrelatedtohardenedgelsofalkaline-activatedsystems[38]. Taking into consideration the reactivity properties of BA and RHA, the valorization of these wastescouldbeachievedthroughtheiruseinalkali-activation. Therefore,ourstudyproducedbinary andternaryhybridsystemsusingBA,RHA,andOPC.Thesampleswereevaluatedbaseduponthe settingtimeofthefreshmaterial,aswellastheirmechanical,microstructural,anddurableproperties. Therefore,thestudysoughttoimprovethetechnologyofself-levelingAAMscreeds. 2. MaterialsandMethods BA,classF,wascollectedfromtheJorgeLacerdathermoelectricplant,inSouthBrazil. Toimprove thereactivityandtheformationofanamorphousphase,theashwasgroundinamillfor450min,and treatedat600◦Cfor1hinamufflefurnace. TheBAwasusedasthemainsourceofaluminosilicate thatproducedalkali-activatedmortars. TheOPCusedrepresentedacementwithoutsupplementary materials,typeIIIaccordingtothedescribedonBrazilianStandardNormalizationNBR5733-91High early strength Portland cement—Specification, and NBR NM 22 Portland cement with material additionspozzolan—Chemicalanalysis—Methodarbitration[39,40].RHAoriginatedfromacontrolled combustioninafluidizedbedatthePileccothermoelectricplantinAlegrete, inSouthBrazil. The RHAwasusedtoreplacetheOPCandincreasethereactivesilicaintheternarymixes. Thechemical compoundswereanalyzedwithanenergydispersiveX-rayspectroscopy(EDX700Hs,Shimadzu, Tokyo,Japan).Thestrengthactivityindex(pozzolanicactivity)wasmeasuredaccordingtothestandard NM22[41].Particlesizedistributionwasdeterminedwithlasergranulometry,usingaMastersizer2000 (MicrotracS3500,Largo,FL,USA).Fourier-transformInfrared(FTIR)spectroscopywascarriedout usinganAgilentCary600(Agilent,SantaClara,CA,USA)atfrequenciesbetween4000and400cm−1. AccordingtoNBR5752,thepozzolanicactivityoftheRHAwasdetermined[42]as122%. The strengthactivityindexwastheratioofthecompressivestrength,withtheadditionandthecompressive strengthofcontrol(withoutRHA).Thestrengthresultsweregreaterthan75%andthustheRHAwas foundsuitableforadditiontoconcrete. Inordertoincreasethesurfacearea,theBAwasground. The grindingtimewasestablishedasatleast6h[43],untilallthegroundmaterialwas<45µm. TheBA wasthenpreparedbygrindingtheparticlesinagrindingmillfor7.5handthenputinamufflefurnace in1hat600◦Cforcalcination. TheBAcalcinationwasperformedinordertoreducethecontentof unburnedmaterial,duetothehighashcontentfoundinBraziliancoal. ThesodiumhydroxidesolutionwaspreparedbydissolvingNaOHpelletswith97%(P.A.)in distilledwater,keepingaconstantmolarconcentrationof14forallmixes. Thesolutionofsodium silicate(SiO =26.5%;Na O=10.6%;H O=62.9%;anddensityof1.39g/cm3)wasmixedwiththe 2 2 2 hydroxidesolution,keepingaproportionof1:2ofhydroxidetosilicate. Mixesofthesolutionswere prepared24hbeforetheproductionofthemortarandhadafinalpHof13. Forthemortarsproduction, naturalfinesanddimension(1.2/0.15)mmwithafinenessmodulusof1.83,andaspecificgravityof 2.60g/cm3 wasused. Themortarsweremixedinmechanicalplanaterymixer,withamixingtime of2minat150rpmand1minat300rpm. Asuperplasticizeradditive,withasolidscontentof0.22, based on polycarboxylate ether, was used to obtain workability (flow value > 250 mm), required forself-levelingmortars. Afterthemortarproduction,heatcuringwasperformedfor24hat80◦C. Samples were then kept in a room with controlled temperature (23 ± 2 ◦C) and relative humidity (RH=60±5%)untilthecompletionofthetests. Materials2018,11,1829 4of22 2.1. PreparingAlkali-ActivatedMortars We studied eight alkali-activated mortars with different mixes. In the binary mixtures, BA waspartiallyreplacedwithOPCinproportionsof0%,2.5%,5%,10%,and30%. Thesemixeswere numbered1(Reference),2,3,4,and5,respectively. Theternarymixtureswerebasedon90%BAand 10%OPC,withtheOPCbeingpartiallyreplacedwithRAHinproportionsof25%,50%,and75%;these wereidentifiedasmixes6,7,and8,respectively. Thebinderandsandratioremainedconstantat1:2 (bymass)andthewater/cementratiowaskeptconstantat0.55(wt%). Acorrectionofdensitywas necessarytoequivalentvolumeinsystemswhichadoptahighervolumeofmineralwithdifferent specificgravity. EachmixtureproportionsandthemolarratioareshowninTable1. Table1.Mixtureproportions. MortarMixProportions(per100gofPrecursor) MolarRatio Mix Samples BA OPC RHA SH1 SS2 Sand SP3 Si/Al Na2O/SiO2 H2O/Na2O 1 OPC0(REF) 100 - - 60 30 200 1.0 2.57 0.27 2 OPC2.5 97.5 2.5 - 60 30 200 1.0 2.60 0.27 3 OPC5 95.0 5.0 - 60 30 200 1.4 2.64 0.27 OPC10/RHA 4 90.0 10.0 - 60 30 200 1.2 2.71 0.28 12.26 0 5 OPC30 70.0 30.0 - 60 30 200 1.4 3.08 0.30 6 RHA25 90.0 7.5 2.5 60 30 200 1.2 2.81 0.27 7 RHA50 90.0 5.0 5.0 60 30 200 1.2 2.91 0.26 8 RHA75 90.0 2.5 7.5 60 30 200 1.2 3.02 0.25 1Sodiumhydroxidesolution.2Sodiumsilicatesolution.3Superplasticizer. 2.1.1. WorkabilityoftheAlkali-ActivatedMortars The workability was measured using a mini slump cone (19 cm × 38.1 cm × 57.2 cm). This test was performed according to a previously described procedure [44]. The measurements were performedimmediatelyafterthemixtureofthemortar,andthespreadingwasnotedbywayoffour perpendicularmeasurements. Theflowvaluewasrecordedastheaveragevalue. Themaintenanceof thespreadingpropertywasassessedoveraperiodof2h,atintervalsof30min. 2.1.2. MechanicalStrengthofAlkali-ActivatedMortars Compressiveandflexuralstrengthtests(Solotest,SãoPaulo,Brazil)wereperformedaftermolding the mortars in prisms with dimensions of 4 × 4 × 16 cm, according to Brazilian Standard NBR 13279[45]. Thesamplesweretestedat1and28daysafteractivation. Anaverageofn=3samples toflexuraltestsandn=6tocompressivetests. Theelasticanddynamicmodulusweredetermined accordingtothestandardmethods,NBR8522[46]andASTMC597[47],respectively. Thedynamic moduluswasobtainedwithPunditLabmodel6.0apparatus(Procec,Zurich,Switzerland),featuringa 20mmdiameterand200kHzfrequency. Accordingtothemanufacturer,theaccuracyoftheequipment was0.001%. Todeterminethewaterabsorptionviacapillarity,weusedcylindricalmortarssamplescasted inamold5cm×10cmdimensions,throughthevariationofawatercolumnbytime,inaMariotte tube[48]. Thesorptivity(absorptionversussquarerootoftime)wasobtained. Thewettingangles ofthemixesweredeterminedthroughtheabsorptionoftwodifferenttypesofliquids: Deionized waterandethylalcohol. TheopenporositywasdeterminedaccordingtotheBrazilianstandardNBR 9778[49]usingsamplesof5cm×10cm,andperformingtheatsampleagesof1and28days. Materials2018,11,1829 5of22 2.1.3. ResistancetoAcidAttack Toassessthemortarsdurability,twotestswereconducted: Theacidattacktestandtheimmersion anddryingtest. Theprismaticsampleswithdimensionof4cm × 4cm × 16cmweretestedafter 28days. Theacidattackwasconductedbyimmersingthesamplesin1Nsolutionsofhydrochloric acid(HCl)andaceticacid(HAc),separately,forfourcyclesofsevendays. Theexperimentdetermined themassloss,acidattackresistance,andcompressionstrength. Thewettinganddryingtestwasalso carriedout28daysafterthemortarsactivation. Theimmersionanddryinginvolved28cycleswhere eachcycleconsistedofkeepingthesamplessaturated,indistilledwater,for16h,followedbydrying inanovenat50◦Cfor8h. 2.1.4. DimensionalChanges The linear shrinkage was measured by drying according to NBR 15261 [50]. Samples with dimensions of 2.5 cm × 2.5 cm × 28.5 cm were used to measure the linear variation. The curling phenomenonwasevaluatedbymeasuringtheverticaldisplacementofamortarsampleplacedina 33cm×33cm×3cmmold,usingfourLVDT(Linearvariabledifferentialtransformers)sensorsat themortarscreeds,atdistancesof2.5and4.5cmfromtheedges,andtwosensorsinthecenterofthe mortars[51]. Mortarsweremixedinportableverticalshaftmixer. Lastly,masslosswasmonitored withaloadingcellplacedunderthescreeds. 2.2. PreparingAlkali-ActivatedPastes Aselectivechemicalattackofalkali-activatedpasteswasperformedafter28daysofheattreatment (24hat80◦C).Duetotheashdissolutioninthealkalineactivation,aconcentrationof1:20HClwas usedtodeterminethequantityofreactedcement,thusevaluatingthedegreeofreaction[6]. The1:20 HClsolutionwaspreparedusingHClwith37%ofdistillateddeionizedwater. Besidesthat,weinvestigatedsettingtimeandmicrostructure. X-raydiffractogramsofthesamples in powder form were obtained using a Phillips X-pert CuKa diffractometer (Philipps Analyticlal XRay, Almelo, The Netherlands, 40 kV, 30 mA), with scan step size of 2◦ s−1, at 2θ angle range in 3–55◦ divergenceslitof1◦ andreceivingslitof0.2-mm. Forthescanningelectronmicroscope(SEM) analysis,sampleparticleswereobservedunderaJeolJSM-6390SL(JEOLUSAInc.,PeabbodyMA, USA) (15.0 kV) and energy-dispersive X-ray spectroscopy (EDS) probe was also performed at the CentralLaboratoryofElectronMicroscopy(LCME-UFSC).AfourierTransformInfraredspectroscopy (FTIR)wasperformedforcharacterizemicrostructure,usingaJASCOFT-IR4200(JASCOCoorporation, Tokyo,Japan)atfrequenciesbetween400and4000cm−1. ThemeasurementswereperformedwithKBr pellets. Forthepastestheproportionsofcementmaterialandalkalineactivator(ratiobinder:solution) andsuperplasticizerwerekeptthesame. ThesettingtimewasdeterminedusingaVicatapparatus (Solotest,SãoPaulo,Brazil)applyingthesamecureconditionsusedformortarcasting. 2.3. MicrostructureAnalysis Allmortarsamplesweresubjectedtomicrostructuralanalysisafter28days,inordertoverify theeffectoftheproportionofOPCintheBAreplacement,aswellastheinfluenceofreactivesilica. Powderedmaterial(<150µm)wasanalyzedwithX-raypowderdiffraction(X-pertsystemPhilips AnalyticalXRay,Almelo,TheNetherlands). FragmentswereusedtoobtainSEMimages(JEOLUSA Inc.,Peabbody,MA,USA),withthesamedevicesusedtothepastesinvestigation. Below,Figure1summarizesourexperimentalprocedure. Materials 2018, 11, x FOR PEER REVIEW 5 of 21 33 cm × 33 cm × 3 cm mold, using four LVDT (Linear variable differential transformers) sensors at the mortar screeds, at distances of 2.5 and 4.5 cm from the edges, and two sensors in the center of the mortars [51]. Mortars were mixed in portable vertical shaft mixer. Lastly, mass loss was monitored with a loading cell placed under the screeds. 2.2. Preparing Alkali-Activated Pastes A selective chemical attack of alkali-activated pastes was performed after 28 days of heat treatment (24 h at 80 °C). Due to the ash dissolution in the alkaline activation, a concentration of 1:20 HCl was used to determine the quantity of reacted cement, thus evaluating the degree of reaction [6]. The 1:20 HCl solution was prepared using HCl with 37% of distillated deionized water. Besides that, we investigated setting time and microstructure. X-Ray diffractograms of the samples in powder form were obtained using a Phillips X-pert CuKa diffractometer (Philipps Analyticlal XRay, Almelo, The Netherland, 40 kV, 30 mA), with scan step size of 2° sec−1, at 2θ angle range in 3°–55° divergence slit of 1° and receiving slit of 0.2-mm. For the scanning electron microscope (SEM) analysis, sample particles were observed under a Jeol JSM-6390SL (JEOL USA Inc, Peabbody MA, USA) (15.0 kV) and energy-dispersive X-ray spectroscopy (EDS) probe was also performed at the Central Laboratory of Electron Microscopy (LCME-UFSC). A fourier Transform Infrared spectroscopy (FTIR) was performed for characterize microstructure, using a JASCO FT-IR 4200 (JASCO Coorporation, Tokyo, Japan) at frequencies between 400 and 4000 cm−1. The measurements were performed with KBr pellets. For the pastes the proportions of cement material and alkaline activator (ratio binder:solution) and superplasticizer were kept the same. The setting time was determined using a Vicat apparatus (Solotest, São Paulo, Brazil) applying the same cure conditions used for mortar casting. 2.3. Microstructure Analysis All mortar samples were subjected to microstructural analysis after 28 days, in order to verify the effect of the proportion of OPC in the BA replacement, as well as the influence of reactive silica. Powdered material (<150 µm) was analyzed with X-ray powder diffraction (X-pert system Philips Analytical XRay, Almelo, The Netherland). Fragments were used to obtain SEM images (JEOL USA Inc, Peabbody MA, USA), with the same devices used to the pastes investigation. Materials2018,11,1829 6of22 Below, Figure 1 summarizes our experimental procedure. Figure1. Experimentalprocedureschema: (a)mini-slumpflowandself-levellingmortar(spread); (b)linearshrinkagebartestswithbinarysamples(up)andternarysamples(down);(c)heatcureof screedsofandcurlingtest. 3. Results 3.1. PrecursorsCharacteristics Below,Table2summarizestheresultsobtainedforthechemicalandphysicalcharacteristicsof theprecursorminerals. Table2.Chemicalandphysicalcharacteristicsoftheprecursormaterials. TotalOxide (wt%) SiO2 Al2O3 Fe2O3 CaO K2O SO3 TiO2 LOI1 BA 40.82 37.46 5.71 1.73 5.20 0.29 1.90 6.67 RHA 92.00 ND2 0.17 0.60 1.72 ND2 0.11 3.50 ParticleSize SpecificGravity BlaineFineness D10(µm) D50(µm) D90(µm) (g/cm3) (m2/kg) BA 1.545 8.597 22.733 2.43 715.5 OPC 3.561 15.822 60.120 3.05 500.5 RHA 2.166 10.497 41.208 2.12 1174.0 1Lossonignition;2ND:Notdetected. TheBAandRHAcontained>40%reactiveSiO andthecontentofvitreousphasewas>50%[6,7]. 2 TheaverageparticlesizeswerebelowthediameterofOPC—8.59µmand10.49µmforBAandRHA, respectively. ThespecificareasoftheRHAandBAasheswerecomparedwithOPC(500.5m2/kg), withvaluesof1174and715.5(m2/kg),respectively. Figure2showstheX-raydiffractogramofBA aftertreatmentandtheRHA.RHAashcontainsmainlyamorphousmaterials,withcrystallinephases ofcristobalite(PDF#39-1425)andquartz(PDF#46-1045). TheRHAparticleshadanirregularshape andwerelamellarandfibrous,withpresenceofelongatedparticles,duetotheunburnedmaterial, Materials 2018, 11, x FOR PEER REVIEW 6 of 21 Figure 1. Experimental procedure schema: (a) mini-slump flow and self-levelling mortar (spread); (b) linear shrinkage bar tests with binary samples (up) and ternary samples (down); (c) heat cure of screeds of and curling test. 3. Results 3.1. Precursors Characteristics Below, Table 2 summarizes the results obtained for the chemical and physical characteristics of the precursor minerals. Table 2. Chemical and physical characteristics of the precursor materials. Total Oxide wt (%) SiO2 Al2O3 Fe2O3 CaO K2O SO3 TiO2 LOI 1 BA 40.82 37.46 5.71 1.73 5.20 0.29 1.90 6.67 RHA 92.00 ND 2 0.17 0.60 1.72 ND 2 0.11 3.50 Particle Size Specific Gravity Blaine Fineness D10 (µm) D50 (µm) D90 (µm) (g/cm3) (m2/kg) BA 1.545 8.597 22.733 2.43 715.5 OPC 3.561 15.822 60.120 3.05 500.5 RHA 2.166 10.497 41.208 2.12 1174.0 1 Loss on ignition; 2 ND: Not detected. The BA and RHA contained > 40% reactive SiO2 and the content of vitreous phase was > 50% [6,7]. The average particle sizes were below the diameter of OPC—8.59 µm and 10.49 µm for BA and RHA, respectively. The specific areas of the RHA and BA ashes were compared with OPC (500.5 m2/Kg), with values of 1174 and 715.5 (m2/kg), respectively. Figure 2 shows the X-ray diffractogram of BA after treatment and the RHA. RHA ash contains mainly amorphous materials, with crystalline phases of cristobalite (PDF#39-1425) and quartz (PDF#46-1045). The RHA particles had an irregular shape and were lamellar and fibrous, with presence of elongated particles, due to the unburned material, with lengths of around 150 µm (Figure 3). BA mainly contains quartz crystals, mullite, and Mheatmeriaatlsit2e0,1 w8,1it1h,1 m82o9stly spherical particles and unburnt carbon (loss on ignition (LOI) = 6.67%). 7of22 The grinding of the BA particles promoted a reduction in the internal porosity, the formation of irregular particles, and the presence of microspheres (Figure 3a). The RHA particles exhibit irregular, withlengthsofaround150µm(Figure3). BAmainlycontainsquartzcrystals,mullite,andhematite, lamellar, and fibrous materials related to the presence of unburned material (Figure 3b). withmostlysphericalparticlesandunburntcarbon(lossonignition(LOI)=6.67%). C Q - Quartz M Q M - Mullite H - Hematite C - Cristobalite Q Q Q C Rice Husk Ash Q M H M H MQ M M M Q Bottom Ash 5 10 15 20 25 30 35 40 45 50 55 60 65 70 2θ (degrees) Materials 2018, 11, x FOR PEER RFiEgVuIrEeW2 . XRDpatternsofbottomashandricehuskash. 7 of 21 Figure 2. XRD patterns of bottom ash and rice husk ash. (a) (b) (c) FFiigguurree 33.. SSiinngguullaarr eelleeccttrroonn mmiiccrroossccooppee ((SSEEMM)) iimmaaggeess:: ((aa)) BBoottttoomm aasshh aafftteerr ggrriinnddiinngg aanndd ccaallcciinnaattiioonn aatt 660000 °◦CC ((××1510500)0;) (;b(b) )RRiciec ehhuusksk asahsh (R(RHHAA) )(×(×50500)0; )(;c)( cR)HRHA A(×(1×50105)0. 0). FTThIeRg srpinedctirnogscoofpthye wBaAs fpaacriltiitcaletesdp urosmingo taend Aagreildeunct tCioanryin 60th0e ininsttreurnmaelnpto arto sfriteyq,uthenecfioersm baettiwoneeonf 4ir0r0e0g uanladr p40a0rt iccmle−s1,. aFnigdutrhee 4p srhesoewnsc ethoef mFTicIRro srpeshuelrtess f(oFri gthuer em3ian)e.rTahlse (RtrHeaAtepda artnicdl eusnetxrheaibteitdi rBrAeg uanladr, RlaHmAel)l.a Tr,haen wdifidber obuansdm laotceartieadls aret l3a6te0d0 ttoo t3h2e00p rcemse−n1 cfoero bfoutnhb suarmnepdlems watearsi adlu(eF itgou trhee3 Ob)-.H bond. The band FcTloIRses ptoe c1t6ro0s0c ocmpy−1w caosrrfeascpiloitnadteedd utosi nthgea nanAgguillaern tdCefaorrym6a0t0ioinn sotrfu tmhee nHt-aOt-fHre qbuoenndc ieins btheetw reaewn m40a0t0eraianlds. 4T0h0ec mba−n1d. Fpirgeuserent4 asth 1o2w00s tahnedF 8T0I0R crmes−u1 wltsafso druthe etom tihnee rsatrlset(cthreinatge dviabnradtiuonnt roefa atemdoBrpAhaonuds SRiH-OA-S).i,T whheiwchi dsehobwanedd lao hcaigtehdera cto3n6t0e0ntto to3 2R0H0Acm d−u1e ftoor itbso ttrhansasmmipttleasncwea. sThdeu esmtoaltlh beaOnd-H clboosen dto. T50h0e cbman−1d foclro BseAt ow1a6s0 d0ucem t−o1 ccroyrsrteasllpinoen dSei-dOt-oSit.h Tehaen bgaunladr adte f5o6r0m–5a5ti0o ncmo−f1 tchaenH b-eO a-Hssobcoiantdedin wtihteh rtahwe omctaathereidarlso.nT phreebseanntd inp rtehsee nmtualtli1t2e.0 0Thaen dBA80 h0ecamt −tr1eawtmasednut eatt o60t0h e°Cst rinetccrheiansgedv itbhrea tiniotnenosfitaym oof rtphheo Ou-s HSi -bOa-nSdi, awndhi tchhes pheoawk eadt 1a5h51ig chmer−1c aopnpteenartetdo bReHcaAusdeu teheto stirtosntrgaensts mviibtrtaanticoen. wThase tshme aCll-Ob abnodndcl.o seto 500cm−1forBAwasduetocrystallineSi-O-Si. Thebandat560–550cm−1canbeassociatedwiththe octahedronpresentinthemullite. TheBAheattreatmentat600◦CincreasedtheintensityoftheO-H bandandthepeakat1551cm−1appearedbecausethestrongestvibrationwastheC-Obond. 3450 560 1635 800 3450 1635 1090 470 560 800 1635 3450 1090 470 800 1090 470 4000 3500 3000 2500 2000 1500 1000 500 Wavenumber (cm-1) BA untreated BA treated RHA Figure 4. Fourier-transform Infrared (FTIR) spectra for precursor minerals: Bottom ash (BA) untreated, BA treated, after grinding and calcination at 600 ºC (1 h); Rice husk ash (RHA), from commercial supplier. 3.2. Workability of the Alkali-Activated Mortars We obtained hybrid alkali-activation mortars with self-leveling properties. All mixtures were tested with a superplasticizer based on polycarboxylate, except for OPC 10, which was also prepared with an additive based on naphthalene. There was a trend toward greater flowability for mixes with lower OPC content. The self-leveling properties were acquired with 1.0–1.4% of superplasticizer. Materials 2018, 11, x FOR PEER REVIEW 7 of 21 (a) (b) (c) Figure 3. Singular electron microscope (SEM) images: (a) Bottom ash after grinding and calcination at 600 °C (×1500); (b) Rice husk ash (RHA) (×500); (c) RHA (×1500). FTIR spectroscopy was facilitated using an Agilent Cary 600 instrument at frequencies between 4000 and 400 cm−1. Figure 4 shows the FTIR results for the minerals (treated and untreated BA and RHA). The wide band located at 3600 to 3200 cm−1 for both samples was due to the O-H bond. The band close to 1600 cm−1 corresponded to the angular deformation of the H-O-H bond in the raw materials. The band present at 1200 and 800 cm−1 was due to the stretching vibration of amorphous Si-O-Si, which showed a higher content to RHA due to its transmittance. The small band close to 500 cm−1 for BA was due to crystalline Si-O-Si. The band at 560–550 cm−1 can be associated with the octahedron present in the mullite. The BA heat treatment at 600 °C increased the intensity of the O- Materials2018,11,1829 8of22 H band and the peak at 1551 cm−1 appeared because the strongest vibration was the C-O bond. 3450 560 1635 800 3450 1635 1090 470 560 800 1635 3450 1090 470 800 1090 470 4000 3500 3000 2500 2000 1500 1000 500 Wavenumber (cm-1) BA untreated BA treated RHA Figure 4. Fourier-transform Infrared (FTIR) spectra for precursor minerals: Bottom ash (BA) untreFaigtuedre, B4.A Ftorueraiteerd-tr,aanfstfeorrmgr iInndfrianrgeda n(FdTIcRa)l csinpaectitoran faotr 6p0r0ec◦uCrso(1r hm);inReriacles: hBuostktoams has(Rh H(BAA)), from untreated, BA treated, after grinding and calcination at 600 ºC (1 h); Rice husk ash (RHA), from commercialsupplier. commercial supplier. 3.2. WorkabilityoftheAlkali-ActivatedMortars 3.2. Workability of the Alkali-Activated Mortars Weobtainedhybridalkali-activationmortarswithself-levelingproperties. Allmixtureswere We obtained hybrid alkali-activation mortars with self-leveling properties. All mixtures were testedwithasuperplasticizerbasedonpolycarboxylate,exceptforOPC10,whichwasalsoprepared tested with a superplasticizer based on polycarboxylate, except for OPC 10, which was also prepared withanadditivebasedonnaphthalene. Therewasatrendtowardgreaterflowabilityformixeswith with an additive based on naphthalene. There was a trend toward greater flowability for mixes with lower OPC content. The self-leveling properties were acquired with 1.0–1.4% of superplasticizer. lower OPC content. The self-leveling properties were acquired with 1.0–1.4% of superplasticizer. OPC5requiredthesamesuperplasticizercontentasOPC30sample,duetotheirlowerworkability. Self-l evelingpropertieswerealsoobtainedforallternarymixesbyadding1.2%ofadditive(bymass) tothebinder. Noexudationorsegregationwasobservedinthefinalmini-slumptest. Mortarsamplesthatreachedaninitialflowvaluewere250mm. After30min,thebinarymix spreadswere: 280,225,and225forOPC2.5,OPC5andOPC10,respectively. Forternarymixes,the spreadsafter30minwere225±25mm. Reductionsofmorethan60%,comparedtoinitialflowvalue, occurredafter60min. The final setting after heat alkali-activation were 18 to 20 min, a rapid hardening to self-levelingproduct. 3.3. MechanicalStrengthandDurabilityoftheAlkali-ActivatedMortar Table3summarizesthestrengthvaluesandphysicalparameters(sorptivity,porosity,andwet angle)foundatday1andday28. ThedevelopmentofcompressivestrengthforthedifferentamountsofaddedOPCindicatethat themeanvaluesdidnotdiffersignificantly. Lowstrengthwasnotedforthesamplescontainingonly BA,duetothelowdegreeofreactivityobservedintheselectiveacidattackonthepastes. ForOPC0 andOPC2.5,therewasnosignificantdifferenceinthestrengthforlaterages. Whencomparingthe averagevaluesofOPC2.5andOPC5,nosignificantdifferenceafteronedayofcuringwasrevealed. Therefore,itwasverifiedthathigherlevelsofcementinfluencedthedevelopmentofthecompressive strength. Theflexuralstrengthresultsforthebinarysamplesdemonstratedsimilarbehavior,with valuesfivetimeslowerthanthosewiththecompressivestrength. Therewasadelaytothereactionof bottomashes(OPC0). WhenRHAwasusedasreplacementmaterialinthemixOPC10inproportionsof0%,25%,50% and75%,therewasareductioninthestrength[12]. TheelongatedRHAparticlesindicatealower Materials2018,11,1829 9of22 degreeofashdissolution. ThereductionintheamountofCaOpresentalsodecreasedthereactivity. HighercompressivestrengthvaluescanbeobservedwithloweramountsofRHA,indicatingthat RHAhadlowreactivity. Theflexuralstrengthwaslowerthanthatofthebinariesmixes,duetothe degreeofreaction. Table3. Strengthsvaluesandphysicalparametersat1and28daysformixesofordinaryPortland cement(OPC)andRHA. Mixes Compressive FlexuralStrengh Sor√ptiviy WetAngle(◦) OpenPorosity E(GPa) Strengh(MPa) (MPa) (cm min) Day1 28Days Day1 28Days Day1 28Days Day1 28Days Day1 28Days 28Days 1.92 2.58 1.35 1.60 OPC0 0.0158 0.0159 89.76 89.76 25.30 22.74 — (1.12) (0.13) (0.20) (0.22) 21.27 21.59 4.61 7.38 OPC2.5 0.0205 0.021 80.84 80.84 20.05 15.22 14.68 (1.64) (2.40) (0.12) (0.29) 22.34 23.22 3.89 7.98 OPC5 0.0361 0.031 76.04 76.04 22.91 14.08 14.78 (1.31) (2.40) (0.18) (0.20) 26.01 28.13 4.86 7.75 OPC10 0.0409 0.0084 88.65 88.65 15.80 15.62 16.05 (1.69) (2.21) (0.29) (0.49) 28.00 33.46 4.67 7.53 OPC30 0.0445 0.0215 85.28 85.28 15.89 12.17 18.73 (1.10) (2.08) (0.49) (0.23) 11.46 20.14 3.47 4.05 RHA25 0.0363 0.0306 89.95 89.92 12.21 10.9 13.91 (0.68) (2.11) (0.21) (0.29) 6.13 14.30 2.56 3.60 RHA50 0.0471 0.0197 89.91 89.97 13.20 12.22 12.62 (0.25) (1.52) (0.12) (0.11) 5.76 12.74 2.48 3.26 RHA75 0.0369 0.0149 89.82 89.98 14.62 12.93 10.51 (0.24) (0.62) (0.13) (0.32) Itwasfoundthatatthebinarymixesdynamicmodulusat28days,highervalueswereobtained byincreasingOPCcontentinthemixes; forternarymixes,thevaluesdecreasedwithhigherRHA content. Dynamicmodulusofelasticitywas10.5to18.5GPa. Forallmixes,theinitialsorptivityafterday1washigherthanthatobservedafter28days. Withan increaseintheOPCcontentinthebinarymixesafteralkaline-activation,anincreaseintheabsorption canbeobservedafterday1ofaging,whichwasassociatedwithincreasedporosity. After28days, weobservedthatforthelowerlevelsofBAreplacementinOPC(OPC2.5andOPC5samples)there washighersorptivitywhencomparedwithOPC10andOPC30. Thiswasduetotheformationof (N,C)-A-S-H gels in the activations when a higher OPC content was present, thus resulting in the fillingofthepores. In addition, we observed that all ternary mixes exhibited higher sorptivity when compared withthereferencesample. Similarbehaviorwasevidentafter1and28dayswithareductioninthe sorptivity. RHA75showedlowsorptivity,primarilyduetotheinertnatureoftheRHA,whichshowed lowerreactivityandfilledthecapillarypores. Everysamplehadahigheropenporosityafterday1, comparedwiththeresultafter28days. AnincreaseintheOPCcontentsresultedinloweropenporosityandabsorption,duetoanincrease inthe(N,C)-A-S-Hgelformation. Inaddition,theternarymixesshowedloweropenporosityand absorption,whichweattributedtothefillereffectduetopresenceofRHA. Durability Figure5showsthatthemasslosswashigherforthebinarymixescomparedwiththeternary. This isduetotheattackontheOPC,the(N,C)-A-S-Hgel,andthealkalineactivationBAgel(N-A-S-H).Itcan benotedthatthemixescontaining2.5%and5%ofOPCshowedasignificantdegreeofdegradation, indicatingthatOPCcontentaffectedtheresistanceoftheacidattack. InthemortarscontainingRHA, thedegreeofdegradationwaslower. Comparatively,theresultsshowthatthePortlandcementmortar (PCM),withthesamew/cratio,degradedmoreaggressivelythantheAAMs. FortheAAM,nosigns ofsurfacedeteriorationwerefoundinthevisualinspectionoftheprisms. TheOPCspecimenswere severelydeterioratedafterthe28cycles. Materials 2018, 11, x FOR PEER REVIEW 9 of 21 For all mixes, the initial sorptivity after day 1 was higher than that observed after 28 days. With an increase in the OPC content in the binary mixes after alkaline-activation, an increase in the absorption can be observed after day 1 of aging, which was associated with increased porosity. After 28 days, we observed that for the lower levels of BA replacement in OPC (OPC 2.5 and OPC 5 samples) there was higher sorptivity when compared with OPC 10 and OPC 30. This was due to the formation of (N,C)-A-S-H gels in the activations when a higher OPC content was present, thus resulting in the filling of the pores. In addition, we observed that all ternary mixes exhibited higher sorptivity when compared with the reference sample. Similar behavior was evident after 1 and 28 days with a reduction in the sorptivity. RHA 75 showed low sorptivity, primarily due to the inert nature of the RHA, which showed lower reactivity and filled the capillary pores. Every sample had a higher open porosity after day 1, compared with the result after 28 days. An increase in the OPC contents resulted in lower open porosity and absorption, due to an increase in the (N,C)-A-S-H gel formation. In addition, the ternary mixes showed lower open porosity and absorption, which we attributed to the filler effect due to presence of RHA. Durability Figure 5 shows that the mass loss was higher for the binary mixes compared with the ternary. This is due to the attack on the OPC, the (N,C)-A-S-H gel, and the alkaline activation BA gel (N-A-S- H). It can be noted that the mixes containing 2.5% and 5% of OPC showed a significant degree of degradation, indicating that OPC content affected the resistance of the acid attack. In the mortars containing RHA, the degree of degradation was lower. Comparatively, the results show that the Portland cement mortar (PCM), with the same w/c ratio, degraded more aggressively than the AAMs. MFaoterr itahlse2 A01A8,M11,, 1n8o29 signs of surface deterioration were found in the visual inspection of the prism10s.o Tf2h2e OPC specimens were severely deteriorated after the 28 cycles. 14 14 a) b) 12 12 %)10 %)10 ( ( ss 8 ss 8 o o s L 6 s L 6 s s a 4 a 4 M M 2 2 0 0 0 5 10 15 20 25 30 0 5 10 15 20 25 30 Time (days) Time (days) PCM OPC2.5 OPC5 OPC10 OPC30 PCM OPC2.5 OPC5 OPC10 OPC30 c)14 d)14 12 12 %)10 %)10 ss ( 8 ss ( 8 o o s L 6 s L 6 s s a 4 a 4 M M 2 2 0 0 0 5 10 15 20 25 30 0 5 10 15 20 25 30 Time (days) Time (days) RHA0 RHA25 RHA50 RHA75 RHA0 RHA25 RHA50 RHA75 FFiigguurree5 5.. MMaassss lloossss ooff bbiinnaarryy aanndd mmoorrttaarrssd duurriinngga acciidda attttaacckk:: ((aa)) BBiinnaarryy aafftteerr HHAAcc eexxppoossuurree;;( (bb))B Biinnaarryy aafftteerrH HCClle exxppoossuurree;;( c(c))T Terenrnaaryrya aftfetrerH HAAcce xepxpoosusurer;e(; d(d))T eTrenranrayrya fatfetreHr HCClel xepxopsousruer.e. FFiigguurree6 6s shhoowwsst thheer reessuultltssa afftteerrt thheeH HAAcca annddH HCClle exxppoossuurree.. TThheefl fleexxuurraall ssttrreennggtthh wwaass ssttrroonnggllyy aafffefeccteteddb ybyth tehaec iadciadtt aactkta,cdku, edtuoet hteo fothrme afotiromnaotfiocnra cokf scirnatchkes mino rtthaer, masosrhtoawr, nasin sFhiogwurne 6ina .FHigouwreev 6ear,. fHoroawlleavlekra,l finoer -aalcl taivlkaatelidnes-aamctpivleast,endo ssaimgnpilfiecsa, tnivoe sicgonmifpicraetsisvive ecostmrepnrgetshsilvoes sstwreansgotbhs leorsvse wda(Fs iogbusreer6vbe)d. Materials 2018, 11, x FOR PEER REVIEW 10 of 21 (Figure 6b). 100 92 90 90 96 89 81 82 83 80 67 73 67 67 76 62 63 %) 60 52 k ( 40 20 0 PCM OPC 2.5 OPC 5 OPC 10/ OPC 30 RHA 25 RHA 50 RHA 75 RHA 0 Sample HAc attack HCl attack (a) 100 99 95 99 99 98 98 99 98 86 90 77 80 68 70 60 %) 50 k ( 40 30 20 10 0 PCM OPC 2.5 OPC 5 OPC 10/ OPC 30 RHA 25 RHA 50 RHA 75 RHA 0 Sample HAc Attack HCl Attack (b) FiguFriegu6r.e R6e. sRisetsaisntacnecteo toa caidcida tatattcakck:: (a(a)) FFlleexxuurraall ssttrreennggtthh; ;(b(b) )CComomprpersesisvseiv setrsetnrgetnhg. tKh .vKaluvea l(u%e): ( %): K=K(( R= e(s(Risetsainstcaenacet 2at8 2d8a dyasy−s-RReessisitsatanncec eafateftre cryccylec)l/eR)e/sRisetasnisctea natc e28a dta2y8sd) ×a y1s0)0.× 100. The results of the wetting and drying tests carried out on the mortars are shown in Table 4. It can be observed that 28 cycles of wetting and drying did not cause degradation of the PCM samples. The alkali-activation had an influence on the flexural strength, mainly in the case of binary mixes. Only the alkali-activated sample containing the highest percentage of Portland cement showed lower results after wetting and drying cycles. Table 4. Rate of degradation of binary and ternary mortars of alkali-activated mortars. Sample Kf Remark kc Remark PCM −0.016 No effect −0.149 No effect OPC 2.5 0.333 Influence −0.107 No effect OPC 5 0.431 Influence −0.436 No effect OPC 10 0.249 Influence −0.056 No effect OPC 30 0.375 Influence 0.095 Influence RHA 25 0.170 Influence −0.125 No effect RHA 50 0.033 Influence −0.011 No effect RHA 75 0.002 Influence −0.033 No effect K value: K = ((Resistance at 28 days-Resistance after cycle)/Resistance at 28 days). 3.4. Dimensional Changes 3.4.1. Linear Shrinkage The results for linear shrinkage variation are shown in in Figure 7. We found that binary samples had a lower linear shrinkage than ternary samples. In binary mixes, high contents of OPC raised the linear shrinkage, due to the amount of Portland paste, which increased gel formations. RHA was

Description:
Luís Urbano Durlo Tambara Júnior , Malik Cheriaf and Janaíde Cavalcante Rocha *. Laboratory of Waste .. To determine the water absorption via capillarity, we used cylindrical mortars samples casted in a mold 5 cm × 10 cm
See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.