外文翻譯--混合性能單軸構(gòu)件式攪拌機(jī) 英文版.pdf

08855,BlenderAcetaminophenpowderSomeer.Iwithrespecttotwoaxes:ahorizontalaxis(tumblingmotion),andacentralsymmetryaxis(spinningisconductedonmixingperformanceofpowdersandtheeffectofcriticalinametallthelongativeltypeisthetumblingblender,wheregrainsowbydevice,mixingandsegregationmechanismsinthesedevicesarenotmicronizedAcetaminophen.Weuseextensivesamplingtocharacter-PowderTechnology196(2009)17ContentslistsavailableatScienceDirectPowderTechnologysfullyunderstoodandthedesignofblendingequipmentislargelybasedonempiricalmethods.Tumblersarethemostcommonbatchmixersinindustry,andalsonduseinmyriadofapplicationasdryers,kilns,coaters,millsandgranulators48.Whilefree-owingmaterialsinrotatingdrumshavebeenextensivelystudied9,10,cohesivegranularowsinthesesystemsarestillnotcompletelyunderstood.Littleisknownabouttheeffectoffundamentalparam-izemixingbytrackingtheevolutionofAcetaminophenhomogeneityusingaNearInfraredspectroscopydetectionmethod.AftermaterialsandmethodsaredescribedinSection2,resultsarepresentedinSection3,followedbyconclusionsandrecommendations,whicharepresentedinSection4.eterssuchasblendergeometry,speed,lllevloadingpatternandaxisofrotationoncohesivepowdersorthescalingrequirementsHowever,conventionaltumblers,rotatingallshareanimportantcharacteristic:whileCorrespondingauthor.Tel.:fax:+1E-mailaddress:(F.J.Muzzio).0032-5910/$seefrontmatter2009ElsevierB.V.Alldoi:10.1016/j.powtec.2009.06.008acombinationofgravityisaverycommonrotationonthemixingperformanceofafree-owingmatrixofFastFlolactoseandAvicel102,containingamoderatelycohesiveAPI,andvesselrotation.Althoughthetumblingblender1.IntroductionParticleblendingisarequiredstepspanningtheceramic,food,glass,pharmaceuticalsindustries.Despitemixing(orperhapsbecauseofit),comparmechanismsinvolved13.Acommonandaxisofrotation.InthisworkAcetaminophenisusedastheactivepharmaceuticalingredientandtheformulationcontainscommonlyusedexcipientssuchasAvicelandLactose.Themixingefciencyischaracterizedbyextractingsamplesafterpre-determinednumberofrevolutions,andanalyzingthemusingNearInfraredSpectroscopytodeterminecompositionaldistribution.Resultsshowtheimportanceofprocessvariablesincludingtheaxisofrotationonhomogeneityofpowderblends.2009ElsevierB.V.Allrightsreserved.varietyofapplicationsurgical,polymers,andhistoryofdrysolidsylittleisknownoftheofbatchindustrialmixerdirectionofrotationisfast,mediatedbyaconvectivemixingprocess,mixinginthehorizontal(axial)direction,drivenbyadispersiveprocess,isoftenmuchslower.Inthispaper,weexperimentallyinvestigateanewtumblingmixerthatrotateswithrespecttotwoaxes:ahorizontalaxis(tumblingmotion),andacentralsymmetryaxis(spinningmotion).Weexaminetheeffectsoflllevel,mixingtime,loadingpatternandaxisofRelativestandarddeviationNIRfundamentalparametersincludingblendergeometry,speed,lllevel,presenceofbafes,loadingpattern,Mixermotion).AdetailedstudyComparingmixingperformanceofuniaxialAmitMehrotra,FernandoJ.MuzzioDepartmentofChemicalandBiochemicalEngineering,RutgersUniversity,Piscataway,NJ,abstractarticleinfoArticlehistory:Received17February2009Receivedinrevisedform30May2009Accepted14June2009Availableonline27June2009Keywords:PowdermixingCohesionThedynamicsinvolvedintheorystillremainsunderdeveInmanyindustries,includingmixers”.Tumblingmixersaresomenumberofrevolutions.coneblendersandbinblenders.mediatedbyaconvectivemixingprocess,isoftenmuchslowjournalhomepage:www.elel,presenceofbafes,mixingperformanceofofthedevices.aroundahorizontalaxis,homogenizationinthe7324452581.rightsreserved.andbiaxialbinblendersUnitedStatesmixingremainsatopicofinterestformanyresearchers；howevertheloped.Mostofthemixersarestilldesignedandscaleduponempiricalbasis.pharmaceutical,themajorityofblendingisperformedusing“tumblinghollowcontainerswhicharepartiallyloadedwithmaterialsandrotatedforcommonexamplesincludehorizontaldrummixers,v-blenders,doubleInallthesemixerswhilehomogenizationinthedirectionofrotationisfast,process,mixinginthehorizontal(axial)direction,drivenbyadispersiventhispaper,/locate/powtec2.MaterialsandmethodsThematerialsusedinthestudyarelistedinTable1,alongwiththeirsizeandmorphology.Acetaminophenisblendedwithcom-monlyusedexcipientsandisusedasatracertoevaluatethedegreeofhomogeneityachievedasafunctionofnumberofrevolutions.Acetaminophenisoneofthedrugsmostwidelyusedinmixingstudies,andAvicelandLactosearecommonlyusedpharmaceuticalTheexperimentalplanusedinthisstudyisasfollows:Filllevel:blender160%Filllevel:blender260%,70%,80%Loadingpattern:blender1sidesideloading,topbottomloadingLoadingpattern:blender2sidesideloading,topbottomloadingSpeed:blender115rpm,20rpm,25rpmSpeed:blender2rotational/spinning:15/7.5rpm,20/10rpm,30/15rpmSamplingtime:blender1,blender27.5,15,30,60,120revolutions3.ResultsThehomogeneityindexusedistheRSD,whereCistheconcentrationofeachindividualsample,C_istheaverageconcentra-tionofallsamplesandnisthetotalnumberofsamplesobtainedatagivensamplingtime.RSD=SC；whereS=nPC2PC2nn1sWeexaminetheeffectoflllevelonmixingperformance.2A.Mehrotra,F.J.Muzzio/PowderTechnology196(2009)17excipients.IntheinterestofbrevitytheirSEMimagesarenotincludedinthispaper,butcanbefoundin“HandbookofPharmaceuticalexcipients”.2.1.NearinfraredspectroscopyAcetaminophenhomogeneitywasquantiedusingnearinfraredspectroscopy.Acalibrationcurvewasconstructedforapowdermixturecontaining(inaverage)35%avicelPH102,62%lactoseand3%acetaminophen.Nearinfrared(NIR)spectroscopycanbeausefultooltocharacterizeacetaminophen.SamplesarepreparedbykeepingtheratioofAviceltolactoserandomizedinordertominimizeeffectsofimperfectblendingofexcipientsduringtheactualexperimentsontheaccuracyoftheresults.TheRapidContentAnalyzerinstrumentmanufacturedbyFOSSNIRSystems(SilverSpring,MD)andVisionsoftware(version2.1)isusedfortheanalysis.Thesamplesarepreparedbyweighing1gofmixtureintoseparateopticalscintillationvials；(KimbleGlassInc.Vineland,NJ)usingabalancewithanaccuracyof0.01mg.Near-IRspectraarecollectedbyscanningintherange11162482nminthereectancemode.Partialleastsquare(PLS)regressionisusedincalibrationmodeldevelopmentusingthesecondderivativemathematicalpretreatmenttominimizetheparticlesizeeffects.AsshowninFig.1,excellentagreementisachievedbetweenthecalibratedandpredictedvalues.2.2.Binblendersusedinthisstudy:uni-axialblender(Blender1),bi-axialblender(Blender2)Duetoitswidespreaduse,acylindricalblender1withacapacityof30Lischosenasareferenceblenderinthestudy.AsshowninFig.2,thisblenderhasacircularcrosssectionandtapersatthebottom.Itcanbeusedwithorwithoutbafes,whicharemountedonaremovablelid.Inthisstudyalltheexperimentsareconductedwithouttheuseofbafes.Mixingperformanceinthisdeviceisusedtoprovideabase-lineforevaluatingthemixingperformanceofanewlydevelopedblender2withacapacityof40L,whichisalsocylindrical,inordertodeterminetheeffectofdualaxisofrotationonmixingperformance.TheblendershowninFig.2(b)hastwoaxisofrotation.Thespinningrateofprecessionrelativetothecentralaxisofsymmetryisgearedtobehalfofthatoftherateofrotationaroundthehorizontalaxis.2.3.ExperimentalmethodTwotypesofinitialpowderloadingusedintheexperiments:topbottomloadingandsideloading,whichareschematicallyrepre-Table1Materialsstudiedinthispaper.NameSizeandmorphologyVendor,City,StateFast-FloLactose100,sphericalForemostfarms,Newark,NJAvicelPH102Microcrystallinecellulose90,needle-likeFMC,Rothschild,WIAcetaminophen40,needle-likeMallinckrodt,StLouis,MOsentedinFig.3.Toavoidagglomeration,theAPI,acetaminophen,wasdelumpedpriortoloadingitintotheblenderbypassingitthrougha35meshscreen.Inordertocharacterizemixingperformance,agroovesamplerwasusedtoextractsamplesfromtheblendersat7.5,15,30,60,120revolutions.Thethiefwascarefullyinsertedinthebin,andacorewasextractedateachpointofinsertion(each“stab”)minimizingperturbationtothepowderbedremainingintheblender.Approxi-mately7samplesaretakenfromeachthiefstab,andatotalofvestabsareusedateachsamplingtime,asshowninFig.4soatotalof35samplesaretakenateachsamplingpoint.PreviouslytherehavebeenstudiesontheeffectoflllevelintheBohlebinblender,GallaybinblenderandV-blenderanddoubleconeblender1113.Alltheaforementionedblendershaveonlyoneaxisofrotation,thereforetheobjectiveofthisstudyistoexaminehowdualaxisimpactmixingperformancesathighlllevels.Toavoidrepetition,studiesforlllevelarenotconductedforbinblender1.ResultsavailablefromapreviousstudyusingMgStasatracershowedthatmixinginauni-axialblendersloweddownquitedramaticallyasthelllevelexceeded70%.Moreover,resultsforacetaminophencanbeassumedtobesimilartothoseobtainedinpreviousworkbyMuzzioetal.11,13,forasingleaxisrectangularbinblender11,whichhaveshownthatevenafterfewhundredrevolutionshomo-geneityachievedwitha80%lllevelisverypoorascomparedto60%lllevel.Toexaminetheeffectoflllevelinadualaxisblender,experi-mentswereperformedinblender2withthetop-bottomloadingpatternforarotationalspeedof15rpmandwithspinningspeedof7.5rpm.Thelllevelsexaminedare60%,70%and80%respectivelyandsamplesaretakenafter7.5,15,30,60,120revolutions.TypicalresultsareshowninFig.5,whichshowstheRSDvs.numberofblenderrevolutions.Asexpectedfornon-agglomeratingmaterials,thecurvesFig.1.NearInfrared(NIR)spectroscopyvalidationcurve.Theequationusedtopredictacetaminophenconcentrationisvalidatedbytestingsampleswithknownamountsofacetaminophenconcentration.Theyaxisrepresentstheconcentrationcalculatedfromtheequationandthexaxisrepresentstheactualconcentration.Thusaperfectlystraightlineat45wouldrepresentthebestcalibrationmodel.Eachpointonthegraphrepresentsasinglesample.Theconcentrationofacetaminophenexaminedhererangesfrom0to8%.Fig.2.Pictorialrepresentationof(a)binblender1and(b)binblender2showingthecorrespondingaxisofrotation.Fig.3.Schematicoftheloadingpatternusedinthestudy.Intopbottomloading,AvicelisloadedrstintotheblenderfollowedbyLactoseontopofitandnallyAcetaminophenisuniformlysievedover.InsidesideloadingavicelisplacedatthebottomandthenAcetaminophenisonlysievedonlyinhalfpartoftheblenderandissandwichedbetweenlactoseandAvicel.Fig.4.(a)Thiefsampler(b)topviewofthesamplingpositionscheme.Fig.5.Mixingcurvesfordifferentlllevelsinblender2.TheRSDofacetaminophenisplottedasafunctionofnumberofrevolutions.Theloadingpatternintop-bottomandtheblenderrotationalspeedis15rpmwithspinningspeedof7.5rpm.3A.Mehrotra,F.J.Muzzio/PowderTechnology196(2009)17d4A.Mehrotra,F.J.Muzzio/PowderTechnology196(2009)17showarapidlydecayingregion.Theslopeofthecurvesinthisregion,insemi-logarithmiccoordinates,isusedtodenethemixingrate.Thecurvesthenlevelofftoaplateauthatindicatesthemaximumdegreeofhomogeneitythatisachievableintheblenderforagivematerial.Similartopreviousstudieswithothertumblingblendersweob-servethatblendingperformanceisadverselyaffectedbyincreasinglllevels.AsshowninFig.5,thecurvefor80%llperformsmorepoorlythanthosefor60%and70%ll；aslllevelincreases,RSDcurvesdecaymoreslowly,signifyingaslowermixingprocess.However,theeffectisnotaspronouncedasinotherbinblendersandafteraboutonly100revolutions,thesameplateau(thesameasymptoticblendhomo-geneity)isachievedforallthreelllevels.Next,theeffectofrotationalspeedisinvestigatedintheblender1Fig.6.Mixingcurvesfortop-bottomloadingexperimentswith60%lllevel.RSDisplotteblender1,whilesolidlinesrepresentdatapointsfromtheblender2.withoneaxisofrotationandiscomparedtotheblender2withdualrotationaxis.Experimentswereconduc