TheEvaluationofHygroscopicInertiaandItsImportancetotheHygrothermalPerf - PDF document

TheEvaluationofHygroscopicInertiaandItsImportancetotheHygrothermalPerformanceofBuildingsNunoM.M.RamosandVascoPeixotodeFreitasAbstractHeatingandventilatingarefundamentalactionsforthecontrolofhumidityintheindoorenvironment,butthehygroscopicinertiaprovidedbythematerialsthatcontacttheinsideaircanbeacomplementforthatcontrol.ThehygroscopicbehaviorofthewallsandceilingÞnishingmaterials,aswellasfurnitureandtextilesinsidethedwellings,deÞnestheirhygroscopicinertia.Reducingthepersistenceofhighrelativehumidityvaluesinsidebuildingsisessentialforthecontrolofmouldgrowthonmaterialsurfaces,thatcanotherwisecausedegradationandbringaboutsocialandeconomicalproblemsfortheusers.AsthehygroscopicinertiaconceptcanbeverydifÞculttoapproachforbuildingdesigners,adeÞnitionofdailyhygroscopicinertiaclassesispresented,basedonnumericalandlaboratoryworkonthissubject.Anoutlineofasimplemethod,usingthoseclasses,thatallowsfortheevaluationofthereductionofmouldgrowthpotentialassociatedtoaconÞgurationofinsideÞnishesisproposed.TheextensiveexperimentalcampaignaimingthecharacterizationofthemoisturebufferingcapacityofinteriorÞnishingsystemandtheassessmentofaroomÕshygroscopicinertiaisdescribed.TheMBVÑMoistureBufferValueisevaluatedfordifferentrevetments.TheassessmentofhygroscopicinertiaatroomlevelisimplementedusingaßuxchamberdesignedspeciÞcallyforthisexperiment.Adailyhygroscopicinertiaindex,,isdeÞnedusingMBVasabasisfortheassessmentofmaterialscontributiontothebufferingcapacityofaroom.Thecorrelationbetweenthatindexandpeakdampeningisprovedusingthepresentedexperimentalresults.Systematicsimulationofthesetofdynamicexperimentsoftransientmoisturetransferin N.M.M.Ramos(V.P.deFreitasLFCBuildingPhysicsLaboratory,CivilEngineeringDepartment,FacultyofEngineering,UniversityofPorto,4200-465Porto,Portugale-mail:nuno.ramos@fe.up.ptV.P.deFreitase-mail:vpfreita@fe.up.ptJ.M.P.Q.Delgado(ed.),HeatandMassTransferinPorousMediaAdvancedStructuredMaterials13,DOI:10.1007/978-3-642-21966-5_2,Springer-VerlagBerlinHeidelberg2012 thehygroscopicregionispresented;allowingtoverifyingandcorrectingthemodelingassumptionsandthebasicdatausedinsimulations,andconcludeonthemosteffectivestrategiestoconductthistypeofsimulations.1IntroductionThevariationofinsideRelativeHumidity(RH)isinßuencedbythemoistureexchangebetweenairandbuildingelements.Therelevanceofthatexchangeislinkedtotheactivemoisturebuffercapacitypresentinaroom,whichcanbeidentiÞedwithitshygroscopicinertia.Relativehumiditytheairinsidebuildingscanhaveaninßuenceonthermalcomfort,ontheperceptionofindoorairquality,usersÕhealth,materialsdurabilityandenergyconsumption.Thisdependencyhasbeenestablishedbysciencebutthecommonuserwillnotalwaysrecognizeit.MouldgrowthonbuildingelementÕssurfaces,ontheotherhand,iseasilyassociatedwiththepersistenceofhighRHlevelsevenbyusersandcanbetiednotonlytodurabilitybutalsotoPAQandTheworkofseveralresearchershasalreadydemonstratedthebeneÞtsfrominsiderelativehumidityvariationcontrolprovidedbyhygroscopicmaterialsls7,11],andtheInternationalEnergyAgency(IEA)researchproject,IEA-Annex41contributedtoadeeperunderstandingofthatprocess.Laboratoryexperimentsareawayofdemonstratingandquantifyingthatcapacityandcanbeimplementedinthreedifferentlevels.Atmateriallevel,internationalstandardsalreadysupportthedeterminationofthebasicpropertiesthatconditionmoisturestorageperformance,suchassorptionisotherms(ISO12571[])andvapourpermeability(ISO12572[]).ApropertydeÞnedasMBV,MoistureBufferValue,wasproposedby[],allowingforadirectexperimentalmeasureofthemoistureaccumulationcapacityofamaterialundertransientconditions.Atelementlevel,whereseveralmaterialscanbecombinedbytheirapplicationindifferentthicknesses,MBVcanalsobeappliedasanexperimentalmeasureofeachspeciÞcelementconÞgurationmoistureaccumulationcapacity.Atroomlevel,theauthorsbelievethatalaboratorymeasurementoftheactivemoisturebuffercapacityshouldbedirectlylinkedtotheRHpeakdampeningpromotedbytheroomÕsinteriorconÞguration,comparedtothepeaksinthesameroomwithoutanyactivehygroscopicsurfaces.TherelationbetweenthethreelevelsofexperimentalmeasurementsisexploredusingnumericalsimulationofelementandroombehaviourusingthematerialpropertiesmeasuredintheÞrstlevel.ThistextalsodeÞnesdailyhygroscopicinertiaclassesandproposestheirimplementationasaneasywayofincludingthebuildingmaterialsmoisture26N.M.M.RamosandV.P.deFreitas storagecapacityinßuenceonRHvariationandmouldgrowthriskanalysis.AproperselectionofinteriorÞnishescanobviouslybeneÞtfromthatinclusion.2MaterialProperties2.1MaterialsIntheseexperiments,anoptionwasmadeinusingspecimensofcommoncom-mercialmaterialsusedÞnishingsystemsforwallsandceilings.Theexperimentswerethereforeperformedinspecimensof,gypsumplaster(1200kg/m)asbasematerial,eithernakedorcombinedwithacoating.Thistypeofgypsumisblendedinfactoryand,afteradditionofwater,canimmediatelybeapplied.Thecoatingisalsocommerciallyavailableand,therefore,formulationisunknown.Itwasknown,however,thatitwascomposedof25acrylicprimerand50mvinylÞnishinglayer.Theprocedureforthepreparationofthegypsumplasterspecimenstriedtoreplicatetheconditionsthatareusedinpractice.Thedryplasterpowder(2kg)wasmechanicallymixedwithwater(1dm),during5min,toproduceahomo-geneousmass.Aftercastinginwoodframesfor24h,thespecimensweredriedintheair.2.2SorptionIsothermMostbuildingmaterialsarehygroscopic,whichmeansthattheyadsorbvapourfromtheenvironmentuntilequilibriumconditionsareachieved.Thisbehaviourcanbedescribedbysorptioncurvesoverahumidityrangeof0Ð95%RH.ThesorptionisothermsrepresenttheequilibriummoisturecontentsofaporousmaterialasafunctionofrelativehumidityataspeciÞctemperature.TheexperimentswereperformedinaccordancewithISO12571standard.Atatemperatureof23C,fourofÞveavailablerelativehumidityambienceswereusedinthecharacterizationofeachofthethreebasematerials.EachambiencewasobtainedindesiccatorsusingaspeciÞcsaturatedsaltsolution(NH4Cl-77%,NH4Cl-84%andKNO3-91%)orinaclimaticchamber(33and50%).Thegypsumplasterspecimenshaddimensionsof606mm.Thespecimenswereinitiallydriedatambienttemperature,during30days,indesic-catorscontainingCaCl,guaranteeingarelativehumiditybelow0.5%.Forsorptionmeasurements,thetestspecimenisplacedconsecutivelyinaseriesoftestenvironments,withrelativehumidityincreasinginstages,untilequilibriumisreachedineachenvironment.Equilibriumbetweenmoisturecontentandrelativehumidityhasbeenreachedwhensuccessiveweightings,TheEvaluationofHygroscopicInertia27 attimeintervalsofatleastoneweek,showadifferenceinmasslesserthan0.1%.Thestartingpointforthedesorptionmeasurementswasabove91%RH.Whilemaintainingaconstanttemperature,thespecimenisplacedconsecutivelyinaseriesoftestenvironments,withrelativehumiditydecreasinginstages,untilequilibriumisreachedineachenvironment.Withtheobjectivetogaintime,differentspecimenswereusedforthedifferentambiencesinthesorptionphase.Finally,thespecimensweredriedattheappropriatetemperaturetoconstantmass.Fromthemeasuredmasschanges,theequilibriummoisturecontent(),ateachtestcondition,couldbecalculatedandthesorption/desorptionisothermdrawn(seeFig.2.3VapourPermeabilityVapourpermeabilitywasdeterminedforsamplesofthebasematerial,bothnakedandcombinedwiththecoating.ThetestswereconductedaccordingtoISO12572standard.Foreachcombinationofbasematerialandcoating,threepermeabilityvaluesweredeÞned,correspondingtothreedifferentrangesofRHdifferencesacrossthesample.Priortotesting,allthespecimensarepreconditionedinaclimaticchamberatCand50%RH,foraperiodlongenoughtoobtainthreesuccessivedailydeterminationsoftheirweightlesserthan0.5%.Afterstabilisation,thespecimensareplacedincupswithasaturatedsaltsolutionbelowthebottomsurfaceofthespecimen.Thesidesofthespecimenswerecoveredwithvapourimpermeabletape.Duetothedimensionsofthecups,thedimensionsofthesamplescorrespondedto21011mm.Thechangeofweightofthecupwasmeasuredperiodically,withaprecisionof0.1mg,onanelectronicbalanceuntilthesteadystatewasobtained.ThesketchpresentedinFig.showstherelativehumidityproÞlesadmittedforthepaintedsamplesduringthepermeabilityexperiments,inaccordancewiththetheorypresentedbelow. Fig.1Sorptionisothermsobtainedforgypsumplaster28N.M.M.RamosandV.P.deFreitas Thevapourpermeability,,isamaterialpropertydeÞnedasthetransportcoefÞcientforvapourdiffusioninaporousmaterialsubjectedtoavapourpressuregradient.Thepermeabilitycanbecalculatedusing, isthemassßuxdensityandisthesamplethicknessexposedtoavapourpressuregradientOtherpropertiesderivedfromvapourpermeabilitycanbeused,asthetransportcoefÞcientunderavapourconcentrationgradient:vapourresistancefactor,orvapourdiffusionthickness,Althoughaconstantvalueisderivedforaftereachpermeabilitytest,itÕswellestablishedthatthispropertyisRHdependant,Todeterminethesefunc-tionsforthetestedmaterials,thepermeabilitytypeEq.,anditsadaptedformforvalue(3),proposedbyGalbraithetal.[],wasusedandtheregressionswerecarriedoutusingtheLevenbergÐMarquardtmethodandSPSS14.0program. TheempiricalconstantsofEq.werederivedbytheregressionmethoddescribedabovefornakedspecimensasAandAForthecoatedsamples,however,amorecomplexmethodologywasappliedintheresultsanalysis.Withtheknowledgeofthebasematerialsfunction,assumingÞxedvaluesforandacceptingthatforeachtestasrepresentedinFig.,itwaspossibletoobtainthecoefÞcientsforthefunctionofthecoatingappliedonthegypsumbasematerial.TheresultswereA1,A2showstheresultsobtainedforthemeasurementresultsforthevapourpermeabilityoftheunpaintedspecimensandsdvalueoftheappliedpaint. chambersolution d,ar,intd,gd,p 2 3 4 5Plaster Paint Fig.2RelativehumidityproÞlesduringthevapourpermeabilityexperimentsTheEvaluationofHygroscopicInertia29 3MoistureBufferValueTheMBVexperiments,asdescribedin[],proposeacyclicclimaticexposurewhichconsistsof8hofhighrelativehumidity,followedby16hoflowrelativehumidity.Thistesttriestoreplicatethecycleseeninbedrooms.ForthespeciÞctestsdescribedinthisarticle,lowvaluewasÞxedat33%RHandthehighvalueat75%RH,foraconstanttemperatureof23C,whichisthebasictestconÞgurationproposedintheprotocol.Thecycleswererepeateduntilthespecimenweightoverthecyclevariedlessthan5%fromdaytoday.Thetestswereconductedinaclimatechamberensuringagoodcontrollevelofthetestconditions.Allthesamplestestedwereputintothechamberatthesametime.ThreesimilarsamplesweretestedforeachconÞguration.Eachsamplewasputonabalancewhenitwaslikelytohavereachedastablemassvariationoverthecycle.Withthisprocedureitwaspossibletotestalargenumberofsamples.Thebalancewasconnectedtoacomputerallowingforacontinuousrecordofthesamplemassvariation.Thesampleswereplacedhorizontallyonthebalance.Thebackofthesampleswaspreviouslytreatedwithepoxypaintandthefouredgeswerecoveredwithaluminiumtape,allowingvapourtransferonlyinthemainface. Fig.3Vapourpermeabilityofgypsumplasterandsdvalueofappliedpaint30N.M.M.RamosandV.P.deFreitas ThestablecycleforeachconÞgurationispresentedinFig..Thistypeofexperimentisinterestinginthewayitprovidesaneasyassessmentofthetransientbehaviourofabuildingelement.Justbywatchingthecurves,theeffectofpaintingiseasilyhighlighted.4FluxChamberTests4.1TestFacilityThetestfacilityisasmallcompartmentwhereÞnishingmaterialscontributiontohygroscopicinertiacanbeevaluated.Todoso,aßuxchamberwhereRHcanßoatfreelyasafunctionofboundaryconditionsandamountofmoisturebufferingwasdeveloped.Astrictcontroloftheheat,airandmoisturebalancesofthatchamberwasthereforecrucial.Accordingtothisidea,severalguidelinesweredeÞned:thebaseelementshouldbeasmallsizechamberwithcontrolofmoistureandairßuxes;thewholesetshouldhavetemperaturecontroland;temperatureandRHshouldbecontinuouslymonitored.Figuredisplaysaschemeofthetestfacility,followingtheseguidelines.Theßuxchamberwasbuiltinsideanexistingclimaticchamber(Fig.).Thischamberhasacapacityforcontrollingtemperatureintherange15Ð35CandRHintherange30Ð90%.ThatcontrolcanbedoneusingÞxedvaluesorusingpro-grammablecyclesincludingvariationofoneorbothparameters.Acontinuouslogoftheactualvaluesisregisteredinacomputer.Thesizeoftheßuxchamber(Fig.),tobestoredinside,correspondstoaboxwithavolumeof(1500584)mm.Thinkingofaregularbedroomwith2.7)m,thevolumescalefactorisaround1/70.Byplacingalltheelementsinsideaclimaticchamber,astricttemperaturecontrolwassecured.Theventilationsystemusesapump(Fig.)controlledbyßowmeters(Fig.)thatextractsairintwopointsinsidetheboxandaninleton Fig.4MassvariationstablecycleinMBVexperimentswithgypsumplasterbasedmaterialsTheEvaluationofHygroscopicInertia31 Fig.5Fluxchamberscheme Fig.6Climaticachamber32N.M.M.RamosandV.P.deFreitas topallowsfortheairtogetinand,atthesametime,preventspressuredifferences.Theairthatenterstheboxcomesdirectlyfromtheclimaticchamber,andthereforeitscharacteristicsareknown.Theairßuxvaluecorrespondstoarangeoftheairexchangerate(ach)of0.26Ð17hThetemperatureandhumidityoftheairbeingsuckedinsideareknown,sincethewholesetisinsidetheclimaticchamber.Alsoforthatreason,inÞltrationthroughtheopeningsdoesnÕtaffecttheoverallbalancesofheat,airandmoisture.Themonitoringsystem(Figs.)iscomposedofasetoftemperatureandRelativeHumiditysensorsconnectedtoadatalogger.Thedataloggeris Fig.7Fluxchamber Fig.8AirpumpTheEvaluationofHygroscopicInertia33 Fig.9Flowmeters Fig.10Monitoringsystem34N.M.M.RamosandV.P.deFreitas Fig.11Datalogger Fig.12TemperatureandrelativehumiditysensorsTheEvaluationofHygroscopicInertia35 connectedtoacomputerallowingtokeeptrackofresultsandstoretheminahard-drive.4.2TestResultsThetestsperformedintheßuxchamberreportedinthistext,consistedofthedeÞnitionofthestabledailyRHcycleforahygrothermalscenario.TheselectedscenariowasdeÞnedassumingthenumberofairchangesperhour,Rph,of0.5handavapourproductionof2g/h,during8hinthedailycycle.AsthetemperatureofthesystemwasÞxedat23C,thevapourproductionwasobtainedwiththeRHvariationoftheclimaticchamberbetween40and80%RH.Usingthosescenarios,differentcombinationsofsampleswereplacedinsidetheßuxchamber,resultingindifferentRHcycles.ForeachconÞguration,thedailyhygrothermalcycleisrepeateduntiltheßuxchamberRHfallsinastablecycle.ThedifferenttestconditionsusedarespeciÞedinTableTheresultsofthetestsarepresentedinTableandFig.displaystheRHvariationinsidetheßuxchamberforthetestedcombinations.ForquantiÞcationofthetestresults,thedifferencebetweentheaverageRHandtheRH90thpercentile,,isused.TheaverageRHvalueobtainedforeachstablecycleshowedasmallvariation,demonstratingthehighcontrolleveloftheexperiments.TheseresultsclearlyillustratetheapplicationoftheßuxchamberinmeasuringtheactualRHdampeningcausedbythepresenceofdifferentlevelsofmoisturebufferingincontactwithinsideair.TheobservedvariationshighlightthecontributionofdifferentelementstohygroscopicinertiaanditseffectonRHpeakdampening.TestHI1resultedina Table1FluxchambertestsconÞgurationTestPeriod(h)ClimaticchamberFluxchamberC)RH(%)Rph(h)SamplesHI10Ð823800.5Ð8Ð242340HI20Ð823800.50.75m8Ð242340HI30Ð823800.50.75m8Ð242340 Table2FluxchambertestsEnsaioRH(%)RH(%)RHHI154.673.218.6HI254.363.99.6HI354.766.812.136N.M.M.RamosandV.P.deFreitas differencebetweenpeakandaverageof18.6%RH.Theintroductionintheßuxchamberofaporousmaterial,nakedgypsumplaster,inHI2testresultedinadifferencebetweenpeakandaverageof9.6%,acleareffectofhygroscopicinertia.Thesametypeofmaterial,butwithacoatingapplied,correspondingtotestHI3,resultedinlesspeakdampening,adifferencebetweenpeakandaverageof12.1%RH.Thisresultclearlydemonstrateshowacoatinginßuencesmaterialscontri-butiontohygroscopicinertia.5HygroscopicInertiaClassesAmethodofbringingthehygroscopicinertiaconceptclosertopractitionershasbeendevelopedandexperimentallyevaluatedby[].Thebasicidea,illustratedin,istohaveapredictiontoolthatcanestablisharelationbetweenthedampeningoftheRHvariationinaroomanditshygroscopicitylevel,whichismainlydependantonthesurfaceÞnishingmaterialsandfurnishing.Accordingtotheprinciple,ahygroscopicinertiaindexwasdeÞnedasasinglenumber,representingthehygroscopicinertiaofaroomandthatcancorrelatetotheexpectedreductionoftheRHßuctuation.ItwasdecidedthatthisindexshouldconcentrateonlyondailycyclesanditshouldbederivedfromroomconÞgurationandknownmaterialproperties.TheMBVÑMoistureBufferValuewastheselectedmaterial/elementpropertythusactingasabaseforthatindexdeÞnition.Theproposeddailyhygroscopicinertiaindex,,isdeÞnedby[]asafunctionofMBV,accordingtoexpression),whereMoisturebuffervalueofelement(g/(m.%RH));ofelementMoisturebuffervalueofcomplexelement(g/%RH);ImperfectmixingreductioncoefÞcient(=airexchangerate(roomvolume(mVapourproductionperiod().ThecanbeunderstoodastheroomMBV,homogenizedtoairrenovationconditionsandvapourproductionperiodvariations. PniCr;iMBViSiþPmjCr;jMBVobj;jNVTG! gm3%HR4Þ Fig.13RHvariationinsidetheßuxchamber(fc)andtheclimaticchamber(cc)TheEvaluationofHygroscopicInertia37 TheevaluationofaRHvariationcurveintimeshouldbebasedonasinglenumber,keepinginmindtheprincipledescribed.TheAMDRparameterwasdeÞnedaccordingtoEq.,whereHRistheaveragerelativehumidityvariationand standsforthedailyaverageofthe90thpercentileoftherelativehumidityvariation.TheindexreferstothebasescenarioofaroomwithouthygroscopicityandidentiÞesascenariounderstudyforthatsameroom.TheAMDRparametercanthereforebeinterpretedasrelativedailyaverageamplitudeofaRHvariationofaroomhygroscopiccon-Þguration.ThisparameterisinterestingsincetheaverageRHvariationinlong-termanalysiswillnotbeaffectedbydailyhygroscopicinertia.AMDR HR90HRm Theselectedparameterswereprovenby[]tobeconnectedbyEq..TheresultingcurvesupportsthedeÞnitionofdailyhygroscopicinertiaclasses,accordingtoFig. 0,00,20,40,60,81,0 CLASS I CLASS II CLASS III CLASS IV Fig.15GraphicrelationbetweenAMDRand RHit Hygroscopic inertia indexRHi peak-avg difference III Classes No hygroscopicity Hygroscopic room Fig.14HygroscopicinertiaclassesdeÞnitionprinciple38N.M.M.RamosandV.P.deFreitas AMDR TheadoptedclassessupportanewclassiÞcationofbuildingelementscontri-butiontothehygroscopicinertiaofaroom,basedontheirMBVandonaratiobetweenroomvolumeandareaofapplicationof0.7(Fig.6NumericalSimulation6.1NumericalModelAdecisionwasmadetonumericallysimulatethebehaviourofaroomandthemoisturebuffercapacityofthematerialsusedintheroomsimulation.Thispro-videddatathatcanillustratethedesiredrelationbetweenhygroscopicinertiaandroomconÞguration.TheauthorschosetouseforthesesimulationsthesoftwareprogramHAM-Tools[].TheInternationalBuildingPhysicsToolbox,isasoftwarelibraryspe-ciallyconstructedforHAMsystemanalysisinbuildingphysics.AspartofIBPT,HAM-ToolsisopensourceandpubliclyavailableontheInternet.Thelibrarycontainsblocksfor1DcalculationofHeat,AirandMoisturetransferthroughbuildingmaterials.ThetoolboxisconstructedasamodularstructureofthestandardbuildingelementsusingthegraphicalprogramminglanguageSimulink.Allmodelsaremadeasblockdiagramsandareeasilyassembledinacomplexsystemthroughthewell-deÞnedcommunicationsignalsandports.6.2MBVSimulationsAnassemblyofHAM-ToolsmoduleswasusedforsimulatingtheMBVexperi-ment.Thesimulationswereperformedusingthevirtualspecimenslistedin.Thedataforthechosenmaterialsusedinthesimulationswasretrievedfrom[].Thedifferent(convectivewatervapourtransfercoefÞcient)valuesused Fig.16BuildingelementclassiÞcationasafunctionoftheircontributiontoaroomÕshygroscopicinertiaTheEvaluationofHygroscopicInertia39 representthepossibilityofcoatingswithdifferentadditionalvapourresistancesappliedinthegypsumboardspecimens.InFig.,themassvariationofallthespecimensinthesteadycycleispre-sented.TheMBVforeachspecimenispresentedinTable.Aswecansee,thisnumber,whenassociatedtothetestedelements,providesthemeanstocomparethem.Butasitwassaidbefore,thisnumberisusedaheadforhygroscopicinertiaanalysis.6.3RoomSimulationsAdifferentHAM-ToolsmodulesassociationallowedforthesimulationofaroomÕshygrothermalbehaviourinyearlycycles.Thevirtualroomis2.5m,withoneexteriorwall,containinga1mwindow,facingsouth.Thesurroundingroomsareassumedtohavesimilarconditionsoftem-peratureandrelativehumidityasthesimulatedroom.TheclimateconditionsweredeÞnedforLisbon,usingMeteonormsoftware.TheinsidetemperaturewasallowedtoßoatbetweenTminand28C,andRHwasallowedtoßoatbelow90%.Theventilationrateisconstantandthevapourproductiontakesplacebetween0and8h,withaconstantvalue.Onwallsandceilingtheadmittedmaterialwasgypsumboardandontheßoorspruce.Tablesdescribetheconditionsthatwerechangedforeachsimulation.Itcanbeeasilyinferredthattheroomcon-ÞgurationintheÞrstlineofTablestandsfortheroomwithnohygroscopicinertia,thereferenceroom.PartialresultsfromsimulationsSG1andSG2arepresentedinFigs.illustratingthetestedscenarios. 04812162024 W (kg/m2) Specimen2 Specimen 3 Specimen 4 Fig.17Moisturecontentvariationofallthespecimensinastablecycle Table3VirtualspecimenÕsSpecimenMaterial(s/m)Thickness(m)1Gypsumboard2e-80.012Gypsumboard2e-90.013Gypsumboard2e-100.014Spruce2e-80.0140N.M.M.RamosandV.P.deFreitas Thetemperaturevariations,presentedinFig.showedalmostnoinßuenceofhygroscopicinertia.Therelativehumidityvariation,ontheotherhand,ishighlyinßuencedbyhygroscopicinertia,asitcanbeseeninFig..Thatresultisalsovisibleinvapourpressurevariation,presentedinFig..Thatinßuencehowever,isonlyimportantonpeaklevel.Theaveragevalueswilltendtobethesame,withorwithouthygroscopicinertia.ThatisquiteclearonFig.wheremonthlyvaluesarecloseandasperiodsaremoreextended,thatdifferencetendstobeevensmaller. Table4MBVforthefourspecimenstestedSpecimen1234)0.09710.03660.00450.0832 Table5parametersadoptedinSimulationsTmin(C)G(g/h)N(hSG1ÐSG2181001.0SG11ÐSG12181501.0SG15ÐSG16151001.0SG19ÐSG20211001.0SG23ÐSG24181000.67SG27ÐSG28181000.33 Table6RoomconÞgurationsSimulationsWallsCeilingFloorArea(m(s/m)Area(m(s/m)Area(mSG:1-11-15-19-23-27342e-1212.252e-1212.252e-12SG:2-12-16-20-24-28342e-812.252e-812.252e-12 Fig.18ResultsfromsimulationsSG1andSG2ÑtemperaturevaluesfromOctobertoMarchTheEvaluationofHygroscopicInertia41 020406080100120140160180 Fig.19ResultsfromsimulationsSG1andSG2ÑrelativehumidityvaluesfromOctoberto Fig.20ResultsfromsimulationsSG1andSG2ÑvapourpressurevaluesfromOctobertoMarch Fig.21ResultsfromsimulationsSG1andSG2Ñrelativehumidityaverage42N.M.M.RamosandV.P.deFreitas 6.4HygroscopicInertiaAnalysisTheapplicationofthehygroscopicinertiaclassesmethodallowsforthedeÞnitionofparametersAMDRandcorrespondingtothesimulationscenarios.ThevaluesobtainedarepresentedinTable,andrevealthatthescenarioshygro-scopicalyactivewouldbeplacedinclassIII-IV.6.5MouldGrowthRiskAssessmentSeveralauthorshavestudiedtherelationshipbetweenwateractivityinasubstrateandthedevelopmentofmouldonitssurface(e.g.[]).Themoulddevelopmentundertransientconditionsoftemperatureandhumidityishighlycomplex.Ref-erence[]proposedamodelbasedontheTOW(timeofwetness)concepttosolvethisproblem.TOWisthequotientbetweenthetimeperiodwherethesurfaceRHisabove80%andthetotaldurationoftheperiodunderanalysis.Usingthatmodelinseverallaboratorytests,therewasevidencethatforTOW0.5theriskformouldgrowthisverylow.UsingtheTOWconceptwhensimulatingaroomÕshygrothermalbehaviouritispossibletodeÞnethenumberofdayswithwith0.5,ndnd0.5,asasimplisticindicatorofthemouldgrowthrisk.Thisapproachlosesaccuracyifthesimulationsindicateactualsurfacecondensationandtheanalysistoolisunabletotreatthatprocesswithhighprecision.Additionally,theauthorsuseanotherparameter,nd,representingthenumberofdayswhensurfacecondensationwasdetected.Theobjectiveoftheseindicatorsisnottoaccuratelyestimatemouldgrowthrisk,butrathertocomparehygrothermalscenarios.Usingtheseparametersintheabovesimulatedscenariosandimposinganadditionalconditionoftheexistenceofaratherextremethermalbridge,deÞnedby theresultsforeachscenarioareaspre-sentedinFig.ThisresultshowstheimportanceofhygroscopicinertiaandthebeneÞtsthatcanderivefromasssRHvariationwithimportantpeakreduction.Adesignmethodforthepreventionofmouldgrowthcanbederivedfromthisanalysis.Themethoditselfcanhavedifferentlevelsofcomplexity.Thebasicideais: Table7Parametersforhygroscopicinertiaanalysis.%RH))AMDR(%)SG:1-11-15-19-23-2701.0SG20.4380.23SG120.4380.24SG160.4380.24SG200.4380.23SG240.6530.18SG281.3260.12TheEvaluationofHygroscopicInertia43 deÞnetheriskassociatedwiththeroomÕsRHvariationforthefourclassesofhygroscopicinertia;selecttheadequatevalueanddeÞnetheroomÕsrenderingstoprovidethat7ConclusionsThistextpresentsresearchthatallowsforthefollowingconclusions:Moisturebufferingtestswereconductedatelementlevel.RenderingsÞnishedwithdifferentcoatingsweretestedforMBVdetermination,allowingforbuffereffectcomparison.TheÞnishingcoatinghasarelevantinßuenceonthateffect. 20406080100160180SG1 - SG2SG11 - SG12SG15 - SG16SG19 - SG20SG23 - SG24SG27 - SG28max(ndcond, nd�TOW0,5) (days) ref Fig.22Mouldgrowthrisk Fig.23Therelevanceofbalancebetweenhygroscopicinertiaandsurfaceprotection44N.M.M.RamosandV.P.deFreitas Hygroscopicinertiatestswereconductedinaßuxchamberthatrepresentsasmallscaleroom.TheinßuenceofdifferentelementsonRHvariationwasobtained.Theexperimentalevidenceofdailyhygroscopicinertiawasdemon-ThedeÞnitionofdailyhygroscopicinertiaclassesbasedonaroomÕsindexallowsforthepredictionoftheRHvariationamplitude;TheMBVpropertycanbeusedasanindicatorofanelementÕscontributiontotheroomÕshygroscopicinertia;Mouldgrowthriskislowerforhighervaluesofhygroscopicinertia,admittingthesamecompositionofthesurfaceÕsÞnalrendering;ThedesignandselectionofinteriorÞnishesinpracticecanbeneÞtfromtheproposedapproachtothehygroscopicinertiaconcept.ButthebalancebetweentheMBVsuppliedbyasurfaceanditsprotectionagainstbiologicaldefacement)mustbeachieved.1.Adan,O.:OnthefungaldefacementofinteriorÞnishes.Ph.D.thesis,EindhovenUniversityofTechnology(1994)2.Galbraith,G.,MClean,R.,Guo,J.:Moisturepermeabilitydatapresentedasamathematicalrelationship.Build.Res.Inf.(3),157Ð168(1998)3.ISO12571:2000:HygrothermalperformanceofbuildingmaterialsandproductsÑDeterminationofhygroscopicsorptionproperties(2000)4.ISO12572:2001:HygrothermalperformanceofbuildingmaterialsandproductsÑDeterminationofwatervapourtransmissionproperties(2001)5.Kalagasidis,A.:HAM-Tools:anintegratedsimulationtoolforheat,airandmoisturetransferanalysesinbuildingphysics,DepartmentofBuildingTechnology,BuildingPhysicsDivision,ChalmersUniversityofTechnology,Gothemburg,Sweden(2004)6.Kumaran,M.:Heat,airandmoisturetransferthroughnewandretroÞttedinsulatedenvelopeparts(Hamtie),IEAANNEX24(1996)7.PadÞeld,T.:Theroleofabsorbentbuildingmaterialsinmoderatingchangesofrelativehumidity.Ph.D.thesis.DepartmentofStructuralEngineeringandMaterials,Lyngby,TechnicalUniversityofDenmark150(1998)8.Ramos,N.:Theimportanceofhygroscopicinertiainthehygrothermalbehaviourofbuildings(inPortuguese).Ph.D.thesis,DepartmentofCivilEngineering,FEUP,Porto,Portugal(2007)9.Rode,C.,Peuhkuri,R.,Mortensen,L.,Hansen,K.,Time,B.,Gus-Tavsen,A.,Svennberg,K.,Arfvidsson,J.,Harderup,L.,Ojanen,T.,Ahonnen,J.:Moisturebufferingofbuildingmaterials,ReportBYG-DTUR-126,DepartmentofCivilEngineering,DTU,Lyngby,Denmark(2005)10.Sedlbauer,K.:Predictionofmouldfungusformationonthesurfaceofandinsidebuildingcomponents.Ph.D.thesisÑreport,FraunhoferInstituteforBuildingPhysics,Germany(2001)11.Simonson,C.,Salonvaara,M.,Ojanen,T.:Theeffectofstructuresonindoorhumidity-possibilitytoimprovecomfortandperceivedairquality.IndoorAir,243Ð251(2002)TheEvaluationofHygroscopicInertia45