USF Libraries
USF Digital Collections

Advancement and optimization of an electrospray injection based in-vacuum patterning system for macromolecular materials

MISSING IMAGE

Material Information

Title:
Advancement and optimization of an electrospray injection based in-vacuum patterning system for macromolecular materials
Physical Description:
Book
Language:
English
Creator:
Stark, Andreas
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Electrospray ionization
Ion focussing
Ion funnel
Cytochrome C
Faraday cup
Dissertations, Academic -- Electrical Engineering -- Masters -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: Electrospray ionization is a technique widely used in mass spectrometry. Almost every material, specifically large molecules like proteins or polymers can be ionized directly out of solution. During the ionization process molecules are not fragmented. In this work a prototype apparatus for creating three-dimensional patterns in a ultra high vacuum environment using an electrospray ion source was optimized for higher ion currents hence deposition rate by improving the core component of the apparatus, an electrodynamic ion funnel. The major improvements are a redesigned heated vacuum inlet, modified gas flow inside the ion funnel because of sealing the ion funnel against perpendicular gas flow and a better measurement setup for the transmitted current. The transmission of the ion funnel was improved from 25% to 82% resulting in ion currents of up to 7nA (500pA before advancements) focused through the ion funnel. At this rate an area of 1 cm² can be coated with a molecular monolayer of Cytochrome C in 64 minutes.
Thesis:
Thesis (M.S.E.E.)--University of South Florida, 2008.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Andreas Stark.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 67 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002001013
oclc - 319421649
usfldc doi - E14-SFE0002534
usfldc handle - e14.2534
System ID:
SFS0026851:00001


This item is only available as the following downloads:


Full Text

PAGE 1

AdvancementandOptimizationofanElectrospray InjectionBasedIn-VacuumPatterningSystemfor MacromolecularMaterials by AndreasStark Athesissubmittedinpartialfulllment oftherequirementsforthedegreeof MasterofScienceinElectricalEngineering DepartmentofElectricalEngineering CollegeofEngineering UniversityofSouthFlorida MajorProfessor:RudySchlaf,Ph.D. MatthiasBatzill,Ph.D. JingWang,Ph.D. DateofApproval: May20,2008 Keywords:electrosprayionization,ionfocussing,ionfunnel,cytochrome c,faradaycup Copyright2008,AndreasStark

PAGE 2

ACKNOWLEDGEMENTS IwouldliketothankDr.RudySchlaffortheopportunityofwriting thisthesis,mycommitteemembersDr.MatthiasBatzillandDr.Jing Wangfortheirsupportandtime,Dr.MarkAnthonyandDr.Martin Beerbomfortheirsupport,FlorianKaiserandMatthewHollandfortheir help,theUSFmachineshopforitsexcellentwork,allgroupmembers andEEDepartmentstafortheirhelpandsupportandChuckfromthe Coeeshopfortherstwrappingfoil.IalsowouldliketothankRomina Wiedmannforhergreatsupportandlendingmeherearsformyproblems andmyfamilywhomadeitallpossible.

PAGE 3

TABLEOFCONTENTS LISTOFFIGURESiv ABSTRACTvii 1INTRODUCTION1 1.1Objectiveofthisthesis2 1.2Electrosprayionization4 1.3MoleculeWriterprototype5 1.4Experimentalsteps7 2ELECTROSPRAYIONIZATION8 2.1Otherionizationtechniques8 2.2Electrospray9 3MOLECULEWRITERPROTOTYPEAPPARATUS15 3.1Overview15 3.2Electrosprayunit17 3.3Vacuuminterface18 3.4Firstvacuumchamberandinsidefunnel18 i

PAGE 4

3.5SecondvacuumchamberandFaradaycup21 3.6Currentrequirementsfortheelectrospraydeposition22 4RESULTSANDDISCUSSIONFIRSTPROTOTYPEAPPARATUS24 4.1Resultsfrompreviouswork24 4.2Characterizingtheexistingprototypesetup:dropletevaporationproblems26 4.2.1Pressurevariation26 4.2.2Distancevariation28 4.2.3Temperaturevariation-Eectsofpositiveandnegativeionmodeandthedierencebetween20miland 30milcapillary31 4.3Discussion34 5REDESIGNOFTHEPROTOTYPEAPPARATUS36 6RESULTSANDDISCUSSIONREDESIGNEDPROTOTYPE APPARATUS41 6.1Commencementoftheredesignedprototype41 6.2Improvingionfunneltransmissionthroughmodiedairow46 6.2.1Gas-owsimulations46 6.2.2Firstexperimentwiththewrappedionfunnel47 6.2.3EectoftheDCvoltagegradientinsidethefunnel49 6.2.4Eectofthecapillaryposition50 ii

PAGE 5

6.2.5Summaryoftherstexperiments51 6.3ExcludingmeasurementerrorsattheFaradaycup52 6.3.1Eectofthecapillarylength53 6.3.2Eectofthecapillarydiameter54 6.3.3Dierencesbetweensolutemolecules55 6.3.4Investigationoftheeectsofwrappingthefunneland extendingtheFaradaycup57 7CONCLUSION60 LISTOFREFERENCES62 iii

PAGE 6

LISTOFFIGURES Figure1.1:Electrosprayinjectionprocess,obtainedfrom[8].3 Figure1.2:Productionof3DstructureswiththeMoleculeWriter, obtainedfrom[9].5 Figure2.1:Electrospray:AhighvoltageisappliedtotheES needleontheleftwhilepumpingasolutionof macromoleculesthroughit.10 Figure2.2:Cone-jetmodesofelectrosprayionization,obtained from[2].13 Figure3.1:OriginaldesignofESIpatterningprototypesystem,obtainedfromAnthonyCascio.16 Figure3.2:Vacuuminterface:Acapillaryismountedonthe angeofthevacuumchamberusingaSwagelok tubetting.19 Figure3.3:Schematicoftheionfunnel.19 Figure3.4:Oldandredesignedionfunnel.20 Figure4.1:Pressuredependenciesatdierenttemperatures.27 Figure4.2:Distancedependenciesatdierenttemperatures.30 iv

PAGE 7

Figure4.3:Negativevspositiveionmodewitha20milcapillary.32 Figure4.4:Comparisonofa20miland30milcapillaryin negativeionmode.33 Figure5.1:Newprototypesystem,obtainedfromFlorianKaiser: Amorecompactandeasiertoreassembledesign waschosen.36 Figure5.2:Cross-sectionofthenewionfunnel,obtainedfrom FlorianKaiser.38 Figure5.3:Newheaterdesign,obtainedfromFlorianKaiser.39 Figure6.1:Measurementerrorfunnelresistors:Theparallelresistanceoftheanalogoscilloscopeisinthe magnitudeofthefunnelresistorsandinducesa largemeasurementerror.42 Figure6.2:Firsttestrunswithnewprototype:Brilliantblue dyemoleculessoluteinmethanolweresprayed.44 Figure6.3:Simulatedairowinsidethevacuumchamber: Highverticalvelocityleftmovesionstothe pumpexit.47 Figure6.4:Funnelwrappedinplasticfoil.48 Figure6.5:Funnelwrappedinplasticfoiltoimproveairow: Theresultsarecomparabletootherresearchgroups.49 v

PAGE 8

Figure6.6:Comparisonofacapillaryushwiththeionfunnelentranceaandsticking10mmintothefunnelb.50 Figure6.7:ComparisonofcurrentintotheFaradaycupfor 8.5cmand10cmcapillary.54 Figure6.8:Comparisonbetween8.5cmlongcapillaries.55 Figure6.9:ComparisonbetweenGly-Glnandcytochromec.56 Figure6.10:Eectsofwrappingthefunnelandextendingthe Faradaycupoverfunnelfrequency.58 Figure6.11:Eectsofwrappingthefunnelandextendingthe Faradaycupoverfunnelamplitude.59 vi

PAGE 9

AdvancementandOptimizationofanElectrospray InjectionBasedIn-VacuumPatterningSystemfor MacromolecularMaterials AndreasStark ABSTRACT Electrosprayionizationisatechniquewidelyusedinmassspectrometry.Almosteverymaterial,specicallylargemoleculeslikeproteinsor polymerscanbeionizeddirectlyoutofsolution.Duringtheionization processmoleculesarenotfragmented.Inthisworkaprototypeapparatusforcreatingthree-dimensionalpatternsinaultrahighvacuumenvironmentusinganelectrosprayionsourcewasoptimizedforhigherion currentshencedepositionratebyimprovingthecorecomponentofthe apparatus,anelectrodynamicionfunnel.Themajorimprovementsarea redesignedheatedvacuuminlet,modiedgasowinsidetheionfunnel becauseofsealingtheionfunnelagainstperpendiculargasowanda bettermeasurementsetupforthetransmittedcurrent.Thetransmission oftheionfunnelwasimprovedfrom25%to82%resultinginioncurrentsofupto7nApAbeforeadvancementsfocusedthroughtheion funnel.Atthisrateanareaof1cm 2 canbecoatedwithamolecular monolayerofCytochromeCin64minutes. vii

PAGE 10

1INTRODUCTION Manyapplicationsforacontrolledassemblyofmacro-molecularmaterialstocomplex3Dstructuresexist.Theseareforexampletheproductionofelectronicdevicesandbiomedicalsensors.Butthemainproblem istocontroltheexactandthree-dimensionaldepositionofthosemacromoleculeswithintheaccuracyofnanometersinahighvacuumenvironment.Thevacuumisnecessaryasfewimpurityatomscanaectthe functionalityofnanometer-sizedevices.Alsonotexactdepositiondueto limitedaccuracycanaectthedesiredfunctionalityofadevice. TheapproachofDr.RudySchlaf'sresearchgrouptoproduce3Dstructuresofmacro-moleculesistogenerateabeamofionizedmacro-molecules ionbeambyanelectrosprayionsourceseesections1.2and2.2ina vacuumchamberanduseelectricalandmagneticeldstocontrolthis beamsimilartotheelectronbeaminaTVset.Aprototypeofa3D patterningdeviceformacro-molecularmaterialsMoleculeWriter,[9] iscurrentlybeingdeveloped.Uptonowthereexistsnomethodofgenerationanionbeamwithoutdestroyingmacro-molecularmaterialswithin avacuumenvironment.Hencetherearethreemainproblemstosolve: 1

PAGE 11

Generateabeamofionizedmacro-moleculesandinjectitintoa vacuumchamber Minimizelossesbetweenionsourceandtarget Controltheionbeamwithhighaccuracy 1.1Objectiveofthisthesis Intheworkdoneduringthisthesisthemainfocuswasonsolving thersttwoproblems.Togenerate3Dstructuresconsistingofmacromoleculeseconomicallythemaincostfactoristheproductiontimeper device.Themainlimitingfactoristhedepositionrate.Thehigherthe depositionrate,thefastermacro-moleculescanbedepositedandmore 3Dstructurescanbeproducedpertime.Onlywithahighioncurrent ionspertimeahighdepositionratecanbeachieved.Sotherearetwo processestobeoptimized: Outputfromtheionsource Lossbetweenionsourceanddepositiontarget Toachievehighperformance,highoutputandlowlossisneeded.A highoutputfromtheionsourceleadstoahighiondensitywithinthe beam.Butthisiscontrarytoahighperformancebecausealltheions havethesamepolarity,theyrepeleachotherandsothebeamisspread outandlossincreases. 2

PAGE 12

Duringtheworkforthisthesistheexistingprototypesystemwascharacterizedtogainbetterunderstandingoftheelectrosprayprocessandthe losssources.Thereafterthemainfocuswasonrunningaredesignedprototypesystem,specicallyaredesignedionfunneltoimprovetheion currentintothesecondvacuumchamberofthesystemandreacheconomicaldepositiontimes. astrongelectriceldextractsdropletsfromTaylor conecontainingsoluteions M )]TJ/F55 7.9701 Tf 6.254 -2.813 Td [(;mostcounterions L remainincapillary bsolventevaporates,droplets shrink,charge densityincreases csoluteionsare ejectedfromthe droplets;solventis removedinpumping stages dsolutemolecules enterthepreparation chamberandformthin lm Figure1.1:Electrosprayinjectionprocess,obtainedfrom[8]. 3

PAGE 13

1.2Electrosprayionization Usuallymacro-molecules,specicallybiomoleculeslikepeptides,proteinsorDNAdeoxyribonucleicacidareonlyavailableinsolution.A techniqueknownfrommassspectrometrytocreateabeamofionized macro-molecules,iselectrosprayESionization.Electrosprayionization isbasedontheworkofDole1968[5]andwasestablishedbyFenn1984 [35]NobelPrizeinChemistry2002. Figure1.1showstheelectrosprayionizationandinjectionprocess:The desiredmacro-moleculesarepumpedthroughtheelectrospraycapillary. SurfacetensionofthesolutionandtheappliedelectriceldbetweencapillaryandvacuuminletleadtotheformationoftheTaylorconedetailedexplanationinsection2.2,wherepositiveandnegativeionsare separated.AtthetipoftheTaylorconemainlyuniformdropletsconsistingofunchargedsolventmoleculesandionizedmacro-moleculesare emittedduetoincreasinginstabilitywithincreasingdistancefromthe capillaryend. Duringtheightunchargedsolventevaporatesanddropletsgetsmaller andchargedensityonthesurfaceofthedropletincreases.Aftershrinking toacertainradiusRayleighlimit[23]thedropletgetsinstablebecause ofthehighchargedensityanddisintegrateintosmallerdropletsasthe Coulombrepulsionforcesexceedthesurfacetensionthatholdthedroplets together. 4

PAGE 14

Thesmalldropletsandionsarethenguidedbytheelectriceldand airowintothevacuumchamber. 1.3MoleculeWriterprototype Prof.Dr.Schlaf'sgrouphasdesignedaprototypeapparatustorealizetheobjectiveofpatterning3Dstructuresofmacro-molecules.The MoleculeWriter[8]prototypeisdescribedindetailinchapter3. Figure1.2:Productionof3DstructureswiththeMoleculeWriter,obtainedfrom [9].Solutionscontainingmacro-moleculesareionizedthroughtheelectrosprayES device.Theresultingbeamofionizedmacro-moleculesisfocusedbyionopticsand masslteredtoextractthedesiredmolecules.Beforehittingasubstratesurface, thebeamcanbedeectedsimilartotheelectronbeaminaTVsettoproduce3D patternslikenano-compositelayersandsmallcoatedareasonsensorarrays. 5

PAGE 15

Figure1.2showstheconceptoftheMoleculeWriter.Adistribution unitselectsasolutioncontainingthedesiredmacro-molecules.Thissolutionispumpedintothe electrosprayunit ,whereabeamofionized, gas-phasemacro-moleculesiscreatedasdescribedinsection1.2.Until thispointtheprocesstakesplaceatatmosphericpressure.Nowthebeam passesthe vacuuminterface thatcanbeacapillaryororice. Dierential pumpingstages removeremainingsolventfromthemacro-moleculesand reducethepressurestepwisefromatmosphericpressureto10 )]TJ/F8 9.9626 Tf 7.749 0 Td [(3 till10 )]TJ/F8 9.9626 Tf 7.749 0 Td [(9 mBar 1 1 million till 1 1 trillion ofatmosphericpressure. Thebeamisfocusedbyan ionfunnel seesection3.4tominimize lossesduetothenaturalspreadingoutofthebeam.Ourionfunnelconsistsofastackofmetalplateslenseswithconcentricholesofdecreasing diameter,separatedbyspacers.AcombinationofDCconstantvoltage andRFhighfrequencysignalisappliedtothelenses.TheelectricRF eldcreatedbythissignalfocusestheionbeam. Afterbeingfocusedtheionbeamismass-lteredandcanbedeected bythebeamdeectortocontrolthedepositionpositiononthesubstrate surface.Thisallowstocreate3Dpatternsofmacro-moleculessuchas bio-assays 1 andmoleculesintegratedintonano-scaledevices. 1 Devicetodeterminethestrengthorbiologicalactivityofasubstance,suchasadrugorhormone, bycomparingitseectswiththoseofastandardpreparationonacultureoflivingcellsoratest organism.Wikipedia 6

PAGE 16

1.4Experimentalsteps Aseriesofexperimentalstepsandsimulationswasmadetoanalyze andminimizelossesatthevacuuminterfaceandtheionfunnel.Allthe experimentalstepsarefullydescribedinchapters4to6andtheresults arediscussed.Thesehavebeen: Characterizingtheexistingprototypeandidentifyingthemainloss sources Redesigningtheprototypesystemtogetherwithallothergroup members 2 Commissioningthenewprototypedesign Optimizingtheelectrosprayandionfunnelparameterstoreachand exceedresultsobtainedfromotherresearchwithsimilardesigns 2 Prof.Dr.RudySchlaf,Dr.MarkAnthony,Dr.MartinBeerbom,FlorianKaiser,Matthew Holland 7

PAGE 17

2ELECTROSPRAYIONIZATION 2.1Otherionizationtechniques Thereisanalphabetofionizationtechniques,butnotallofthemare usefulforthepurposetocreate3Dpatternsoflargemoleculeswithout destroyingthem.Thedesiredionizationtechniqueshouldalsoprovidea highioncurrent,sothattheproductionofpatternscanbeaccomplished inashortertime.Itshouldalsobeabletoionizeabroadspectrum ofinorganicandorganicmaterials,specicallybiomoleculeswhichare usuallyavailableonlyinsolution. Mostoftheionizationtechniquesareknownfrommassspectrometry wherethemolecularcompositionofasampleormacromoleculesisanalyzed.TheseareelectronionizationEI,chemicalionizationCI,atmosphericpressurechemicalionizationAPCI,fastatombombardment FAB,elddesorptionFD,matrix-assistedlaserdesorption/ionization MALDI,atmosphericpressurephotoionizationAPPI,electrosprayionizationESI,desorptionelectrosprayionizationDESI,glowdischarge GD,inductivelycoupledplasmaICP,microwaveinducedplasmaMIP, 8

PAGE 18

thermosprayionizationTS,directanalysisinrealtimeDARTand laserdiodethermaldesorptionLDTD. Mostofthemarenotusefulforourpurposebecausetheyrequire eithergasphaseorsolidstatemoleculesorfragmentthemoleculesduring theionizationprocess.Electrosprayhasemergedasthebestionization techniqueforbiomoleculesinasolution.Ithasahigheciencyand almostnofragmentationoccurs. 2.2Electrospray JohnB.Fennwasawardedthe2002Nobelpriceinchemistryforhis workontheelectrosprayionizationtechnique.Hefoundoutduringhis workthatwaspublishedin1984[35]thatusingelectrosprayionization thereisnoevidenceofanyfragmentationordecompositionofeither solventorsolutespecies.Soluteinthiscasemeansthemacromolecule insolution=solventthatisionizedduringtheelectrosprayprocess.In [35]healsostatesthatthesoluteions[...]alwaysappearsinglyinany clusters,sothemacromoleculesaretransformedintosingleions.But Fennalsoassumedpossiblecondensationduringthefree-jetexpansion inthevacuumchamber,aproblemthatalsooccurredduringthework forthisthesis. Theelectrosprayionizationprocessstartswithmacromoleculesbeeing ionizedinasolution.Thenthesolutionispumpedthroughametal 9

PAGE 19

Figure2.1:Electrospray:AhighvoltageisappliedtotheESneedleontheleftwhile pumpingasolutionofmacromoleculesthroughit.AtthetipoftheTaylorconea beamofionizeddropletsisemitted.Coulombrepulsionforceandtheelectriceld widenupthebeamtothesocalledplume. capillaryESneedle.Dependingonthesolventandthemacromolecule eitherthepositiveionmode,whereapositivevoltageisappliedtothe ESneedle,orthenegativeionmode,whereanegativevoltageisapplied, arepossible.Inthiscaseonlythenegativeionmodeisexplained,the positiveionmodeissimilar,justwithoppositechargesandvoltages. WithahighnegativevoltageappliedtotheESneedlethecations positivelychargedinthesolutionmigratetowardsthemetalwallofthe needleandtheanionsnegativelychargedmigrateawayfromittowards thecounterelectrode,inthiscaseametalcapillaryororiceatthe vacuumchamberentrance.Theelectriceldstrengthattheneedletip withaplanarcounter-electrodecanbeapproximatedas[12,17,21]: E needle = 2 V needle r needle ln d=r needle .1 10

PAGE 20

where V needle istheappliedvoltage, r needle istheouterradiusoftheneedle and d isthedistancebetweenneedletipandcounter-electrode.Atypical valuefor E needle forasuccessfulelectrosprayis 10 6 )]TJ/F21 14.3462 Tf 12.51 0 Td [(10 7 V m [12,21].Ifthe electriceldisstrongenough,theforceontheanionsfromtheelectric eldiscounterbalancedbythesurfacetensionofthesolutionforminga coneofliquid,calledTaylorcone[31]attheneedleend. Usingamodied[30,33]formoftheHendricksequation[21]theelectrospraycurrent i es canbeapproximatedas: i es = H f n S E c .2 where H isaconstantdependingonthesurfacetensionandthedielectric constantofthesolvent, f istheow-rate, S isthespecicconductivity and E c istheelectriceldfromequation2.1.Experimentalwork[29, 30]showedthattheexponents n and canbeapproximatedwith thosepredictedintheHendricksequation.FernandezdelaMoraand Loscertales 1 determinedintheircaseacurrentproportionaltothesquare rootofelectricalconductivity n =1 = 2 andow-rate =1 = 2 .Asa generalstatementitcanbesaidthat i es increasesiftheconductivityof thesolutionincreasesoriftheow-rateisraised[32,33]. AtthetipoftheTaylorconeajetofliquidemerges.Ifthecharge densityinsidethisjetistoohigh,itbreaksupintodroplets.Thisbreak1 obtainedfrom[3] 11

PAGE 21

upincreasesthesurfaceareaperchargedensityandsoastablecondition isreestablished.Thedropletsizeincreaseswithdecreasingconductivity andinproportionto flow )]TJ/F22 14.3462 Tf 14.835 0 Td [(rate 2 = 3 [6].Toproducesmalldropletsa smallow-rateandahighconductivityisneeded. Theresultingnesprayofdropletsofthesamepolarityisforcedtowardsthecounterelectrodebytheelectriceldsandinourcaseair owintothevacuumchamber.Duringtheightsolventevaporates, butthedropletchargeremainsconstant.Whentheelectrostaticrepulsionforcebetweenthechargesbecomesequaltotheforcefromthesurfacetensionwhichisholdenthedroplettogether,dropletssplitupinto smallerdroplets.ThisprocessiscalledCoulombssion.Atthispoint theRayleighstabilitylimitisreachedthatcanbeexpressedwiththe Rayleighequation[12,23]: q R =8 e 0 R 3 1 = 2 .3 where q R istheexcesschargeonthedropletofradius R isthesurface tensionand e 0 isthepermittivityofvacuum.ThroughtheCoulombssionprocesschargedensityandsotherepulsionforcebetweenthecharges decreasesundertheRayleighlimitandtheresultingdropletsarestable again. TheCoulombssionprocessrepeatsseveraltimesuntilonlyonesingle macromoleculeisexistentineachdroplet.Asthenalstepinthetran12

PAGE 22

sitionfromdropletstoagas-phaseionstwomodelsexist:Thecharge residuemodelCRMsuggestthatelectrospraydropletsrepeatseveral evaporationanddisintegrationcyclesuntilonlysmalldropletscontainingonlyoneionizedmacro-moleculeexist.Throughevaporationofthe remainingsolventmoleculesagas-phaseionremains.IntheionemissionmodelIEMitisassumedthatsingleionsareemittedfromlarger droplets.Theseionsreducethechargedensityintheemitterdropletand reestablishstabilityintheemittingdroplet.ResultswhereFernandezde laMorafoundusuallymultiplychargedions[7]indicatetheCRMfor largemoleculesover3300DaDa=1u MassofasingleHydrogen atom.WangandCole[4,34]foundoutthatahighersolventpolarity leadstomoremultiplychargedions.AndKebarleetal.[13]foundoriginallynegativeionschargedpositivelyandviceversa.Thisindicatesthat theanalytemoleculesarechargedbyexcesschargespresentinthenal droplets,nomatteroftheirinitialcharge. Figure2.2:Cone-jetmodesofelectrosprayionization,obtainedfrom[2].Withthe rightvoltageappliedaperfectTaylor-coneformsandanesprayofdropletsisproduceda.Increasingthevoltageleadstoao-centeredbconeandthentomulti-jet modesc,d. 13

PAGE 23

M.CoupeauandB.Prunet-Foch[2]didanextensiveseriesofexperimentstodeterminetheinuencefactorsforelectrosprayionization. Theyfoundoutthatastableelectrospraywiththeformationofaspray ofdropletsofnearlyidenticaldiametersdependsontheconductivityof thesolution,appliedvoltage,ow-rate,capillarygeometryandcapillary wettability.Alltheseparameterscanonlybevariedwithinanarrow range,otherwisetheperfectTaylorconegure2.2aisnotpossible andnotusefulmulti-jetmodesgure2.2c,dorlargeunevendroplets areresulting. Toconcludethischapter,itistosaythatelectrosprayionizationis comparedtootherionizationmethods`softer'[3].Evencomplexeswith weaknon-covalentbindingsthatexistinsolutioncanbestudiedingasphase[18,19,22].Alltheseabovelistedpropertiesmakeelectrospray ionizationaveryversatiletechniquetoionizeallkindsofmolecules,but specicallylargecomplexesoforganicmolecules. 14

PAGE 24

3MOLECULEWRITERPROTOTYPEAPPARATUS 3.1Overview ThemaingoalofthedescribedelectrosprayionizationESIprototype apparatusistodeposit3Dpatternsoflargemoleculeslikeproteinsona surface.Beingableofthat,complex3Dstructuresinananometerrange canbeachievedopeningthepathfortotallynewfutureapplications.The approachistogenerateabeamofchargedmoleculesionbeamthatcan becontrolledbyelectrostaticforcessimilartotheelectronbeamina TV.Thetargetofthebeamwherethemoleculesaredepositedhasto beinahighvacuumtoavoidimpuritiesthatwilldisturbthedeposited structure'sproperties.Theionbeamiscreatedatatmosphericpressure, becauseelectrospraydoesnotworkinvacuum,andpassesseveraldierentialpumpingstagesvacuumchamberswithdecreasingpressuresand ltering,focusinganddeectionmechanismsuntilitreachesthetarget chamberwherethesubstrateislocatedataultrahighvacuum. Figure3.1showstheoriginaldesignoftheESIprototypeapparatus. Inthe injectionchamber theionbeamisgeneratedbytheelectrospray 15

PAGE 25

Figure3.1:OriginaldesignofESIpatterningprototypesystem,obtainedfromAnthonyCascio. processatatmosphericpressureasdescribedinchapter2.Nitrogengasis usedtoacceleratetheformationofionsfromchargeddroplets.Theions areguidedthroughacapillaryintotherstvacuumchamberincludingan ionfunnel .Withtherapiddropinpressureandduetorepulsionofthe ionsofthesamepolarityfromeachothertheionbeamspreadsout.The ionfunnelfocusesthebeamandpreventsalossofionsatthechamber walls. Afterbeingfocusedinthefunneltheionsenterthesecondvacuum stage.Thisstageiscalled collisioncell andconsistsofaRFradio frequencyquadrupole.Onlyionswithacertainmass-to-chargem/z ratiocanpassthroughthequadrupoleandallotherionscollidewiththe 16

PAGE 26

wallsofthequadrupoleandaresolteredout.Theremainingionswith thedesiredmass-to-chargeratiopassthe ionextractionchamber and anelectrostaticanalyzerchamber,whereonlyionswithadesiredspeed canpassthrough.Fromtheretheionsgetintothenalstagewhere thecomponentsfor beamscanning deectionofthebeam, blanking shuttingthebeamonandoand nalfocusing arebeforetheyhit the targetsubstrate 3.2Electrosprayunit Themostimportantpartoftheelectrosprayunitistheelectrospray needle.Dimensionsoftheusedneedleareanouterdiameterof0.47mm attheneedletipand0.72mmatthestraightpartoftheneedle.The innerdiameterincreasesfrom0.13mmattheneedletipto0.15mm.For exactaligningoftheneedle,itismountedonaxyz-stageadjustableby micrometers.AStanfordResearchSystemsPS350highvoltagepower supplythatcansupplyavoltagefrom-5kVto5kVbutonlycurrentsup to5mAisconnectedtotheneedle. DuringthedevelopmentprocessasolutionofacetonitrileanddeionizedwatercontainingCoomassieBrilliantBluedyemoleculeswasused. Thedyemoleculesarerelativelylargewithmolecularmassof 833Da andresultinbluedepositionpatterns,thatmakeiteasytodetectloss. Electricalconductivityofthesolutioncanbecontrolledbytheamountof deionizedwateraddedandwasduringtheexperimentsatabout40 Sfor 17

PAGE 27

optimumelectrosprayperformance.Flow-rateofthesolutionwascontrolledusingaprecisionsyringepumpfromHarvardApparatus.During theexperimentstheow-ratewastypically0.3mL/min. Todetectastableelectrosprayavideocamerawithamagnication of10wasused.Additionallyagreenlaserbeamwasdirectedtothe areainfrontoftheTaylorconetoverifythatanesprayandnotlarge non-uniformdropletsormulti-jetswereproduced. 3.3Vacuuminterface Originallytheinterfacebetweentheelectrosprayunitandtherst vacuumchamberwasimplementedusingacapillary.Thetypicalcapillary usedfortheexperimentswas20cmlongwithaninnerdiameterof0.5mm. ItwasmountedonthevacuumchamberopeningusingaSwageloktube tting.Thisconstructionallowseasyandfastexchangeofthecapillary forcleaningpurposes.Insidethevacuumchamberthecapillaryishold inplacebytheheaterconstruction.Aplexiglasschamberaroundthe electrosprayneedleandthecapillarypreventsairturbulenceandcanbe lledwithheatednitrogengastoacceleratesolventevaporation. 3.4Firstvacuumchamberandinsidefunnel Intherstvacuumchamber,alsocalledrstdierentialpumping stage,atypicalpressureof 1Torrispresentduetothelimitedpowerof thevacuumpumpandconstantairowintothechamber.Mainpartof 18

PAGE 28

Figure3.2:Vacuuminterface:Acapillaryismountedontheangeofthevacuum chamberusingaSwageloktubetting. therstvacuumchamberistheinsideionfunnel.Theionbeamenteringthischamberthroughacapillaryororiceiswidenedupduetothe rapidexpansioninthechamberandCoulombrepulsionforces.Tofocus thediuseionsbacktoacollimatedionbeamtheinsidefunnelhasbeen designedbuildinguponpreviousworkofJulianetal.[11]. Figure3.3:Schematicoftheionfunnel.Ionsentertheionfunnelfromtheleftand getfocusedthroughtheDCextractionlensontherightside. AsinesignalintherangeofseveralkHztoMHzisappliedtothe evenandoddlensesRF1andRF2.ThesignalonRF2is180degree phase-shiftedcomparedtoRF1.Thisfactresultsinastrongelectriceld intheouterregionsofthefunneldrivingtheionstowardsthemiddle 19

PAGE 29

andanearlyeldfreeregioninthemiddleofthefunnel.Duetothe cone-shapetheeldfreeregiongetssmallerindiametertowardstheend ofthefunnelfocusingthediuseionsintoonespotatthefunneloutlet. AstaticelectriceldissuperimposedtheRFeld.ThereforeaDC voltageisconnectedto V 1 and V 2 .Throughtheresistorsbetweenthe funnellensesaeldgradientpushingtheionstowardsthefunneloutlet isgenerated.ThecapacitiesbetweenRFvoltageconnectorandlenses separateRFandDCvoltagefromeachother.Lastlensoftheinside funnelistheDCextractionlens.ADCvoltageappliedtothislenscan controltheamountofpassingionsandworkasashutter. aOldionfunnel bRedesignedionfunnel Figure3.4:Oldandredesignedionfunnel.Theionsenterbothfunnelsontheleft endexitthroughtherightside. Concludingthissectionthemaintasksoftheinsidefunnelaregeneratingacollimatedionbeamoutofthediuseionsinthevacuumchamber andlteringoutunchargedparticlesthatarenotusefulforthefurther 20

PAGE 30

process.Unchargeddropletsorsolventmoleculesarenotinuencedby theelectriceldandsotheyhitthelensesoftheionfunnelandarelteredout.ConnectingtheDCoutletstoapicoamperemeterallowsto measuretheioncurrenthittingthefunnel. 3.5SecondvacuumchamberandFaradaycup Asthepressurecouldbereducedfromatmosphericpressure 750 Torrto1Torrintherstdierentialpumpingstage,anadditionalturbo pumpinthesecondvacuumchamberisabletoreducethepressuredown to15mTorr.Thispressurereductioniscrucialforthedepositionofions inthisprototype:Thelowerthepressurethelongeristhemeanfree pathoftheionsandthelongeristhedistancetheionscanightwithout bumpingintoothermoleculeswheretheygetdeectedandtheirvelocity decreases.Ifthemeanfreepathistooshortandthereisnoadditional forceacceleratingtheionstheystopanddonotreachthedeposition substrate. Intheactualdevelopmentstatetheonlypartinthischamberisa Faradaycup:Ametalcylinderclosedattheendisconnectedtoapicoamperemetertomeasuretheioncurrentaftertheinsidefunnel.These measurementswereusedtocalculatetheeciencyoftheinsidefunnel andtooptimizethevoltagesoftheinsidefunnel. 21

PAGE 31

3.6Currentrequirementsfortheelectrospraydeposition Inordertodeterminethenecessaryioncurrenttomakedepositions inafairlyshorttime,considerthefollowingexample: Assumingtheareaof1cm 2 shallbecoatedwith1monolayerofcytochromec,howlongwouldittaketodothedepositionwithanion currentof1,2,4and8nA?Withthesimplifyingassumptionthattheirregularshapedproteincytochromecisaspherewithadiameterof3.1nm [20],thenumberofmoleculespercm 2 canbecalculated: molecules cm 2 = 1 cm 2 : 1 10 )]TJ/F8 9.9626 Tf 7.749 0 Td [(7 cm 2 =1 : 04 10 13 .1 Themeasuredelectricalcurrentcanbeconvertedintomolecularcurrent moleculespertimeusingthefollowingformula: Molecularcurrent = electricalcurrent chargestate elementarychargee .2 AccordingtoIavaroneetal.[10],whousedasolutionsimilartothe oneusedinthelaterexperiments,themostintensechargestateforcytochromecsprayedina47:50:3water:methanol:aceticacidsolutionin positiveionmodeis16 + withamaximumof19 + andaminimumcharge stateof11 + .Usingequations3.1and3.2anelectricalcurrentof1nA atachargestateof16 + correlatesto 0 : 39 10 9 molecules = secondnA: 0 : 78 10 9 ;4nA: 1 : 56 10 9 ;8nA: 3 : 13 10 9 .Thesemolecularcurrentsequal 22

PAGE 32

depositionstimesof7.4hnA,3.7hnA,1.85hnA,56minnA. Sothegoalistogetahighcurrentoflargemoleculesatalowcharge stateinordertomakefastandeconomicaldepositionsandpatterns. 23

PAGE 33

4RESULTSANDDISCUSSIONFIRSTPROTOTYPE APPARATUS 4.1Resultsfrompreviouswork Previouswork[28]disclosedverylargeimprovementpossibilitiesin theusedelectrospraysetup.Alossof99%oftheemittedelectrospray currentwascreditedfourmajorsources: Ionizationeciencyduringtheelectrosprayprocess-toolargedroplets Expandingplumebetweenelectrosprayneedleandvacuum Highlossinsidethevacuuminterfacecapillary Loweciencyoftheinsidefunnel Duringthisworkitwastriedtoimprovethelowperformanceofthe electrospraysystemwiththreemeasures:Heatedgasstreamingfrom capillarytoelectrosprayneedleandcapillaryheatingwereimplemented toimprovethedropletevaporationratethusresultinginsmallerdroplets orideallygas-phaseionsinsidetheionfunnelchamber.Theloweciency oftheinsideionfunnelcouldbeimprovedfrom10%to45%resulting inaFaradaycupcurrent I FC of930pA. 24

PAGE 34

SinceahighioncurrentintotheFaradaycupresultsinashortdepositiontimeitwastriedasanextsteptoincreasethecurrent I Funnel intothe ionfunnel.Replacingthe15cmlongcapillarybya1.5mmthickconeshapedoricethecurrentintotheionfunnel I Funnel couldbeincreased from2nAto100nAwhichis50%ofthetotalemittedelectrospray current.Despitethehighcurrentintensityitwasimpossibletofocusany oftheionfunnelcurrent I Funnel intotheFaradaycup,probablybecause ahighspacechargeeldinsidethefunnelandlargedropletsbecauseof themissingheating. Thethirdmeasurewastoimplementanelectrostaticionfunnelbetweenelectrosprayionsourceandinletorice.Itwasbuiltsimilarlytoa designofSafetal.[24]tofocustheexpandingelectrosprayplumeintothe inletcapillary.Theoutsidefunnelshouldenablelargerdistancesbetween electrospraysourceandoriceresultinginsmallerdropletsorgas-phase ionsatthevacuuminletorice-largeorfrozendropletsweresuspected tobethereasonforthelowionfunnelperformance[28].Itwaspossible tomaintainanioncurrent I Funnel of2.8nA,butthemaximumcurrent intotheFaradaycup I FC was300pAor10%. Theresultsgainedfromthispreviousworkindicateahighimportance ofamoderateheatinginsideavacuuminterfacecapillarythatresults insmalldropletsorevenfullyevaporatedgas-phasesoluteionsinside theionfunnelchamber.Increasingthedistancebetweenelectrospray 25

PAGE 35

sourceandvacuuminletusinganelectrostaticionfunnelemergedtobe impracticalandinecientcomparedtoaheatedcapillaryincombination withaheatedgasstream. 4.2Characterizingtheexistingprototypesetup:droplet evaporationproblems Buildingontheresultsofthepreviouswork,anextensivesetofexperimentswasdonetocharacterizetheelectrospraydepositionprototype apparatus.Theseexperimentsshouldfocusspecicallyonthedroplet evaporationproblemwhichissuspectedtobethemaincauseforthelow performanceofthevacuuminterfacecapillaryandtheelectrodynamic ionfunnelinsidetherstvacuumchamber.Threemainparameterswere variedduringtheexperiments: Pressureinsidethevacuumchamber Distancebetweenelectrosprayneedleandcapillary Temperaturearoundtheinletcapillary 4.2.1Pressurevariation Todeterminethebestpressuretooperatetheionfunnelthreeexperimentswereperformedduringwhichthepressureintherstvacuum chamberwasvariedusingthegate-valvebetweenchamberconnectionand vacuumpump.Thetemperaturewasvariedusingthetwohalogenlights. Brilliantbluedyewassprayedinasolutionwith9:1methanol:water,con26

PAGE 36

ductivity35 S,usingthestandardmetalneedleagainsta30mil 1 inner diameter,15cmlongstainlesssteelcapillaryataowrateof1 L/min. Theresultsoftheseexperimentsaredisplayedingure4.1.Thetotal aTotalcurrentintoionfunnel bCurrentfocusedintoFaradaycup cIonfunneltransmission Figure4.1:Pressuredependenciesatdierenttemperatures.Whereasthetotalcurrentintothevacuumchamberdoesnotshowasignicantpressuredependency,the transmittedcurrentthroughtheionfunnelisverydependentonthepressureand showsanarrowpeakaround1Torr. currentintotheionfunnelI F remainsnearlyconstantoverthepressure rangefrom0.71to6Torr.Heatingto80 atadistanceof4mmbe1 1mil=1/1000inch=0.0254mm 27

PAGE 37

tweenelectrosprayneedleandcapillaryincreasesthecurrentbyafactor oftwocomparedtoroomtemperature.Increasingthedistanceto12mm resultsinalowertotalcurrentintothechamberwhichcanbeexplained bytheexpandingelectrosprayplume.Theelectrospraycurrentsimply spreadsoutoveralargerareaandmoremoleculesgetneutralizedatthe chamberwallsaroundtheinletcapillary. UnlikethetotalcurrentthetransmittedcurrentintotheFaradaycup I FC hasaverystrongpressuredependency.Forallmeasuredtemperaturesanddistancesthereisaverynarrowpeakaroundthepressureof1 Torr.Thisisconsistentwithresultsfrompreviouswork[28],butother researchgroupshavedierentexperiences.Julianetal.[11]couldobserve aworkingrangeofasimilarionfunnelfromseveraltorrto50millitor. SinceitisourgoaltomaximizethetotalcurrentintotheFaradaycup thebestconditionsareapressureof1Torrandelectrospraywithasmall distancemmbetweenneedleandinletcapillaryatatemperatureof 80 .Theseconditionsresultinamaximumcurrentof430pAinthe Faradaycup. 4.2.2Distancevariation Afterdeterminingthebestpressuretooperatetheionfunnelthebest distancebetweenelectrosprayneedleandinletcapillaryshouldbefound. Thisdistanceisaveryimportantfactorforthetransformationprocess fromchargedmicrodropletstofreegas-phaseions.Atatmosphericpres28

PAGE 38

surethemeanfreepathis68nm,resultingin 60000collisionsbetween themicrodropletandairmoleculesalongthe4mmdistancebetween needleandcapillary.At1Torrpressureinsidetherstchamberthe distancehastobe760 aslongasatambientpressuretoresultinthe samenumberofcollisions.Thenumberofcollisionsinuencestheevaporationspeedofthemicrodropletandsotheionformationtime.With a760 highermeanfreepathinsidetherstvacuumcamberalmostall evaporationhastooccuroutsidethechamberandinsidethecapillary. Inpreparationofadistanceexperimenttheneedlewasmovedtoa positionushwiththecapillaryopeningwiththehelpofthemicroscope cameracapillaryinnerdiameterislargerthanneedleouterdiameter. Arulerwasadjustedtothezeropositionontheapparatustabletoallow accuratedistancemeasurementsattheneedlexture.Thetotalcurrent intothevacuumchamberI F andthecurrentintotheFaradaycupI FC weremeasuredatatemperatureof22 ,50 and80 .Brilliant bluedyewassprayedinasolutionwith9:1methanol:water,conductivity 35 S,usingthestandardmetalneedleagainsta30milinnerdiameter and15cmlongstainlesssteelcapillaryataowrateof1 L/min.The pressureintheionfunnelchamberwaskeptconstantat1Torr. Figure4.2presentstheresultsofthedistanceexperiments.Thetotal currentintotheionfunneloverdistanceisalmostthesameforatemperatureof50 and80 withamaximumcurrentof3.75nAand3.89 29

PAGE 39

aTotalcurrentintoionfunnel bCurrentfocusedintoFaradaycup cIonfunneltransmission Figure4.2:Distancedependenciesatdierenttemperatures.Atadistancelarger than8-10mmbetweenelectrosprayneedleandcapillaryheatingshowsnoadditional inuenceondropletevaporation. nAat2mmdistance.Itdropsalmostexponentiallywithincreasingdistance.Atroomtemperaturethecurrentdecreasesbetween1and3mm, thenincreasesuntil8mm.Fromtherethecurrentatroomtemperature isequaltothecurrentswithheatinginvolved. TheioncurrentsintotheFaradaycupfollowsalmostexactlythesame shapesofthetotalcurrents.At50 and80 thereisagainanexponentialdecreasewithincreasingdistanceanat20 thecurrentincreases 30

PAGE 40

between1and8mmdistanceandissimilartothetwoothercurvesbetween10and30mm.Thetransmissionthroughtheionfunnelremains fairlyconstantwithvaryingdistance,exceptforthecurrentatroomtemperature,wherethetransmissionincreaseslinearlybetween1and6mm andthenremainsconstant. Theseexperimentsdemonstratethatthedistancebetweenelectrospray needleandvacuuminletcapillaryaswellasthetemperatureplayan importantroleinthedropletevaporationprocess.Specicallyforshort distancesheatcanimprovedropletevaporationremarkably.Atadistance around8-10mmcollisionswithairmoleculeshavecontributedsomuchto theevaporationprocessthatheatinghasnoadditionaleect.Butbecause oftheexpandingelectrosprayplumeashortdistanceisdesired,where alargerpercentageofthemoleculesgetstransferredintothevacuum chamber,thoughheatingisneededtosupportsolventevaporationfrom thedroplets. 4.2.3Temperaturevariation-Eectsofpositiveandnegative ionmodeandthedierencebetween20miland30mil capillary Afterdeterminingtheimportanceofheatingandaclosedistancebetweenelectrosprayneedleandcapillaryfrompreviousexperiments,three majorexperimentsweredonetodeterminetheinuenceofheatingin combinationwithnegativeandpositiveionmodeand20and30milinner 31

PAGE 41

diametervacuuminletcapillaries.Inallexperimentsthestandardbrilliantbluedyesolutionsolvent9:1methanol:water,conductivity35 S wassprayedwiththestandardmetalneedleataowrateof1 L/min. Thedistancebetweenneedleandcapillarywasheldconstantat4mm andallcapillarieshadalengthof15cm.Thepressureinsidetheion funnelchamberwasconstantat1Torr. aNegativeionmode bPositiveionmode Figure4.3:Negativevspositiveionmodewitha20milcapillary.Innegativeion modethebesttransmissionisobservedbetween33-56 .Theverylowandnot temperaturedependentFaradaycupcurrentinpositiveionmodeindicatesthatonly negativechargestatesexistsforbrilliantbluedyein9:1methanol:watersolution. Figure4.3showsthecomparisonofa20milversusa30milinner diametercapillary.Thetotalcurrentabsolutevaluesisalmostconstant at1.2nAfornegativeionmodeand1.0nAforpositiveionmode.In positivespraymodethecurrentintotheFaradaycupI FC showedno temperaturedependencyandwasconstantattheverylowvalueof40-50 pA,resultinginatransmissionofabout5%. 32

PAGE 42

InnegativeionmodehoweverI FC showedastrongtemperaturedependency.Atemperatureof33-56 increasedtheFaradaycupcurrent from100pAatroomtemperatureto180pAandthefunneltransmission from10to15%.Highertemperaturesthan80 letthetransmitted currentI FC decreaseto80-90pA,atransmissionof7%. Thisresultindicatesthatthebrilliantbluedyemoleculesarenotpositivelychargedinoursolutionandthecurrentinpositiveionmodeonly consistsofsolventmoleculeswhichhaveamass-to-chargeratiothatcannotbefocusedbytheionfunnel.Innegativeionmodethenegatively chargeddyemoleculescontributetotheelectrospraycurrentandsoa highertransmissionthroughtheionfunnelisachieved. a20mil b30mil Figure4.4:Comparisonofa20miland30milcapillaryinnegativeionmode. Inthethirdexperimentthebrilliantbluedyesolutionwassprayed againsta30milcapillaryinnegativeionmode.Thedierencesbetween 20and30milcapillaryinnegativeionmodearedisplayedingure4.4. 33

PAGE 43

ThetotalcurrentintotheionfunnelI F increasesrapidlybetween22and 43 .Over43 I F followsalinearincreasewithtemperatureresulting inamaximumcurrentof6nAat134 .ThetransmittedcurrentI FC followsasimilarinclinebetween22and33 from170to500pAand remainsnearlyconstantbetween43and94 Thisexperimentshowsthatalargercapillarydiameterleadstoa higherandmoretemperaturedependenttotalandtransmittedcurrent. Butothergroups,e.g.Kimetal.[14],observedtransmittedcurrentsover 3nAaftertheionfunnel. 4.3Discussion Theresultsofthisseriesofexperimentshaverevealedastrongpressure dependencyoftheusedionfunnel.Whereasotherexperiences[1,11] showedthatsimilarionfunnelswereabletooperateatawiderpressure range.Thisproblemcouldberelatedtounfavorablegasstreaminside thevacuumchamberandproblemsinthedropletevaporationprocess. Recondensationofdroplets[35]orfrozendropletsduetotherapidgas expansioninsidethevacuumchamberweresuspectedtocounteractthe ionformationanddecreasetheionfunneleciency. Thistheoryissupportedbytheresultsofthedistancevariationand heatingexperiments.Increasingthedistancebetweenelectrosprayneedle andinletcapillaryandmoderateheatingresultinhighertransmission 34

PAGE 44

throughtheionfunnel.Shaeretal.[26]couldobservethatsolvent relatedcurrentandlargedropletsareeectivelylteredoutbytheion funnel.Heatingtheinletcapillaryonlyfromtheoutsidewithhalogen lampsalsocreatesalargetemperaturegradientoverthecapillarythat couldincreaserecondensationandlossinsidethecapillary. Thelowoverallperformanceoftheprototypecongurationcompared toothersystemleadtoaredesignoftheprototypewhichisdescribedin thefollowingchapter. 35

PAGE 45

5REDESIGNOFTHEPROTOTYPEAPPARATUS Figure5.1:Newprototypesystem,obtainedfromFlorianKaiser:Amorecompact andeasiertoreassembledesignwaschosen.Thevacuuminterfaceisacapillaryhold byasolidaluminumheaterblock.Intherstvacuumchambercontainstheionfunnel atapressureof3-6Torr.ThesecondvacuumchambercontainstheFaradaycupto measuretransmittedcurrentthroughthefunnel.Thischamberwillbereplacedwith aquadrupoleionguideinthenextdesigncycle. Duetothelowperformanceoftheoldprototypesetupandspecically thecapillaryheatingandtheionfunnelamajorredesignwasconsidered. Themainrequirementswereamoreuniformheatingofthecapillaryand 36

PAGE 46

astrongerionfunneltohandlehigherspacechargewithlessinuenced ofthegasowinsidethechamber.Figure5.1showsthenewprototype apparatusdesign.Theionfunnelchamberwascustommade,adjusted tothelengthofthenewionfunnel.SPECSTechnologiesCorporation,Sarasota,FL.NW100ISOangesareusedasthelargediameter interconnectsandKF25Kwikangesareusedforallconnectionsperpendiculartothechamberaxis.Tottheheaterblockthefrontange isequippedwithaKF40ange.HermeticsealedBNCconnecterswere usedinthefrontangeforelectricalconnectionoftwoDCvoltagesand twoRFsignalstothePCboard. ThenewionfunnelwasdesignedsimilarlytothefunnelthatBelov etal.[1]presented.Themaindierencetotheoldionfunnelisthe smallerdistancebetweenthefunnellensesandthehighermanufacturing precision.Thesmallerdistancebetweenthelensesresultsinahighereld strengthneartheinsideedgeofthefunnel.Asaresulttherestoringforce onionsintheouterregionsofthefunnelislargercomparedtotheoldion funnel.Thisresultsinalargerpressurerangeandahighertransmission specicallyforspacechargedominatedionbeams.Figure5.2showsthe funneldrawing. Thenewionfunnelconsistsof100lenselectrodesmadeof0.5mmthick stainlesssteel.Thelensesarealignedandholdbyfourceramicrodsand separatedby0.5mmthickteonwashers.Thelensesareenclosedbya 37

PAGE 47

Figure5.2:Cross-sectionofthenewionfunnel,obtainedfromFlorianKaiser.The straitpartconsistsof55lenseswitha25mmdiameterholeinthecenter.Overthe next45lensestheholediameterdecreaseslinearfrom25mmto2mm.Eachlenshas athicknessof0.5mmandthelensesareseparatedby0.5mmthickteonwashers. Thelensesarealignedby4ceramicrods.andenclosedbyastainlesssteelfront-and endplate.ElectricalconnectiontothelensesismadethroughBNCfeedthroughsin thefrontangeandacircuitboardthatconnectsviasteelwiresspot-weldedtothe lenses. stainlesssteelfront-andendplate.Thefrontplateisboltedtothefront angeofthevacuumchamberandtheendplatewithasurroundingo-ring sealstherstvacuumchamberagainstthesecondone. RFandDCsignalsareappliedtothelensplatesusingacircuitboard. ThisboardcontainsthesameRCnetworkdescribedingure3.3.All funnellensesareconnectedinserieswith0.5M resistors.Theseresistors createalinearlydecreasingelectricalpotentialinthefunnelaccordingto thevoltagedividerrule.Thispotentialdropdrivestheionstowardsthe endofthefunnel.Changingthevoltagedierencebetweenfunneltop 38

PAGE 48

andbottomtheionvelocitycanbeadjustedhigherpotentialdierence meanshighervelocity. Specialfocuswasonanewheaterdesignthatshouldresultinamore uniformtemperaturedistributionacrossthestainlesssteelcapillary.It alsoshouldbeoptimizedforshortercapillarylengthasthecapillaryis suspectedtoberesponsibleforthelargestpartoflostions. Figure5.3:Newheaterdesign,obtainedfromFlorianKaiser. Thecurrentheaterdesignisshowningure5.3.Itconsistsofacustom madealuminumblockwithtwoholesforcartridgeheatersuppercavity incrosssection,oneforathermocoupletocontrolthetemperaturein theblockandthecenterholeforthecapillary.Thevacuuminletcapillary UpchurchScientic,OakHarbor,WAisholdinplaceandsealedbya swagelockttingFloridaFluidSystemTechnologies,Sunrise,Flthatis placedatthevacuumendoftheblocktoallowshortercapillaries.The endofthettingisushwiththefunnelentrance.Thermalandelectrical insulationfromthevacuumchambermetalisimplementedbyacustom 39

PAGE 49

madeteonsurrounding.Foreaseofusetheheaterissealedwithaviton o-ringandisholdinplacebythepressuredierencebetweenatmospheric pressureandvacuumchamberpressurenodierencecouldbeobserved usingaKwikangeKFclamp,MDCVacuumProducts,Hayward,CA. Toexchangeorcleanthecapillaryonlythevacuumpumpshavetobe turnedoandtheheaterblockcanbeslidout. Temperaturecontroloftheheaterblockisimplementedbyusinga Eurotherm2132PID-controller.Athermocouplethatcanbeslidinto theheaterblockprovidestemperaturereadingandtwo175WattcartridgeheatersOMEGAEngineering,INC.,Stamford,CTenablefast heatingover200 .HeatersandPIDcontrollerare110Vpoweredand theheatingpoweriscontrolledusingasolidstaterelayconnectedtothe puls-width-modulationportofthecontroller.Theheaterworksexcellent andisabletoreachatemperatureof200 withinabout2minutesand canholditwithin5 40

PAGE 50

6RESULTSANDDISCUSSIONREDESIGNED PROTOTYPEAPPARATUS 6.1Commencementoftheredesignedprototype AfterphysicallybuildingthenewprototypeaseriesofelectricalmeasurementswasmadetoensureproperfunctionalityoftheRC-network onthemanufacturedcircuitboard,functionalityofthenewlydesigned heaterandproperfeedthroughconnections.All100capacitorvalueswere measuredandcapacitorsoutof10%tolerancewerereplaced.Toexclude designandmanufacturingerrorsintheprintedcircuitboardthefullyassembledandmountedfunnelwasmeasuredusingananalogoscilloscope. Theresultsofthismeasurementandaschematicareshowningure 6.1.Initialmeasurementsdidnotshowtheexpectedlinearpotential decrease.Sinceall100resistorsareconnectedinseriesthevoltagedrops 1 = 100 oftheinputvoltageovereachresistoraccordingtothevoltage dividerrule.Sothevoltage U real atthex th lenscanbecalculatedto: U real = )]TJ/F22 14.3462 Tf 14.346 0 Td [(x 100 U input .1 41

PAGE 51

aSchematicofthefunnelresistors bMeasured,calculatedandrealfunnellensDCvoltages Figure6.1:Measurementerrorfunnelresistors:Theparallelresistanceoftheanalog oscilloscopeisinthemagnitudeofthefunnelresistorsandinducesalargemeasurementerror.Calculationthevoltagesaftereachresistorwiththeinnerresistance oftheoscilloscopeinparallelresultsincorrectvaluesthatcanbeveriedbythe measurement. Allresistorshaveanominalvalue R of0.5M .Forthemeasurement aanalogoscilloscopewithainternalresistance R internal of1M witha 10xprobeisconnectedtothex th funnellensinparalleltotheresistors. Becauseofthe10xprobetheinternalresistanceis10M whichisinthe magnitudeofthetotalfunnelresistanceof50M andsosignicantlydistortsthemeasurement.Thedistortedvoltage U calculated canbecalculated andcomparedtothemeasuredvoltage U measured : U measured = xR xR + R parallel U input ;R parallel = )]TJ/F22 14.3462 Tf 14.346 0 Td [(x R R internal )]TJ/F22 14.3462 Tf 14.346 0 Td [(x R + R internal .2 Measurementandcalculationareconsistenthencethevoltagedropsover theresistorsislinearandthecorrectfunctionofthecircuitboardis proven. 42

PAGE 52

Asarsttestasolutionofbrilliantbluedyeinmethanolwitha conductivityof80 S/cmwassprayedwiththestandardmetalneedle througha20milinnerdiameter,15cmlongcapillaryinnegativeion modeatadistanceof3mm.Thetemperaturewasincreasedfrom20 to200 Figure6.2a.Withincreasedtemperaturelessvolumeows intothevacuumchamberbecauseofgasexpansion.Whilethepressure insidechamber1droppedlinearfrom1.71Torr to1.44Torr thecurrentalsodecreasedfrom2nAto1.5nA.Thisindicatesthatthe gasowthroughthecapillaryisthedominantfactorintheion/droplet transportintothevacuumchamber.Butlimitedbypumpingpowerand themaximumoperationpressureoftheionfunnela30mildiameter turnedouttobeagoodcompromiseforagoodcurrentintothevacuum chamber. Totakeadvantageofthenewheaterdesigna11cmlong20mildiametercapillarywasusedforthesecondexperimentshowningure6.2 b.Thetotalcurrentintotheionfunnelwas3.0nAat24 and3.1 nAat200 .Overtherst28minutesthecurrentintotheFaradaycup increasedalmostlinearovertimewithnofunnelvoltageconnectedfunneloating.Thenthetemperaturewasslowlyincreasedto200 at 53minutes.TheFaradaycupcurrentdecreasedwithhighertemperature andremainedconstantagainafterthetemperaturewaskeptconstantat 200 :56-1:06. 43

PAGE 53

aTotalfunnelcurrentwithbluedye:methanolsolution bCurrentintotheFaradaycupovertime.Funnel currentwas3nA. Figure6.2:Firsttestrunswithnewprototype:Brilliantbluedyemoleculessolute inmethanolweresprayed.Currentinadecreaseswithincreasingtemperature.In bthecurrentintotheFaradaycupwasmeasuredwiththefunnelnotconnected. Aself-chargingeectcanbeobserved,butthecurrentdecreasedafterheatingwas startedat28minutes. TheincreasingcurrentintotheFaradaycupindicatesaslightself chargingeectoftheionfunnel.Chargeddropletshitthefunnellenses andresidethereforacertaintime.Anequilibriumisreachedwhenthe staticelectriceldfromtheresidingchargesisstrongenoughtodeect theions/dropletsinsidethefunnel.Otherresearchgroupse.g.Kremer etal.[15]triedtousethiseecttocreateapassiveionfunnelforthe useatatmosphericpressure. Inaseriesofshortexperimentsthebestparametersforthefunnel parametersDCvoltageoninletlens,DCvoltageonsmallestlens,amplitudeofthesineRFsignalandfrequencyoftheRFsignalshouldbe optimizedforhighestcurrentintotheFaradaycup.Thesolutionwas changedfrombluedyeinmethanoltoa0.5mg/mlsolutionoftheamino 44

PAGE 54

acidGly-Glnina50:50:1methanol:water:aceticacidtobeabletoachieve resultscomparabletoliteraturevalueslikeKimetal.[14].Itisproblematictospraysolutionswithahighwaterpercentagewiththeusedmetal needlebecauseofthehighsurfacetensionofwater.Forthatreasona switchtonano-electrospraywasnecessary.Nano-electrosprayrefersto aprocesswherethesolutionissprayedataow-ratebetween200and 500nL/minthroughaneedlewithatipdiameteroftypically5-100 m [25].Becauseofthehigherionizationeciencyofnano-electrospraylower voltages-2000Vinsteadof3000-5000Vcanbeused. Theseexperimentsdemonstratedcurrentsbetween7and10nAinto thefunnelwitha10cmlong,20mildiametercapillary,buttheionfunnel wasnotabletofocusanyofthiscurrentintotheFaradaycup.Butwith a30mildiameter,15cmlongcapillaryupto500pAofafunnelcurrent of1.4nAcouldbefocusedatatemperatureof140 .Astheseresults arenotnearlyconsistentwithliteraturevaluesitwastriedtomodifythe airowinsidetheionfunnelchambertoreduceturbulenceandgas-ow perpendiculartothefunnelaxisasthiswassuspectedtodeecttheions towardsthevacuumpump. 45

PAGE 55

6.2Improvingionfunneltransmissionthroughmodiedair ow 6.2.1Gas-owsimulations Gas-owinsidetheionfunnelchamberwassuspectedtoeectthe transmissioneciencynegatively.Toexaminetheseeectstheionfunnelincludingcircuitboardswaswrappedintoplasticfoil.Theplastic wrapwasplacedunderthesealingo-ringattheexitendofthefunnelso thatthegasinsidethefunnelcouldonlyexitbacktothefunnelinletor throughthefunnelexitintothesecondvacuumchamber.Tocompare theeectofthewrappingasimulationwithasimpliedmodelofthe funnelwasmadeusingthesoftwareCOMSOLMultiphysics.Forsimplicationatmosphericpressurewasassumedatthecapillaryexitand theoutowatthepumpwassetto250L/min.Thegaswasmodeled asanincompressibleuidwithonlylaminarow.Thosesimplications certainlyeecttheaccuracyoftheresults,butwerenecessarybecauseof thecomplexityofthesimulation. Figure6.3showsthecomparisonbetweenunwrappedandwrapped ionfunnel.Insidetheunwrappedionfunnelallmoleculesareexposed toverticalvelocitiesof30m/sinthecenterandupto80m/sinthe outerregionswhereasallionsthatgetinsidethewrappedionfunnelare onlyexposedtotheelectricaleld.Inbothversionstherapidexpansion 46

PAGE 56

Figure6.3:Simulatedairowinsidethevacuumchamber:Highverticalvelocity leftmovesionstothepumpexit.Thiseectisnotpresentinthewrappedfunnel right.Thewhiteareasinsidethefunnelrepresentvelocitiesoutsidethescaleonthe right. afterthecapillarycanbeobservedwhiteblobsin45degreeangleat thecapillaryend.Thisrapidexpansionandspacechargearethemain factorsinwideninguptheionbeamandmaketheuseofafocussing elementinsidethevacuumchambernecessary. 6.2.2Firstexperimentwiththewrappedionfunnel Asarstexperimentwiththewrappedversionoftheionfunnel0.5 mg/mlGly-Glnina50:50:1methanol:water:aceticacidconductivity140 Swassprayedwithaowrateof300nL/min.A30mildiameter,10 cmlongcapillarywasslowlyheatedfromroomtemperatureto200 Theendofthecapillarywasushwiththerstlensofthefunnel.A 47

PAGE 57

sineRFsignalwithanamplitudeof100VVpeaktopeakanda frequencyof230kHzwasappliedtothefunnellensesandtheDCvoltage gradientinsidethefunnelwas10V/cm. Figure6.4:Funnelwrappedinplasticfoil.Duetoimprovedairowacurrentof1 nAcanbemeasuredattheFaradaycupwas500pAbefore. Theshort30milcapillaryresultsinarelativelyhighioncurrentinto thefunnelof5.5-6.3nAandthefunnelisabletofocusupto1nAinto theFaradaycup.Amaximumintransmissionoccursatatemperature of150 whereitreaches17.5%.Thisresultisencouragingtoproceed withthewrappedfunnel,butitisstillfarfromliteraturevalues,where transmissionsbetween60and80%areaccomplished.Itstillhastobe determinedifthereisaproblemwiththeparameterslikefunnelfrequency, amplitude,DCgradient,temperatureandcapillarylengthorifthereis aproblemwiththedesignofthefunnelitself. 48

PAGE 58

6.2.3EectoftheDCvoltagegradientinsidethefunnel Furtheroptimizationwastriedbychangingthevoltagegradientthat drivestheionstowardsthefunnelexit.Atatemperatureof200 the peaktopeakvoltageU RFpk )]TJ/F11 9.9626 Tf 7.749 0 Td [(pk wasincreasedfrom0VnoRFsignal to250Vmaximumratingoftheusedcapacitorsandthecurrentinto theFaradaycupwasmeasured.Figure6.5ashowsthecurrentinto aFunnelwrappedintoplasticfoil bBelovetal.[1] Figure6.5:Funnelwrappedinplasticfoiltoimproveairow:Theresultsarecomparabletootherresearchgroups. theFaradaycupwithaDCgradientof10and18V/cm.Outofatotal currentof5.3nAintothefunnel1.3nAV/cmand1.4nAV/cm couldbetransmitted.Bothcurrentsareverycloseandsaturateintoa maximumabove200Vpeak-to-peakvoltage.Theshapeoftherecorded curveissimilartocomparablesetupsbwiththedierencethatthe amplitudeneededtoreachthemaximumtransmissionishigherinour setup.Buttheseresultindicateaproperdesignoftheionfunneland 49

PAGE 59

encouragetofurtheroptimizationtoreachhighertransmissionsupto80 %andcurrentsofupto7nAintotheFaradaycup. 6.2.4Eectofthecapillaryposition Thereaftertheimportanceofthepositionofthecapillarywasstudied. A30mildiameter,10cmlongcapillarywasmountedrstushwiththe rstfunnellensthensticking10mmintotheionfunnel.Itwassuspected thationsgetreectedatthefunnelentrancesincethepotentialthereis higherthanthecapillarypotentialifthesinewaveonthelensreachesa maximum.Theresultsaredisplayedingure6.6. aFlushcapillary bCapillarysticking10mmintofunnel Figure6.6:Comparisonofacapillaryushwiththeionfunnelentranceaand sticking10mmintothefunnelb.Bettertransmissionisachievedifthecapillary sticksintothefunnelbecausethereisnopotentialhilltoovercome. Thecurrentsintotheionfunnelareverysimilarinshapeandthe littleosetcanbeexplainedbyaslightvariationintheelectrospray parameters,specicallythedistanceneedle-capillary.Butwhereasthe transmittedcurrentreachesitssaturationat1nAwiththeushcapillary, 50

PAGE 60

itcontinuouslyincreaseswithtemperatureandreachesamaximumof1.5 nAat200 .Highertemperaturescannotbeusedbecauseoflimitations ofrubberandplasticelements.Thetransmissionisabout1.5timeshigher withthecapillarystickingintothefunnelcomparedtoaushcapillary alignmentwiththerstfunnellens. Thiseectcanbeexplainedbytheeldenergythattheionentering thefunnelhastoovercome.Theionhasacertainkineticenergyafterit leavesthecapillary.Togetintotheionfunnelthekineticenergyhasto belargerthentheeldenergycreatedbythepotentialdierencebetween capillaryandfunnel.Sointhebestcasethesinesignalappliedtothe rstfunnellensisattheminimumvalueabout-100Vwith200Vpeakto-peakRFsignal.Thenthepotentialoftherstfunnellenssumof RFsignal+DCvoltageappliedtofunnelentranceislowandthereisa forcemovingtheiontowardsthefunnelentrance.Butifthesinesignal attherstfunnellensisatthemaximumvaluetypically+100Vthere isarepellingforceontheion.Thisexplanationisasimplicationofthe verycomplexinteractionbetweengasow,DCeldandRFeldbetween capillaryandionfunnel,thatrequiresextensivecomputersimulationfor fullunderstanding. 6.2.5Summaryoftherstexperiments Resultingfromthisrstseriesofexperimentswiththeredesignedion funnelrstconclusionscanbedrawn.Modifyingtheairowinsidethe 51

PAGE 61

ionfunnelbycreatingagas-tightenclosurereducesdisturbinggasow perpendiculartothefunnelaxisandincreasesiontransmission.The DCgradientinsidethefunneldoesnotseemtoplayanimportantrole, butthemeasuredtransmittedcurrentoverpeak-to-peakRFvoltageis comparabletoliterature,evenifhigherabsolutevoltagesareneededfor thisfunneldesign.WhiletheRFsignalappliedtothefunnellensesis intendedtofocusthediuseionsenteringthefunneltowardsthefunnel centeritcanalsorepelionsbeforereachingthefunnel.Thiseectdoes notshowupinmeasurementsofthetotalcurrentintotheionfunnel becausethereallDCandRFvoltagesaredisconnected.Unfortunately thetotalcurrentintothefunnelcannotbemonitoredwiththeexisting equipmentwhilerunningduetothemeasurementequipmentandsetup. Theseresultsindicateaproperdesignoftheionfunnelbutthetransmissionvaluesaretoolowcomparedtoliterature. 6.3ExcludingmeasurementerrorsattheFaradaycup Resultingfromthepreviousexperimentsaproblemwiththeabsolute transmissionvaluesemergeswhereasthecourseofmeasuredtransmission curvesissimilartoliterature.Thisindicateseitheradesignerrorora measurementerror,e.g.notthetotalcurrentismeasuredintheFaraday cuporthesignalgetslostbetweenFaradaycupandpicoampere-meter. UntilnowtheFaradaycupiscenteredinthesecondvacuumchamber about8cmawayfromtheionfunnelexit.Sothereisapossibilitythat 52

PAGE 62

amajorpartofthespace-chargedominatedionbeammissestheareaof approximately1cm 2 oftheFaradaycup.Toinvestigatethisbehavior furthertheFaradaycupwasextendedsothattheionbeamcouldbe measuredinadistanceofabout5mmbehindtheionfunnelexit.Ifnot otherwisenoteda8.5cmlong,30milinnerdiametercapillarywasused. Itwasextending1mmintotheionfunnelandstickingout1mmatthe atmosphericsideoftheheaterblock.Anano-electrosparyneedlewithan innertipdiameterof30 mwastospraythesolutionataowrateof300 nL/mininadistanceof3-4mmofthecapillary. 6.3.1Eectofthecapillarylength TherstexperimentafterextendingtheFaradaycuptothefunnel exitoricewasmeasuringthetransmittedcurrentwithdierentcapillary lengths.Bothcapillarieshadainnerdiameterof30mil.The10cmlong capillarywassticking10mmintotheionfunnelwhereasthe8.5cmlong onewasalignedush.Atemperatureof150 wasusedtooptimize dropletevaporation. Theresultisshowningure6.7.Bothcapillarylengthsresultinan almostconstanttransmissionfrom150to800kHzsimilartotheresultsof otherresearchgroups.Atotalcurrentof5.75nAintotheionfunnelwas measuredusingthe10cmcapillaryresultinginamaximumtransmission of46%.Theshorter8.5cmcapillaryresultsinlesslossandleadsalso toamaximumtransmissionof45%ofthehighertotalcurrentof6.6nA. 53

PAGE 63

Figure6.7:ComparisonofcurrentintotheFaradaycupfor8.5cmand10cmcapillary. Totaltransmittedcurrentisbetterforshortercapillaries. 6.3.2Eectofthecapillarydiameter Previousexperimentswith20mildiametercapillarieswerenotsuccessfulconcerningtransmittedcurrent.Anotherexperimentusingthe shortestpossiblelengthof8.5cmwassetupsinceresultsfromthecapillarylengthexperimentshowbettertransmissionforthislength.20and 30milcapillarieswerecomparedatdierentcapillarytemperatures. Whereasthetotalcurrentintothefunnelusingthe8.5cmcapillary decreasedfrom7nAat24 to2.7nAat150 and2.4nAat200 increasedthetotalcurrentusinga30milcapillaryfrom5.6nAto6.6 nAand7.2nAat200 .Thehighesttransmittedcurrentwasmeasured0.8nA%at200 withthe20mildiameterand3.5nA %withacapillarydiameterof30mil.Thisexperimentindicatesthat thereisaproblemusinga20milcapillaryintheusedsetup.Collisions 54

PAGE 64

a20milID8.5cmcapillary b30milID8.5cmcapillary Figure6.8:Comparisonbetween8.5cmlongcapillaries.aAtleastsomecurrent getstransmittedthroughthefunnelnotransmissionforlonger20milcapillaries. bHeatingsupportsdropletevaporationandimprovestransmissionbyafactorof3. andre-condensationatthecapillarywallsissuspectedtocreatelarge dropletsinsteadofassistingdropletevaporation.Theshorterlengthand theimprovedmeasurementthroughtheFaradaycupextensiondemonstratethatthe20milcapillarycanbeused,butworksveryinecientin theusedexperimentalsetupcomparedtoa30milinnerdiametercapillary. 6.3.3Dierencesbetweensolutemolecules Withimprovementsingasowandmeasurementsetuptheredesigned prototypeiscomparabletosetupsusedbyotherresearchgroups.GlyGlnwascomparedtocytochromecusingtheoptimumparametersthat areatemperatureof200 anda8.5cmlong30milcapillary.Firstthe 55

PAGE 65

transmittedcurrentwasmeasuredoverfrequencyatapeak-to-peakRF voltageof80Vandthenthepeak-to-peakRFvoltagewasincreasedat thebestfrequencytogetthebestpossibletransmission. aFrequencyspectrum bAmplitudespektrumatoptimumfrequency Figure6.9:ComparisonbetweenGly-Glnandcytochromec.Thetransmittedcurrent issimilarupto230kHzfunnelfrequency.WhereasthecurrentusingGly-GlnMass: 203u,chargestate 1 + doesnotincreaseforlargerfrequenciesreachescytochromec currentMass:12500u,chargestate 11 + )]TJ/F21 11.9552 Tf 12.221 0 Td [(24 + asaturationatabout300kHzand 5nA.Withcytochromecaremarkabletransmissionof82%isreachedatU pk )]TJ/F38 7.9701 Tf 6.587 0 Td [(pk = 150V Figure6.9ashowsthetransmittedcurrentsoverfrequency.Both currentsarealmostidenticalforfrequenciesunder180kHz.Forhigher frequenciesthetransmittedcurrentusingGly-Glnreachesasaturation at3.2nAwhereasthecurrentusingcytochromecreachesitssaturation atafrequencyof350kHzat5nA.ForGly-Glnthetotalmaximumtalkinggure6.9bintoconsideration-isreachedatacurrentof3.5 nA.Thisindicatesstrongspacechargeinsidetheionfunnelcomparedto cytochromec.Theheavierandhigherchargedcytochromecionsresult inalowerspacechargeandallowcurrentsupto7.2nAoratransmission 56

PAGE 66

of82%.Thisresultexceedscomparableresultsfromliterature[14]where transmissionup60%arereached. Thisexcellentresultprovesaproperdesignoftheionfunneland showstheinuenceofgasowinsidetherstpumpingstage.Butitalso revealsthattheionbeamspreadsoutagainafterbeingfocusedthrough theionfunnel.Theinuencesofgasowandspreadingoutbeamare investigatedfurtherinthenextsubsection. 6.3.4Investigationoftheeectsofwrappingthefunneland extendingtheFaradaycup ModifyingthegasowandextendingtheFaradaycupcurrentsupto 7.2nAcouldbetransmittedthroughtheionfunnel.Thelastexperiment wasdonetogetfurtherunderstandingofthecompositionoftheions beamandtheinuenceofmodifyingthegasow. Thegures6.10and6.11showthecoherencebetweenwrappedand unwrappedfunnelandregularandextendedFaradaycup.Ifthefunnel isunwrappedtheshapeofthegraphissimilar,buttheintensityisalot lowertoptobottom.IftheFaradaycupisnotextended,theformer broadfrequencyresponsedegradestoanarroweroneandtheintensityis lowerlefttoright. Thisindicatesthattheionbeamconsistsofdierentsizemolecules/ dropletswithdierentchargestatesand/orvelocities.Theionfunnel 57

PAGE 67

Figure6.10:EectsofwrappingthefunnelandextendingtheFaradaycupoverfunnel frequency.0.5mg/mlGly-Glnin50:50:1methanol:water:aceticacidwassprayed atatemperatureof150 DCgradient18V/cm,U pk )]TJ/F38 7.9701 Tf 6.587 0 Td [(pk 80V,I Funnel 6 : 6 )]TJ/F21 11.9552 Tf 12.7 0 Td [(8 : 2 nA.Topleft:52%transmissionat300kHz.Topright:narrowpeakaround150 kHz,25.5%transmission.Bottomleft:Samespectrumastopleft,butonly31% transmission.Bottomright:Similargraphtotopright,butonly9%transmission. Thisindicatesamassdistributionofdroplets/ionsinsidethefunnelandfrequency dependenttransmissionofthedierentmasses. transmitsdependingonthefrequencynotallofthechargedparticles,at leastnotaU pk )]TJ/F11 9.9626 Tf 7.749 0 Td [(pk of80V.Figure6.9showsthatlargervoltagesincrease thetransmission.Afterthefunnelexittheionbeamspreadsoutagain, buttheatlowerfunnelfrequenciesparticelsaretransmittedthroughthe funnelthatspreadoutatalowerspeedhighercurrentintoFaradaycup whereasathigherfrequenciesparticlesaretransmittedthatspreadout 58

PAGE 68

faster.Ifthefunnelisnotwrapped,thegasstreamtowardsthepump exitaectsalldierentparticlesequallywhichisindicatedbythesame shapeofthegraphsandthelowerintensityoftheunwrappedone. Figure6.11:EectsofwrappingthefunnelandextendingtheFaradaycupoverfunnel amplitude. Ingure6.11theexperimentwasrepeatedattheoptimumfrequenciesforeachcase.Thetransmittedcurrentovertheamplitudeconrm theresultsfromthepreviousexperiment.Thisexperimentconcludesthe workforthisthesis.Theresultsbuildastablebaseforfurtherdevelopmentsofthesystem.Itiscrucialforthenextdesignphasetoattachthe quadrupoleionguideascloseaspossibletotheionfunnelexit,because otherwisethebeamwillbespreadouttoofarandadditionallossoccurs. 59

PAGE 69

7CONCLUSION Previoustothisworkioncurrentsupto930pAcouldbetransmitted throughtheionfunnelintothesecondpumpingstageoftheprototype system.Theexistingprototypesystemwasthoroughlycharacterizedand improvementsweretriedwithoutmajorsuccess.Problemsinthedroplet toionevaporationprocesswereassumedtoberesponsibleforthelow performanceoftheionfunnel.Heatingthecapillaryusinghalogenlights wastriedincombinationwithelectricalheatingonthevacuumsideof thecapillarytoimprovethedropletevaporation.Dierentcapillarydiametersandspraymodesweretried,aswellasparameteroptimization inoperationpressureandspraydistance. Afteraredesignphasethenewprototypecontaininganewdesigned, improvedionfunnelandanewcapillaryheaterwerecommenced.Afterelectricalchecksofthefunnelitwastriedtondthebestoperating parametersforthesystem.Itwastriedtodecreasetheinuenceofthe gasowfromthevacuumchamberinlettothepumpexitbywrapping thefunnelintofoil.Simulationsofthegasowweremade.Transmissioncouldbeimprovedoveralargefrequencyrangecomparabletoother 60

PAGE 70

publications.Butthoughsimilarshapedgraphscouldbemeasured,the absolutevalueswerenotconsistentwithliterature.ExtendingtheFaradaycuptotheexitoriceoftheionfunnelsolvedthisproblem. Toconcludethisworkitistomentionthatcurrentsupto7.2nAcould betransmittedthroughtheionfunnel.Atransmissioneciencyof82 %wasreachedmodifyingtheairow.Crucialfortheionfunneltransmissioneciencyisthedropletevaporationprocess.Nano-electrospray isnecessaryduetoitshighionizationeciencyandtheresultingsmall dropletsizes.Uniformheatingacrosstheinletcapillarysupportsthe dropletevaporationprocessandpreventscondensationorfrozendroplets. Togetherwiththemodiedgasowtheheateristhemostimportantelementdecreasingloss. Inthenextdesignphaseanionguidewillbedesigned.Aquadrupole elementcanbeusedtofocus,lterandguidetheionsafterthefunnel exitintohighervacuumchambers.Thehighermeanfreepathinthose regionssimpliesionopticdesignsinceknownandestablisheddesign methodscanbeused.Timesintheregionofonehourfordepositingone monolayeracross1cm 2 arepossible.Multi-capillary/Multi-emitterion sourcescouldbeusedtoimprovethecurrentintothevacuumchamber anddecreasingdepositiontimesfurther. 61

PAGE 71

LISTOFREFERENCES [1]M.E.Belov,M.V.Gorshkov,H.R.Udseth,G.A.Anderson,A.V. Tolmachev,D.C.Prior,R.Harkewicz,andR.D.Smith.Initialimplementationofanelectrodynamicionfunnelwithfouriertransform ioncyclotronresonancemassspectrometry. JournaloftheAmerican SocietyforMassSpectrometry ,11:19,2000. [2]M.CloupeauandB.Prunet-Foch.Electrostaticsprayingofliquids incone-jetmode. JournalofElectrostatics ,22:135,1989. [3]R.B.Cole.Sometenetspertainingtoelectrosprayionizationmass spectrometry. JournalofMassSpectrometry ,35:763,2000. [4]R.B.ColeandA.K.Harrata.Solventeectonanalytechargestate, signalintensity,andstabilityinnegativeionelectrospraymassspectrometry;implicationsforthemechanismofnegativeionformation. JournaloftheAmericanSocietyforMassSpectrometry ,4:546 556,1993. [5]M.Dole,R.L.Hines,L.L.Mack,R.C.Mobley,L.D.Ferguson, andM.B.Alice.Gasphasemacroions. Macromolecules ,1:96, 1968. 62

PAGE 72

[6]J.FernandezdelaMora,J.Navascues,F.Fernandez,andJ.RosellLlompart.Generationofsubmicronmonodisperseaerosolsinelectrosprays. J.Aerosol.Sci ,21:5673,1990. [7]J.FernandezdelaMora,M.Labowsky,andJ.B.Fenn.Acontinuum modelforionevaporationfromadrop:eectofcurvatureandcharge onionsolvationenergy. AnalyticaChimicaActa ,406:105, 2000. [8]DrRudySchlaf'sGroupHomepage.Electrospraythinlmdepositionofmacro-molecularmaterialsinvacuum[online].[accessed21st april2008].availablefromtheworldwideweb: < http://rsl.eng.usf.edu/pages/researchelectrospray.html > ,042008. [9]DrRudySchlaf'sGroupHomepage.Moleculewriter[online].[accessed21stapril2008].availablefromtheworldwideweb: < http://rsl.eng.usf.edu/pages/researchmolwriter.html > ,042008. [10]AnthonyT.Iavarone,JohnC.Jurchen,andEvanR.Williams.Effectsofsolventonthemaximumchargestateandchargestatedistributionofproteinionsproducedbyelectrosprayionization. Journalof theAmericanSocietyforMassSpectrometry ,11:976,2000. [11]R.R.Julian,S.R.Mabbett,andM.F.Jarrold.Ionfunnelsfor themasses:Experimentsandsimulationswithasimpliedionfunnel. JournaloftheAmericanSocietyforMassSpectrometry ,16: 1708,2005. 63

PAGE 73

[12]P.KebarleandY.Ho.Electrosprayionizationmassspectrometry: Fundamentals,instrumentationandapplications.pages3.John Wiley&Sons:NewYork,1997. [13]P.Kebarle,D.B.Hager,N.J.Dovichi,andJ.Klassen.Droplet electrospraymassspectrometry. AnalyticalChemistry ,66:3944 3949,1994. [14]T.Kim,A.V.Tolmachev,R.Harkewicz,D.C.Prior,G.Anderson,H.R.Udseth,R.D.Smith,T.H.Bailey,S.Rakov,andJ.H. Futrell.Designandimplementationofanewelectrodynamicion funnel. Anal.Chem ,72:224755,2000. [15]E.P.Kremer,G.J.Evans,andR.E.Jervis.Anovelmethodforthe collimationofionsatatmosphericpressure. JournalofPhysicsD: AppliedPhysics ,39:5008,2006. [16]B.LinandJ.Sunner.Iontransportbyviscousgasowthrough capillaries. JournaloftheAmericanSocietyforMassSpectrometry 5:873,1994. [17]L.B.Loeb,A.F.Kip,andG.G.Hudson.Pulsesinnegativepointto-planecorona. PhysicalReview ,60:714,1941. [18]J.A.Loo.Studyingnoncovalentproteincomplexesbyelectrospray ionizationmassspectrometry. MassSpectrometryReviews ,16: 1,1997. 64

PAGE 74

[19]J.A.LooandR.R.O.Loo.Electrosprayionizationmassspectrometry:Fundamentals,instrumentationandapplications.pages 385.JohnWiley&Sons:NewYork,1997. [20]SimonPapadopoulos,KlausD.Jurgens,andGerolfGros.Protein diusioninlivingskeletalmusclebers:Dependenceonproteinsize, bertype,andcontraction. Biophys.J. ,79:2084,2000. [21]R.J.PfeiferandC.D.Hendricks.Parametricstudiesofelectrohydrodynamicspraying. AIAAJ ,6:496,1968. [22]B.N.Pramanik,P.L.Bartner,U.A.Mirza,Y.H.Liu,andA.K. Ganguly.Electrosprayionizationmassspectrometryforthestudyof non-covalentcomplexes:anemergingtechnology. JournalofMass Spectrometry ,33:911,1998. [23]JohnWilliamStruttLordRayleigh.Ontheequilibriumofliquid conductingmasseschargedwithelectricity. Phil.Mag. ,14:184,1882. [24]R.Saf,T.E.Hamedinger,T.Steindl,J.Albering,andS.Rentenberger.Directpatterningoffunctionalmaterialsviaatmosphericpressureiondeposition. 2005NSTINanotechnologyConferenceand TradeShow ,pages467,2005. [25]AndreaSchmidt,MichaelKaras,andThomasDlcks.Eectofdifferentsolutionowratesonanalyteionsignalsinnano-esims,or: whendoesesiturnintonano-esi? JournaloftheAmericanSociety forMassSpectrometry ,14:492,2003. 65

PAGE 75

[26]S.A.Shaer,A.Tolmachev,D.C.Prior,G.A.Anderson,H.R. Udseth,andR.D.Smith.Characterizationofanimprovedelectrodynamicionfunnelinterfaceforelectrosprayionizationmassspectrometry. Anal.Chem. ,71:29574,1999. [27]A.P.Snyder,AmericanChemicalSocietyDivisionofAnalyticalChemistry,andAmericanChemicalSocietyMeeting. Biochemical andBiotechnologicalApplicationsofElectrosprayIonizationMass Spectrometry .AmericanChemicalSociety,1995. [28]AStark. PerformanceOptimizationofa3DPatterningDevicefor Macro-MolecularMaterials .Studienarbeit,UlmUniversity,2007. [29]L.TangandP.Kebarle.Eectoftheconductivityoftheelectrosprayedsolutionontheelectrospraycurrent.factorsdetermininganalytesensitivityinelectrospraymassspectrometry. AnalyticalChemistry ,63:270915,1991. [30]L.TangandP.Kebarle.Dependenceofionintensityinelectrospray massspectrometryontheconcentrationoftheanalytesintheelectrosprayedsolution. AnalyticalChemistry ,65:36548,1993. [31]G.I.Taylor.Disintegrationofwaterdropsinanelectricaleld. Proc. R.Soc ,280:383,1964. [32]G.J.VanBerkel.Electrosprayionizationmassspectrometry:Fundamentals,instrumentationandapplications.pages65.John Wiley&Sons:NewYork,1997. 66

PAGE 76

[33]G.J.VanBerkelandF.Zhou.Characterizationofanelectrosprayion sourceasacontrolled-currentelectrolyticcell. AnalyticalChemistry 67:2916,1995. [34]G.WangandR.B.Cole. J.Am.Soc.Mass.Spectrom. ,7:1050,1996. [35]M.YamashitaandJ.B.Fenn.Electrosprayionsource.another variationonthefree-jettheme. TheJournalofPhysicalChemistry 88:4451,1984. 67


xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 002001013
003 fts
005 20090427082104.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 090427s2008 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002534
035
(OCoLC)319421649
040
FHM
c FHM
049
FHMM
090
TK145 (Online)
1 100
Stark, Andreas.
0 245
Advancement and optimization of an electrospray injection based in-vacuum patterning system for macromolecular materials
h [electronic resource] /
by Andreas Stark.
260
[Tampa, Fla] :
b University of South Florida,
2008.
500
Title from PDF of title page.
Document formatted into pages; contains 67 pages.
502
Thesis (M.S.E.E.)--University of South Florida, 2008.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
3 520
ABSTRACT: Electrospray ionization is a technique widely used in mass spectrometry. Almost every material, specifically large molecules like proteins or polymers can be ionized directly out of solution. During the ionization process molecules are not fragmented. In this work a prototype apparatus for creating three-dimensional patterns in a ultra high vacuum environment using an electrospray ion source was optimized for higher ion currents hence deposition rate by improving the core component of the apparatus, an electrodynamic ion funnel. The major improvements are a redesigned heated vacuum inlet, modified gas flow inside the ion funnel because of sealing the ion funnel against perpendicular gas flow and a better measurement setup for the transmitted current. The transmission of the ion funnel was improved from 25% to 82% resulting in ion currents of up to 7nA (500pA before advancements) focused through the ion funnel. At this rate an area of 1 cm¨§ can be coated with a molecular monolayer of Cytochrome C in 64 minutes.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
590
Advisor: Rudy Schlaf, Ph.D.
653
Electrospray ionization
Ion focussing
Ion funnel
Cytochrome C
Faraday cup
690
Dissertations, Academic
z USF
x Electrical Engineering
Masters.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.2534