ThesinglemostimportantadvanceintheuseofcentrifugalforcetoseparateBIOLOGicallyimportantsubstances.wasthecouplingofmechanics,opticsandmathematicsbyT.SvedbergandJ.W.Williamsinthe1920"s.Theyinitiatedthemathematicsandadvancedtheinstrumentation.1toapointwhereitwaspossIBLetoprovethatproteinswerelargemoleculesthatcouldbeweighedinacentrifuge.2.Inhonorofthatwork,thevalueforamolecule"s(ororganelle"s)sedimentationvelocityinacentrifugalfieldisknownasitsSvedbergconstantorSvalueforshort. TheinstrumentationhasprogressedquitefarsincetheearlyworkofSvedbergandWilliams.Today,anytechniqueemployingthequantitativeapplicationofcentrifugalforceisknownasultracentrifugation.Thedesignofthebasicinstruments,therotorsandtheopticalsystemsformeasurementaretooextensivetocoverinthisvolume.Forourpurposes,wewillconcentrateontwotypesofrotor,andafewselectedparameterstobemeasured. Rotorsforacentrifugeareeitherfixedangles,swingingbuckets,continuousflow,orzonal,dependinguponwhetherthesampleisheldatagivenangletotherotationplane,isallowedtoswingoutonapivotandintotheplaneofrotation,designedwithinletandoutletportsforseparationoflargevolumes,oracombinationofthese.FigureF.1demonstratesthecharacteristicsofeachofthese. Fixedanglesgenerallyworkfaster;substancesprecipitatefasterinagivenrotationalenvironment,ortheyhaveanincreasedrelativecentrifugalforceforagivenrotorspeedandrADIus.Theyalsohavefew(orno)movingpartsontherotoritselfandthushavevirtuallynomajormechanicalfailures,otherthanpotentialmetalstress,whichallrotorsundergo.Theserotorsarethework-horseelementsofacelllaboratory,andthemostcommonisarotorholding8centrifugetubesatanangleof34°Cfromthevertical(suchastheSorvallSS-34rotorortheBeckmanJA-20).FigureF.1presentsacross-sectionaldiagramoftheSorvallSS-34. Swingingbucketrotors(alsoknownashorizontalrotors)havetheadvantagethatthereisusuallyacleanmeniscusofminimumarea.Inafixedanglerotor,thematerialsareforcedagainstthesideofthecentrifugetube,andthenslidedownthewallofthetube.Thisactionistheprimaryreasonfortheirapparentfasterseparation,butalsoleadstoabrasionoftheparticlesalongthewallofthecentrifugetube.Foraswingingbucket,thematerialsmusttraveldowntheentirelengthofthecentrifugetubeandalwaysthroughthemediawithinthetube.Sincethemediaisusuallyaviscoussubstance,theswingingbucketappearstohavealowerrelativecentrifugalforce,thatisittakeslongertoprecipitateanythingcontainedwithin.If,however,thepointofcentrifugationistoseparatemoleculesororganellesonthebasisoftheirmovementsthroughaviscousfield,thentheswingingbucketistherotorofchoice.Moreover,ifthereisadangerorscrapingoffanoutershellofaparticle(suchastheoutermembraneofachloroplast),thentheswingingbucketistherotorofchoice.Mostcommonclinicalcentrifuges1haveswingingbuckets.Sincethebucketsareeasytointerchange,thistypeofrotorisextremelyversatile.Itsmajordrawbackisthenumberofmovingpartswhicharepronetofailurewithextendeduse. Nearlyallcellbiologylaboratorieswillhaveseveralexamplesoffixedangleandhorizontalrotors.Whilethesamplevolumesoftheserotorscanbesignificant,theyarelimiting.Toovercomethislimitation,acontinuousflowcentrifugecanbeused.Limnologistsoftenemploysuchadevicetoseparateplanktonfromgallonsoflakewater.Cellbiologistsemployzonalrotorsforthelargescaleseparationofparticlesondensitygradients.Zonalrotorscancontainupto2litersofsolutionandcanworkwithtissuesamplesmeasuredinounces(orevenpounds).Therotorsarebroughtuptoabout3000RPMwhileempty,andthedensitymediaandtissuesareaddedthroughspecializedports.Thistypeofrotorhasadistinctpreparativeadvantageoverthegradientcapacityofmoretypicalrotors. Forbiologicalwork,thetubesaredividedintofunctionalgroups,madeofregularglass,Corexglass,nitrocellulose,orpolyallomer.Regularglasscentrifugetubescanbeusedatspeedsbelow3,000RPM,thatisinastandardclinicalcentrifuge.Abovethisspeed,thexgforceswillshattertheglass. AspecialhighspeedglasswiththetradenameofCorex(CorningGlassWorks)hasbeendevelopedtohandlespeedsupto15-18,000RPM.Thesetubescanbeusedinmostroutinecellorganellepreparations,if,andonlyif,theproperadaptersarealsousedwithinthecentrifugerotors.Thesetubesarerelativelyexpensive(about$3.50each)andshouldneverbeusedforanypurposeotherthanthecentrifuge.Anytubeswithscratchesorchipsshouldbedisposedofimmediately.Thesehigh-speedglasstubeswillshatterabove18,000RPM. Forworkinthehigherspeedranges,centrifugetubesaremadeofplasticornitrocellulose.Preparativecentrifugetubesaremadeofpolypropylene(sometimespolyethylene)andcanwithstandspeedsupto20,000RPM.Thesetubesshouldbecarefullyexaminedforstressfracturesbeforeuse.Atubewithafracturewillholdfluidsbeforecentrifugation,butthecrackswillopenundercentrifugalforce. Nitrocelluloseareinexpensiveandusedformostultracentrifugation.Theyaremeanttobeusedonlyonceandthendiscarded.Repeateduseincreasesthechanceoftubecollapseduetointernalmolecularstresswithinthetubewalls.Thereisnowaytopre-determinethis,soitisbesttoalwaysuseanewtubeforultracentrifugation.Nitrocellulosealsobecomeslessflexiblewithage,andthepurchasedateforalltubesshouldbenoted.Tubesolderthan1yearshouldbediscarded.Acentrifugetubeisinexpensivewhencomparedtothelossoftimeandmaterialsforatypicalultracentrifugerun. Polyallomertubesarere-usable,moreexpensive,andslippery.Moleculeswillslidedownthewallsofthesetubesmoreeasily,andthusarethetubesofchoiceforprecipitationcentrifugations.Theyarealsomorechemicallyinert. Moderndayultracentrifugescangenerateforcesinexcessof300,000timesthatofgravity,forcessufficienttoovercometheverycohesionofmostmolecules(includingthemetaloftherotor).Theforceisusuallygivenassomevaluetimesthatofgravity. Thecentrifugalforceisdependentupontheradiusoftherotationoftherotor,thespeedatwhichitrotates,andthedesignoftherotoritself(fixedangle,vsswingingbucket).Rotorspeedanddesigncanbeheldconstant,buttheradiuswillvaryfromthetopofacentrifugetubetothebottom.Ifameasurementfortheradiusistakenasthemid-point,orasanaverageradius,andallforcesaremathematicallyrelatedtogravity,thenoneobtainsarelativecentrifugalforce,labeledasxg.Centrifugationproceduresaregivenasxgmeasures,sinceRPMandotherparameterswillvarywiththeparticularinstrumentandrotorused.RelativeCentrifugalForceisaconstantthatisindependentoftheapparatusused. FigureF.2presentsaNomogramforcalculationofR.C.F.foragivenradiusandRPM.Asimpleformulaforcalculatingthisvalueis: Thedifficultywithusingtheformulaisestablishingthevalueforr.Typically,therearethreervaluesgiven(bythemanufacturer)forarotor:themaximum,minimumandaverager.Thesecorrespondtothedistancesfromthecenterofrotationtothebottom,topandmiddleofthesampletube. Ifthedensityandviscosityofthemediumareknown,aswellasthedensityofagivenparticle,thenthetimeneededtocompletelysediementaparticlecanbedeterminedbytheformula: WorkingwiththemaximumRCFforeachrotor,theconversioncanbemadewiththefollowingequation: Thisequationcanbealteredsomewhattobeevenmoreuseful.Theefficiencyofrotorscanbecomparedforanygiventask.Eachrotorisgivenavaluefork(theclearingconstant,orclearingfactor),whichisanestimateofthetime(inhours)requiredtopelletaparticleofknownsedimentationcoefficientatthemaximumspeedoftherotor.4Thelowerthevalueofk,theshorterthetimerequiredtosedimentagivenparticle.Therelationshipisgivenby Itispossibletocomputek,butitiseasiesttousethemanufacturersvalues,whichcanbelookedupintablessuchasthoseinTableF.1. Usingtabulatedkvalues,however,thecomparisonofrotorsiseveneasier,astheequationabovebecomes Bysubstitutingthisequation,youcanalsodeterminethetimeofcentrifugationforanychangeinrotorspeed.5 ForaSorvallSS-34,forexample,at20,000RPM(k=402),thetimerequiredtosedimentaparticleof100S,inwaterat20°Ciscalculatedby: TableF.2presentsthekvaluesfortheSorvallSS-34rotorandfortheSorvallGSArotor.Sincethesetworotorswillbeusedthroughouttheremainderofthecourse,referenceshouldbemadetothistableinanyfuturecentrifugationwork,whenanotherrotorissubstituted. RotorFailure: Allrotorsaresubjecttostressandwithtimewillundergometalfatigue.Thisisagiven,andconsequently,adetailedhistoryoftherotoruseshouldbekept.Thisisusuallynotdonewithclinicalcentrifuges,butisanabsoluteforanultracentrifugerotor. EVERYUSEOFANULTRACENTRIFUGEROTORMUSTBERECORDEDINTHECENTRIFUGELOG.ABSOSLUTELYNOEXCEPTIONS! Afteraperiodofuse,eachrotorwillinturnbederated,thatisitsmaximumRPMwillbelowered.TheBeckmanrotorsmaycontainopticalspeedcontrolringsattheirbase-besuretheyarepresent,andcleanbeforeuse.Thesedeviceswillstrobiscopicallymonitorthemaximumspeedthatarotorcanbeusedat.Theyarereplacedastherotorisderated. Byfarthemostcommoncauseofrotorfailureiscorrosionstress.Salts,highlyalkalinedetergentsandofcoursecorrosiveacidsandalkali"swillcausedecompositionofthecoatingsonaluminumrotors,whichinturnwillconcentratestressandeventuallyresultincracksandtotalrotorfailure.Titaniumrotorsaremorecorrosionresistant,butmoreexpensive.Ultracentrifugerotorsareexpensive(inexcessof$6,000eachonaverage)andcanbepotentiallyhazardous.Attheforcesgeneratedinanultracentrifuge,arotorfailureistheequivalentofasmallbomb. Note:Ifproperlybalancedandused,therotorshouldacceleratesmoothlyandwithaconstantchangeinthepitchofthemotorsound.Anyvibrations,orunusualsoundsshouldcausethecessationofoperationIMMEDIATELYbytheoperator.NEVERleavethecentrifugeuntilyouarecertainthatithasreacheditsoperatingspeedandisfunctioningproperly.Allrotorsgothroughaminorvibrationphasewhentheyfirststart.Therewillbeaminorflutterwhentherotorreachesthisvibrationpoint-donotconfusethiswithaseriousvibrationcausedbyimbalance.Ifindoubt,haltthecentrifugeandgetassistance.ROTORS
FigureF.1.CrosssectionofSorvallSS-34rotor.ROTORTUBES
Inusingeitherafixedangleorswingingbucketrotor,itisnecessarytocontainthesampleinsometypeofholder.Continuousandzonalrotorsaredesignedtobeusedwithoutexternaltubes.RELATIVECENTRIFUGALFORCE
FigureF.2.NomogramforR.C.F.RCF=1.12r(RPM/1000)
wherer=radiusinmillimetersRPM=revolutionsperminuteT=((D-L)/(D+L))*(N/(d
(g-p)S
))whereT=timeinminutesD=radialdistanceincmforr
L=radialdistancetomeniscusN=viscosityofthefluidmediumg=densityofthefluidmediump=densityoftheparticletosedimentd=diameteroftheparticleincms.S=rotationalvelocityinRPMROTORCOMPARISONS
Whenseparatingaparticle,itisconvenienttobeabletocompareonerotor"sdesigncharacteristicswiththoseofadifferingrotor.Aproceduremaybegivenascentrifugingfor20minutesat15,000RPMinaSorvallSS34rotor.TheproblemthenbecomesoneofhowdoyouequatethattoaBeckmanJA-20rotor?t
=(t
xRCF
)/RCF
wheret
=runtimeneededfortheBeckmanrotort
=runtimespecifiedintheprocedureRCF
=RCFofJA-20rotoratmaximumspeedRCF
=RCFspecifiedintheprocedure(calculatedifprocedureinRPM)t=k/s

¸
k=clearingfactorexpressedinhour-Svedbergss
¸
=SedimentationCoefficientinwaterat20°CexpressedinSvedbergsk=(253,300)[ln(r
/r
)]/(Rotorangularvelocity/1000)
t
=k
t
/k
wherethetandkvaluesarethetimeandkfactorforeachrotor.t=402/100or4.02hours(4hrs,1min)
ROTORSAFETY
Thereareanumberofsafetyprecautionsthatmustbeadheredtowhenusinganycentrifugeandrotor.THEREFORETHEFOLLOWINGRULESMUSTBEOBSERVED.